Oxford Handbook of The Five Factors Model

The Oxford Handbook of the Five Factor Model
OXFORD LIBRARY OF PSYCHOLOGY
EDITOR-IN-CHIEF
Peter E. Nathan
AREA EDITORS
Clinical Psychology
David H. Barlow
Cognitive Neuroscience
Kevin N. Ochsner and Stephen M. Kosslyn
Cognitive Psychology
Daniel Reisberg
Counseling Psychology
Elizabeth M. Altmaier and Jo-Ida C. Hansen
Developmental Psychology
Philip David Zelazo
Health Psychology
Howard S. Friedman
History of Psychology
David B. Baker
Methods and Measurement

Todd D. Little
Neuropsychology
Kenneth M. Adams
Organizational Psychology
Steve W. J. Kozlowski
Personality and Social Psychology

Kay Deaux and Mark Snyder
Editor in Chief PETER E. NATHAN
Edited by
Thomas A. Widiger
Oxford University Press is a department of the University of Oxford. It furthers the University’s objective
of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade
mark of Oxford University Press in the United Kingdom and certain other countries.
Published in the United States of America by Oxford University Press

198 Madison Avenue, New York, NY 10016, United States of America.
© Oxford University Press 2017
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or
transmitted, in any form or by any means, without the prior permission in writing of Oxford University
Press, or as expressly permitted by law, by license, or under terms agreed with the appropriate reproduction
rights organization. Inquiries concerning reproduction outside the scope of the above should be sent to the
Rights Department, Oxford University Press, at the address above.
You must not circulate this work in any other form and you must impose this same condition on any
acquirer.
CIP data is on file at the Library of Congress

ISBN 978–0–19–935248–7
eISBN 978–0–19–067953–8
SHORT CONTENTS
Oxford Library of Psychology
About the Editor
Contributors
Contents
Chapters
Index
OXFORD LIBRARY OF PSYCHOLOGY
The Oxford Library of Psychology, a landmark series of handbooks, is published

by Oxford University Press, one of the world’s oldest and most highly respected
publishers, with a tradition of publishing significant books in psychology. The
ambitious goal of the Oxford Library of Psychology is nothing less than to span a
vibrant, wide-ranging field and, in so doing, to fill a clear market need.
Encompassing a comprehensive set of handbooks, organized hierarchically,
the Library incorporates volumes at different levels, each designed to meet a
distinct need. At one level is a set of handbooks designed broadly to survey the
major subfields of psychology; at another are numerous handbooks that cover
important current focal research and scholarly areas of psychology in depth and
detail. Planned as a reflection of the dynamism of psychology, the Library will
grow and expand as psychology itself develops, thereby highlighting significant
new research that will have an impact on the field. Adding to its accessibility
and ease of use, the Library will be published in print and, later on,
electronically.
The Library surveys psychology’s principal subfields with a set of handbooks
that capture the current status and future prospects of those major subdisciplines.
This initial set includes handbooks of social and personality psychology, clinical
psychology, counseling psychology, school psychology, educational psychology,
industrial and organizational psychology, cognitive psychology, cognitive
neuroscience, methods and measurements, history, neuropsychology, personality
assessment, developmental psychology, and more. Each handbook undertakes a
review of one of psychology’s major subdisciplines with breadth,
comprehensiveness, and exemplary scholarship. In addition to these broadly
conceived volumes, the Library also includes a large number of handbooks
designed to explore in depth more specialized areas of scholarship and research,
such as stress, health and coping, anxiety and related disorders, cognitive
development, or child and adolescent assessment. In contrast to the broad
coverage of the subfield handbooks, each of these latter volumes focuses on an
especially productive, more highly focused line of scholarship and research.
Whether at the broadest or most specific level, however, all of the Library
handbooks offer synthetic coverage that reviews and evaluates the relevant past
and present research and anticipates research in the future. Each handbook in the
Library includes introductory and concluding chapters written by its editor to
provide a roadmap to the handbook’s table of contents and to offer informed
anticipations of significant future developments in that field.
An undertaking of this scope calls for handbook editors and chapter authors
who are established scholars in the areas about which they write. Many of the
nation’s and the world’s most productive and best-respected psychologists have
agreed to edit Library handbooks or write authoritative chapters in their areas of
expertise.
For whom has the Oxford Library of Psychology been written? Because of its
breadth, depth, and accessibility, the Library serves a diverse audience,
including graduate students in psychology and their faculty mentors, scholars,
researchers, and practitioners in psychology and related fields. Each will find in
the Library the information they seek on the subfield or focal area of psychology
in which they work or are interested.
Befitting its commitment to accessibility, each handbook includes a
comprehensive index, as well as extensive references to help guide research.
And because the Library was designed from its inception as an online as well as
a print resource, its structure and contents will be readily and rationally
searchable online. Furthermore, once the Library is released online, the
handbooks will be regularly and thoroughly updated.
In summary, the Oxford Library of Psychology will grow organically to
provide a thoroughly informed perspective on the field of psychology, one that
reflects both psychology’s dynamism and its increasing interdisciplinarity. Once
it is published electronically, the Library is also destined to become a uniquely
valuable interactive tool, with extended search and browsing capabilities. As you
begin to consult this handbook, we sincerely hope you will share our enthusiasm
for the more than 500-year tradition of Oxford University Press for excellence,
innovation, and quality, as exemplified by the Oxford Library of Psychology.
Peter E. Nathan
Editor-in-Chief
Oxford Library of Psychology
ABOUT THE EDITOR
Thomas A. Widiger
Thomas A. Widiger is the T. Marshall Hahn Professor of Psychology at the
University of Kentucky. He has published extensively on personality and
personality disorders, in over 500 articles and chapters. He currently serves as
co-editor of the Annual Review of Clinical Psychology and editor of Personality
Disorders: Theory, Research, and Treatment. In 2010, he received the
Distinguished Scientist Award from the Society for the Science of Clinical
Psychology; in 2013, he was given the Joseph Zubin Award by the Society for
Research in Psychopathology; and in 2013, the Senior Investigator Award by the
North American Society for the Study of Personality Disorders.
CONTRIBUTORS
Nadia Al-Dajani
Department of Psychological Clinical Science
University of Toronto Scarborough
Toronto, Ontario, Canada
Timothy A. Allen
Department of Psychology
University of Minnesota
Minneapolis, Minnesota
Jüri Allik
University of Tartu
Tartu, Estonia
R. Michael Bagby
Departments of Psychology and Psychiatry
Centre for Addiction and Mental Health
University of Toronto
Toronto, Canada
Paul T. Costa Jr.
Behavioral Medicine Research Centers
Department of Psychiatry and Behavioral Science
Duke University School of Medicine
Durham, North Carolina
Cristina Crego
University of Kentucky
Lexington, Kentucky
Marleen De Bolle
Ghent University
Ghent, Belgium
Barbara De Clercq
Ghent University
Ghent, Belgium
Filip De Fruyt
Ghent University
Ghent, Belgium
Sarah S. W. De Pauw
Ghent University
Ghent, Belgium
Boele de Raad
Faculty of Behavioral and Social Sciences
University of Groningen
Groningen, the Netherlands
David S. DeGeest
Department of Management and Marketing
Faculty of Business
Hong Kong Polytechnic University
Kowloon, Hong Kong
Colin G. DeYoung
University of Minnesota
Minneapolis, Minnesota
Howard S. Friedman
University of California, Riverside
Riverside, California
Marieke C. Gartner
Philadelphia Zoo in Philadelphia
Pennsylvania, USA
Whitney L. Gore
Lexington, Kentucky
Tara M. Gralnick
Department of Psychology and Neuroscience
Dalhousie University
Halifax, Nova Scotia, Canada
William G. Graziano
Department of Psychological Sciences
Purdue University
West Lafayette, Indiana
Ashley C. Helle
Oklahoma State University
Stillwater, Oklahoma
Joshua J. Jackson
Department of Psychological and Brain Sciences
Washington University at St. Louis
St. Louis, Missouri
Amber M. Jarnecke
Purdue University
Margaret L. Kern
Melbourne Graduate School of Education
The University of Melbourne
Melbourne, Australia
Benjamin B. Lahey
University of Chicago
Chicago, Illinois
Robert R. McCrae
Gloucester, Massachusetts
Boris Mlačić
Ivo Pilar Institute of Social Sciences
Zagreb, Croatia
Stephanie N. Mullins-Sweatt
Oklahoma State University
Stillwater, Oklahoma
Ericka Nus
University at Buffalo
Buffalo, New York
Brian P. O’Connor
University of British Columbia
Kelowna, British Columbia, Canada
Joshua R. Oltmanns
Lexington, Kentucky
Ralph L. Piedmont
Department of Pastoral Counseling and Spiritual Care
Loyola University, Maryland
Baltimore, Maryland
Anu Realo
University of Warwick
Coventry, United Kingdom
William Revelle
Northwestern University
Evanston, Illinois
Brent W. Roberts
University of Illinois Urbana-Champaign
Champaign, Illinois
Thomas E. Rodgerson
Department of Pastoral Counseling and Spiritual Care
Loyola University Maryland
Baltimore, Maryland
Stephanie L. Rojas
Lexington, Kentucky
Douglas B. Samuel
Purdue University
Scott E. Seibert
Department of Management & Organizations
Tippie College of Business
The University of Iowa
Iowa City, Iowa
Leonard J. Simms
Buffalo, New York
Susan C. South
Purdue University
Angelina R. Sutin
Laboratory of Personality and Cognition at the National Institute on Aging
Baltimore, Maryland
Jennifer L. Tackett
Department of Clinical Psychology
University of Houston
Houston, Texas
Renée M. Tobin
Illinois State University
Normal, Illinois
Amanda A. Uliaszek
Personality, Psychopathology, and Psychotherapy Laboratory
University of Toronto Scarborough
Toronto, Canada
Alexander Weiss
School of Philosophy, Psychology and Language Sciences
The University of Edinburgh
Edinburgh, Scotland, United Kingdom
Thomas A. Widiger
Lexington, Kentucky
Trevor F. Williams
Buffalo, New York
Joshua Wilt
Northwestern University
Evanston, Illinois
Aidan G. C. Wright
University of Pittsburgh
Pittsburgh, Pennsylvania
CONTENTS
1. Introduction
Thomas A. Widiger
Section 1 • The Five Factor Model

2. The NEO Inventories as Instruments of Psychological Theory
Paul T. Costa Jr. and Robert R. McCrae
3. Neuroticism
Jennifer L. Tackett and Benjamin B. Lahey
4. Extraversion
Joshua Wilt and William Revelle
5. Openness
Angelina R. Sutin
6. Agreeableness and the Five Factor Model
William G. Graziano and Renée M. Tobin
7. Conscientiousness
Joshua J. Jackson and Brent W. Roberts
Section 2 • Construct Validity

8. Robustness
Brian P. O’Connor
9. Universal and Specific in the Five Factor Model of Personality
Jüri Allik and Anu Realo
10. The Lexical Foundation of the Big Five Factor Model
Boele de Raad and Boris Mlačić
11. Factor Analytic Support for the Five Factor Model
Aidan G. C. Wright
12. Childhood Personality and Temperament
Sarah S. W. De Pauw
13. Animal Personality
Alexander Weiss and Marieke C. Gartner
14. Behavior and Molecular Genetics of the Five Factor Model
Amber M. Jarnecke and Susan C. South
15. Personality Neuroscience and the Five Factor Model
Timothy A. Allen and Colin G. DeYoung
Section 3 • Applications
16. Assessment of the Five Factor Model
Leonard J. Simms, Trevor F. Williams, and Ericka Nus
17. The Five Factor Model of Personality in Business and Industry
Scott E. Seibert and David S. DeGeest
18. Health Psychology
Margaret L. Kern and Howard S. Friedman
19. Cross-Over Analysis: Using the Five Factor Model and NEO Personality
Inventory-3 for Assessing Compatibility and Conflict in Couples
Ralph L. Piedmont and Thomas E. Rodgerson
20. Five Factor Model and Personality Disorder
Thomas A. Widiger, Whitney L. Gore, Cristina Crego, Stephanie L. Rojas,
and Joshua R. Oltmanns
21. Axis I Disorders
R. Michael Bagby, Amanda A. Uliaszek, Tara M. Gralnick, and Nadia Al-
Dajani
22. The Five Factor Model of Personality and Consequential Outcomes in
Childhood and Adolescence
Filip De Fruyt, Barbara De Clercq, and Marleen De Bolle
23. Clinical Utility of the Five Factor Model
Stephanie N. Mullins-Sweatt, Douglas B. Samuel, and Ashley C. Helle
Section 4 • Conclusions
24. A Five Factor Discussion
Thomas A. Widiger
Index
Introduction
Thomas A. Widiger
Abstract
This book concerns the Five Factor Model (FFM) of general personality structure.
It brings together much of the research literature on the FFM and demonstrates its
potential applications across a wide range of disciplines and concerns. The book
is organized into four sections: the first section explores the FFM and its domains,
the second focuses on matters and issues concerning the construct validity of the
FFM, the third discusses applications of the FFM to a variety of social and
clinical issues, and the fourth summarizes the book’s interesting points and
considers potential implications. Topics range from Neuroticism and Extraversion
to Openness, Agreeableness, and Conscientiousness. The book also considers the
universality of the FFM, the factor analytic support, childhood temperament and
personality, animal personality, behavior and molecular genetics, personality
neuroscience, personality disorders, adult psychopathology, and child
psychopathology.
Key Words: Five Factor Model, personality structure, Neuroticism, Extraversion,
Conscientiousness, childhood temperament, animal personality, molecular
genetics, personality neuroscience, personality disorders
John, Naumann, and Soto (2008) asserted, in their chapter within the widely
cited text edited by John, Robins, and Pervin (2008) on personality theory and
research, that “after decades of research, the field has now achieved an initial
consensus on a general taxonomy of personality traits, the ‘Big Five’ personality
dimensions” (p. 116). This proclamation was perhaps overstated. Few areas of
psychology can be said to have achieved consensus within an area of
investigation that was once dominated by critical debate and sharp dispute. Even
within this text, we will observe arguments and findings inconsistent with the
Big Five perspective. Unanimity of opinion, and perhaps even an established
consensus, may not really be achievable. But that is to be expected, if not
embraced. Scientific research is driven largely by differences of opinion and
ongoing debate (Popper, 1963), and the structure of personality remains a richly
productive line of investigation.
Nevertheless, consistent with the assertion of John, Naumann, et al. (2008), it
does seem clearly evident that the Five Factor Model (FFM) of general
personality structure has a singular strength and predominance within
psychology. There are many reasons for this recognition. One is that the FFM
has been richly successful in providing an integrative trait model. Its ready
accommodation of the other predominant models of personality allows it not to
compete with these models. As expressed by John, Naumann, et al. (2008),
“rather than replacing all previous systems, the Big Five taxonomy serves an
integrative function because it can represent the various and diverse systems of
personality description in a common framework” (p. 116). We can recover and
conceptualize most alternative dimensional trait models within the FFM
(O’Connor, 2002, 2005; John, Naumann, et al., 2008; McCrae & Costa, 2003).
We can also move higher up in the hierarchical model, working within a two-
factor, three-factor, or four-factor model from which the FFM can be derived
but, as suggested by Markon, Krueger, and Watson (2005), “the Big Five traits
occupy an important, unique position in the hierarchy” (p. 154).
Authors of reviews of personality trait research repeatedly find that the FFM
has proven to be very fruitful as a means of organizing a respective body of
literature. Examples include the trait literature concerning temperament (Caspi,
Roberts, & Shiner, 2005; Shiner & Caspi, 2003), temporal stability (Roberts &
Del Vecchio, 2000), gender (Feingold, 1994), health psychology (Friedman &
Kern, 2014; Hampson & Friedman, 2008; Segerstrom, 2000), positive and
negative life outcomes (Ozer & Benet-Martinez, 2006; Roberts, Kuncel, Shiner,
Caspi, & Goldberg, 2007), personality disorder (Clark, 2007; Widiger & Trull,
2007), and even animal species behavior (Weinstein, Capitanio, & Gosling,
2008). The relative ease with which the FFM has been useful for such purposes
reflects in part its coherence and comprehensiveness, as well as its naturally
compelling inherent structure.
The FFM structure is comfortably aligned with how the trait language is itself
organized, particularly within the English language. It is relatively easy to
conceptualize personality trait literature in terms of the FFM in part because the
organization of trait terms within the English language is itself consistent with
the FFM. It may indeed be the case (De Raad et al., 2010) that the first three
domains of the FFM (i.e., Extraversion, Agreeableness, and Conscientiousness)
are more unequivocally universal across languages than the latter two (i.e.,
Neuroticism and Openness), but it is also apparent that all five domains are
necessary to fully account for personality trait research. Imagine the
conceptualization of personality disorders and psychopathology more generally
without the domain of Neuroticism. Neuroticism may not be as universal as
Extraversion, but it would appear to be more than sufficiently universal. In any
case, if our thoughts are in terms of the English language, then it is quite natural
to conceptualize personality in a manner consistent with the FFM.
A further strength of the FFM is that it is atheoretical. There are certainly
theoretical models for the existence, etiology, and structure of the FFM (McCrae
& Costa, 2003; Wiggins, 1996a), but the FFM structure is itself theoretically
neutral. This substantially facilitates the ability for alternative theoretical models
of personality (e.g., neurobiological, interpersonal, cognitive, and
psychodynamic) to work together within the same model of personality
structure. There is no necessary presumption as to how a person might have
developed, for instance, a disposition for angry hostility, anxiousness, warmth,
callousness, suspiciousness, or excitement seeking. Persons from alternative
theoretical perspectives, such as the neurobiological (DeYoung et al., 2010), the
interpersonal (Wiggins & Trapnell, 1996), and the psychodynamic (Stone,
2013), have worked comfortably within the FFM.
Wiggins (1996a) edited a text that provided five alternative theoretical models
for the FFM, with chapters by Digman (1996), Saucier and Goldberg (1996),
McCrae and Costa (1996), Wiggins and Trapnell (1996), Hogan (1996), and
Buss (1996). “These essays [were] meant to illustrate (1) the diversity of
theoretical perspectives that are currently being brought to bear on the FFM of
personality and (2) the opportunities the FFM can provide for communication
and the sharing of ideas among some of the major figures in contemporary
personality research, investigators whose sub-disciplinary ‘boundaries’ might
otherwise have proven less permeable to such exchanges” (Wiggins, 1996b, p.
viii). The current text will hopefully serve a similar function by informing one
area of FFM investigation (e.g., research on Neuroticism, childhood antecedents,
or universality) of all that is going on within another area of FFM investigation
(e.g., research on Extraversion, adult psychopathology, and assessment). To
date, there has not been a text that brings together in one location all that is
known about the FFM from different research programs.
The strength of a theoretical model is also evidenced by its ability to survive
critical review (Meehl, 1978; Popper, 1963). The FFM, as a predominant model
of general personality structure, has certainly been subjected to tough critical and
empirical tests (e.g., Block, 1995, 2010; Shedler & Westen, 2004). It has not
only survived (e.g., DeYoung, 2010b; John & Naumann, 2010; McCrae, 2010),
it has arguably thrived, amassing a rich and extensive body of empirical
research.
Indeed, a major strength of the FFM is the considerable body of empirical
research in support of its validity that has now accumulated over the past few
decades. A primary purpose of this text is to bring much of this research
literature together within one location, as well as demonstrate its potential
application across a wide array of disciplines and concerns. There are other
notable texts devoted to the FFM, including, as noted earlier, the overview of
alternative theoretical models edited by Wiggins (1996). De Raad and Perugini
(2002) co-edited a text devoted simply to alternative measures for its assessment.
The mere existence of this book was itself a testament to the importance and
impact of the FFM, in that simply its assessment warrants an entire text. De
Raad (2000) devoted a text to its lexical foundation. McCrae and Allik (2002)
co-edited a text devoted to the cross-cultural and cross-language application of
the FFM, documenting well its universal appeal and relevance. Finally, of
course, McCrae and Costa (2003) composed a text on the FFM, summarizing the
vast body of FFM research conducted by themselves and other investigative
teams, presenting therein their own particular theoretical model for its
understanding.
The current text is organized into four sections. The first section concerns the
FFM and its domains, including a chapter covering Neuroticism by Jennifer
Tackett and Benjamin Lahey (Lahey, 2009; Tackett, 2006), Extraversion by
Joshua Wilt and William Revelle (Revelle, Wilt, & Condon, 2011; Wilt &
Revelle, 2009), Openness by Angelina Sutin (McCrae & Sutin, 2009; Sutin,
Beason-Held, Resnick, & Costa, 2009), Agreeableness by William Graziano and
Renée Tobin (Graziano & Eisenberg, 1997; Graziano & Tobin, 2009, 2013), and
Conscientiousness by Joshua Jackson and Brent Roberts (Jackson et al., 2010;
Roberts, Chernyshenko, Stark, & Goldberg, 2005; Roberts, Jackson, Fayard,
Edmonds, & Meints, 2009). I believe readers unfamiliar with this literature will
be struck and impressed by the sheer volume of theory and research devoted to
each particular domain of the FFM. There is indeed a vast body of research
devoted simply to Neuroticism, Extraversion, Openness, Agreeableness, and
Conscientiousness, and their many facets. I am truly grateful to the authors of
these chapters for providing these overviews, as the task was quite daunting.
Introducing this section, as well as perhaps the book itself, is an overview of
the FFM by Paul Costa and Jeff (Robert) McCrae. Their richly productive and
internationally collaborative laboratory was highly instrumental in the
recognition, growth, and development of the FFM. I was myself introduced to
the FFM by Dr. Paul Costa. At the time I was serving as the Research
Coordinator for the American Psychiatric Association’s fourth edition of the
Diagnostic and Statistical Manual of Mental Disorders (DSM-IV; American
Psychiatric Association, 1994), with an office in New York City. I had
collaborated on a number of projects with Allen Frances, Chair of DSM-IV. One
such product was a review paper on personality disorders, arguing that the
diagnosis and classification of the APA personality disorders should be informed
by the research literature within psychology, including the research on
dimensional trait models (e.g., Widiger & Frances, 1985). At that time, I was
attempting to describe and differentiate the APA personality disorders with
respect to the interpersonal circumplex (IPC), but was nevertheless having some
difficulty adequately understanding the borderline, obsessive–compulsive, and
schizotypal personality disorders with respect to the IPC (Widiger & Frances,
1985; Widiger & Hagemoser, 1997). Paul Costa called me on the phone and
asked if he could come for a visit. I agreed, and he introduced me to the FFM. In
the course of this meeting it became readily apparent that the additional domains
of Neuroticism, Conscientiousness, and Openness (i.e., beyond the interpersonal
domains of Agreeableness and Extraversion) addressed extremely well the
difficulties I was having with respect to the coverage of the borderline,
obsessive–compulsive, and schizotypal personality disorders, respectively. Dr.
Costa and I subsequently developed a productive collaboration (e.g., Costa &
Widiger, 1994; Widiger & Costa, 1994). It is with considerable pleasure that the
first chapter is by Paul Costa and Robert McCrae.
The second section of the text is devoted to matters and issues concerning the
construct validity of the FFM. Both junior and senior investigators were invited
to provide chapters. Authors were free to express their own views and opinions,
and it should become apparent that the authors are not simply advocates for the
FFM. A variety of topics could be covered. Hopefully the reader will find that
the key points of concern were indeed addressed, and in a manner that was
objective, fair, thorough, and accurate. It is also apparent that each of the authors
addressed key methodological, analytic, conceptual, and substantive issues that
warrant further research and attention.
The second section begins with an overview of the robustness of the FFM by
Brian O’Connor (O’Connor 2002, 2005), followed by the universality of the
FFM by Jüri Allik and Anu Realo (Allik, 2006; Allik, Realo, & McCrae, 2013;
McCrae & Allik, 2002), the lexical foundation by Boele de Raad and Boris
Mlačić (De Raad, 2000; De Raad et al., 2010; De Raad & Perugini, 2002; Mlačić
& Goldberg, 2007; Mlačić & Ostendorf, 2005), the factor analytic support by
Aidan Wright (Wright & Simms, 2014; Wright & Zimmermann, 2015),
childhood temperament and personality by Sarah De Pauw (De Pauw &
Mervielde, 2010; Mervielde & De Pauw, 2012), animal personality by
Alexander Weiss and Marieke Gartner (Gartner & Weiss, 2013; Weiss, Adams,
Widdig, & Gerald, 2011; Weiss, Gartner, Gold, & Toinksi, 2013), behavior and
molecular genetics by Amber Jarnecke and Susan South (South & DeYoung,
2013; South & Krueger, 2014; South, Reichborn-Kjennerud, Eaton, & Krueger,
2012), and ending with personality neuroscience by Timothy Allen and Colin
DeYoung (DeYoung, 2010a; DeYoung et al., 2010). It is evident from this series
of chapters that there is an incredible body of research that supports and informs
the validity of FFM personality structure. There are certainly gaps and
differences of opinion, but the FFM is clearly a model of personality structure
that is very heavily researched.
The third section of the text concerns applications of the FFM to a variety of
social and clinical issues. The FFM is not just an abstract, academic model of
personality structure. FFM personality traits have been associated with a wide
array of important, consequential life outcomes (Ozer & Benet-Martinez, 2006;
Roberts, Kuncel, Shiner, Caspi, & Goldberg, 2007). Much of these outcomes
and more are considered in some depth within this section.
Section Three begins with an overview of the assessment of the FFM by
Leonard Simms, Trevor Williams, and Ericka Simms (Simms et al., 2011;
Wright & Simms, 2014). Central to any application of the FFM will be its proper
assessment. This chapter is followed by a chapter on the application of the FFM
to business and industry by Scott Seibert and David DeGeest (DeGeest &
Schmidt, 2010; Seibert & Kraimer, 2001; Zhao & Seibert, 2006), health
psychology by Margaret Kern and Howard Friedman (Friedman, 2007;
Friedman & Kern, 2014; Kern & Friedman, 2010), marital and family therapy by
Ralph Piedmont and Thomas Rodgerson (Costa & Piedmont, 2003; Piedmont,
1998; Piedmont & Rodgerson, 2013), personality disorders by Thomas Widiger,
Whitney Gore, Cristina Crego, Stephanie Rojas, and Joshua Oltmanns (Widiger
& Costa, 2013; Widiger, Crego, & Oltmanns, 2015; Widiger, Samuel, Mullins-
Sweat, Gore, & Crego, 2012), adult psychopathology by Michael Bagby,
Amanda Uliaszek, Tara Gralnick, and Nadia Al-Dajani (Bagby, Quilty, &
Ryder, 2008a; Bagby et al., 2008b; Costa, Bagby, Herbst, & McCrae, 2005), and
child psychopathology by Filip De Fruyt, Barbara De Clercq, and Marleen De
Bolle (De Bolle, Beyers, De Clercq, & De Fruyt, 2012; De Bolle & Tackett,
2013; De Clercq & De Fruyt, 2012; De Fruyt & De Clercq, 2013, 2014). The
section ends with a discussion of the concept of clinical utility and its relevance
to the application of the FFM to personality disorder conceptualization and
treatment by Stephanie Mullins-Sweatt, Douglas B. Samuel, and Ashley C. Helle
(Mullins-Sweatt, 2013; Mullins-Sweatt & Lengel, 2012; Mullins-Sweatt, Lengel,
& DeShong, 2015; Samuel & Widiger, 2006).
The fourth and final section of the book is a concluding chapter by myself. In
this section I do my best to summarize interesting points from each chapter and
discuss potential implications. Hopefully I represent each well, as I am very,
very grateful for the willingness of these friends and colleagues to provide their
chapters, and their outstanding success in having done so.
References
American Psychiatric Association. (1994). Diagnostic and statistical manual of mental
disorders (4th ed.). Washington, DC: Author.
Allik, J. (2006). Personality dimensions across cultures. In T. A. Widiger, E. Simonsen,
P. J. Sirovatka, & D. A. Regier (Eds.), Dimensional models of personality disorders.
Refining the research agenda for DSM-V (pp. 117–132). Washington, DC: American
Psychiatric Association.
Allik, J., Realo, A., & McCrae, R. R. (2013). Universality of the five-factor model of
personality. In T. A. Widiger & P. T. Costa (Eds.), Personality disorders and the five
factor model of personality (3rd ed., pp. 61–74). Washington, DC: American
Psychological Association.
Bagby, R. M., Quilty, L. C., & Ryder, A. C., (2008a). Personality and depression. The
Canadian Journal of Psychiatry, 53, 14–25.
Bagby, R. M., Quilty, L. C., Segal, Z. V., McBride, C. C., Kennedy, S. H., & Costa, P. T.
(2008b). Personality and differential treatment response in major depression: A
randomized controlled trial comparing cognitive-behavioural therapy and
pharmacotherapy. Canadian Journal of Psychiatry, 53, 361–370.
Block, J. (1995). A contrarian view of the five-factor approach to personality description.
Psychological Bulletin, 117, 187–215.
Block, J. (2010). The five-factor framing of personality and beyond: Some ruminations.
Psychological Inquiry, 21, 2–25.
Buss, D. (1996). Social adaptation and five major factors of personality. In J. S. Wiggins
(Ed.), The five-factor model of personality. Theoretical perspectives (pp. 180–207).
New York, NY: Guilford Press.
Caspi, A., Roberts, B. W., & Shiner, R. L. (2005). Personality development: Stability and
change. Annual Review of Psychology, 56, 453–484.
Clark, L. A. (2007). Assessment and diagnosis of personality disorder: Perennial issues
and an emerging reconceptualization. Annual Review of Psychology, 57, 227–257.
Costa, P. T., Bagby, R. M., Herbst, J. H., & McCrae, R. R. (2005). Personality self-
reports are concurrently reliable and valid during acute depressive episodes. Journal of
Affective Disorders, 89(1), 45–55.
Costa, P. T., & Piedmont, R. L. (2003). Multivariate assessment: NEO PI-R profiles of
Madeline G. In J. S. Wiggins (Ed.), Paradigms of personality assessment (pp. 262–
280). New York, NY: Guilford Press.
Costa, P. T., & Widiger, T. A. (Eds.), (1994). Personality disorders and the five-factor
model of personality. Washington, DC: American Psychological Association.
De Bolle, M., Beyers, W., De Clercq, B., & De Fruyt, F. (2012). General personality and
psychopathology in referred and nonreferred children and adolescents: An investigation
of continuity, pathoplasty and complication models. Journal of Abnormal Psychology,
121, 958–970.
De Bolle, M., & Tackett, J. L. (2013). Anchoring bullying and victimization in children
within a Five-Factor-Model-based person-centered framework. European Journal of
Personality, 27(3), 280–289.
De Clercq, B., & De Fruyt, F. (2012). A five-factor model framework for understanding
childhood personality disorder antecedents. Journal of Personality, 80, 1533–1563.
De Fruyt, F., & De Clercq, B. (2013). Childhood antecedents of personality disorder: A
five factor perspective. In T. A. Widiger & P. T. Costa (Eds.), Personality disorders
and the five-factor model of personality (3rd. ed., pp. 43–60). Washington, DC:
American Psychological Association.
De Fruyt, F., & De Clercq, B. (2014). Antecedents of personality disorder in childhood
and adolescence: Toward an integrative developmental model. Annual Review of
Clinical Psychology, 10, 449–476.
DeGeest, D. S., & Schmidt, F. L. (2010). The impact of research synthesis methods on
industrial–organizational psychology: The road from pessimism to optimism about
cumulative knowledge. Research Synthesis Methods, 1(3–4), 185–197.
De Pauw, S. S. W., & Mervielde, I. (2010). Temperament, personality and developmental
psychopathology: A review based on the conceptual dimensions underlying childhood
traits. Child Psychiatry & Human Development, 41(3), 313–329.
De Raad, B. (2000). The Big Five personality factors: The psycholexical approach to
personality. Seattle, WA: Hogrefe & Huber.
De Raad, B., Barelds, D. P. H., Levert, E., Ostendorf, F., Mlacic, B., Blais, L. D., …
Katigbak, M. S. (2010). Only three factors of personality description are fully
replicable across languages: A comparison of 14 trait taxonomies. Journal of
Personality and Social Psychology, 98, 160–173.
De Raad, B., & Perugini, M. (Eds.). (2002). Big five assessment. Bern, Switzerland:
Hogrefe & Huber.
DeYoung, C. G. (2010a). Personality neuroscience and the biology of traits. Social and
Personality Psychology Compass, 4(12), 1165–1180.
DeYoung, C. G. (2010b). Toward a theory of the Big Five. Psychological Inquiry, 21,
26–33.
DeYoung, C. G., Hirsh, J. B., Shane, M. S., Papademetris, X., Rajeevan, N., & Gray, J. R.
(2010). Testing predictions from personality neuroscience: Brain structure and the big
five. Psychological Science, 21, 820–828.
Digman, J. M. (1996). The curious history of the five-factor model. In J. S. Wiggins
(Ed.), The five-factor model of personality. Theoretical perspectives (pp. 1–20). New
York, NY: Guilford Press.
Feingold, A. (1994). Gender differences in personality: A meta-analysis. Psychological
Bulletin, 116, 429–456.
Friedman, H. S. (2007). Personality, disease, and self-healing. In H. S. Friedman & R. C.
Silver (Eds.), Foundations of health psychology (pp. 172–199). New York, NY: Oxford
University Press.
Friedman, H. S., & Kern, M. L. (2014). Personality, well-being, and health. Annual
Review of Psychology, 65, 719–742.
Gartner, M. C., & Weiss, A. (2013), Scottish wildcat (Felis silvestris grampia) personality
and subjective well-being: Implications for captive management. Applied Animal
Behaviour Science, 147, 261–267.
Graziano, W. G., & Eisenberg, N. (1997). Agreeableness: A dimension of personality. In
R. Hogan, J. Johnson, & S. Briggs (Eds.), Handbook of personality psychology (pp.
795–824). San Diego, CA: Academic Press.
Graziano, W. G., & Tobin, R. M. (2009). Agreeableness. In M. Leary & R. Hoyle (Eds.),
Handbook of individual differences in social behavior (pp. 46–61). New York, NY:
Guilford Press.
Graziano, W. G., & Tobin, R. M. (2013). The cognitive and motivational foundations
underlying agreeableness. In M. D. Robinson, E. R. Watkins, & E. Harmon-Jones
(Eds.), Handbook of cognition and emotion (pp. 347–364). New York, NY: Guilford
Press.
Hampson, S. E., & Friedman, H. S. (2008). Personality and health. A lifespan
perspective. In O. P. John, R. W. Robins, & L. A. Pervin (Eds.), Handbook of
personality. Theory and research (3rd. ed., pp. 770–794). New York, NY: Guilford
Press.
Hogan, R. (1996). A socioanalytic perspective on the five-factor model. In J. S. Wiggins
(Ed.), The five-factor model of personality. Theoretical perspectives (pp. 163–179).
Jackson, J. J., Wood, D., Bogg, T., Walton, K. E., Harms, P. D., & Roberts, B. W. (2010).
What do conscientious people do? Development and validation of the behavioral
indicators of conscientiousness (BIC). Journal of Research in Personality, 44(4), 501–
511.
John, O. P., & Naumann, L. P. (2010). Surviving two critiques by Block? The resilient
Big Five have emerged as the paradigm for personality trait psychology. Psychological
Inquiry, 21, 44–49.
John, O. P., Naumann, L. P., & Soto, C. J. (2008). Paradigm shift to the integrative Big
Five trait taxonomy: History, measurement, and conceptual issues. In O. P. John, R. R.
Robins, & L. A. Pervin (Eds.), Handbook of personality. Theory and research (3rd. ed.,
pp. 114–158). New York, NY: Guilford Press.
John, O. P., Robins, R. R., & Pervin, L. A. (Eds.). (2008). Handbook of personality.
Theory and research (3rd. ed.). New York, NY: Guilford Press.
Kern, M. L., & Friedman, H. S. (2010). Why do some people thrive while others succumb
to disease and stagnation? Personality, social relations, and resilience. In P. S. Fry & C.
L. M. Keyes (Eds.), Frontiers of resilient aging (pp. 162–184). New York: Cambridge
University Press.
Lahey, B. B. (2009). Public health significance of neuroticism. American Psychologist,
64, 241–256.
Markon, K. E., Krueger, R. F., & Watson, D. (2005). Delineating the structure of normal
and abnormal personality: An integrative hierarchical approach. Journal of Personality
and Social Psychology, 88, 139–157.
McCrae, R. R. (2010). The place of the FFM in personality psychology. Psychological
Inquiry, 21, 57–64.
McCrae, R. R., & Allik, J. (Eds.). (2002). The five-factor model of personality across
cultures. New York, NY: Kluwer Academic.
McCrae, R. R., & Costa, P. T. (1996). Toward a new generation of personality theories:
Theoretical contexts for the five-factor model. In J. S. Wiggins (Ed.), The five-factor
model of personality. Theoretical perspectives (pp. 51–87). New York, NY: Guilford
Press.
McCrae, R. R., & Costa, P. T. (2003). Personality in adulthood. A five-factor theory
perspective (2nd ed.). New York, NY: Guilford Press.
McCrae, R. R., & Sutin, A. R. (2009). Openness to experience. In M. Leary & R. Hoyle
(Eds.), Handbook of individual differences in social behavior (pp. 257–273). New
Meehl, P. E. (1978). Theoretical risks and tabular asterisks: Sir Karl, Sir Ronald, and the
slow progress of soft psychology. Journal of Consulting and Clinical Psychology, 46,
806–834.
Mervielde, I., & De Pauw, S. S. W. (2012). Models of child temperament. In M. Zentner
& R. L. Shiner (Eds.), Handbook of temperament (pp. 21–40). New York, NY:
Guilford Press.
Mlačić, B., & Goldberg, L. R. (2007). An analysis of a cross-cultural personality
inventory: The IPIP Big Five factor markers in Croatia. Journal of Personality
Assessment, 88, 168–177.
Mlačić, B., & Ostendorf, F. (2005). Taxonomy and structure of Croatian personality-
descriptive adjectives. European Journal of Personality, 19, 117–152.
Mullins-Sweatt, S. N. (2013). Conceptual and empirical support for the clinical utility of
five factor model personality disorder diagnoses. In T. A. Widiger & P. T. Costa, Jr.
(Eds.), Personality disorders and the five-factor model of dersonality (3rd ed., pp. 311–
324). Washington, DC: American Psychological Association.
Mullins-Sweatt, S. N., & Lengel, G. J. (2012). Clinical utility of the five-factor model of
personality disorder. Journal of Personality, 80, 1615–1639.
Mullins-Sweatt, S. N., Lengel, G. J., & DeShong, H. L. (2015). The importance of
considering clinical utility in the construction of a diagnostic manual. Annual Review of
Clinical Psychology.
O’Connor, B. P. (2002). A quantitative review of the comprehensiveness of the five-
factor model in relation to popular personality inventories. Assessment, 9, 188–203.
O’Connor, B. P. (2005). A search for consensus on the dimensional structure of
personality disorders. Journal of Clinical Psychology, 61, 323–345.
Ozer, D. J., & Benet-Martinez, V. (2006). Personality and the prediction of consequential
outcomes. Annual Review of Psychology, 57, 401–421.
Piedmont, R. L. (1998). The revised NEO Personality Inventory: Clinical and research
applications. New York, NY: Plenum Press.
Piedmont, R. L., & Rodgerson, T. E. (2013). Crossover analysis: Using the five factor
model and Revised NEO Personality Inventory to assess couples. In T. A. Widiger &
P. T. Costa (Eds.), Personality disorders and the five-factor model of personality (3rd
ed., pp. 375–394). Washington, DC: American Psychological Association.
Popper, K. R. (1963). Conjectures and refutations: The growth of scientific knowledge.
New York, NY: Basic Books.
Revelle, W., Wilt, J., & Condon, D. M. (2011). Individual differences and differential
psychology: A brief history and prospect. In T. Chamorro-Premuzic, S. von Stumm, &
A. Furnham (Eds.), The Wiley-Blackwell handbook of individual differences (pp. 3–38).
New York, NY: Wiley-Blackwell.
Roberts, B. W., Chernyshenko, O. S., Stark, S., & Goldberg, L. R. (2005). The structure
of conscientiousness: An empirical investigation based on seven major personality
questionnaires. Personnel Psychology, 58, 103–139.
Roberts, B. W., & DelVecchio, W. F. (2000). The rank-order consistency of personality
traits from childhood to old age: A quantitative review of longitudinal studies.
Roberts, B. W., Jackson, J. J., Fayard, J. V., Edmonds, G., & Meints, J. (2009).
Conscientiousness. In M. Leary & R. Hoyle (Eds.), Handbook of individual differences
in social behavior (pp. 369–381). New York, NY: Guilford Press.
Roberts, B. W., Kuncel, N. R., Shiner, R., Caspi, A., & Goldberg, L. R. (2007). The
power of personality: The comparative validity of personality traits, socioeconomic
status, and cognitive ability for predicting important life outcomes. Perspectives on
Psychological Science, 2, 313–345.
Samuel, D. B., & Widiger, T. A. (2006). Clinicians’ judgments of clinical utility: A
comparison of the DSM-IV and five-factor models. Journal of Abnormal Psychology,
115, 298–308.
Saucier, G., & Goldberg, L. R. (1996). The language of personality: Lexical perspectives
on the five-factor model. In J. S. Wiggins (Ed.), The five-factor model of personality.
Theoretical perspectives (pp. 21–50). New York, NY: Guilford Press.
Segerstrom, S. C. (2000). Personality and the immune system: Models, methods, and
mechanisms. Annals of Behavioral Medicine, 22, 180–190.
Seibert, S. E., & Kraimer, M. L. (2001). The five-factor model of personality and career
success. Journal of Vocational Behavior, 58, 1–21.
Shedler, J. & Westen, D. (2004). Dimensions of personality pathology: An alternative to
the Five-Factor Model. American Journal of Psychiatry, 161, 1743–1754.
Shiner, R. L., & Caspi, A. (2003). Personality differences in childhood and adolescence:
Measurement, development, and consequences. Journal of Child Psychology and
Psychiatry, 44, 2–32.
Simms, L. J., Goldberg, L. R., Roberts, J. E., Watson, D., Welte, J., & Rotterman, J. H.
(2011). Computerized adaptive assessment of personality disorder: Introducing the
CAT-PD project. Journal of Personality Assessment, 93(4), 380–389.
South, S. C., & DeYoung, N. J. (2013). Behavior genetics of personality disorders:
Informing classification and conceptualization in DSM-5. Personality Disorders:
Theory, Research, and Treatment, 4, 270–283.
South, S. C., & Krueger, R. F. (2014). Genetic strategies for probing conscientiousness
and its relationship to aging. Developmental Psychology, 50, 1362–1376.
South, S. C., Reichborn-Kjennerud, T., Eaton, N. R., & Krueger, R. F. (2012). Behavior
and molecular genetics of personality disorders. In T. A. Widiger (Ed.), The Oxford
handbook of personality disorders (pp. 143–165). New York, NY: Oxford University
Press.
Stone, M. H. (2013). Treatment of personality disorders from the perspective of the five-
factor model. In T. A. Widiger & P. T. Costa (Eds.), Personality disorders and the five
Sutin, A. R., Beason-Held, L. L., Resnick, S. M., & Costa, P. T. (2009). Sex differences
in the resting-state neural correlates of openness to experience among older adults.
Cerebral Cortex, 19, 2797–2802.
Tackett, J. L. (2006). Evaluating models of the personality-psychopathology relationship
in children & adolescents. Clinical Psychology Review, 26, 584–599.
Weinstein, T. R., Capitanio, J. P., & Gosling, S. D. (2008). Personality in animals. In O.
P. John, R. W. Robins, & L. A. Pervin (Eds.), Handbook of personality (3rd ed., pp.
328–348). New York, NY: Guilford Press.
Weiss, A., Adams, M. J., Widdig, A., & Gerald, M. S. (2011). Rhesus macaques (Macaca
mulatta) as living fossils of hominoid personality and subjective well-being. Journal of
Comparative Psychology, 125, 72–83.
Weiss, A., Gartner, M. C., Gold, K. S., & Toinksi, C. S. (2013). Extraversion predicts
longer survival in gorillas: An 18-year longitudinal study. Proceedings of the Royal
Society B-Biological Sciences, 280, 20122231.
Widiger, T. A., & Costa, P. T. (1994). Personality and personality disorders. Journal of
Abnormal Psychology, 103, 78–91.
Widiger, T. A., & Costa, P. T. (Eds.). (2013). Personality disorders and the five-factor
model of personality (3rd ed.). Washington, DC: American Psychological Association.
Widiger, T. A., Crego, C., & Oltmanns, J. R. (2015). The validation of a classification of
psychopathology. Psychological Inquiry, 26, 272–278.
Widiger, T. A., & Frances, A. (1985). The DSM-III personality disorders: Perspectives
from psychology. Archives of General Psychiatry, 42, 615–623.
Widiger, T. A., & Hagemoser, S. (1997). Personality disorders and the interpersonal
circumplex. In R. Plutchik & H. R. Conte (Eds.), Circumplex models of personality and
emotions (pp. 299–325). Washington, DC: American Psychological Association.
Widiger, T. A., Samuel, D. B., Mullins-Sweat, S., Gore, W. L., & Crego, C. (2012).
Integrating normal and abnormal personality structure: The five-factor model. In T. A.
Widiger (Ed.), Oxford handbook of personality disorders (pp. 82–107). New York,
NY: Oxford University Press.
Widiger, T. A., & Trull, T. J. (2007). Plate tectonics in the classification of personality
disorder: Shifting to a dimensional model. American Psychologist, 62, 71–83.
Wiggins, J. S. (Ed.). (1996a). The five-factor model of personality. Theoretical
perspectives. New York, NY: Guilford Press.
Wiggins, J. S. (1996b). Preface. In J. S. Wiggins (Ed.), The five-factor model of
personality. Theoretical perspectives (pp. vii–xi). New York, NY: Guilford Press.
Wiggins, J. S., & Trapnell, P. D. (1996). A dyadic-interactional perspective on the five-
factor model. In J. S. Wiggins (Ed.), The five-factor model of personality. Theoretical
perspectives (pp. 88–162). New York, NY: Guilford Press.
Wilt, J., & Revelle, W. (2009). Extraversion. In M. R. Leary & R. H. Hoyle (Eds.),
Handbook of individual differences in social behavior (pp. 27–45). New York, NY:
Guilford Press.
Wright, A. G. C., & Simms, L. J. (2014). On the structure of personality disorder traits:
Conjoint analyses of the CAT-PD, PID-5, and NEO-PI-3 trait models. Personality
Disorders: Theory, Research, and Treatment, 5(1), 43–54.
Wright, A. G. C., & Zimmermann, J. (2015). At the nexus of science and practice:
Answering basic clinical questions in personality disorder assessment and diagnosis
with quantitative modeling techniques. In S. Huprich (Ed.), Personality disorders:
Assessment, diagnosis, and research. Washington, DC: American Psychological
Association.
Zhao, H., & Seibert, S. E. (2006). The big five personality dimensions and entrepreneurial
status: A meta-analytical review. Journal of Applied Psychology, 91(2), 259–271.
SECTION 1
The Five Factor Model

The NEO Inventories as Instruments of Psychological
Theory
Paul T. Costa Jr. and Robert R. McCrae
Abstract
This chapter reviews the contribution of the NEO Inventories and the Five Factor
Model to progress in personality psychology since Loevinger’s 1957 essay.
Personality structure is now viewed as a complex hierarchy of continuously
distributed attributes; the content of this hierarchy consists of traits and their
manifestations as needs, habits, and so on. The chapter also introduces the duality
principle, according to which personality measures must be understood as both
collections of characteristic adaptations and proxy measures of basic tendencies.
Finally, the chapter considers the status of Five Factor Theory, a general theory of
personality intended to account for research findings stimulated by the discovery
and assessment of the Five Factor Model.
Key Words: history of psychology, personality assessment, personality theory,
personality structure, traits
Reading Loevinger’s 1957 monograph, “Objective tests as instruments of

psychological theory,”1 is a bit like watching a documentary recreation of the
evolution of mammals. In the film, we see recognizable prototypes of mice,
horses, and monkeys, but the primordial landscape in which they move is alien
to us. In Loevinger’s brilliant essay we can see the emerging form of modern
psychometrics—the centrality of construct validity, the requirement (yet
unnamed) of both convergent and discriminant validity, the importance of
structural models—but all within the context of a psychology dominated by
behaviorism and psychoanalysis. Loevinger took a realist stance with regard to
the existence of traits, but said nothing about which traits are important; of
today’s familiar domains only introversion and neuroticism are mentioned, both
in passing. Because she wished to construe test responses as instances of
behavior, Loevinger endorsed dichotomous item response formats and rejected
Likert scales, which she saw as requiring judgment on the part of the respondent
—a dubious business in those days before the cognitive revolution. Finally, she
adopted unquestioningly the prevailing opinion that “self-reports concerning
personality traits are subject to such massive distortion as to make them virtually
worthless as direct measurements of personality traits” (Loevinger, 1957, p.
646).
The psychological landscape has changed dramatically over the past 60 years,
due in large part to research on, and using, objective personality tests. Loevinger
would not have been surprised by this, because the fundamental premise of her
monograph was that the enterprise of psychological assessment is intrinsically
bound up with the development of psychological theory. Concepts and broader
theories guide the initial development of tests; psychometric evaluations of the
test lead to the refinement, or replacement, of conceptual interpretations; these in
turn lead to revised measures. In this chapter we focus on how this dialectic has
shaped both the NEO Inventories (McCrae & Costa, 2010) and our current
understanding of personality.
Of course, the interplay of data and theory did not proceed in the orderly
fashion that we will lay out here. In developing the NEO Inventories we
frequently relied on intuitions that became codified as explicit theoretical
premises only many years later, and revisions of the instrument sometimes
lagged behind our conceptualizations—for example, we anticipated that there
would be facet scales for the Agreeableness and Conscientiousness domains
many years before we specified and developed them. Readers interested in a
chronological history can consult other chapters (e.g., Costa & McCrae, 2008).
Metatheoretical Premises Underlying the NEO Inventories

After several years of analyzing personality data (e.g., Costa, Fozard, McCrae,
& Bossé, 1976; Costa & McCrae, 1978) from the ongoing Normative Aging
Study (NAS; Bell, Rose, & Damon, 1972), we decided to create our own
omnibus personality questionnaire, because (like everyone else who embarks on
this project) we believed we could offer something better than the available
instruments. Implicit in that decision was a set of assumptions about human
nature (cf. Hjelle & Siegler, 1976) that we eventually articulated in terms of the
premises of knowability, variability, proactivity, and rationality (McCrae &
Costa, 1996). Knowability refers to the belief that human personality is an
appropriate object for scientific research, not an ineffable metaphysical entity.
This premise distinguishes psychologists from theologians, poets, and mystics.
Variability asserts that people’s personalities differ, and we chose to align
ourselves with differential psychologists instead of those who—with equal
legitimacy—seek universals of human nature. These two metatheoretical
premises are hardly controversial, but in the intellectual climate of the 1970s, the
premises of proactivity and rationality most certainly were.
Proactivity: The Reality of Traits

By proactivity we mean the idea that people are, in substantial measure, the
origins of their own behavior. They are not empty organisms that merely react to
their history of reinforcements, nor are they tabula rasa that take on the character
imprinted by their family and the larger culture. They have an essential nature
(whether innate, as in the classical concept of temperament, or acquired, as in
Allport’s 1937 notion of the functional autonomy of motives) that not only
responds to the press of the environment, but also actively seeks out congenial
circumstances. In short, people have traits. In the years following Loevinger’s
1957 article, that belief had been challenged, most famously by Mischel’s (1968)
critique. A generation of psychologists had begun to assume that traits were
mere attributions, social constructions with no objective basis in fact. In this
view behavior was determined by the situation, and by the ways that people
construed or misconstrued it. A self-report on a personality questionnaire might
tell us something about how people thought of themselves, or how they wished
to present themselves, but it told us nothing about what they were really like,
because there was nothing to tell.
Creating a personality inventory in the 1970s was first and foremost a wager
that this view was wrong. In retrospect this was a rather safe bet: Lay persons
since antiquity and psychologists from Galton to Loevinger had believed that
people could be characterized by their enduring dispositions. Yet so strong was
the antitrait zeitgeist that we remained on the defensive for many years. We were
emboldened to proceed chiefly because of a remarkable finding. We had
inherited data collected years before at the NAS, and when we readministered
the personality questionnaires, we found stability coefficients as high as .84
(Costa & McCrae, 1977a). As researchers hoping to uncover the origins of the
profound developmental changes that everyone assumed must characterize
adulthood, these results were disconcerting: If traits were fictions, why were
they so durable? If they were real, why did they not vary in response to
biological aging, midlife crises, epigenetic growth, and stressful life events? In
place of the intriguing models of adult metamorphosis we had briefly entertained
(Costa & McCrae, 1976), we began to think that William James had been right
when he asserted that “by the age of thirty, the character is set like plaster”
(1981/1890, p. 126).
Yet stability in self descriptions was not conclusive proof of the existence of
traits; there were many alternative explanations for high retest correlations,
including enduring response sets and a crystallized self-concept. We conducted a
number of studies to rule out these possibilities (Costa, McCrae, & Arenberg,
1983; McCrae & Costa, 1982). But more direct evidence was needed—evidence
that went beyond the questionnaire responses of the individual. Funder and
colleagues (e.g., Funder & Sneed, 1993) pursued one form of validation, based
on correlating trait scores with behavior observed in the laboratory. We choose a
different approach: Cross-observer agreement. The lay conception of traits
includes the assumption that traits are usually observable: Everyone can see that
“John” is sociable and “Jane” is diligent. Of course, even laypersons
acknowledge that there are differences of opinion, and that some people may on
occasion conceal their true selves from everyone else. Still, substantial
agreement among observers ought to be the rule. In the 1970s this very basic
requirement for the existence of traits had been challenged, most notably by
Fiske (1978). He believed that significant agreement could be obtained if at all
only by aggregating across many observers, and that the resulting scores “would
still have little conceptual value and would not refer to anything in the
observable world. They are hardly more meaningful than the statement that the
average women is .17 pregnant” (p. 109).
It was this need for empirical evidence on the degree of cross-observer
agreement on personality traits that was behind our decision to create the NEO
Inventories in two forms, S (for self-reports) and R (for observer ratings). Some
earlier research (Jackson, 1974) had validated self-reports on questionnaire
scales using global observer ratings, but that approach had certain limitations,
including the unreliability of the single-item criterion. Our idea was to create
parallel forms, using the same items phrased in first-person or third-person.
Although Form R was initially created to validate Form S, we realized almost
immediately that the same data would allow us to use Form S to validate Form
R, and provide a second instrument. We quickly came to use both versions in the
same study in order to provide multimethod replication of findings—for
example, showing that the associations of personality traits with measures of
psychological well-being were not mere artifacts of shared self-report methods
(Costa & McCrae, 1984).
In 1980 we administered Form S of the 180-item NEO Inventory (assessing
six facets for each of the three domains of Neuroticism, Extraversion, and
Openness) to members of the Augmented Baltimore Longitudinal Study of
Aging (BLSA; Shock et al., 1984), and 6 months later we administered Form R
to their spouses. Results supported our hypotheses, with cross-observer
correlations ranging from .30 to .58 (N = 281, all ps < .001) for the 18 facet
scales (McCrae, 1982). These were very substantial correlations indeed for a
generation that had come to believe that trait correlations never exceeded the
“.30 barrier.” Evidence of consensual validation was again crucial in expanding
the three-domain NEO Inventory into the five-domain NEO Personality
Inventory (NEO-PI; Costa & McCrae, 1985), because the two new domains,
Agreeableness and Conscientiousness, were highly evaluative, and might have
turned out to be nothing more than different evaluative biases in self-reports.
Instead, correlations among peer raters and between mean peer ratings and self-
reports (rs = .28–.47, ps < .01) confirmed that these two factors were substantive
characteristics of individuals (McCrae & Costa, 1987). Cross-observer
agreement has subsequently been replicated in studies in adolescent (McCrae,
Costa, & Martin, 2005), psychiatric (Yang et al., 1999), and cross-cultural
(McCrae et al., 2004) samples. Combined with steadily accumulating evidence
that traits, as assessed by instruments such as the NEO Inventories, are heritable
(Jang, Livesley, & Vernon, 1996) and predictive of consequential life outcomes
(Ozer & Benet-Martínez, 2006), these data demonstrate that traits are real
features of persons. People have a real nature; they understand it, and others
perceive it; it endures over decades; and it proactively helps shape people’s
lives.
Rationality: The Interpretation of Self-Reports

If responses to a questionnaire are merely samples of behavior, then the manifest
content of the item need bear no relation to the trait construct. That was the
rationale for the development of the Minnesota Multiphasic Personality
Inventory (MMPI; Hathaway & McKinley, 1943), whose scales were purely
empirical. It was a rationale needed at a time when psychologists had a profound
distrust of what people said, particularly about themselves. Psychoanalytic
notions of defense had taught that blind spots in self-perceptions were the rule
rather than the exception, and unconscious distortions were suspected not only
with regard to Oedipal conflicts and latent homosexuality, but also about
mundane topics such as neatness (an anal obsession?) and cheerfulness (reaction
formation against feelings of maternal rejection?). We might hope to find items
that distinguished groups high and low on a given trait, but the content would
presumably be so disguised by defenses that attempting to understand the trait by
reading the items would be as futile as trying to read a coded message without
knowing the cipher.
Our approach to item writing was very different from that adopted by the
authors of the MMPI. Our view was shaped primarily by our experience factor
analyzing responses to a wide variety of questionnaires that had been
administered to members of the NAS, such as the Cornell Medical Index (CMI)
and the Strong Vocational Interest Blank (SVIB). Factor analysis shows which
items covary, and the factor analyst must then make sense of the factors that are
identified. It soon became abundantly clear to us that what the items in almost
any factor chiefly shared was manifest content. The CMI, for example, had an
Irritability factor with items concerning anger and frustration and a Depression
factor with items concerning sadness and loneliness (Costa & McCrae, 1977b).
The SVIB had a factor contrasting preference for sociable occupations such as
YMCA Secretary and Social Worker with task-oriented occupations such as
Architect and Farmer—a factor that was strongly correlated with self-report
measures of Extraversion (Costa, Fozard, & McCrae, 1977). It was as if the
respondents read and understood the items, and responded to them in a
consistent and meaningful manner. That is what we mean by rationality, and it
led us to write straightforward items for the NEO Inventories that embodied the
constructs as we understood them. A few years later this decision was vindicated
by analyses of the items of the MMPI itself (Wrobel & Lachar, 1982).
Of course, the fact that scale responses made rational sense did not, in itself,
mean that they accurately described the respondent’s traits. That was a question
of construct validity, and hundreds of studies of NEO Inventory scales, using as
criteria other measures, different observers, and alternative methods such as
sentence completions (McCrae & Costa, 1980, 1988a), behavioral observations
(Sneed, McCrae, & Funder, 1998), and clinical judgments (Mullins-Sweatt,
Jamerson, Samuel, Olson, & Widiger, 2006), have provided supportive
evidence. The NEO Inventories Bibliography
(http://www4.parinc.com/WebUploads/samplerpts/NEO%20Biblio%202011_1.pdf
documents a vast nomological network that attests not only to the construct
validity of the scales of the NEO Inventories, but more broadly to the rationality
of respondents and the wisdom of Allport’s advice: If you want to know what
people are like, ask them.
Our commitment to rationality did not, of course, mean that we thought
people were perfectly logical or absolutely consistent. For centuries philosophers
have pointed to common fallacies in reasoning, and modern social psychology
has made error in judgments one of its main concerns (Ross, 1977). We meant
that respondents were reasonably rational, and that their responses were
interpretable if we psychologists were equally reasonable. Do people tend to
exaggerate their strengths and minimize their weaknesses? Of course. But we
make allowance for that when we used normed tests, in which general evaluative
biases are corrected. Do people sometimes misunderstand an item? Surely they
do; that is why we ask a series of questions to assess any single trait. Do people
—or some people, at least—have a bias to agree (or disagree) with almost any
statement? Acquiescent tendencies are well documented (McCrae, Herbst, &
Costa, 2001), so we constructed an instrument with balanced keying, in which
that bias is neutralized. Funder (1993) has made the point that error and accuracy
are not logical opposites; we can be reasonably accurate without being error-
free. Conversely, personality assessors (including clinicians) must always bear in
mind that test scores never reveal the absolute truth about the individual; they
are a useful approximation.
The Structure of Personality

The Five Factor Model (FFM) is a model of personality trait structure, and most
surveys of the history of this topic address nothing more than the question of
“How many factors are needed to describe personality traits?,” comparing older
sixteen-factor models (Cattell, Eber, & Tatsuoka, 1970) or three-factor models
(Eysenck & Eysenck, 1975) with the FFM, and perhaps mentioning rival
contemporary models with more (Ashton et al., 2004) or fewer (De Raad &
Peabody, 2005) factors. In 1957 many more possibilities were still in the
running, and Loevinger coined the term structural validity to refer to the
correspondence between the structure of the test and the intrinsic structure of the
attributes it measured. Allport’s (1961) notion of personal dispositions and
Kelly’s (1955) theory of personal constructs essentially claimed that there were
as many structures of personality as there were persons. Guttman’s (1954) view
of measurement in terms of scalability implied a structure in which trait
manifestations were strictly cumulative. Although such a scalable structure has
some desirable psychometric properties, there is no evidence that it is useful for
assessing human personality.
In this section we review more current controversies. The bulk of the evidence
suggests that personality attributes are best viewed as continuous dimensions,
not discrete types or categories; that traits are hierarchically ordered, with
multiple levels of specificity; and that the structure is neither simple nor neatly
circumplex. Personality assessment instruments ought to be compatible with this
understanding of structure.
Traits and Types

As an alternative to the usual rating format in which each individual’s response
to an item was compared to the responses of others, Stephenson (1953) proposed
the Q-sort method, in which the respondent compares items to other items to
identify which are most and least descriptive. In personality research, this
method was most fruitfully employed by Block (1961), who painstakingly
assembled a set of items intended to cover the entire range of personality
features. Each individual had a score on each of the 100 items of Block’s
California Q-Set, and it was, of course, possible to factor these items in a sample
of persons, just as it was possible to factor responses to items on the NEO
Inventories. McCrae, Costa, and Busch (1986) reported such an analysis; it
provided early evidence for the comprehensiveness of the FFM by showing that
Block’s enumeration of personality features could be mapped by the FFM
factors. The identification of FFM-like factors in the California Q-Set was
subsequently replicated by Lanning (1994), although he also reported additional,
smaller factors.
But the Q-sort methodology also lent itself to what appeared to be a very
different kind of analysis. Two respondents with similar patterns of Q-sort
responses would seem to be people with similar personalities, and person
similarity could be quantified by correlating the two sets of 100 items. Such
correlations could be computed for each pair of individuals in a sample, and the
Q-correlation matrix could then be subjected to factor analysis. People who had
high loadings on a given Q-factor could be seen as forming a type, a group of
individuals with a similar personality profile. To some researchers, this appeared
to be a particularly fruitful way to conceptualize personality structure. If we
could identify the type to which a person belonged, we knew immediately the
entire configuration of their traits, just as knowing that an animal is a bird at
once alerts us to the fact that the animal has feathers, lay eggs, is warm-blooded,
and probably flies. Furthermore, this person-centered approach to personality (in
contrast to the variable-centered approach adopted by the FFM) seemed to be
more dynamic. What mattered in a Q-sort was not whether someone was
agreeable or conscientious in an absolute sense, but whether someone was
relatively more agreeable than conscientious, or vice versa. When faced with a
situation that could elicit either agreeable or conscientious behavior, the Q-sort
results seemed to predict which would prevail.
In the 1990s such considerations led to renewed interest in the use of Q-sorts
and the investigation of types. Early results (Asendorpf & van Aken, 1999;
Robins, John, Caspi, Moffitt, & Stouthamer-Loeber, 1996) suggested that when
types were based on the full range of personality attributes found in the FFM,
three types reliably emerged, usually called Resilient, Undercontrolled, and
Overcontrolled—types that Block himself had proposed years earlier. These
findings provided a structural alternative to the standard FFM: The two models
covered identical ranges of personality traits, but organized them in very
different ways.
Enthusiasm for this approach quickly waned, however, as a number of
limitations of types became apparent:
1. Types were not cleanly separated. Instead of discreet clusters of individuals, persons
were smoothly distributed across the cluster space, with most people falling between
cluster centers—for example, equally resembling a Resilient and an Undercontrolled type.
2. Types did not offer any predictive advantage. In fact, head-to-head comparisons
showed that considerably more information was contained in the five factor scores on
which types were based than in the three types themselves (Asendorpf, 2003). If types
were the natural cleavages of distinct kinds of people, the opposite effect would be
expected.
3. Subsequent attempts at replication, encouraged by a Special Issue of the European
Journal of Personality, provided very mixed support (Costa, Herbst, McCrae, Samuels, &
Ozer, 2002; McCrae, Terracciano, Costa, & Ozer, 2006). Types resembling Resilient,
Undercontrolled, and Overcontrolled were found in some samples and with some
statistical methods, but they were far from robust.
4. The argument that person-centered approaches are inherently more dynamic than
variable-centered approaches is specious, because they are interchangeable. It is possible,
for example, to easily create a person-centered profile from the familiar variable-centered
NEO Personality Inventory profile by simply ranking the individual’s T-scores. For
example, according to an expert rater, Jean-Jacques Rousseau had T-scores of 61 on
Tender-Mindedness, 45 on Altruism, and 32 on Trust (McCrae, 1996). We might
conclude from this that he was sympathetic with his fellow human beings in principle, but
not particularly generous to the people around him, possibly because he had a deep
distrust of them. This “dynamic” interpretation is made by comparing facet scores within
a profile, but it is based on the usual normed scores of a variable-centered approach. A Q-
sort that showed a higher ranking for Tender-Mindedness than for Trust would yield the
same information.
5. In fact, person-centered and variable-centered approaches turn out to be statistically
equivalent, as shown by McCrae, Terracciano, Costa, and Ozer (2006). When five person
factors were extracted from standardized responses to the California Q-Set, they could be
perfectly aligned with the usual FFM. The “types” discovered in this way were simply
neurotic versus stable, extraverted versus introverted, open versus closed, agreeable
versus antagonistic, and conscientious versus undirected collections of people.
Q-sort types are syndromes or configurations of many different traits, but the
term type is also used to refer to discrete (e.g., bimodal) distributions in a single
trait dimension. Most psychologists assume, with good reason, that traits are
more or less normally distributed. We need only examine the distribution of
scores on almost any personality scale to see the familiar bell shape.
Occasionally theorists have argued that this apparent continuity masks an
underlying discontinuity. The Myers-Briggs Type Indicator (MBTI; Myers &
McCaulley, 1985), for example, is based on the premise that people can be
divided into two discreet classes on each of four indices—for example, into
introverts and extraverts. These discrete classes are then configured into 16
qualitatively distinct types. However, empirical studies have failed to find
support for either the dichotomous classes or the MBTI types (McCrae & Costa,
1989b; Reynierse, 2012).
The distinction between dimensional and categorical structures has figured
most prominently in conceptualizations of normal versus abnormal personality
traits. Normal personality traits are usually viewed as having a continuous
distribution, whereas abnormal traits are often seen as qualitatively different
from normal functioning, just as having polio is qualitatively different from not
having polio. One of the most important applications of the FFM has been in
understanding personality disorders (Widiger & Costa, 2013), and one of the
clearest results of research has been the demonstration that personality disorders
are not qualitatively distinct categories (Livesley, 2001). Instead, individuals
who are diagnosed with personality disorders typically show extremely high or
low levels of a trait or a set of traits, together with problems in living that are
associated with these traits (McCrae, Löckenhoff, & Costa, 2005). The alleged
structural difference between normal and abnormal is largely illusory, at least
with regard to personality structure.
Levels of the Trait Hierarchy
The FFM is a hierarchical model, in which related groups of narrow and
specific traits combine into broader and more general traits. Almost all research
in support of the FFM structure has consisted of factor analyses of narrow traits;
the truly remarkable finding in this body of literature is that five very similar
factors emerge regardless of the narrow traits selected, as long as they constitute
a wide sample of personality attributes. This is true of English language
adjectives (Goldberg, 1990), California Q-Set items (Lanning, 1994),
psychological needs (Costa & McCrae, 1988), measures of temperament
(Angleitner & Ostendorf, 1994), facets of the Turkish translation of the Revised
NEO Personality Inventory (NEO-PI-R; Gülgöz, 2002), or standard personality
scales assessing normal and abnormal traits (Markon, Krueger, & Watson,
2005). This consistent finding has led to the current consensus on the value of
the FFM, to its description as “the most scientifically rigorous taxonomy that
behavioral science has” (H. Reis, personal communication, April 24, 2006), and,
indeed, to this Handbook. Although some researchers still argue for more
(Ashton et al., 2004) or fewer (De Raad & Peabody, 2005) factors, almost
everyone agrees that the FFM is a reasonably close approximation to the true
structure of personality, an “adequate taxonomy of personality attributes”
(Norman, 1963).
Curiously, this is the only feature of the hierarchy of traits on which there is
agreement. There is an ongoing controversy about whether the five factors
themselves are orthogonal, and thus the topmost level of the hierarchy, or
correlated, thus constituting yet higher levels. There is no doubt that most
measures of the five factors (e.g., adjective scales, domains of the NEO
Inventories) show intercorrelations, which usually (although not always; see
Costa & McCrae, 1992a) contrast Agreeableness and Conscientiousness with
Neuroticism, and link Extraversion and Openness. Digman (1997), who first
pointed this out, called the former second-order factor Socialization and the
latter Personal Growth. DeYoung, Peterson, and Higgins (2002) preferred the
labels Stability and Plasticity. The question is whether the correlations between
scales that lead to these superfactors are substantive or merely artifacts—in
particular, evaluative biases of negative and positive valence, respectively. If
evaluative biases generate the intercorrelations, they ought not to be seen across
raters, and some studies have supported that prediction (Biesanz & West, 2004).
However, the issues are extraordinarily complicated (see also the chapter by
Wright), and even elaborate designs involving self-reports and observer ratings
of monozygotic and dizygotic twins yield somewhat ambiguous results (McCrae
et al., 2008). It appears that correlations among measures of the five factors are
certainly inflated by method biases, if not entirely created by them, so any real
higher-order factors that may exist are extremely subtle, and presumably of
limited explanatory value. For the NEO Inventories, orthogonal factor scores
consistently show higher convergent and discriminant cross-method validity than
do correlated scores (McCrae & Costa, 1989a; McCrae et al., 2008). For
practical purposes, it may be reasonable to assume that the five factors are truly
independent.
The final step in this upward progression is the proposal that Stability and
Plasticity are themselves correlated, leading to a single, general factor of
personality (GFP; van der Linden, te Nijenhuis, & Bakker, 2010), combining the
more desirable pole of each factor. This GFP is, of course, related to a wide
variety of criteria, because its constituents are related to many outcomes.
Whether anything is gained by adding them together is questionable. Of course,
if the FFM is truly orthogonal, then the general factor must be pure artifact. In a
sophisticated study of 1,615 German twins for whom self-reports and two peer
ratings were available, Riemann and Kandler (2010) concluded that “the
conjecture of a meaningful GFP is not supported in the analysis of variance
shared by self- and peer reports and controlled for method specific effects” (p.
271; see also Ferguson, Chamorro-Premuzic, Pickering, & Weiss, 2013).
More fruitful controversies are found at lower levels of the hierarchy. The
simplest strategy would be to subdivide each of the five broad domains into two
subdomains, called aspects by DeYoung, Quilty, and Peterson (2007). If the six
facets of each NEO-PI-R domain in the normative sample are factored, a single
factor has an eigenvalue greater than 1.0, suggesting no natural subdivision. But
a number of subdivisions have been suggested, either on rational grounds or
based on factor analyses of a larger pool of facets than is provided by the NEO
Inventories. Costa and McCrae (1980), for example, noted that the facets of
NEO Extraversion can be conceptually classified as interpersonal (E1: Warmth,
E2: Gregariousness, and E3: Assertiveness) and temperamental (E4: Activity,
E5: Excitement Seeking, and E6: Positive Emotions), and McCrae and Costa
(1987) pointed to inhibitive and proactive aspects of Conscientiousness. Glisky,
Tataryn, Tobias, Kihlstrom, and McConkey (1991) divided Openness into
Imaginative Involvement and Liberalism. Ashton and Lee (2005) have proposed
a six-factor model that is (from our point of view) chiefly distinguished from the
FFM by subdividing a broad Agreeableness factor into narrow Agreeableness
and Honesty/Humility. The six-factor model of Jackson and Tremblay (2002)
divides Conscientiousness into Methodicalness and Industriousness. Caprara,
Barbaranelli, Borgogni, and Perugini (1993) created two scales for each of the
five factors, based on an examination of Italian trait adjectives. For example,
they offered Emotional Control and Impulse Control as aspects of Neuroticism
(or rather its polar opposite, Emotional Stability).
One empirical study (DeYoung et al., 2007) dissected all five domains using a
joint factor analysis of NEO-PI-R facets and the facets of the Abridged Big Five
Circumplex (AB5C; Goldberg, 1999). In the latter instrument, each domain was
represented by a pure indicator and eight blends (e.g., Openness has pure
Openness, Agreeable Openness, and Disagreeable Openness facets, along with
six other blends). The aspects of DeYoung and colleagues (2007) bear only a
partial resemblance to previous suggestions. For example, their subdivision of
Agreeableness into Compassion and Politeness corresponds to the Agreeableness
and Honesty/Humility factors of Ashton and Lee (2005), and their Volatility and
Withdrawal aspects of Neuroticism appear to resemble the Impulse Control and
Emotional Control factors of Caprara and colleagues (1993). However, the
Enthusiasm aspect of DeYoung and colleagues (2007) combines E6: Positive
Emotions not with other temperamental facets of Extraversion, but with E1:
Warmth and E2: Gregariousness, and they classify both the inhibitive C3:
Dutifulness and the proactive C4: Achievement Striving as parts of an
Industriousness aspect. There does not yet seem to be a compelling reason to
adopt any existing classification at the aspect level (see also Judge, Rodell,
Klinger, Simon, & Crawford, 2013).
Much more research has been conducted at the next lower level, facets.
Although many instruments measure only the five factors (De Raad & Perugini,
2002), there is clear evidence that assessment of more specific traits is
worthwhile to increase predictive and discriminant validity (Paunonen &
Ashton, 2001; Reynolds & Clark, 2001). In developing the NEO Inventories, we
reviewed the personality literature in an attempt to identify the most important
traits to represent each factor. Gorsuch (1974) had recommended that six
variables be included to clearly define each factor in an analysis, so we identified
six facets per domain (Costa & McCrae, 1980; Costa, McCrae, & Dye, 1991).
We wrote items based on our rational analyses of the facet constructs, and used
item factor analyses with targeted rotations to select items that showed
convergent and discriminant validity with respect first to other factors and then
to other facets within their domain—a fundamental application of the dialectic
between theory and data that Loevinger had espoused. Subsequent analyses in
independent samples confirmed that the facets of the NEO Inventories do in fact
divide up the items in meaningful ways (McCrae & Costa, 2008a). Other studies
have shown convergent and discriminant validity for NEO facets across
observers and with respect to other measures (Costa & McCrae, 1992b).
The facets of the NEO Inventories thus constitute one possible and useful way
to specify narrow traits within the five domains, but they are clearly not the only
way. For example, Watson and Clark (1997) combined Warmth and
Gregariousness into an Affiliation facet; Cheung and colleagues (Cheung et al.,
2008) proposed a Diversity facet for Openness. At present there does not seem to
be any natural way to carve up the domains in the way that the five factors
consistently carve up the personality sphere.
What we can do is compare different proposals to determine the extent to
which they agree, and to what new facet-level constructs they point. Table 2.1
offers a mapping of facet-level constructs across instruments and studies. The
first column lists the facets of the NEO Inventories, together with an Other row
for each domain. The second column shows corresponding scales in the Sixteen
Personality Factor Questionnaire (16PF; Conn & Rieke, 1994), and the third
column shows scales from the Eysenck Personality Profiler (EPP; Eysenck,
Barrett, Wilson, & Jackson, 1992). For these two inventories, scales are aligned
with the NEO facet with which they are most strongly correlated (rs = .38 to .81,
ps < .001). The remaining columns in Table 2.1 report efforts to facet three of
the domains. Watson and Clark (1997) suggested facets of Extraversion based on
a review of the literature. Cheung and colleagues (2008) identified potential
facets of Openness in Chinese culture, and Woo and colleagues (2014) factored
scales from several inventories to suggest facets of Openness. Roberts, Bogg,
Walton, Chernyshenko, and Stark (2004) based their facets of Conscientiousness
on an analysis of trait adjectives, whereas Roberts, Chernyshenko, Stark, and
Goldberg (2005) factored Conscientiousness-related scales from personality
inventories. The alignment of these traits with NEO facets in Table 2.1 is
conceptual, guided where possible by empirical findings in these studies—for
example, Cheung and colleagues reported that the Social Sensitivity scale,
intended as a facet of Openness, in fact loaded chiefly on an Extraversion factor.
There appears to be some degree of consensus across studies. For example,
O6: Values is similar to Tolerance and to low Dogmatism, Conventionality, and
Traditionalism. C2: Order has its counterparts in Perfectionism, Obsessive,
Orderliness, and Order. Only six of the 30 NEO facets are not seconded in any
other system: N5: Impulsiveness, N6: Vulnerability, O3: Feelings, A3: Altruism,
A6: Tender-Mindedness, and C5: Self-Discipline. Inspection of the Other rows
in Table 2.1 suggests that it might be desirable to add a Social Sensitivity facet
to Extraversion, Diversity and Formalness facets to Openness, Interpersonal
tolerance to Agreeableness, and Punctuality, Responsibility, and Virtue facets to
Conscientiousness.
It is also noteworthy that the alternatives to the NEO system of facets do not
agree among themselves. Punctuality emerges as a facet of Conscientiousness
when adjectives are considered, but not personality scales. Diversity appears as a
facet in the Chinese study, but not in an American study of Openness scales.
Thus, the structure of personality at the level of facets appears to be ill-defined;
some system of facets is needed and useful, but which system is chosen is to
some extent arbitrary.
De Young and colleagues (2007) noted in passing that there may be a level of
trait attributes below facets, corresponding operationally to individual items. For
example, a hostility scale may include one item that taps anger and another that
taps bitterness. Anger and bitterness are both relevant to hostility, but they are
not identical. McCrae (2015) has proposed the term nuance to describe
characteristics at this lowest level, and argued that facet scales ought to sample a
wide range of nuances, instead of seeking to maximize internal consistency
through the selection of items with redundant content. If items express different
nuances of a facet they must have item-specific variance, and Mõttus, McCrae,
Allik, and Realo (2014) have demonstrated in a large Estonian sample that this
item-specific variance is consensually valid. In that study, item residuals were
created by partialling out the relevant facet score; residuals in self-reports were
then correlated with residuals in observer ratings. All 240 items in the Estonian
NEO-PI-3 showed small but statistically significant cross-observer agreement on
residuals.
Table 2.1. Correspondence of Facet-Level Scales

NEO-PI-R Facet 16PF EPP Scaleb E Facetc O Facetd O C Facetf
Scalea Facete
N1: Anxiety Apprehension Anxious,
Hypochondriacal
N2: Angry Hostility Tension
N3: Depression –Emotional Unhappy,
stability Dependent
N4: Self- Inferior, Guilty
Consciousness
N5: Impulsiveness
N6: Vulnerability
Other N
E1: Warmth Affiliation
E2: Gregariousness –Self-reliance, Sociable Affiliation
Warmth
E3: Assertiveness Social Assertive Ascendance
boldness,
Dominance
E4: Activity Active Energy
E5: Excitement Liveliness Sensation Venturesomeness
Seeking seeking
E6: Positive Positive
Emotions affectivity
Other E Social
sensitivity
O1: Fantasy Abstractedness
O2: Aesthetics Sensitivity –Tough-minded Aesthetics Aesthetics
O3: Feelings
O4: Actions Openness to Novelty
change
O5: Ideas –Practical Divergent Intellectual
thinking efficiency,
Curiosity,
Depth
O6: Values –Dogmatic Tolerance –
Conventionality
Other O Diversity Formalness
A1: Trust –Vigilance, –
Privateness
A2: –Manipulative
Straightforwardness
A3: Altruism
A4: Compliance –Aggressive
A5: Modesty
A6: Tender-
Mindedness
Other A Interpersonal
tolerance
C1: Competence Decisiveness
C2: Order Perfectionism Obsessive Orderliness
C3: Dutifulness –Irresponsible Reliability
C4: Achievement Ambitious Ambition Industriousness
Striving
C5: Self-Discipline
C6: Deliberation Rule- –Impulsive, – Impulse control
consciousness Risk-taking,
Expressive
Other C Punctuality
Note : Scales preceded by a minus sign are inversely related. NEO-PI-R = Revised NEO Personality
Inventory. 16PF = 16PF, Fifth Edition. EPP = Eysenck Personality Profiler.
a Conn and Rieke (1994).
b Costa and McCrae (1995b).
c Watson and Clark (1997).
d Cheung et al. (2008).
e Woo et al. (2014).
f Roberts, Bogg, Walton, Chernyshenko, and Stark (2004).
g Roberts, Chernyshenko, Stark, and Goldberg (2005).
Classical models of scale reliability assume that items are determined solely
by the trait the scale measures and random error. But if items measure different
nuances of a trait, there is also item-specific variance. One implication is that
retest reliability should be higher than internal consistency, because item-
specific variance detracts from internal consistency but contributes to retest
reliability. And in fact, retest reliability is usually higher than internal
consistency (McCrae, 2015). Because item-specific variance is shared by
observers (Mõttus et al., 2014), retest reliability ought to be a better predictor of
cross-observer agreement on facet scores than is internal consistency, and that is
indeed the case (McCrae, Kurtz, Yamagata, & Terracciano, 2011). But McCrae
and colleagues also showed that retest reliability is a better predictor of facet
heritability, suggesting by the same logic that nuances of facets are also
themselves heritable—a hypothesis that has not yet been tested, but had been
proposed (Harkness & McNulty, 2002). Much remains to be learned about
nuances of personality.
The Shape of the Hierarchy

The FFM is often described as a taxonomy of traits, but that is true only in a
very loose sense. In biological taxonomy, for example, each species is included
exclusively in a single genus, and each genus in a single family. Werewolves
aside, there are no animals that are part canine and part primate. Personality
traits, however, are often related to more than a single factor. E3: Assertiveness,
for example, is related to low Neuroticism, low Agreeableness, and high
Conscientiousness as well as Extraversion. This is a well-established fact,
supported by data from many cultures (McCrae & Costa, 1997). In factor
analytic terms, this means that personality does not show simple structure, and
structural equation models that impose simple structure do not fit the data
(McCrae, Zonderman, Costa, Bond, & Paunonen, 1996).
One alternative to simple structure is circumplex structure, and there is ample
evidence that interpersonal traits show such a circular order (Wiggins &
Broughton, 1985). E3: Assertiveness is a disagreeable form of Extraversion,
whereas E1: Warmth is an agreeable form. Some efforts have been made to
assess traits by systematically crossing each pair of factors (Goldberg, 1999), but
many combinations are poorly represented in trait descriptive adjectives. Judging
from the pattern of loadings in the NEO-PI-R, there do not appear to be any
neurotic facets of Openness, nor any stable facets. In short, personality traits, at
least at the facet level, appear to show a rather messy hybrid of simple and
circumplex structures. This structure can be adequately represented by a factor
matrix, provided we pay attention to secondary and tertiary loadings as well as
primary loadings.
Sixty years after Loevinger’s essay we have a much clearer idea of the
structure of personality traits. There are many trait-like attributes—stable,
observable, and heritable—that are all more or less normally distributed. They
vary in breadth, from subtle nuances up to the five broad factors, and possibly
beyond. They are clustered hierarchically through their associations with each
other, and all (or at least most of them) are influenced by one or more of the
FFM factors. Correspondingly, the NEO Inventories have items expressing
nuances, each scored on a five-point Likert scale that recognizes the normal
distribution of attributes; facets that load on one or more of the five factors; and
factor scores that represent the essential orthogonality of the five. Scores are
interpreted in terms of five levels (from Very Low to Very High) to emphasize
the dimensional nature of traits. In a nod to circumplexity, there is even a
provision for interpreting pairs of factors, which jointly define personality styles
(Costa & Piedmont, 2003). It appears that the NEO Inventories show a fair
degree of structural validity.
The Content of Personality and the Duality Principle

The success of the FFM is due in large measure to its comprehensiveness: It
encompasses most, if not all, of the attributes of the personality sphere.
Establishing comprehensiveness is a difficult task. Teams of mathematicians
around the world spent decades discovering sporadic symmetry groups,2 and
then proving that there are exactly 26 of them, and that no more will ever be
found (Ronan, 2006). Psychologists have a more difficult task, because there is
in principle no way to prove that any proposed model is absolutely
comprehensive: Any day, a hitherto undiscovered trait may pop up. Instead,
psychologists must diligently seek out new traits and compare different lists to
see what, if anything, is missing. Our present knowledge has evolved through
successive approximations in which conceptions, assessment instruments, and
empirical findings have led to their mutual evolution. For example, it is well
known that the FFM can be traced to the lexical hypothesis, which asserts that
all important traits have been codified in natural language terms (John,
Angleitner, & Ostendorf, 1988). Adjective scales were created to represent the
five lexical factors and were then correlated with other trait models, including
the early three-factor NEO model (McCrae & Costa, 1985). Such research
showed that lexical Factor V, sometimes called Intellect, was closely akin to
Openness to Experience. But Openness is a broad and subtle factor, and it soon
became clear that its facets are not adequately represented in English language
trait terms (McCrae, 1990). In that sense, the lexical hypothesis is wrong—some
important traits are not codified in natural language adjectives—but it had
provided a crucial step in the right direction.
The BLSA proved to be an exceptionally valuable resource in this quest. As
part of a broad and long-term biomedical and psychosocial study, several
hundred volunteers (and their spouses and friends) agreed to complete
personality questionnaires. Over the course of several years they provided data
on most of the major personality inventories in use, from the MMPI to the
Revised Interpersonal Adjective Scales (Wiggins, Trapnell, & Phillips, 1988).
Correlations of these instruments with each other and with standard measures of
the FFM, including the NEO Inventories, provided persuasive evidence that the
FFM is reasonably comprehensive. This does not mean that personality traits can
be reduced to five scores. Because traits are hierarchically structured, narrow
traits include both variance that is common to the five factors and variance that
is specific. The FFM is not a set of five factors; it is the complete matrix of five
factors and the many lower-level variables that define them.
It is of course true that not all individual difference variables fall within even
this broad definition of the FFM. There is an assortment of isolated variables that
seems to be essentially independent of all five factors (Saucier & Goldberg,
1998), and proposals have been made for a few additional broad factors,
including spirituality (Piedmont, 1999), interpersonal relatedness (Cheung et al.,
2001), and gender diagnosticity (Lippa, 1991). It is not yet clear whether these
are best thought of as new dimensions of personality traits, or whether they can
be incorporated into the FFM (perhaps a somewhat broader FFM than the NEO
Inventories assess). For example, some forms of spirituality, such as the self-
transcendence scales of the Temperament and Character Inventory (Cloninger,
Przybeck, Svrakic, & Wetzel, 1994), load on an Openness factor (McCrae,
Herbst, & Costa, 2001).
Or perhaps they do not belong in the category of personality at all.
Historically, intelligence has sometimes been included in the personality sphere
—indeed, one of the 16PF scales is a measure of intelligence—and early
interpretations of the Intellect factor identified it with intelligence (Goldberg,
1981). But it is now well established that Openness is only weakly related to
intelligence or academic achievement (Noftle & Robins, 2007), so if intelligence
is to be included in the personality sphere, it is best regarded as a sixth factor
(McCrae, 1994). However, most psychologists exclude abilities from personality
per se.
The Nature of Traits

The distinction between dispositions and abilities raises the deeper question of
what the attributes included in the FFM really are. That issue was far from
resolved when the NEO Inventories were first developed. At that time, a number
of terms were in common use to describe individual differences in personality,
including trait, need, motive, habit, temperament, personality disorder, and folk
concept. In the wake of Mischel’s (1968) critique, trait was perhaps the least
popular of these. Much careful thought has gone into distinguishing among these
kinds of attributes. For example, Pervin (1994) argued that traits were overt
patterns of behavior, whereas motives were deep-rooted tendencies that guided
behavior in flexible ways. Allport (1937) had distinguished adaptive traits from
expressive traits. Some traits were relational and some were temperamental;
some were adaptive and some were maladaptive. Yang and colleagues (2014)
argued that situations,3 behaviors, and explanations are the key ingredients of
traits. These are all meaningful conceptual distinctions, but how they would map
onto the FFM was unclear in the 1980s. The NEO Inventories are often said to
assess “emotional, interpersonal, experiential, attitudinal, and motivational
characteristics” (Costa & McCrae, 2014, p. 231). Do these five terms apply
respectively to the five factors, such that Neuroticism consists of emotional
traits, Extraversion consists of interpersonal traits, and so on? If so, it would
provide a perfect alignment of conceptual and empirical classifications of traits.
But research relating the FFM to alternative measures of personality led to
two less tidy conclusions. First, it established that regardless of the
conceptualizations and labels used, most individual differences assessed in
personality questionnaires fit within the FFM. The California Psychological
Inventory (CPI; Gough, 1987), for example, was intended to assess folk
concepts, not traits—yet CPI scales were substantially related to NEO-PI factors
(McCrae, Costa, & Piedmont, 1993). The Personality Research Form (PRF;
Jackson, 1984) measures Murray’s needs, but a joint factor analysis of the PRF
and NEO-PI factors showed that all those needs loaded on FFM factors (Costa &
McCrae, 1988). The MBTI was designed to assess Jungian attitudes and
functions, but its four scales corresponded directly and substantially to four of
the five FFM factors (McCrae & Costa, 1989b). At least in one important sense,
the conceptual distinctions among different kinds of individual difference
variables are all moot: They all refer to aspects of the same FFM, and can be
considered generically as traits.
The second conclusion is that there is no neat one-to-one correspondence
between kinds of traits and their empirical clustering into factors. Many
emotional traits (such as characteristic levels of anger, sadness, and guilt) fall
within the Neuroticism domain, but positive emotions form a part of
Extraversion, and love and sympathy are emotions tied to Agreeableness.
Traditionally, Extraversion and Agreeableness have been considered
interpersonal factors, but in fact all five factors have interpersonal implications
(Costa & McCrae, 2010), perhaps most notably Openness, which affects
friendships, marriage, social attitudes, and political behavior (McCrae, 1996).
Conscientious people are highly motivated, but each of the five factors is
associated with a distinct set of Murray’s needs (Costa & McCrae, 1988): All
traits have implications for motivation. The jaunty step of extraverts is an
expressive trait, but so is the erect posture of the conscientious (Back, Schmukle,
& Egloff, 2009).
This insight was quantified by Pytlik Zillig, Hemenover, and Dienstbier
(2002), who asked raters to judge the degree to which NEO-PI-R items had
affective (including motivational), behavioral, and cognitive characteristics.
Conscientiousness items were chiefly behavioral (70.2%) and Neuroticism items
were chiefly affective (67.9%), but all five domains had some content from all
three categories.
What does this say about the nature of FFM traits? One interpretation is that
they are very deep and abstract entities that can become manifest in a variety of
ways—expressive, emotional, motivational, and so on. McCrae and Costa (2003,
p. 25) proposed a correspondingly abstract definition of traits as “dimensions of
individual differences in tendencies to show consistent patterns of thoughts,
feelings, and actions.” This definition could apply to goals, habits, attitudes, and
so on. Conversely, questionnaire items about goals, habits, or attitudes can be
employed as indicators of an underlying trait, as Pytlik Zillig and colleagues
(2002) showed.
One important conceptual difference between traits and trait indicators such as
goals, habits, and attitudes is that the latter are usually contextualized. The goals
of a twenty-first century American woman might include buying a hybrid car,
getting a law degree, or amassing a thousand friends on Facebook—none of
which would have been goals for a sixteenth century Hungarian peasant. The
daily routines of Japanese are not the daily routines of Bedouins. But traits are
abstract enough to be similar now and yesteryear, here and anywhere.
The Duality Principle

These considerations—along with a number of empirical findings—led to the
formulation of a fundamental distinction among psychological variables.
McCrae and Costa (1996) proposed that traits and their indicators constituted
two entirely different categories (or levels, as McAdams, 1996, called them):
Traits are basic tendencies, whereas goals, habits, attitudes, relationships, and so
on, are characteristic adaptations. A similar distinction in cognitive psychology
is seen between aptitude and achievement, and in fact McCrae and Costa argued
that abilities, like personality traits, are basic tendencies, whereas skills are
characteristic adaptations. The distinction thus applies broadly to contrast innate
psychological potentials with their acquired and contextualized realizations.
The crucial distinction in understanding the content of personality, then, is not
between motives and habits, or between expressive and adaptive traits, but
between abstract dispositions (clustered into five broad domains) and the myriad
of acquired adaptations that express them. Unfortunately, this clear conceptual
distinction cannot be cleanly operationalized. There is no direct way to assess
psychological potentials; we can know them only by their concrete
manifestations.4 And these manifestations are characteristic adaptations that are
also inevitably shaped by the environmental context, and are thus necessarily
imperfect indicators of the trait.
Quantum physics teaches that objects have a dual nature: They are in a sense
simultaneously waves and particles. A similar duality principle applies to trait
indicators: They are at once characteristic adaptations and proxy measures of
basic tendencies. Consider a vocabulary test. It can be considered a sample of
knowledge, acquired in a particular family and educational context; but it can
also be considered a measure of general intelligence, our abstract capacity for
learning. No one inherits an understanding of the words frenetic or ciekawość,
but vocabulary test scores are highly heritable in typical samples. Vocabulary
tests have a dual nature.
The same is true for many of the individual difference variables relevant to
personality, and the manifest content of items the NEO Inventories use to assess
underlying traits consists of characteristic adaptations. “I prefer jobs that let me
work alone” nominally concerns vocational interests, but it can be used to assess
the trait of Gregariousness. Even the item “I am dominant, forceful, and
assertive” is not a direct assessment of the trait of Assertiveness; it is an
assessment of the respondent’s self-concept with respect to assertiveness, which
is an acquired adaptation, and one that can be modified by the environment—at
least briefly (Swann & Hill, 1982).
It is essential to understand the dual nature of trait indicators in order to make
sense of some assertions about trait measures. How can situations, behaviors,
and explanations be “key ingredients” (Yang et al., 2014) of traits when traits are
highly heritable, and situations, behaviors, and explanations clearly are not?
Situations are not even part of the person. We would prefer to say that situations,
behaviors, and explanations are some of the ingredients of indicators of heritable
traits.
Not all characteristic adaptations are relevant to traits (regional dialects, for
example, have little to do with personality), and some are better suited than
others to use as trait indicators. Items that implicitly aggregate across occasions
(“I always enjoy meeting new people”) are usually better than highly
contextualized items (“I enjoy meeting new people in train stations”), because
the influences of the specific circumstances are neutralized in the former. One of
the reasons for creating personality scales with many different, nonredundant,
items is that by sampling diverse manifestations of a trait, the total variance in
the scale becomes more reflective of the underlying trait and less influenced by
the specific contextually influenced manifestations.
A TAXONOMY OF INDIVIDUAL DIFFERENCES

It may be useful to put these ideas of different kinds of individual differences
into a broader perspective. After all, people also differ in age, wealth, religion,
and physical attractiveness, and all these differences are enormously important,
but they are not what we mean by personality—in fact, they are not even
intrinsically psychological variables. Figure 2.1 presents an outline of different
classes of individual differences. In this scheme, age and physical attractiveness
would be classified as biomedical variables and wealth and religion as cultural
variables, with both under the heading extrapsychological. Among psychological
variables, we draw the distinction between adaptations and tendencies, and
personality dispositions are nested within tendencies. This figure does not, of
course, indicate anything about causal relations among these variables. The
personality literature is replete with their correlations, but very few of these
variables can be experimentally manipulated, so causal interpretations require a
judicious consideration of a broad array of findings. We turn next to one attempt
to make sense of the big picture.
Theoretical Implications of FFM Research

Characterizing the structure and content of personality attributes can be done
simply by analyzing self-report responses to personality questionnaires
administered on a single occasion. But to understand the workings of personality
—its origins and development, its implications for human life—it is necessary to
go far beyond that simple correlational design. Longitudinal, cross-cultural,
epidemiological, and behavior genetic studies are all necessary. No single
investigator or team of investigators could amass sufficient data in one career to
learn all that is needed. Fortunately, the widespread adoption of the FFM has
meant that in the past 30 years hundreds of researchers around the world have
conducted thousands of studies that cumulate within its framework. As a result,
we know a great deal about how personality operates, and with that knowledge
we can formulate a new generation of personality theories.5
Figure 2.1. A sketch of a taxonomy of individual difference variables. Adapted with
permission from “The Place of the FFM in Personality Psychology,” by R. R. McCrae,
2010, Psychological Inquiry, 21, p. 59. Copyright 2010 by Taylor & Francis.
Established Facts
In the discussion of proactivity, we noted extensive research showing that
knowledgeable observers agree substantially on the description of FFM traits in
individuals, a phenomenon most easily explained by assuming that all these
sources are reasonably accurate. Different raters agree not only on the broad
factors, but also on facets (McCrae, Costa, et al., 2005) and the specific variance
within facets (McCrae & Costa, 1992), and on items and the specific variance
within items (Mõttus et al., 2014). A rich variety of personality attributes
appears to be an intrinsic aspect of each person.
The rank-ordering of these traits is remarkably stable, at least in adulthood.
Retest correlations of .80 are not uncommon over intervals of 10 (Costa,
McCrae, & Arenberg, 1980) or 20 years (Terracciano, Costa, & McCrae, 2006).
Because the scales used to assess personality at each time are subject to retest
unreliability, these observed values must be conservative estimates of the true
stability. These high values seem to characterize all adults, regardless of the
intervening life experience. For example, Costa, Metter, and McCrae (1994)
examined the medical records of 273 BLSA participants who had self-reports or
observer ratings of personality twice, across intervals of from 3 to 7 years.
Participants were divided into two groups, those who had maintained or
improved their health status over the interval (N = 175) and those who had
shown minor or major declines in physical health (N = 98). The median retest
correlation across all methods and factors was .79 for the first group and .80 for
the second group. Physical health changes do not seem to affect personality traits
(unless, of course, the changes are in the brain, as in Alzheimer’s disease)
(Siegler et al., 1991).
There are notable changes in the mean levels of traits between adolescence
and adulthood, and changes continue at a reduced rate throughout much of
adulthood. In general, people decrease in Neuroticism, Extraversion, and
Openness, and increase in Agreeableness and Conscientiousness (McCrae,
Martin, & Costa, 2005). Most of the data on age differences comes from cross-
sectional studies of American samples, and from these studies alone it would be
impossible to divine the origin of age differences. They might be cohort effects,
reflecting the successive impacts of different historical eras in which respondents
grew up; or they might be cultural effects, a distinctively American way of
growing older. Fortunately, there have also been longitudinal studies, less
affected by cohort effects, and cross-cultural studies that permit comparisons of
personality development in different contexts. Such studies show that cross-
sectional age differences are very similar to longitudinal age changes
(Terracciano, McCrae, Brant, & Costa, 2005), implying that historical eras leave
little trace on personality. Cross-cultural comparisons show that age differences
are similar in almost all cultures (McCrae et al., 1999), implying that personality
development is not shaped by culture.
Or at least they are not shaped by features of culture that vary across nations
(see also the chapter by Allik and Realo). Roberts, Wood, and Smith (2005) have
argued that personality maturation (specifically, the decline in Neuroticism and
the increases in Agreeableness and Conscientiousness that are typically seen
between adolescence and full adulthood) is motivated by social expectations.
According to their Social Investment Theory, individuals become more mature
because they are pressured to do so by a society that requires stability,
cooperation, and diligence in the adults who are responsible for raising a family
and contributing useful work to the community. This suggests the hypothesis
that maturation should be accelerated in those cultures (usually less developed)
that require young men and women to join the adult world of work and family at
an earlier age. Two large-scale tests of that hypothesis have been offered; they
provided little (Bleidorn et al., 2013) or no (De Bolle et al., 2012) support. The
rate of maturation does not seem to be driven by social investment.
Although cultures vary dramatically in their prescription of sex roles, sex
differences in personality traits appear to be very similar everywhere, and this is
true for self-reports of traits (Costa, Terracciano, & McCrae, 2001) as well as
observer ratings (McCrae, Terracciano, & 78 Members of the Personality
Profiles of Cultures Project, 2005). Even stereotypes of sex differences in
personality are shared around the world (Löckenhoff et al., 2014). Women
everywhere see themselves and are seen by others as being higher in Anxiety,
Vulnerability, and Openness to Aesthetics; men are consistently higher in
Assertiveness, Excitement Seeking, and Openness to Ideas.
Cross-cultural comparisons of age and sex differences are possible only if the
same personality traits are found across cultures. After all, it would make no
sense to ask how, say, Openness develops in a culture in which that construct did
not make any sense. But there is by now ample evidence that the same
personality traits are found everywhere. The NEO Inventories have been
translated into dozens of languages, and although an occasional item resists
meaningful translation, that process has been relatively easy. Bilingual
psychologists readily find equivalent expressions for NEO items in languages as
diverse as Croatian, Korean, and Telugu. Furthermore, when these translations
are administered to indigenous samples, they typically yield the same factor
structure seen in American samples (McCrae, Terracciano, & 78 Members of the
Personality Profiles of Cultures Project. 2005).
The NEO Inventories have been used in a number of large-scale studies of
heritability (see also the chapter by Jarnecke and South), with results confirming
the general finding from behavior genetic studies that traits are substantially
heritable but show little evidence of shared environmental effects (Plomin &
Daniels, 1987)—that is, effects due to experiences children growing up in the
same family share, such as parental role models, diet, religious training, schools,
and neighborhoods. These environments may well affect the adjustment of the
children when they are children, but seem to leave little or no enduring effect
that shapes adult personality. There have been five notable contributions to the
behavior genetics literature from studies of the NEO Inventories:
1. Because the FFM is comprehensive, it is now possible to assert that essentially all
personality traits are heritable.
2. Because the NEO Inventories assess both domains and facets, it has been possible to
test the hypothesis that facets, as well as domains, are heritable, and that the heritability of
facets is due in part to the heritability of the specific variance in each facet (Jang,
McCrae, Angleitner, Riemann, & Livesley, 1998). There is reason to hypothesize that the
specific variance in nuances of each facet is also heritable (McCrae, 2015). People are not
born with a few innate temperaments that are then differentiated by experience into the
wealth of traits seen in adults; instead, the wealth of traits is already implicit in the
infant’s genome.
3. Because the NEO Inventories have been translated into a number of different
languages, cross-cultural studies now provide evidence that heritability is universal, as is
the differential heritability of facets. For example, Openness to Aesthetics is one of the
most heritable traits in Canada, Germany, Japan, and Sardinia (mean h2 = .47);
Deliberation is one of the least heritable in all four countries (mean h2 = .28).
4. Because the structure of the NEO Inventories is well established, it is possible to
compare the phenotypic structure to the genetic structure of personality. Here research has
shown that the phenotypic structure mirrors the genotypic structure, whether assessed
through classic twin studies (Yamagata et al., 2006) or studies of non-twin family
members (Pilia et al., 2006). There is also some evidence that the phenotypic structure is
due solely to the genotypic structure, rather than to influences from the nonshared
environment (McCrae, Jang, Livesley, Riemann, & Angleitner, 2001).
5. There are also some interesting, if tentative, suggestions of genetic influence from
NEO Inventory studies of variance in populations. Borkenau, McCrae, and Terracciano
(2013) reported that males show higher variance than females for traits in the domains of
Extraversion, Openness, Agreeableness, and Conscientiousness. One possible mechanism
is inheritance through the X-chromosome: Because males have only one copy, any
personality-related allele on it will be fully expressed, whereas effects in females are
likely to be tempered by the presence of alleles on two X-chromosomes. Costa and
colleagues (2007) showed that personality trait variance was reduced in Sardinia
compared to mainland Italy. Sardinia is a founder population, relatively isolated from
other gene pools, and over the course of time, the phenomenon of genetic drift reflects the
permanent loss of rare alleles from the pool. In consequence, there is greater genetic
uniformity in Sardinia, which could account for the decreased variance in personality
traits.
A Theoretical Synthesis
What do these facts imply for an understanding of the origins and functioning
of personality—or at least personality traits? In one sense, the findings are
mostly negative. In general, adult personality is not shaped by parental guidance
(McCrae & Costa, 1988b), childhood environment, patterns of culture, or
historical experiences shared by a generation. Personality trait levels in adults
change gradually, but the rate of change is not predicted by features of culture.
Individual differences are barely changed by intervening life events or changes
in physical health. This litany of facts contradicts the basic premises of almost
all the classic theories of personality—psychoanalytic, behavioral, and
humanistic—and demands new theories of personality, several of which have
been offered (McCrae, 2011a).
What does shape personality? Genes, which contribute to traits at all levels of
the hierarchy; sex and aging; brain pathology, including Alzheimer’s disease and
major depression; pharmaceutical agents, including antidepressants (Costa,
Bagby, Herbst, & McCrae, 2005) and perhaps hallucinogens (MacLean,
Johnson, & Griffiths, 2011); malnutrition in infancy (Galler et al., 2013). In
short, personality appears to be in substantial measure biologically based. It is,
of course, not new to suggest that there is some temperamental or constitutional
element to personality; many theories of personality nod to such a possibility.
What is radically new is the idea that in a certain sense, biology is essentially the
only determinant of traits. That is one of the basic tenets of Five Factor Theory
(FFT; McCrae & Costa, 2008b), a general theory of personality proposed to
account for research findings derived from studies of the FFM.
Two caveats must immediately be appended to the assertion that personality
traits rest on an exclusively biological basis. The first is that FFT is a theory; it is
intended to provide a broad-brush account of human nature and conduct, and,
like every theory in the social sciences, it will certainly have exceptions. Indeed,
identifying the particular environmental conditions that create exceptions may be
the most fruitful way to gain a deeper understanding of personality. The second
caveat is that the claim applies only to a particular conception of what traits are,
namely, basic tendencies. The concrete manifestations of traits in habits,
preferences, attitudes, and relationships are unquestionably shaped by personal
experience, but these—by the definitions of FFT—are not traits, but
characteristic adaptations.
The conceptual distinction between basic tendencies and characteristic
adaptations allows a causal interpretation of the data supplied by FFM research:
Basic tendencies (including traits at all levels of the hierarchy) are based in
biology, whereas characteristic adaptations result from the interplay of basic
tendencies with external influences—the interaction of the organism with the
environment.
This conception, represented schematically in Figure 2.2, did not emerge fully
formed. Our early longitudinal research had convinced us that personality was
better construed as an independent variable than as a dependent variable.
Personality was not shaped by adult life (as many theories of adult development
had held); instead, personality itself affected the life choices people made and
their responses to changing circumstances. We enshrined that insight in the title
of our first book, Emerging Lives, Enduring Dispositions (McCrae & Costa,
1984). We certainly did not understand at the time where traits came from, and
the basic idea that adult traits are best viewed as causes rather than consequences
of life circumstances would have been consistent with many different accounts
of early personality development.
The full-scale development of FFT coincided with our early cross-cultural
research, beginning in the 1990s. Culture is, in a sense, the ultimate
environment, a more-or-less consistent package of language, customs, kinship
patterns, and religious beliefs. Anthropologists such as Ruth Benedict (1934) had
argued persuasively that personality was little more than the internalization of
culture, but early suggestions of the cross-cultural generalizability of the FFM
(Liu, 1991) encouraged us to begin to think of personality traits as universal
expressions of human nature. We could reconcile that view with the indisputable
fact that people from different cultures had different habits, beliefs, values, and
so on, because we already knew how to separate traits from their changeable
manifestations. FFT and Figure 2.2 were finalized by the mid-1990s (McCrae &
Costa, 1996), with only minor clarifications and additions in later statements
(McCrae & Costa, 2008b). In the intervening time we wrote a series of articles
using FFT to illuminate topics such as trait explanations (McCrae & Costa,
1995), cross-cultural psychology (Allik & McCrae, 2002), and adult
development (McCrae & Costa, 2003).
Figure 2.2. A representation of the Five Factor Theory personality system. Core
components are in rectangles; interfacing components are in ellipses.
Adapted from McCrae and Costa (2008).
Theories are supposed to account for known facts, as FFT generally does, but
they are also supposed to generate novel predictions. The most important tests of
FFT conducted since it was proposed are from cross-cultural studies. We had
preliminary evidence that the structure of personality was the same in some other
cultures, but subsequent work showed that the FFM could be identified in a very
wide range of cultures (McCrae, Terracciano, & 78 Members of the Personality
Profiles of Cultures Project, 2005). Fundamentally, FFT asserts that traits are
transcultural, and therefore that most properties of traits ought to be universal. A
series of studies confirmed that prediction with regard to age differences
(McCrae et al., 1999); gender differences (Costa et al., 2001); stereotypes of age
(Chan et al., 2012) and gender (Löckenhoff et al., 2014) differences; cross-
observer agreement (McCrae et al., 2004); self/informant discrepancies (Allik et
al., 2010); and the differential internal consistency, retest reliability, and
heritability of NEO Inventory facet scales (McCrae et al., 2011). It is hard to
imagine an alternative explanation for all these universals.
New Tests of FFT

In one sense, FFT is an attractive target. It makes sweeping claims, many of
which fly in the face of traditional wisdom in psychology. In addition, many
psychologists would prefer a theory that is more congenial to therapeutic and
life-enhancing interventions (McCrae, 2011b). As a result, a number of
researchers have begun to conduct tests of FFT. This is a welcome development;
it will either confirm the theory or show the ways in which it should be
modified.
Kandler and colleagues have conducted several studies examining FFT using
German versions of the NEO Inventories in behavior genetic designs. In one
(Kandler, Bleidorn, Riemann, Angleitner, & Spinath, 2011), they examined
genetic and environmental origins of a set of characteristic adaptations, interests
(e.g., domestic–manual, social–educational, cultural–intellectual) in relation to
FFM domains. They argued that, as characteristic adaptations, interests ought to
be related to personality traits, and that any heritable component in interests
ought to be mediated by FFM traits. They did find associations with personality
(especially Openness), and found that about a third of the genetic influences on
interest could be accounted for by personality—leaving two-thirds unaccounted
for. They noted that interests might also reasonably be caused by “other
genetically-based cognitive and physical abilities and attributes (e.g.,
intelligence, attractiveness, athleticism)” (Kandler et al., 2011, p. 1640)—that is,
other basic tendencies—as well as by more specific personality facets not
examined in their study. This is consistent with FFT.
In a second article (Bleidorn et al., 2010) they examined major life goals—
Agency and Communion—and the FFM, assessed twice over a 5-year interval.
As characteristic adaptations, goals might be assumed to be less heritable and
less stable than traits, and this is just what they found. About one-third of the
variance in goals could be attributed to genetic influences, and much of that
(40% to 70%) was shared with FFM domains, as FFT would expect. However,
there were genetic influences on goals that were not explained by either the five
personality factors or by the 30 NEO facets, as a supplementary analysis
revealed. They have reported similar results for sociopolitical attitudes (Kandler,
Bleidorn, & Riemann, 2012) and religiousness (Kandler & Riemann, 2013) and
concluded that attitudes, interest, and goals might be “discrete elements of the
personality system” (Bleidorn et al., 2010, p. 375) with a genetic basis outside
that of traits.
That is one possibility, but there is another that is compatible with FFT.
According to the duality principle, goals (like attitudes and interests) are both
characteristic adaptations and indicators of basic tendencies. In other words,
Agency and Communion might in one respect be considered facet-level
personality traits. Both are clearly related to the FFM factors (Agency mainly to
Extraversion and Openness, Communion to Extraversion and Agreeableness),
but both also have specific variance that is independent of the five factors. In this
respect, they are no different from other facets of the FFM, such as Excitement
Seeking and Modesty. Such an interpretation would expand the range of FFM
facets without seriously altering the premises of FFT.
A number of researchers have tested the FFT claim that traits are not
influenced by the environment by seeking conditions under which change does
occur. Boals, Southard-Dobbs, and Blumenthal (2015) assessed traits twice in a
large (>1,000) college sample over a 3-month interval; they also ascertained
which of their respondents had experienced a traumatic event in the interval.
Those who had such an experience increased (nonsignificantly) in Neuroticism,
whereas the others decreased (nonsignificantly). Together, these amounted to a
significant interaction effect, with event status accounting for about 1% of the
variance in Time 2 Big Five Inventory (BFI; John, Donahue, & Kentle, 1991)
Neuroticism. Openness and Conscientiousness also increased in those who
reported a life-changing stressful event.
Leikas and Salmela-Aro (2015) examined life changes in 500+ Finns between
ages 20 and 23 years. Events included entering the workforce, university, or a
relationship; onset of a chronic disease; and drug use. Of 30 moderated latent
difference score analyses, three were significant: Incident chronic diseases
increased Neuroticism and retarded the growth of Extraversion, and entry into
the workforce increased Conscientiousness. Cross-lagged path analyses
suggested that entering into the workforce, university, or a relationship led to
higher Conscientiousness scores, as did experiencing a chronic disease; chronic
disease and drug use led to higher Neuroticism, whereas entering the workforce
led to lower Neuroticism. Standardized regression coefficients for these effects
ranged from –.04 to .13.
Jackson, Jonkmann, Lüdke, and Trautwein (2012) examined personality
change in German males who were recruited for military training or opted for
civilian service. Military discipline might be expected to increase
Conscientiousness, but the recruits did not change more than the controls, nor
were there significant differential changes for Neuroticism, Extraversion, or
Openness. However, those who entered military service increased less in
Agreeableness than did their civilian counterparts.
Löckenhoff, Terracciano, Patriciu, Eaton, and Costa (2009) examined NEO-
PI-R scores for nearly 500 urban adults assessed in 1993 and again 8 years later.
A quarter of the sample reported “an extremely horrifying or frightening event”
within 2 years of the second assessment. Significant (p < .01) effects were seen
for Neuroticism and its N2: Angry Hostility facet, which increased, and for O6:
Values and A4: Compliance, which declined. Traumatic events contributed 1%
to 2% to the prediction of Time 2 personality scores.
Sutin, Costa, Wethington, and Eaton (2010) studied the same urban cohort but
focused on a different issue. At Time 2, participants selected the single most
stressful or upsetting experience since baseline, and were then asked if the event
was a turning point in their lives (58% agreed) and if they had learned from the
event (82% agreed). Events themselves did not predict personality change,
except that those who reported problems with their children (6%) became lower
in Agreeableness and A3: Altruism. However, subjective construals of the event
did predict change: Those who saw the event as a turning point increased in
Neuroticism and N4: Self-Consciousness, and those who claimed to have
learned from it increased in Extraversion (and E6: Positive Emotions) and
Conscientiousness (and C5: Self-Discipline). Standardized regression
coefficients ranged from .08 to .11.
It would be fair to describe these as small and scattered effects, and sometimes
contradictory: Boals and colleagues (2015) found life-changing events increased
Openness, whereas Löckenhoff and colleagues (2009) reported that they
decreased Openness to Values. Some trends do appear: There is fairly consistent
evidence that stressful events increase Neuroticism, at least in the short term, and
that entering into adult life roles increases Conscientiousness. However, the
small magnitude of the effects is sobering, especially considering that the
environmental influences—starting work, beginning a relationship, suffering a
horrifying event—are presumably quite powerful in comparison to the usual
therapeutic interventions familiar to psychologists.
Still, it could legitimately be argued that however small, they demonstrate that
the strong form of FFT—the environment has no influence on traits—is wrong.
But it is perhaps too soon to make that claim, because all of these studies are
subject to an important limitation: They relied exclusively on self-reports.
According to the duality principle, test responses can represent characteristic
adaptations as well as basic tendencies, so it is possible that all the changes
reported reflect modifications in the individuals’ self-concepts, not in their true
underlying personality traits. Military training, for example, might encourage
men to believe that they are tough and (slightly) less agreeable than their civilian
counterparts. From the observer’s perspective, however, there might be no
difference in personality.
This objection has some plausibility, because early research failed to find
agreement across raters on personality changes over time (McCrae, 1993;
Watson & Humrichouse, 2006). In a sense, this is not surprising, because change
scores invariably include a great deal of error, and the signal—if there is one—is
likely weak. It is therefore of particular note that Kandler and colleagues (2010)
were able to conduct genetic analyses of the sources of personality change by
combining self-reports and peer ratings in a large-scale twin study. They
concluded that personality stability was largely genetic in origin, but that change
was mainly attributable to environmental factors. The limitation of this study
design is that it says nothing whatever about the nature of the environmental
factors (other than that they were nonshared). It does not allow us to conclude
that traumatic events or work experiences have a real influence on personality,
because the changes attributable to those sources may not overlap with the
consensually validated changes detected in this multimethod study. Still, it does
suggest that it is worthwhile to look for environmental influences on personality
development.
But environmental influences are not quite the same as FFT’s external
influences, because to a behavior geneticist, environmental means anything that
is not genetic. As Kandler and colleagues (2010) acknowledged, this includes
biological influences, which are entirely consistent with FFT. The onset of
Alzheimer’s disease, for example, has notable effects on personality that are not
genetic; if a subset of twins in Kandler’s study had developed that kind of
disorder, its effects would have been classified as environmental.
Where does this leave research on FFT, and in particular on the issue of
environmental influences on personality development? Certainly it is unfinished,
but we already know that the effects of interest are subtle and susceptible to
artifacts. Future designs, whether of naturalistic occurrences or experimental
interventions (Jackson, Hill, Payne, Roberts, & Stine-Marrow, 2012), need to
use large samples, multiple occasions, and multiple methods of personality
measurement. And researchers ought to widen the list of candidate causes to
include bio logical factors (such as disease, diet, and environmental toxins) as
well as social and psychological factors. Just as physicists have been obliged to
build larger and larger accelerators to discover new particles, psychologists are
discovering that the more we know about personality, the more challenging it is
to advance knowledge.
Conclusions
In looking back over our research on the FFM, there is a certain irony. When
we first proposed that traits were highly stable in adulthood, we were met with
skepticism from adult developmentalists and were challenged to consider a host
of alternative explanations, including the possibility that it was merely the
individual’s self-concept that was stable (McCrae & Costa, 1982). Now we find
ourselves in the position of defending FFT, challenging those who argue for
environmental influences on trait development to consider alternative
explanations, including the possibility that it is merely the individual’s self-
concept that changes. This is the nature of scientific progress, and fortunately,
there has been a great deal of it since Loevinger’s 1957 essay.
Acknowledgments
We thank Christian Kandler for helpful comments. Paul T. Costa Jr., and
Robert R. McCrae receive royalties from the NEO Inventories.
Notes
1. The importance of Loevinger’s article is attested to by the fact that Watson also
paraphrased its title in a recent chapter: Watson, D. (2012). Objective tests as instruments of
psychological theory and research. In H. Cooper, P. M. Camic, D. L. Long, A. T. Panter, D.
Rindskopf, & K. J. Sher (Eds.), APA handbook of research methods in psychology. Vol. 1:
Foundations, planning, measures, and psychometrics (pp. 349–369). Washington, DC : American
2. The largest of these groups occupies 196,883 dimensions—a humbling thought for
personality psychologists, who struggle with a mere 5, 6, or 16 dimensions.
3. “Situations” may seem an odd ingredient of traits, but Yang and colleagues (2014)
mean by it the context that evokes the expression of a trait. Trait theorists such as
Tellegen (1991) have long recognized that traits operate conditionally: Extraverts
usually do not chatter on during a ceremonial moment of silence.
4. In principle, we could also assess them through their biological basis, although
present knowledge about genetics and neuroscience is far from adequate for that
purpose. Note, however, that it would still be an indirect assessment, since biology is
categorically distinct from psychological constructs.
5. It is useful to have multiple measures of psychological constructs (e.g., De Raad &
Perugini, 2002) so that artifacts of a particular instrument can be identified or ruled
out. But there are also advantages to the wide use of a single instrument, because
direct comparisons are then feasible. For example, McCrae and colleagues (2011)
used data from scores of different researchers to assess the generalizability across
samples and cultures of the differential internal consistency, retest reliability,
longitudinal stability, heritability, and cross-observer validity of NEO Inventory
facets.
References
Allik, J., & McCrae, R. R. (2002). A Five-Factor Theory perspective. In R. R. McCrae &
J. Allik (Eds.), The Five-Factor Model of personality across cultures (pp. 303–322).
New York, NY: Kluwer Academic/Plenum Publishers.
Allik, J., Realo, A., Mõttus, R., Borkenau, P., Kuppens, P., & Hřebíčková, M. (2010).
How people see others is different from how people see themselves: A replicable
pattern across cultures. Journal of Personality and Social Psychology, 99, 870–882.
Allport, G. W. (1937). Personality: A psychological interpretation. New York, NY: Holt.
Allport, G. W. (1961). Pattern and growth in personality. New York, NY: Holt, Rinehart
& Winston.
Angleitner, A., & Ostendorf, F. (1994). Temperament and the Big Five factors of
personality. In C. F. Halverson, G. A. Kohnstamm, & R. P. Martin (Eds.), The
developing structure of temperament and personality from infancy to adulthood (pp.
69–90). Hillsdale, NJ: Lawrence Erlbaum.
Asendorpf, J. B. (2003). Head-to-head comparison of the predictive validity of
personality types and dimensions. European Journal of Personality, 17, 327–346.
Asendorpf, J. B., & van Aken, M. A. G. (1999). Resilient, overcontrolled, and
uncercontrolled personality prototypes in childhood: Replicability, predictive power,
and trait-type issue. Journal of Personality and Social Psychology, 77, 815–832.
Ashton, M. C., & Lee, K. (2005). Honesty-Humility, the Big Five, and the Five-Factor
Model. Journal of Personality, 73, 1321–1353.
Ashton, M. C., Lee, K., Perugini, M., Szarota, P., De Vries, R. E., Di Blass, L., … De
Raad, B. (2004). A six-factor structure of personality descriptive adjectives: Solutions
from psycholexical studies in seven languages. Journal of Personality and Social
Psychology, 86, 356–366.
Back, M. D., Schmukle, S. C., & Egloff, B. (2009). Predicting actual behavior from the
explicit and implicit self-concept of personality. Journal of Personality and Social
Bell, B., Rose, C. L., & Damon, A. (1972). The Normative Aging Study: An
interdisciplinary and longitudinal study of health and aging. International Journal of
Aging and Human Development, 3, 5–17.
Benedict, R. (1934). Patterns of culture. Boston, MA: Houghton Mifflin.
Biesanz, J. C., & West, S. G. (2004). Towards understanding assessments of the Big Five:
Multitrait-multimethod analyses of convergent and discriminant validity across
measurement occasion and type of observer. Journal of Personality, 72, 845–876.
Bleidorn, W., Kandler, C., Hülsheger, U. R., Riemann, R., Angleitner, A., & Spinath, F.
M. (2010). Nature and nurture of the interplay between personality traits and major life
goals. Journal of Personality and Social Psychology, 99, 366–379.
Bleidorn, W., Klimstra, T. A., Denissen, J. J. A., Rentfrow, P. J., Potter, P. J., & Gosling,
S. D. (2013). Personality maturation around the world: A cross-cultural examination of
Social Investment Theory. Psychological Science, 24, 2530–2540.
Block, J. (1961). The Q-sort method in personality assessment and psychiatric research.
Springfield, IL: Charles C Thomas.
Boals, A., Southard-Dobbs, S., & Blumenthal, H. (2015). Adverse events in emerging
adulthood are associated with increases in Neuroticism. Journal of Personality, 34,
476–494.
Borkenau, P., McCrae, R. R., & Terracciano, A. (2013). Do men vary more than women?
A study in 51 cultures. Journal of Research in Personality, 47, 135–144.
Caprara, G. V., Barbaranelli, C., Borgogni, L., & Perugini, M. (1993). The “Big Five
Questionnaire”: A new questionnaire to assess the Five-Factor Model. Personality and
Individual Differences, 15, 281–288.
Cattell, R. B., Eber, H. W., & Tatsuoka, M. M. (1970). The handbook for the Sixteen
Personality Factor Questionnaire. Champaign, IL: Institute for Personality and Ability
Testing.
Chan, W., McCrae, R. R., De Fruyt, F., Jussim, L., Löckenhoff, C. E., De Bolle, M., …
Terracciano, A. (2012). Stereotypes of age differences in personality: Universal and
accurate? Journal of Personality and Social Psychology, 103, 1050–1066.
Cheung, F., Cheung, S. F., Zhang, J., Leung, K., Leong, F., & Yeh, K.-H. (2008).
Relevance of Openness as a personality dimension in Chinese culture: Aspects of its
cultural relevance. Journal of Cross-Cultural Psychology, 39, 81–108.
Cheung, F. M., Leung, K., Zhang, J.-X., Sun, H.-F., Gan, Y.-Q., Song, W.-Z., & Xie, D.
(2001). Indigenous Chinese personality constructs: Is the Five-Factor Model complete?
Journal of Cross-Cultural Psychology, 32, 407–433.
Cloninger, C. R., Przybeck, T. R., Svrakic, D. M., & Wetzel, R. D. (1994). The
Temperament and Character Inventory (TCI): A guide to its development and use. St.
Louis, MO: Author.
Conn, S. R., & Rieke, M. L. (Eds.). (1994). 16PF Fifth Edition technical manual.
Champaign, IL: Institute for Personality and Ability Testing.
Costa, P. T., Jr., Bagby, R. M., Herbst, J. H., & McCrae, R. R. (2005). Personality self-
Affective Disorders, 89, 45–55.
Costa, P. T., Jr., Fozard, J. L., & McCrae, R. R. (1977). Personological interpretation of
factors from the Strong Vocational Interest Blank scales. Journal of Vocational
Behavior, 10, 231–243.
Costa, P. T., Jr., Fozard, J. L., McCrae, R. R., & Bossé, R. (1976). Relations of age and
personality dimensions to cognitive ability factors. Journal of Gerontology, 31, 663–
669.
Costa, P. T., Jr., Herbst, J. H., McCrae, R. R., Samuels, J., & Ozer, D. J. (2002). The
replicability and utility of three personality types. European Journal of Personality, 16,
S73–S87.
Costa, P. T., Jr., & McCrae, R. R. (1976). Age differences in personality structure: A
cluster analytic approach. Journal of Gerontology, 31, 564–570.
Costa, P. T., Jr., & McCrae, R. R. (1977a). Age differences in personality structure
revisited: Studies in validity, stability, and change. International Journal of Aging and
Human Development, 8, 261–275.
Costa, P. T., Jr., & McCrae, R. R. (1977b). Psychiatric symptom dimensions in the
Cornell Medical Index among normal adult males. Journal of Clinical Psychology, 33,
941–946.
Costa, P. T., Jr., & McCrae, R. R. (1978). Objective personality assessment. In M.
Storandt, I. C. Siegler, & M. F. Elias (Eds.), The clinical psychology of aging (pp. 119–
143). New York, NY: Plenum Press.
Costa, P. T., Jr., & McCrae, R. R. (1980). Still stable after all these years: Personality as a
key to some issues in adulthood and old age. In P. B. Baltes & O. G. Brim, Jr. (Eds.),
Life span development and behavior (Vol. 3, pp. 65–102). New York, NY: Academic
Press.
Costa, P. T., Jr., & McCrae, R. R. (1984). Personality as a lifelong determinant of well-
being. In C. Malatesta & C. Izard (Eds.), Affective processes in adult development and
aging (pp. 141–157). Beverly Hills, CA: Sage.
Costa, P. T., Jr., & McCrae, R. R. (1985). The NEO Personality Inventory manual.
Odessa, FL: Psychological Assessment Resources.
Costa, P. T., Jr., & McCrae, R. R. (1988). From catalog to classification: Murray’s needs
and the Five-Factor Model. Journal of Personality and Social Psychology, 55, 258–
265.
Costa, P. T., Jr., & McCrae, R. R. (1992a). Reply to Eysenck. Personality and Individual
Differences, 13, 861–865.
Costa, P. T., Jr., & McCrae, R. R. (1992b). Trait psychology comes of age. In T. B.
Sonderegger (Ed.), Nebraska Symposium on Motivation: Psychology and aging (pp.
169–204). Lincoln, NE: University of Nebraska Press.
Costa, P. T., Jr., & McCrae, R. R. (2008). The Revised NEO Personality Inventory (NEO-
PI-R). In G. J. Boyle, G. Matthews, & D. Saklofske (Eds.), Sage handbook of
personality theory and assessment (Vol. 2, pp. 179–198). Los Angeles, CA: Sage.
Costa, P. T., Jr., & McCrae, R. R. (2010). The Five-Factor Model, Five-Factor Theory,
and interpersonal psychology. In S. Strack & L. M. Horowitz (Eds.), Handbook of
interpersonal psychology (pp. 91–104). New York, NY: Guilford Press.
Costa, P. T., Jr., & McCrae, R. R. (2014). The NEO Inventories. In R. P. Archer & S. R.
Smith (Eds.), Personality assessment (2nd ed., pp. 223–255). London, England:
Routledge.
Costa, P. T., Jr., McCrae, R. R., & Arenberg, D. (1980). Enduring dispositions in adult
males. Journal of Personality and Social Psychology, 38, 793–800.
Costa, P. T., Jr., McCrae, R. R., & Arenberg, D. (1983). Recent longitudinal research on
personality and aging. In K. W. Schaie (Ed.), Longitudinal studies of adult
psychological development (pp. 222–265). New York, NY: Guilford Press.
Costa, P. T., Jr., McCrae, R. R., & Dye, D. A. (1991). Facet scales for Agreeableness and
Conscientiousness: A revision of the NEO Personality Inventory. Personality and
Costa, P. T., Jr., Metter, E. J., & McCrae, R. R. (1994). Personality stability and its
contribution to successful aging. Journal of Geriatric Psychiatry, 27, 41–59.
Costa, P. T., Jr., & Piedmont, R. L. (2003). Multivariate assessment: NEO-PI-R profiles
of Madeline G. In J. S. Wiggins (Ed.), Paradigms of personality assessment (pp. 262–
Costa, P. T., Jr., Terracciano, A., & McCrae, R. R. (2001). Gender differences in
personality traits across cultures: Robust and surprising findings. Journal of
Costa, P. T., Jr., Terracciano, A., Uda, M., Vacca, L., Mameli, C., Pilia, G., … McCrae,
R. R. (2007). Personality traits in Sardinia: Testing founder population effects on trait
means and variances. Behavior Genetics, 37, 376–387.
De Bolle, M., De Fruyt, F., McCrae, R. R., Terracciano, A, Löckenhoff, C. E., Costa,
P.T., Jr., … other collaborators of the Adolescent Personality Profiles of Cultures
Project (2012, July). Age differences in adolescent personality ratings: A cross-cultural
investigation with implications for the Social Investment Theory. Paper presented at the
16th European Conference on Personality, Trieste, Italy.
De Raad, B., & Peabody, D. (2005). Cross-culturally recurrent personality factors:
Analyses of three factors. European Journal of Personality, 19, 451–474.
De Raad, B., & Perugini, M. (Eds.). (2002). Big Five assessment. Gottingen, Germany:
Hogrefe & Huber Publishers.
DeYoung, C. G., Peterson, J. B., & Higgins, D. M. (2002). Higher-order factors of the
Big Five predict conformity: Are there neuroses of health? Personality and Individual
DeYoung, C. G., Quilty, L. C., & Peterson, J. B. (2007). Between facets and domains: 10
aspects of the Big Five. Journal of Personality and Social Psychology, 93, 880–896.
Digman, J. M. (1997). Higher-order factors of the Big Five. Journal of Personality and
Social Psychology, 73, 1246–1256.
Eysenck, H. J., Barrett, P., Wilson, G. D., & Jackson, C. (1992). Primary trait
measurement of the 21 components of the P-E-N system. European Journal of
Personality Assessment, 8, 109–117.
Eysenck, H. J., & Eysenck, S. B. G. (1975). Manual of the Eysenck Personality
Questionnaire. San Diego, CA: EdITS.
Ferguson, E., Chamorro-Premuzic, T., Pickering, T., & Weiss, A. (2013). Five into one
doesn’t go: A critique of the general factor of personality. In T. Chamorro-Premuzic, S.
von Stumm, & A. Furnham (Eds.), Wiley-Blackwell handbook of individual differences
(pp. 162–186). Oxford, England: Wiley-Blackwell.
Fiske, D. W. (1978). Strategies for personality research. San Francisco, CA: Jossey-Bass.
Funder, D. C. (1993). Judgments as data for personality and developmental psychology:
Error versus accuracy. In D. C. Funder, R. Parke, C. Tomlinson-Keasey, & R.
Widaman (Eds.), Studying lives through time: personality and development (pp. 121–
Funder, D. C., & Sneed, C. D. (1993). Behavioral manifestations of personality: An
ecological approach to judgmental accuracy. Journal of Personality and Social
Galler, J. R., Bryce, C. P., Zichlin, M. L., Waber, D. P., Exner, N., Fitzmaurice, G. M., &
Costa, P. T., Jr. (2013). Malnutrition in the first year of life and personality at age 40.
Journal of Child Psychology and Psychiatry, 54, 911–918.
Glisky, M. L., Tataryn, D. J., Tobias, B. A., Kihlstrom, J. F., & McConkey, K. M. (1991).
Absorption, openness to experience, and hypnotizability. Journal of Personality and
Goldberg, L. R. (1981). Language and individual differences: The search for universals in
personality lexicons. In L. Wheeler (Ed.), Review of personality and social psychology
(Vol. 2, pp. 141–165). Beverly Hills, CA: Sage.
Goldberg, L. R. (1990). An alternative “description of personality”: The Big-Five factor
structure. Journal of Personality and Social Psychology, 59, 1216–1229.
Goldberg, L. R. (1999). A broad-bandwidth, public domain, personality inventory
measuring the lower-level facets of several five-factor models. In I. Mervielde, I.
Deary, F. De Fruyt, & F. Ostendorf (Eds.), Personality psychology in Europe (Vol. 7,
pp. 7–28). Tilburg, The Netherlands: Tilburg University Press.
Gorsuch, R. L. (1974). Factor analysis. Philadelphia, PA: W. B. Saunders.
Gough, H. G. (1987). California Psychological Inventory administrator’s guide. Palo
Alto, CA: Consulting Psychologists Press.
Gülgöz, S. (2002). Five-Factor Model and the NEO-PI-R in Turkey. In R. R. McCrae & J.
Allik (Eds.), The Five-Factor Model of personality across cultures (pp. 167–188). New
York, NY: Kluwer Academic/Plenum Publishers.
Guttman, L. (1954). An outline of some new methodology for social research. Public
Opinion Quarterly, 18, 395–404.
Harkness, A. R., & McNulty, J. L. (2002). Implications of personality individual
differences science for clinical work on personality disorders. In P. T. Costa, Jr. & T.
A. Widiger (Eds.), Personality disorders and the Five-Factor Model of personality
(2nd ed., pp. 391–403). Washington, DC: American Psychological Association.
Hathaway, S. R., & McKinley, J. C. (1943). The Minnesota Multiphasic Personality
Inventory (rev. ed.). Minneapolis, MN: University of Minnesota Press.
Hjelle, L. A., & Siegler, D. J. (1976). Personality: Theories, basic assumptions, research
and applications. New York, NY: McGraw-Hill.
Jackson, D. N. (1974). Personality Research Form manual (rev. ed.). Port Huron, MI:
Research Psychologists Press.
Jackson, D. N. (1984). Personality Research Form manual (3rd ed.). Port Huron, MI:
Jackson, D. N., & Tremblay, P. F. (2002). The Six Factor Personality Questionnaire. In B.
De Raad & M. Perugini (Eds.), Big Five assessment (pp. 353–375). Seattle, WA:
Hogrefe & Huber.
Jackson, J. J., Hill, P. L., Payne, B. R., Roberts, B. W., & Stine-Marrow, E. A. L. (2012).
Can an old dog learn (and want to experience) new tricks? Cognitive training increases
Openness to Experience in older adults. Psychology and Aging, 27, 286–292.
Jackson, J. J., Jonkmann, K., Lüdke, O., & Trautwein, U. (2012). Military training and
personality trait development: Does the military make the man, or does the man make
the military? Psychological Science, 23, 270–277.
James, W. (1981). The principles of psychology (Vol. 1). Cambridge, MA: Harvard
University Press. (Original work published in 1890.)
Jang, K. L., Livesley, W. J., & Vernon, P. A. (1996). Heritability of the Big Five
personality dimensions and their facets: A twin study. Journal of Personality, 64, 575–
591.
Jang, K. L., McCrae, R. R., Angleitner, A., Riemann, R., & Livesley, W. J. (1998).
Heritability of facet-level traits in a cross-cultural twin sample: Support for a
hierarchical model of personality. Journal of Personality and Social Psychology, 74,
1556–1565.
John, O. P., Angleitner, A., & Ostendorf, F. (1988). The lexical approach to personality:
A historical review of trait taxonomic research. European Journal of Personality, 2,
171–203.
John, O. P., Donahue, E. M., & Kentle, R. L. (1991). The “Big Five” Inventory—
Versions 4a and 54. Berkeley, CA: University of California, Berkeley, Institute of
Personality and Social Research.
Judge, T. A., Rodell, J. B., Klinger, R. L., Simon, L. S., & Crawford, E. R. (2013).
Hierarchical representations of the Five-Factor Model of personality in predicting job
performance: Integrating three organizing frameworks with two theoretical
perspectives. Journal of Applied Psychology, 98, 875–925.
Kandler, C., Bleidorn, W., & Riemann, R. (2012). Left or right? Sources of political
orientation: The roles of genetic factors, cultural transmission, assortative mating, and
personality. Journal of Personality and Social Psychology, 102, 633–645.
Kandler, C., Bleidorn, W., Riemann, R., Angleitner, A., & Spinath, F. M. (2011). The
genetic links between the Big Five personality traits and general interest domains.
Personality and Social Psychology Bulletin, 37, 1633–1643.
Kandler, C., Bleidorn, W., Riemann, R., Spinath, F. M., Thiel, W., & Angleitner, A.
(2010). Sources of cumulative continuity in personality: A longitudinal multiple-rater
twin study. Journal of Personality and Social Psychology, 98, 995–1008.
Kandler, C., & Riemann, R. (2013). Genetic and environmental sources of individual
religiousness: The roles of individual personality traits and perceived environmental
religiousness. Behavior Genetics, 43, 297–313.
Kelly, G. A. (1955). The psychology of personal constructs (Vol. 1–3). New York, NY:
Norton.
Lanning, K. (1994). Dimensionality of observer ratings on the California Adult Q-Set.
Journal of Personality and Social Psychology, 67, 151–160.
Leikas, S., & Salmela-Aro, K. (2015). Personality trait changes among young Finns: The
role of life events and transitions. Journal of Personality, 83, 117–126.
Lippa, R. (1991). Some psychometric characteristics of gender diagnosticity measures:
Reliability, validity, consistency across domains, and relationship to the “Big Five.”
Liu, F. Y. (1991). The generalizability of the NEO Personality Inventory to a university
sample in Hong Kong. Unpublished manuscript, Chinese University of Hong Kong.
Livesley, W. J. (2001). Conceptual and taxonomic issues. In W. J. Livesley (Ed.),
Handbook of personality disorders: Theory, research, and treatment (pp. 3–38). New
Löckenhoff, C. E., Chan, W., McCrae, R. R., De Fruyt, F., Jussim, L., De Bolle, M., …
Terracciano, A. (2014). Gender stereotypes of personality: Universal and accurate?
Journal of Cross-Cultural Psychology, 45, 675–694.
Löckenhoff, C. E., Terracciano, A., Patriciu, N. S., Eaton, W. W., & Costa, P. T., Jr.
(2009). Self-reported extreme adverse life events and longitudinal changes in Five-
Factor Model personality traits in an urban sample. Journal of Traumatic Stress, 22,
53–59.
Loevinger, J. (1957). Objective tests as instruments of psychological theory [Monograph
Supplement 9]. Psychological Reports, 3, 635–694.
MacLean, K. A., Johnson, M. W., & Griffiths, R. R. (2011). Mystical experiences
occasioned by the hallucinogen psilocybin lead to increases in the personality domain
of Openness. Journal of Psychopharmacology, 25, 1453–1461.
McAdams, D. P. (1996). Personality, modernity, and the storied self: A contemporary
framework for studying persons. Psychological Inquiry, 7, 295–321.
McCrae, R. R. (1982). Consensual validation of personality traits: Evidence from self-
reports and ratings. Journal of Personality and Social Psychology, 43, 293–303.
McCrae, R. R. (1990). Traits and trait names: How well is Openness represented in
natural languages? European Journal of Personality, 4, 119–129.
McCrae, R. R. (1993). Moderated analyses of longitudinal personality stability. Journal
of Personality and Social Psychology, 65, 577–585.
McCrae, R. R. (1994). Openness to Experience: Expanding the boundaries of Factor V.
European Journal of Personality, 8, 251–272.
McCrae, R. R. (1996). Social consequences of experiential openness. Psychological
Bulletin, 120, 323–337.
McCrae, R. R. (2011a). Personality theories for the 21st Century. Teaching of
McCrae, R. R. (2011b). Personality traits and the potential of positive psychology. In K.
M. Sheldon, T. Kashdan, & M. F. Steger (Eds.), Designing positive psychology: Taking
stock and moving forward. New York, NY: Oxford University Press.
McCrae, R. R. (2015). A more nuanced view of reliability: Specificity in the trait
hierarchy. Personality and Social Psychology Review, 19, 97–112.
McCrae, R. R., & Costa, P. T., Jr. (1980). Openness to experience and ego level in
Loevinger’s sentence completion test: Dispositional contributions to developmental
models of personality. Journal of Personality and Social Psychology, 39, 1179–1190.
McCrae, R. R., & Costa, P. T., Jr. (1982). Self-concept and the stability of personality:
Cross-sectional comparisons of self-reports and ratings. Journal of Personality and
McCrae, R. R., & Costa, P. T., Jr. (1984). Emerging lives, enduring dispositions:
Personality in adulthood. Boston, MA: Little, Brown.
McCrae, R. R., & Costa, P. T., Jr. (1985). Updating Norman’s “adequate taxonomy”:
Intelligence and personality dimensions in natural language and in questionnaires.
McCrae, R. R., & Costa, P. T., Jr. (1987). Validation of the Five-Factor Model of
personality across instruments and observers. Journal of Personality and Social
McCrae, R. R., & Costa, P. T., Jr. (1988a). Age, personality, and the spontaneous self-
concept. Journal of Gerontology: Social Sciences, 43, S177–S185.
McCrae, R. R., & Costa, P. T., Jr. (1988b). Recalled parent-child relations and adult
personality. Journal of Personality, 56, 417–434.
McCrae, R. R., & Costa, P. T., Jr. (1989a). Different points of view: Self-reports and
ratings in the assessment of personality. In J. P. Forgas & M. J. Innes (Eds.), Recent
advances in social psychology: An international perspective (pp. 429–439).
Amsterdam, The Netherlands: Elsevier Science Publishers.
McCrae, R. R., & Costa, P. T., Jr. (1989b). Reinterpreting the Myers-Briggs Type
Indicator from the perspective of the Five-Factor Model of personality. Journal of
Personality, 57, 17–40.
McCrae, R. R., & Costa, P. T., Jr. (1992). Discriminant validity of NEO-PI-R facets.
Educational and Psychological Measurement, 52, 229–237.
McCrae, R. R., & Costa, P. T., Jr. (1995). Trait explanations in personality psychology.
McCrae, R. R., & Costa, P. T., Jr. (1996). Toward a new generation of personality
theories: Theoretical contexts for the Five-Factor Model. In J. S. Wiggins (Ed.), The
Five-Factor Model of personality: Theoretical perspectives (pp. 51–87). New York,
NY: Guilford Press.
McCrae, R. R., & Costa, P. T., Jr. (1997). Personality trait structure as a human universal.
American Psychologist, 52, 509–516.
McCrae, R. R., & Costa, P. T., Jr. (2003). Personality in adulthood: A Five-Factor
Theory perspective (2nd. ed.). New York, NY: Guilford Press.
McCrae, R. R., & Costa, P. T., Jr. (2008a). Empirical and theoretical status of the Five-
Factor Model of personality traits. In G. Boyle, G. Matthews, & D. Saklofske (Eds.),
Sage Handbook of personality theory and assessment (Vol. 1, pp. 273–294). Los
Angeles, CA: Sage.
McCrae, R. R., & Costa, P. T., Jr. (2008b). The Five-Factor Theory of personality. In O.
P. John, R. W. Robins, & L. A. Pervin (Eds.), Handbook of personality: Theory and
research (3rd ed., pp. 159–181). New York, NY: Guilford Press.
McCrae, R. R., & Costa, P. T., Jr. (2010). NEO Inventories professional manual. Odessa,
FL: Psychological Assessment Resources.
McCrae, R. R., Costa, P. T., Jr., & Busch, C. M. (1986). Evaluating comprehensiveness in
personality systems: The California Q-Set and the Five-Factor Model. Journal of
McCrae, R. R., Costa, P. T., Jr., Lima, M. P., Simões, A., Ostendorf, F., Angleitner, A.,
… Piedmont, R. L. (1999). Age differences in personality across the adult life span:
Parallels in five cultures. Developmental Psychology, 35, 466–477.
McCrae, R. R., Costa, P. T., Jr., & Martin, T. A. (2005). The NEO-PI-3: A more readable
Revised NEO Personality Inventory. Journal of Personality Assessment, 84, 261–270.
McCrae, R. R., Costa, P. T., Jr., Martin, T. A., Oryol, V. E., Rukavishnikov, A. A., Senin,
I. G., & Urbánek, T. (2004). Consensual validation of personality traits across cultures.
Journal of Research in Personality, 38, 179–201.
McCrae, R. R., Costa, P. T., Jr., & Piedmont, R. L. (1993). Folk concepts, natural
language, and psychological constructs: The California Psychological Inventory and
the Five-Factor Model. Journal of Personality, 61, 1–26.
McCrae, R. R., Herbst, J. H., & Costa, P. T., Jr. (2001). Effects of acquiescence on
personality factor structures. In R. Riemann, F. Ostendorf, & F. Spinath (Eds.),
Personality and temperament: Genetics, evolution, and structure (pp. 217–231).
Berlin, Germany: Pabst Science Publishers.
McCrae, R. R., Jang, K. L., Livesley, W. J., Riemann, R., & Angleitner, A. (2001).
Sources of structure: Genetic, environmental, and artifactual influences on the
covariation of personality traits. Journal of Personality, 69, 511–535.
McCrae, R. R., Kurtz, J. E., Yamagata, S., & Terracciano, A. (2011). Internal
consistency, retest reliability, and their implications for personality scale validity.
Personality and Social Psychology Review, 15, 28–50.
McCrae, R. R., Löckenhoff, C. E., & Costa, P. T., Jr. (2005). A step towards DSM-V :
Cataloging personality-related problems in living. European Journal of Personality,
19, 269–270.
McCrae, R. R., Martin, T. A., & Costa, P. T., Jr. (2005). Age trends and age norms for the
NEO Personality Inventory-3 in adolescents and adults. Assessment, 12, 363–373.
McCrae, R. R., Terracciano, A., Costa, P. T., Jr., & Ozer, D. J. (2006). Person-factors in
the California Adult Q-Set: Closing the door on personality trait types? European
Journal of Personality, 20, 29–44.
McCrae, R. R., Terracciano, A., & 78 Members of the Personality Profiles of Cultures
Project. (2005). Universal features of personality traits from the observer’s perspective:
Data from 50 cultures. Journal of Personality and Social Psychology, 88, 547–561.
McCrae, R. R., Yamagata, S., Jang, K. L., Riemann, R., Ando, J., Ono, Y., … Spinath, F.
(2008). Substance and artifact in the higher-order factors of the Big Five. Journal of
McCrae, R. R., Zonderman, A. B., Costa, P. T., Jr., Bond, M. H., & Paunonen, S. V.
(1996). Evaluating replicability of factors in the Revised NEO Personality Inventory:
Confirmatory factor analysis versus Procrustes rotation. Journal of Personality and
Mischel, W. (1968). Personality and assessment. New York, NY: Wiley.
Mõttus, R., McCrae, R. R., Allik, J., & Realo, A. (2014). Cross-rater agreement on
common and specific variance of personality traits and items. Journal of Research in
Mullins-Sweatt, S. N., Jamerson, J. E., Samuel, D. B., Olson, D. R., & Widiger, T. A.
(2006). Psychometric properties of an abbreviated instrument of the Five-Factor
Model. Assessment, 13, 119–137.
Myers, I. B., & McCaulley, M. H. (1985). Manual: A guide to the development and use of
the Myers-Briggs Type Indicator. Palo Alto, CA: Consulting Psychologists Press.
Noftle, E. E., & Robins, R. W. (2007). Personality predictors of academic outcomes: Big
Five correlates of GPA and SAT scores. Journal of Personality and Social Psychology,
93, 116–130.
Norman, W. T. (1963). Toward an adequate taxonomy of personality attributes:
Replicated factor structure in peer nomination personality ratings. Journal of Abnormal
Ozer, D. J., & Benet-Martínez, V. (2006). Personality and the prediction of consequential
Paunonen, S. V., & Ashton, M. C. (2001). Big Five factors and facets and the prediction
of behavior. Journal of Personality and Social Psychology, 81, 524–539.
Pervin, L. A. (1994). A critical analysis of current trait theory. Psychological Inquiry, 5,
103–113.
Piedmont, R. L. (1999). Does spirituality represent the sixth factor of personality?
Spiritual transcendence and the Five-Factor Model. Journal of Personality, 67, 985–
1013.
Pilia, G., Chen, W.-M., Scuteri, A., Orrú, M., Albai, G., Deo, M., … Schlessinger, D.
(2006). Heritability of cardiovascular and personality traits in 6,148 Sardinians. PLoS
Genetics, 2, 1207–1223.
Plomin, R., & Daniels, D. (1987). Why are children in the same family so different from
one another? Behavioral and Brain Sciences, 10, 1–16.
Pytlik Zillig, L. M., Hemenover, S. H., & Dienstbier, R. A. (2002). What do we assess
when we assess a Big 5 trait? A content analysis of the affective, behavioral, and
cognitive processes represented in Big 5 personality inventories. Personality and Social
Psychology Bulletin, 28, 847–858.
Reynierse, J. H. (2012). Toward an empirically sound and radically revised type theory.
Journal of Psychological Type, 72, 1–25.
Reynolds, S. K., & Clark, L. A. (2001). Predicting dimensions of personality disorder
from domains and facets of the Five-Factor Model. Journal of Personality, 69, 199–
222.
Riemann, R., & Kandler, C. (2010). Construct validation using multitrait-multimethod-
twin data: The case of a general factor of personality. European Journal of Personality,
24, 258–277.
Roberts, B. W., Bogg, T., Walton, K. E., Chernyshenko, O. S., & Stark, S. E. (2004). A
lexical investigation of the lower-order structure of Conscientiousness. Journal of
Research in Personality, 38, 164–178.
Roberts, B. W., Chernyshenko, O. S., Stark, S. E., & Goldberg, L. R. (2005). The
structure of Conscientiousness: An empirical investigation based on seven major
personality questionnaires. Personnel Psychology, 58, 103–139.
Roberts, B., Wood, D., & Smith, J. L. (2005). Evaluating Five-Factor Theory and social
investment perspectives on personality trait development. Journal of Research in
Robins, R. W., John, O. P., Caspi, A., Moffitt, T. E., & Stouthamer-Loeber, M. (1996).
Resilient, overcontrolled, and undercontrolled boys: Three replicable personality types.
Ronan, M. (2006). Symmetry and the monster: One of the greatest quests of mathematics.
New York, NY: Oxford University Press.
Ross, L. (1977). The intuitive psychologist and his shortcomings: Distortions in the
attribution process. In L. Berkowitz (Ed.), Advances in experimental social psychology
(Vol. 10, pp. 173–220). New York, NY: Academic Press.
Saucier, G., & Goldberg, L. R. (1998). What is beyond the Big Five? Journal of
Shock, N. W., Greulich, R. C., Andres, R., Arenberg, D., Costa, P. T., Jr., Lakatta, E. G.,
… Tobin, J. D. (1984). Normal human aging: The Baltimore Longitudinal Study of
Aging (NIH Publication No. 84–2450). Bethesda, MD: National Institutes of Health.
Siegler, I. C., Welsh, K. A., Dawson, D. V., Fillenbaum, G. G., Earl, N. L., Kaplan, E. B.,
& Clark, C. M. (1991). Ratings of personality change in patients being evaluated for
memory disorders. Alzheimer Disease and Associated Disorders, 5, 240–250.
Sneed, C. D., McCrae, R. R., & Funder, D. C. (1998). Lay conceptions of the Five-Factor
Model and its indicators. Personality and Social Psychology Bulletin, 24, 115–126.
Stephenson, W. (1953). The study of behavior: Q-Technique and its methodology.
Chicago, IL: University of Chicago.
Sutin, A. R., Costa, P. T., Jr., Wethington, E., & Eaton, W. (2010). Turning points and
lessons learned: Stressful life events and personality trait development across middle
adulthood. Psychology and Aging, 25, 524–533.
Swann, W. B., Jr., & Hill, C. A. (1982). When our identities are mistaken: Reaffirming
self-conceptions through social interactions. Journal of Personality and Social
Tellegen, A. (1991). Personality traits: Issues of definition, evidence and assessment. In
W. Grove & D. Cicchetti (Eds.), Thinking clearly about psychology: Essays in honor of
Paul Everett Meehl (Vol. 2, pp. 10–35). Minneapolis, MN: University of Minnesota
Press.
Terracciano, A., Costa, P. T., Jr., & McCrae, R. R. (2006). Personality plasticity after age
30. Personality and Social Psychology Bulletin, 32, 999–1009.
Terracciano, A., McCrae, R. R., Brant, L. J., & Costa, P. T., Jr. (2005). Hierarchical linear
modeling analyses of NEO-PI-R scales in the Baltimore Longitudinal Study of Aging.
Psychology and Aging, 20, 493–506.
van der Linden, D., te Nijenhuis, J., & Bakker, A. B. (2010). The General Factor of
Personality: A meta-analysis of Big Five intercorrelations and a criterion-related
validity study. Journal of Research in Personality, 44, 315–327.
Watson, D., & Clark, L. A. (1997). Extraversion and its positive emotional core. In R.
Hogan, J. Johnson, & S. R. Briggs (Eds.), Handbook of personality psychology (pp.
767–793). New York, NY: Academic Press.
Watson, D., & Humrichouse, J. (2006). Personality development in emerging adulthood:
Integrating evidence from self-ratings and spouse ratings. Journal of Personality and
Widiger, T. A., & Costa, P. T., Jr. (Eds.). (2013). Personality disorders and the Five-
Factor Model of personality (3rd ed.). Washington, DC: American Psychological
Association.
Wiggins, J. S., & Broughton, R. (1985). The interpersonal circle: A structural model for
the integration of personality research. In R. Hogan & W. H. Jones (Eds.), Perspectives
in personality (Vol. 1, pp. 1–47). Greenwich, CT: JAI Press.
Wiggins, J. S., Trapnell, P., & Phillips, N. (1988). Psychometric and geometric
characteristics of the Revised Interpersonal Adjective Scales (IAS-R). Multivariate
Behavioral Research, 23, 119–134.
Woo, S. E., Chernyshenko, O. S., Longley, A., Zhang, Z.-X., Chiu, C.-Y., & Stark, S. E.
(2014). Openness to Experience: Its lower level structure, measurement, and cross-
cultural equivalence. Journal of Personality Assessment, 96, 29–45.
Wrobel, T. A., & Lachar, D. (1982). Validity of the Wiener subtle and obvious scales for
the MMPI: Another example of the importance of inventory-item content. Journal of
Consulting and Clinical Psychology, 50, 469–470.
Yamagata, S., Suzuki, A., Ando, J., Ono, Y., Kijima, N., Yoshimura, K., … Jang, K. L.
(2006). Is the genetic structure of human personality universal? A cross-cultural twin
study from North America, Europe, and Asia. Journal of Personality and Social
Yang, J., McCrae, R. R., Costa, P. T., Jr., Dai, X., Yao, S., Cai, T., & Gao, B. (1999).
Cross-cultural personality assessment in psychiatric populations: The NEO-PI-R in the
People’s Republic of China. Psychological Assessment, 11, 359–368.
Yang, Y., Read, S. J., Denson, T. F., Yiyuan, X., Zhang, J., & Pedersen, W. C. (2014).
The key ingredients of personality traits: Situations, behaviors, and explanations.
Personality and Social Psychology Bulletin, 40, 79–91.
Neuroticism
Jennifer L. Tackett and Benjamin B. Lahey
Abstract
This chapter provides a comprehensive review of the personality domain of
Neuroticism. Neuroticism is not only one of the more salient higher-order
personality domains across different trait models, it also includes great public
health care significance. We begin by describing the domain, including its facets.
We then consider genetic and environmental influences for its development. We
also consider developmental considerations, including evidence for stability and
change across the lifespan. We then turn to the importance of Neuroticism for
predicting consequential outcomes in several relevant domains of functioning:
psychopathology, physical health, and quality of life. We then summarize and end
with suggestions for future directions in research and public health care
application.
Key Words: neuroticism, negative affectivity, negative emotionality, Five Factor
Model, Big Five
The purpose of this chapter is to provide a broad and comprehensive

description of the broad personality trait of Neuroticism (N). In addition to
representing one of the most salient higher-order personality traits across
different trait models, N shows great public health significance because of its
robust predictive validity for many consequential outcomes (Barlow, Sauer-
Zavala, Carl, Bullis, & Ellard, 2014; Lahey, 2009). We begin by describing the
broader N domain, including evidence for genetic and environmental influences
on N. We then briefly discuss developmental considerations of N, including
evidence for stability and change across the lifespan. We then turn to the
relevance of trait N for predicting consequential outcomes in several relevant
domains of functioning: psychopathology, physical health, and quality of life.
We then summarize and end with suggestions for future directions in research
and application.
Description of Neuroticism
Neuroticism is one of the five higher-order domains in the Five Factor Model
(FFM; Costa & McCrae, 1992; Goldberg, 1993). It is typically one of the first
factors extracted within clinical populations and generally reflects individual
differences in tendencies toward negative affect (including sadness, anxiety, and
anger) and individual responses to threat, frustration, or loss (Widiger, 2009).
Neuroticism appears in almost all higher-order dispositional trait models,
sometimes referred to as Negative Affectivity (NA) or Negative Emotionality
(NE) (Rothbart, Ahadi, Hershey, & Fisher, 2001; Tellegen & Waller, 2008).
Although the higher-order domain reflecting tendencies toward negative affect
is broadly consistent across major trait models, more differences exist at the
facet level. To some extent, such differences partially reflect the higher-order
trait model being measured. Specifically, three-factor trait models (Rothbart et
al., 2001; Tellegen & Waller, 2008), which do not typically include a distinct
Agreeableness domain, often incorporate facets of antagonism and aggression
(reflecting disagreeableness) into the broader N/NA/NE domain (Widiger,
2009). Mid-level divisions of N often differentiate subdomains reflecting
irritable versus withdrawn negative affect (De Young, Quilty, & Peterson, 2007;
Shiner & Caspi, 2003). One prominent lower-order facet model is that assessed
by the NEO Personality Inventory-Revised (NEO PI-R; Costa & McCrae, 1992):
Anxiety, Angry Hostility, Depression, Self-Consciousness, Vulnerability, and
Impulsiveness. The NE domain measured by the Minnesota Personality
Questionnaire (MPQ) also includes facets of alienation and stress reactivity
(Tellegen & Waller, 2008). Differences in content assessed at the lower-order, or
facet level likely account for many domain-level differences between measures
(Ormel, Bastiaansen, et al., 2013).
Genetic Influences on Neuroticism

Understanding the causes of N is critical for better understanding the
pathways by which N is related to various outcomes (some of which, including
physical and mental health, are described further in this chapter). Numerous twin
studies consistently indicate that N, like other major personality traits (and,
indeed, most psychological constructs; Turkheimer, 2000), is substantially
heritable (see also Chapter 14 by Jarnecke and South). Specifically,
approximately half the variance in observed, or phenotypic, N, is attributable to
additive genetic influences (Lahey, 2009; Widiger, 2009). The robustness of this
heritability estimate was impressively confirmed in a recent meta-analysis,
which used item response theory methods to harmonize different measurements
of N, resulting in a very large sample size of almost 30,000 twin pairs (van den
Berg et al., 2014). An interesting literature has documented differentiation in
heritability of N by age, with heritability estimates highest (about 50–60%) in
early adolescence and, notably, an absence of sex differences at these
developmental periods (Gillespie, Evans, Wright, & Martin, 2004; Lake, Eaves,
Maes, Heath, & Martin, 2000; Rettew et al., 2006; Viken, Rose, Kaprio, &
Koskenvuo, 1994). Heritability of N appears to decline across adulthood, but
does so more substantially for males, which results in sex differences favoring
higher heritability in females later in adulthood (Eaves, Eysenck, & Martin,
1989; Lake et al., 2000).
With few exceptions (e.g., van den Oord et al., 2008), genome-wide studies of
the association of common single-nucleotide polymorphism with N have been
notably unsuccessful in identifying specific molecular genetic risk factors
(Fullerton et al., 2003; Nash et al., 2004; Shifman et al., 2008). Indeed, a meta-
analysis of genome-wide association data on 17,375 adults across 10 different
samples identified no significant results for N (De Moor et al., 2012). An even
larger study of over 63,000 individuals comprising the Genetics of Personality
Consortium similarly yielded no significant findings for common single-
nucleotide polymorphisms (SNPs; De Moor et al., under review).
One possible reason for the consistent failure to identify significant
associations of N with common SNPs is that a substantial proportion of the
genetic variation in N may not be related to common SNPs. When the
heritability of N was estimated based on similarity among individuals in
common SNPs, the heritability of N was estimated to be 15% (De Moor et al.,
under review), which is substantially lower than heritability estimates from twin
studies of about 50%. One reason for this large discrepancy could be that
estimates of heritability from twin studies can reflect more that “main effects” of
genetic variation, including some types of gene–environment interactions
(Rijsdijk & Sham, 2002), which were not addressed in the molecular genetic
studies. Furthermore, the estimate of 15% heritability could reflect inadequate
coverage of common SNPs or various kinds of sampling biases in these studies.
Nonetheless, this low heritability estimate could indicate that less common SNPs
or even other kinds of genetic variants, such as repeat polymorphisms, are more
important for N. It will be critically important for future work to examine all of
these possibilities.
Candidate gene studies of N have largely focused on the 5-HTTLPR repeat
polymorphism of the serotonin transporter gene based on the well-described role
of the serotonergic system in emotion processing (Leonardo & Hen, 2006). The
5-HTTLPR polymorphism has received particular attention because it is a
functional polymorphism that influences transcriptions, and because this
transporter is a target of psychopharmacological interventions for internalizing
disorders. Meta-analyses suggest small but significant differences on N scores
between individuals with at least one short 5-HTTLPR allele and individuals
with only long alleles (Munafò, Clark, & Flint, 2005; Schinka, 2005; Schinka,
Busch, & Robichaux-Keene, 2004; Sen, Burmeister, & Ghosh, 2004, 2005).
Additional support for 5-HTTLPR comes from neuroimaging research, which
has linked 5-HTTLPR alleles to variation in brain structures (Brown & Hariri,
2006; Hariri et al., 2002; Passamonti et al., 2008) also linked to differences in N
(Canli, 2004; Haas, Omura, Constable, & Canli, 2007; Omura, Constable, &
Canli, 2005).
Other potential candidates that have been linked to N include the 5-HT1A and
5-HTR2A serotonin receptors (Golimbet et al., 2002; Lesch & Canli, 2006) and
the G72 glutamate receptor (Rietschel et al., 2008). Still other hypothesized
candidates have yielded mixed findings (Hettema, An, et al., 2006; Lang et al.,
2005; Sen et al., 2003; Willis-Owen et al., 2005).
Environmental Influences on Neuroticism

Of course, in addition to genetic influences, environmental factors influence
trait N, as well. Behavioral genetic studies typically differentiate two sources of
environmental variance—shared environmental influences, which act similarly
on individuals growing up in the same environment, and nonshared
environmental influences, which serve to differentiate such individuals from one
another. Although environmental influences account for a substantial portion of
the variance in trait N, studies suggest that this may be entirely accounted for by
nonshared environment (Fullerton, 2006; Lake et al., 2000). There are a number
of caveats to such findings, however. Shared environmental influences are more
difficult to detect, even when present, in part because they may operate through
gene–environment interactions (Johnson, 2007; Robinson, 2004). In other words,
shared environmental influences may be moderated by genes, which effectively
increases the heritability of the construct rather than reflecting unique shared
environmental variance.
Very little work has been done on specific types of measured environmental
factors that are likely to impact N. Adult N levels have been associated with
retrospective recall of a variety of parenting and familial factors (e.g., Allen &
Lauterbach, 2007; McCrae & Costa, 1988; Reti et al., 2002; Roy, 2002;
Willemsen & Boomsma, 2007), although prospective designs, ideally that
account for genetic factors as well, are needed to understand potential causal
influences of such factors.
Developmental Considerations
Absolute and differential stability. Early N, as with other major personality
traits, is relatively stable across development (Caspi, 2000; Caspi et al., 2003;
Caspi & Silva, 1995; Nave, Sherman, Funder, Hampson, & Goldberg, 2010).
This stability is largely accounted for by both genetic and nonshared
environmental factors (De Fruyt et al., 2006), and is robust across community
and clinical samples (De Bolle et al., 2009). Absolute, or mean-level, stability
reflects the extent to which average trait levels are stable across time in the
broader population. Meta-analytic work suggests that mean levels of N begin
shifting in adulthood, whereby overall levels in the population decrease as
individuals progress from early to later adulthood (Roberts, Walton, &
Viechtbauer, 2006). Although adult shifts apparently reflect greater maturation
across development, there have been some suggestions that early developmental
changes may reflect a transition from more to less “maturity” during the
developmental transition from childhood to adolescence (Soto, John, Gosling, &
Potter, 2011). Much more work is needed to better understand how trait N levels
might change across different developmental periods.
Another primary type of stability that is typically investigated in research on
personality development is differential, or rank-order, stability, which refers to
the extent to which individuals maintains their overall ranking on a given trait
over time. Rank-order stability is moderate even in childhood, but shows slight
increases across development, with lowest levels of stability manifest for trait N
(Roberts & DelVecchio, 2000). This is true for both youth (De Fruyt et al., 2006;
Tackett et al., 2008) and adults (Hampson & Goldberg, 2006). In measuring any
type of stability, it will be important for researchers to better account for
measurement difficulties for trait N, particularly early in life, and the extent to
which such measurement problems are likely to influence estimates of stability
and change.
Another aspect of personality development that may be especially relevant for
N is personality variability, or the extent to which an individual’s level on a trait
shows variance across situations and across time. This construct is related to
emotional lability or reactivity, which is often considered a facet of N,
particularly in personality disorder trait models (Widiger, 2009). One study
found substantial intraindividual variability in N trajectories from mid to later
adulthood, for example (Mroczek & Spiro, 2003), suggesting that it will be
important to examine within-person fluctuations in N, as well as between-person
and across-person continuity.
Conceptualization of N at younger ages. N is one of the most robust higher-
order personality traits emerging across myriad personality frameworks, and this
holds in childhood and adolescence, as well (see also Chapter 12 by De Pauw as
well as Chapter 22 by De Fruyt, De Clercq, and De Bolle). As noted previously,
developmentally sensitive approaches to conceptualizing individual differences
(e.g., temperament and child personality) have often differed in their
conceptualization of Neuroticism/Negative Emotionality when compared to
adult models, and also when compared to one another (Lahey et al., 2008;
Lahey, Rathouz, Applegate, Tackett, & Waldman, 2010). Developmentally
sensitive approaches to characterizing N often place antagonism/aggression in a
much more prominent role in trait definitions (De Pauw & Mervielde, 2010;
Rothbart & Bates, 2006; Tackett et al., 2012), with potentially more difficulty in
clearly defining more fearful/sad aspects of N in youth, as compared to adults.
This is consistent with work on personality perception, which suggests that more
observable traits will be perceived more accurately than less observable traits
when relying on informants (Tackett, 2011; Vazire, 2010), which is most often
the case when assessing personality in early life. This is also consistent with
literature suggesting that interinformant agreement is higher for externalizing
psychopathology (which is marked by antagonistic features) than for
internalizing psychopathology (which is marked by sadness and fearfulness) in
youth (Achenbach, 2006; Achenbach et al., 1987).
The distinction between fear and irritability aspects of N has been noted by
developmental psychologists as especially relevant for early personality
(Rothbart & Bates, 2006; Shiner, 2010; Shiner & Caspi, 2003). It is clear that
lower-order personality traits are poorly understood in early life, with only
sparse agreement across different frameworks (de Pauw, Mervielde, & van
Leeuwen, 2009; Shiner & Caspi, 2003; Shiner & Tackett, 2014). Notably, these
“mid-level” traits (fear and irritability) are largely represented across different
youth models and are also analogous to the primary mid-level traits identified for
N in adult samples (De Young et al., 2007). It is also possible to link the sizeable
early literature on “difficult temperament” with early manifestations of N.
Specifically, research on “difficult” or “hard to manage” temperament constructs
typically incorporates aspects of negative affectivity alongside physiological
features (e.g., disrupted eating and sleeping patterns) and aspects of disinhibition
(e.g., Bates, Pettit, Dodge, & Ridge, 1998; Davies & Windle, 2001; Olson,
Bates, Sandy, & Schilling, 2002).
Measurement considerations. Existing evidence seems to strongly suggest
that measurement of N in young children is among the most challenging of all
early personality traits (Durbin, 2010; Tackett et al., 2008, 2012). Specifically,
interinformant agreement may be lower for N than for other early personality
traits, particularly those aspects reflecting sadness and internalized negative
emotions (Durbin, 2010; Tackett, 2011). When working with adults, it has been
argued that it is especially important to utilize self-report for such constructs
(Vazire, 2010), but this is of course challenging when working with younger
populations.
There is some indication that examining informant disagreement regarding
levels of N in youth may provide incremental information about consequential
outcomes, such as psychopathology. For example, mother’s and father’s level of
disagreement on their child’s N, after accounting for the child’s overall levels of
N, incrementally predicted their child’s level of internalizing psychopathology
(Tackett, 2011). Such findings indicate that variability in trait N may be reflected
in informant differences, allowing a powerful method for capturing incremental
information in trait N, even when it is more difficult to measure as it is in
younger age groups.
Laboratory-based tasks are often used to elicit and measure temperament traits
in children, but such measures also tend to show low correlations with parent-
reported N (Durbin, 2010). The potential advantages of laboratory tasks,
however, include enhanced sensitivity to context and greater potential to capture
characteristics of stability and change, all of which may be highly relevant to a
thorough understanding of N (Durbin, 2010). Taken together, the challenges
with measuring N, particularly in early life, call for multimethod, multiinformant
approaches, as well as more sophisticated efforts to leverage potential
information inherent in informant/method discrepancies.
Neuroticism and Psychopathology
Basic Associations
N demonstrates consistent and robust associations with mental disorders
across the lifespan including various forms of internalizing psychopathology,
externalizing psychopathology, and personality disorders (Clark & Watson,
1991; Klein, Kotov, & Bufferd, 2011; Tackett, 2006; Widiger & Smith, 2008).
Meta-analytic evidence supports “very large” effect sizes (Cohen’s d > 1.0;
Cohen, 1992) across a range of Axis I disorders (Kotov, Gamez, Schmidt, &
Watson, 2010; Malouff et al., 2005) and “small” to “medium” effect sizes for
Axis II disorders (Cohen’s d = 0.02 for Narcissistic to d = 0.55 for Borderline;
Saulsman & Page, 2004). This is even the case in children and adolescents using
measures of N that do not contain items referring to synonyms and antonyms of
symptoms of psychopathology (Lahey et al., 2008, 2010). Thus, it is clear that N
shows a moderate to strong association with all major forms of psychopathology.
In addition to associations between N and single mental disorders, variance in
N is also associated with greater comorbidity among different disorders (Barlow
et al., 2014; Khan et al., 2005; Lahey, 2009). Individuals with comorbid
diagnoses are at higher risk for a range of consequences, including greater
disorder persistence and severity and higher consumption of high-cost mental
health services (Kessler, Chiu, Demler, Merikangas, & Walters, 2005). Thus,
understanding mechanisms underlying comorbidity is critical. Toward that end,
recent evidence has suggested that a bifactor modeling approach, which
positions a “general factor” accounting for covariance among all specific
disorders, may be a useful way of reconceptualizing the now-common
internalizing–externalizing disorder framework (such that remaining
internalizing and externalizing covariance is captured in specific factors; Caspi
et al., 2014; Lahey et al., 2012). This general factor of psychopathology appears
to largely overlap with trait N, at both phenotypic and etiologic levels (Tackett et
al., 2013), which implicates N as a particularly crucial transdiagnostic
mechanism underlying disparate forms of psychopathology (Barlow, Sauer-
Zavala, Carl, Bullis, & Ellard, 2014).
Causal Mechanisms
There are a variety of potential explanations for personality trait associations
with psychopathology, which have been reviewed elsewhere (Klein et al., 2011;
Lahey, 2009; Littlefield & Sher, 2010; Tackett, 2006; Widiger & Smith, 2008).
Importantly, various explanations for personality–psychopathology associations
largely tend not to be mutually exclusive. Especially in the case of trait N, it is
likely that multiple mechanisms explain links between N and various disorders
in different individuals (Ormel, Jeronimus, et al., 2013). Vulnerability
explanations posit personality traits as independent constructs that increase the
risk for later psychopathology (thus showing prospective associations). Although
sometimes distinguished from one another, we position spectrum and common
cause explanations as largely overlapping concepts, which suggests that
personality traits and mental disorders lay on related spectra, and such spectra
share core etiologic features that transcend personality/psychopathology
boundaries. The spectrum model is largely consistent with a recent shift toward
focusing on transdiagnostic mechanisms in psychopathology research, such as
the Research Domain Criteria (RDoC) project (Insel et al., 2010).
A number of studies have also examined evidence for the predictive utility of
trait N in prospective studies of psychopathology, which is consistent with a
vulnerability model perspective. The most impressive evidence has amassed for
N prospective prediction of later major depression onset (Fanous, Neale, Aggen,
& Kendler, 2007; Kendler, Neale, Kessler, Heath, & Eaves, 1993). Such
research has used impressively large samples, with longitudinal prediction
extending to 25 years later (Kendler, Gatz, Gardner, & Pedersen, 2006). Other
research has similarly demonstrated predictive utility of N for later
schizophrenia (van Os & Jones, 2001) and suicide (Fergusson, Woodward, &
Horwood, 2000).
Evidence supporting a spectrum conceptualization comes from studies
demonstrating shared etiologic factors underlying trait N and various forms of
psychopathology (e.g., Carey & DiLalla, 1994; Fanous, Gardner, Prescott,
Cancro, & Kendler, 2002; Hettema, Neale, Myers, Prescott, & Kendler, 2006;
Mikolajewski, Allan, Hart, Lonigan, & Taylor, 2013; Silberg, Rutter, Neale, &
Eaves, 2001; Stein & Stein, 2008; Tackett et al., 2012, 2013). Research has also
suggested that N may mediate associations between the 5-HTTLPR
polymorphism and depression (Jacobs et al., 2006; Munafò, Clark, Roberts, &
Johnstone, 2006). Similarly, work has identified amygdala overactivation as
common to many different forms of internalizing psychopathology, as well as to
high levels of N (Barlow et al., 2014; Ormel et al., 2013). Thus, common genetic
and neurobiological mechanisms may account for associations between N and
multiple forms of psychopathology. Although cross-sectional correlations
between phenotypic constructs (i.e., measures of N and psychopathology) have
been occasionally used to promote a spectrum framework, such evidence does
not provide a particularly stringent test of the spectrum model and could reflect
other explanatory mechanisms.
Although existing evidence supports both vulnerability/risk and
spectrum/common cause explanations of N–psychopathology associations, it is
often difficult to disentangle these explanations from one another. In particular,
studies examining prospective associations often cannot rule out the possibility
that common causes account for longitudinal prediction, unless such common
causes are simultaneously measured and controlled for. Longitudinal
relationships between early N and later psychopathology may in fact reflect
heterotypic continuity, whereby similar underlying factors (e.g., genetic
influences) are linked to phenotypically distinct constructs at different points in
time. However, other evidence suggests that N does show bidirectional relations
with mental disorder over time, suggesting that N also shows dynamic interplay
with features of psychopathology across the life course.
Work that is especially relevant to this idea demonstrates that underlying
genetic factors may interact with environmental stress to increase the risk for
later disorders. For example, the risk variant of the 5-HTTLPR polymorphism
already discussed as relevant to both N and psychopathology may increase
depression risk primarily in individuals who have also encountered stressful life
events (Caspi et al., 2003; Lotrich & Pollock, 2004; Uher & McGuffin, 2010),
with some evidence for mixed findings (see Rutter, 2008). Importantly, some
evidence also supports such an interaction between 5-HTTLPR status and
stressful life events in the prediction of N (Caspi & Moffitt, 2006), and other
molecular genetic candidates may show similar patterns (e.g., the G72 gene;
Rietschel et al., 2008; Shi, Badner, Gershon, & Liu, 2008). Other research
suggests that 5-HTTLPR status interacts with earlier depressive symptoms to
predict stress generation (Starr, Hammen, Brennan, & Najman, 2012),
implicating possible mechanistic pathways strengthening such associations. Of
course, environmental factors may also buffer individuals at higher
psychobiological risk against the development of psychopathology. For example,
one recent study found that social support moderated the association between
threat-related amygdala reactivity and anxiety (Hyde, Gorka, Manuck, & Hariri,
2011).
There has been extensive work looking at longitudinal, bidirectional
associations between N, stressful life events, and other environmental variables
such as relational conflict, physical health, and occupational status. This
literature has offered support for the directional hypothesis that individuals with
higher N have an increased likelihood of experiencing later negative outcomes
across these domains (Bolger & Zuckerman, 1995; Gleason, Powers, &
Oltmanns, 2012; Gunthert, Cohen, & Armeli, 1999; Hankin, Fraley, & Abela,
2005; Kendler, Gardner, & Prescott, 2002; Magnus, Diener, Fujita, & Pavot,
1993; Suls & Martin, 2005; van Os, Park, & Jones, 2001). The implications of
this are critical, given the extent to which exposure to life events increases the
subsequent risk for psychopathology (Ehring, Ehlers, & Glucksman, 2006;
Kendler et al., 2004; Parslow, Jorm, & Christensen, 2006). One pronounced
example of this effect is evidence suggesting that premarital levels of N predict
subsequent marital dissolution (Donnellan, Conger, & Bryant, 2004; Karney &
Bradbury, 1997; Kelly & Conley, 1987; Roberts et al., 2007; Rogge, Bradbury,
Hahlweg, Engl, & Thurmaier, 2006; Tucker, Kressin, Spiro, & Ruscio, 1998),
alongside evidence that divorce predicts a number of mental and physical health
problems (Hemström, 1996; Ikeda et al., 2007; Lee & Gramotnev, 2007; Lee et
al., 2005; Overbeek et al., 2006; Perreira & Sloan, 2001). There is additional
evidence that high N predicts lower social support (Kendler, Gardner, &
Prescott, 2002, 2006), and that social support partially mediates associations
between N and depression (Finch & Graziano, 2001; Kendler et al., 2002;
Kendler, Gardner, & Prescott, 2006). Such evidence leads to the suggestion that
high N may indirectly promote adverse health outcomes by increasing the
likelihood of divorce and/or decreasing social support networks (Lahey, 2009).
The possibility of these causal pathways linking N, relational functioning, and
psychopathology outcomes can be further understood by examining research on
the associations between N and emotional responding. Research suggests that
individuals higher on N demonstrate more negative affect in response to both
controlled and naturalistic stressors (Larsen & Ketelaar, 1991; Zautra, Affleck,
Tennen, Reich, & Davis, 2005). In addition, individuals high on N may be more
likely to find the experience of negative affect distressing (Barlow et al., 2014)
and more likely to show impaired sympathetic regulation (as indexed by heart
rate variability) in the face of negative emotional challenge (Di Simplicio et al.,
2012). This is consistent with evidence that individuals with higher levels of N
are at increased risk for internalizing psychopathology following exposure to
stressful life events than individuals with lower levels of N who are exposed to
the same events (Fanous et al., 2002; Hutchinson & Williams, 2007; Jacobs et
al., 2006; Kendler et al., 2004; Parslow et al., 2006). This may be, in part, due to
differences in coping, such that individuals with higher levels of N are more
likely to use inefficient forms of coping, such as avoidance (Bolger, 1990), and
to show deficits in disengaging attention (Bredemeier, Berenbaum, Most, &
Simons, 2011). It may also partially reflect a more generalized cognitive
response in individuals high in N to both ambiguous and negative stimuli. For
example, individuals higher in N show an exaggerated neurobiological response
to uncertain stimuli, even in comparison to their heightened response to negative
stimuli, relative to individuals lower in N (Hirsh & Inzlicht, 2008). Other
relevant evidence suggests that individuals who report more negative appraisals
of stressful life events show an increased risk for internalizing disorders (Espejo,
Hammen, & Brennan, 2012). Most work in this domain has investigated
internalizing psychopathology, but shared genetic factors between N and alcohol
use disorders are, similarly, largely accounted for by an individual’s
endorsement of coping motives (Littlefield et al., 2011). More work is needed to
understand the extent to which the vast literature on N-internalizing disorder
associations might generalize to N-externalizing disorder associations.
A recent study directly compared competing temporal ordering of N, an
intermediate cognitive response (anxious arousal), and stressful life events in
predicting later depressive symptoms (Barrocas & Hankin, 2011). Specifically,
the authors tested the hypothesis that early levels of higher N increased
perceived exposure to stressful life events, which then increased anxious arousal,
culminating in higher levels of depressive symptoms, versus an alternative
pathway whereby early levels of higher N increased anxious arousal, which
increased perceived exposure to stressful live events, ultimately increasing
depressive symptoms in a sample of 350 sixth-grade to tenth-grade students. The
authors found support for the initial model, linking higher N → perceived life
stressors → anxious arousal → depressive symptoms, which was not moderated
by gender or age. This type of integrative, longitudinal research design will be
particularly helpful in delineating pathways between N, cognitive processes,
stressful life events, and psychopathology outcomes.
Neuroticism and Physical Health

Basic Associations
In addition to links with mental health, trait N is also associated with both
subjective and objective indicators of physical health (Lahey, 2009). Individuals
higher in N report more somatic complaints without medical support
(Chaturvedi, 1986; Costa & McCrae, 1987; Magee, Heaven, & Miller, 2013;
Powers & Oltmanns, 2013) and demonstrate distorted cognition regarding
symptoms, which results in greater use of medical services (Goubert, Crombez,
& Van Damme, 2004). Such evidence suggests that higher levels of N result in
stronger perceptions of physical health problems, but research has also
demonstrated connections that go beyond subjective concerns. N is associated
with many different types of physical health problems (Brickman, Yount,
Blaney, Rothberg, & De-Nour, 1996; Drossman et al., 2000; Smith &
MacKenzie, 2006; Suls & Bunde, 2005).
Some associations are suggested by indirect evidence, such that many
physical health problems are more common in individuals with internalizing
psychopathology (which, itself, is highly correlated with high N; Currie &
Wang, 2005; Robles, Glaser, & Kiecolt-Glaser, 2005; Sareen, Cox, Clara, &
Asmundson, 2005; Watkins et al., 2006). Such concerns include problems with
cardiac functioning (Barger & Sydeman, 2005), increased mortality (Penninx et
al., 2001; Robles et al., 2005; Simonsick, Wallace, Blazer, & Gerkman, 1995),
and disrupted immune functioning (Maier & Watkins, 1998; Pace et al., 2006;
Robles et al., 2005).
Indirect evidence is complemented by a growing body of work demonstrating
direct associations between N and objective measures of physical health (Smith
& MacKenzie, 2006). Direct associations have been demonstrated between N
and asthma (Huovinen, Kaprio, & Koskenvuo, 2001), atopic eczema (Buske-
Kirschbaum, Geiben, & Hellhammer, 2001), cardiovascular disease (Suls &
Bunde, 2005), and irritable bowel syndrome (Spiller, 2007). Furthermore, such
associations hold even when depression and social support are controlled for
(Bouhuys et al., 2004; Russo et al., 1997). This research provides an especially
stringent test of links between N and physical health, demonstrating incremental
contributions beyond internalizing psychopathology and associations that are not
purely reflective of potential cognitive distortions associated with higher levels
of N.
Causal Mechanisms
Evidence has also emerged regarding the predictive validity of trait N for
physical health problems. Prospective longitudinal studies have demonstrated
links between earlier trait N and longevity in the general population (Smith &
MacKenzie, 2006). One representative study of British adults found a 10%
higher mortality rate for each +1 standard deviation difference in trait N, even
after controlling for multiple covariates (e.g., age, sex, socioeconomic status,
substance use, physical activity, and health status; Shipley, Weiss, Der, Taylor,
& Deary, 2007). Another population-based study demonstrated that individuals
with higher initial levels of N accounted for 33% more deaths (from all causes)
than individuals with lower levels of N, again controlling for a number of key
covariates (Wilson et al., 2005). A longitudinal study of elderly clergy found
evidence suggesting that individuals with high levels of N manifested a nearly
doubled risk of death relative to individuals with lower levels of N (Wilson,
Mendes de Leon, Bienias, Evans, & Bennett, 2004). However, one longitudinal
study of individuals with compromised health did not show incremental
predictive utility of trait N (Weiss & Costa, 2005).
N also shows predictive utility for longevity in populations of individuals
diagnosed with chronic diseases and cancer. For example, higher levels of N
predicted renal deterioration in individuals diagnosed with Type I diabetes
(Brickman et al., 1996) and cardiac disease (Murberg, 2004). Individuals
diagnosed with chronic renal insufficiency who also had high levels of trait N
manifested a 38% higher mortality rate than individuals with lower levels of N,
even when controlling for multiple covariates (Christensen et al., 2002).
Similarly, among individuals treated for cancer, higher levels of N were
associated with a 130% greater death rate than those lower in N (Nakaya et al.,
2006).
Underlying causal mechanisms explaining associations between N and
physical health may be similar to those underlying associations between N and
mental health. For example, it is possible that shared genetic factors account for
covariation between higher levels of N and greater physical health problems.
One twin study found that genetic influences accounted for a substantial portion
of the covariation between N and self-reported somatic complaints (Vassend,
Røysamb, & Nielsen, 2012). In addition, it is likely that the influence of high N
on variables such as stress exposure and social support also increases the risk for
physical health problems, as it does for mental health problems (Contrada,
Cather, & O’Leary, 1999; Holt-Lunstad, Smith, & Uchino, 2008; Smith &
MacKenzie, 2006). Research does suggest that social support is inversely
associated with greater physical health problems (Uchino, 2006). Similarly,
stress is linked to higher levels of physical health problems, as well
(Schneiderman, Ironson, & Siegel, 2005).
There may also be unique causal mechanisms underlying N and physical
health problems that do not apply to mental health problems. For example,
higher levels of N may have an impact on individuals’ physiology, such as their
stress response and aspects affiliated with immunity, such as inflammation
(Futterman, Kemeny, Shapiro, & Fahey, 1994; Gillespie et al., 2004), which in
turn impacts their physical health outcomes (Contrada et al., 1999; Friedman,
2000; Smith & MacKenzie, 2006). A growing literature implicates N as a
moderator in individual’s physiological response to stressors. For example, some
research shows that those higher in N may have sympathetic stress responses
that are greater in magnitude and longer in duration (Norris, Larsen, &
Cacioppo, 2007; Riese et al., 2007; Vogeltanz & Hecker, 1999), higher levels of
morning cortisol (Portella, Harmer, Flint, Cowen, & Goodwin, 2005), higher
levels of daily cortisol output (Nater, Hoppmann, & Klumb, 2010), higher levels
of cardiovascular reactivity (Muth, Koch, & Stern, 2000), and lower levels of
blood pressure recovery following interpersonal stress (Hutchinson & Ruiz,
2011). Furthermore, research has found blunted cortisol reactivity to stress
exposure for individuals higher in N (Oswald et al., 2006; Phillips, Carroll,
Burns, & Drayson, 2005), which may reflect down-regulation of the
hypothalamic–pituitary–adrenocortical axis following prolonged cortisol
elevation (McCleery & Goodwin, 2001; Zobel et al., 2004). However, discrepant
findings have also emerged in this literature, either in finding no evidence of
associations (Hennig et al., 1996; Schommer, Kudielka, Hellhammer, &
Kirschbaum, 1999), or associations opposite to the predicted directions
(LeBlanc, Ducharme, & Thompson, 2004; see Ormel et al., 2013, for another
review of these findings). Methodological limitations prevent clear conclusions
at this point, although further work into moderators such as gender may prove
fruitful (Hennig et al., 1996; Ormel et al., 2013; Oswald et al., 2006).
Research has linked N to disrupted circadian rhythms (Murray, Allen, Trinder,
& Burgess, 2002) and immune system abnormalities (Bouhuys et al., 2004).
Specific studies have identified associations between higher N and increased
leukocyte counts (Daruna, 1996), diminished antibody response to vaccination
(Phillips et al., 2005), disrupted response of natural killer cells to stress (Borella
et al., 1999), disrupted secretory immunoglobulin response following a stressor
(Hennig, Pössel, & Netter, 1996), and telomere attrition, which is a marker of
cellular aging (van Ockenburg, de Jonge, van der Harst, Ormel, & Rosmalen,
2014). Such indicators have direct relevance for immune system functioning and
physical health outcomes, including mortality. One study of infant rhesus
monkeys found that higher N (nervous temperament) monkeys showed a
disrupted pattern of association between increased cortisol levels and
neutrophils, a leukocyte subset, compared to monkeys lower in N (Capitanio,
Mendoza, & Cole, 2011). Such findings may indicate that individuals higher in
N evidence desensitization of immune cells via increased glucocorticoid levels.
The extent to which N moderates physiological reactivity and physical health
may partly be reflective of shared genetic influences on these constructs. For
example, individuals with at least one short allele of 5-HTLPR manifest higher
resting cortisol levels (Jabbi et al., 2007) and a higher incidence of irritable
bowel syndrome (Yeo et al., 2004). Thus, it remains important for future work to
better understand the mechanisms underlying associations between N and
physical health.
A second unique causal pathway to consider is the extent to which individuals
high in N show greater involvement in health-risk behaviors, which in turn leads
to poorer physical health outcomes (Contrada et al., 1999; Smith & MacKenzie,
2006). Individuals higher on N are more likely to smoke nicotine (Breslau,
Novak, & Kessler, 2004; Malouff et al., 2006; Morissette, Tull, Gulliver,
Kamholz, & Zimering, 2007; Terracciano & Costa, 2004) and abuse other
substances such as alcohol and illicit drugs (Larkins & Sher, 2006). Such
behaviors carry with them a very high risk for a number of important physical
health outcomes including cancer and cardiovascular disease, and further impact
mortality rates. Other health-risk behaviors associated with high N include
unprotected or risky sex (Cooper, Agocha, & Sheldon, 2000; Hoyle, Fejfar, &
Miller, 2000; Trobst et al., 2000). Such behaviors carry with them an increased
risk for physical health outcomes such as HIV/AIDS, cervical cancer, and
sexually transmitted disease. Health-risk behaviors may be related to trait N via
coping or avoidance-motivated pathways (Cooper et al., 2000), which are more
likely to be used by individuals with higher levels of N.
Neuroticism and Quality of Life

In addition to associations with maladaptive outcomes, trait N has been
robustly linked to positive outcomes as well (Lahey, 2009). Perhaps the best
studied link is between trait N and quality of life, or subjective well-being
(Boyce, Wood, & Powdthavee, 2013; Ozer & Benet-Martínez, 2006; Steel,
Schmidt, & Shultz, 2008). Low levels of N, also referred to as trait “Emotional
Stability,” are associated with higher marital satisfaction (Gattis, Berns,
Simpson, & Christensen, 2004), greater occupational success (Ozer & Benet-
Martínez, 2006; Roberts, Kuncel, Shiner, Caspi, & Goldberg, 2007), and quality
of life (Arrindell, Heesink, & Feij, 1999; Lynn & Steel, 2006; Ozer & Benet-
Martínez, 2006). Higher levels of N are also linked to social impairment, beyond
what is accounted for by other traits such as extraversion and agreeableness
(Mullins-Sweatt & Widiger, 2010). Low levels of N have been prospectively
linked to lower levels of burnout and emotional exhaustion (Armon, Shirom, &
Melamed, 2012). Levels of N outperform socioeconomic status in predicting
overall well-being and emotional health, and states with lower proportions of
high N individuals show higher levels of well-being outcomes (McCann, 2011).
Similarly, change in personality (particularly N, but also extraversion) shows
greater prediction of changes in life satisfaction than demographic and economic
variables (e.g., household income, employment status) combined (Boyce et al.,
2013). There is also evidence that overlapping genetic influences at least partly
account for associations between N and subjective well-being (Weiss, Bates, and
Luciano, 2008).
Conclusions and Future Directions

The potential utility of trait N in future prevention/intervention efforts relies
on an increased understanding of how and why N is related to the various
outcomes reviewed here. An understanding of the specific underlying
mechanisms resulting in links between N and depression, anxiety, substance use
disorders, diabetes, cancer, cardiovascular disease, and mortality will result in
enhanced knowledge about each condition itself, as well as more nuanced insight
into the potential role trait N might play in prevention and intervention efforts
for each condition (Klein et al., 2011; Lahey, 2009). Early screening efforts
could include measures of trait N on a large-scale basis with relative ease, given
the resource efficiency of administering a trait measure of N (Lahey, 2009). Not
only would such efforts help identify those at potentially high risk for a range of
adverse outcomes, information about levels of N could be incorporated into
more personalized and, ideally, more effective treatments. Such large-scale
efforts would need to be carefully considered regarding the actual incremental
value provided by N above other information already available (e.g., existing
health status) and the potential for false positives and stigmatization (Lahey,
2009).
Personality assessments could also more routinely be administered in clinical
psychological settings prior to treatment planning and commencement, in order
to better understand how existing personality traits such as N may play a role in
an individual’s mental health and potential intervention course (Barlow et al.,
2014; Widiger & Trull, 2007). It might also be possible for large-scale
preventive interventions to focus on directly modifying levels of trait N (Lahey,
2009). Although purely speculative, adaptations of psychotherapeutic and
psychopharmacological interventions may ultimately prove useful for those high
on N in increasing their chances for adaptive outcomes. Of course, such efforts
would require careful consideration of possible iatrogenic effects. Trait N, as
with all broad personality traits, is adaptive in certain contexts and at certain
levels. Those with particularly low trait N may not be served well by attempts to
lower their levels of N even further, and may then fail to benefit from adaptive
features of N when called for.
Current psychological treatments, which have been largely driven by the
fragmented categorical approach embodied in the DSM (American Psychiatric
Association, 2013), may have become overly specialized, focusing on disorder-
specific symptoms and manifestations (Barlow et al., 2014). Thus, interventions
targeting broader measures of underlying N may require a new approach and
fresh way of thinking about intervention mechanisms of change. It is likely that
some aspects of existing interventions can be leveraged toward a dimensional
personality framework, but other aspects may be less relevant and new aspects
may emerge as meaningful. For example, even individuals who benefit from
psychotherapeutic treatment for internalizing disorders show high temporal
stability for N trait levels (Eaton et al., 2011). Contradictory evidence has also
emerged, however, such that trait levels of N showed the highest level of
temporal change following treatment in another sample (Brown, 2007; Brown &
Barlow, 2009). Thus, in addition to more mechanistic investigations of potential
interventions, clear differentiation between state and trait levels of N will be
needed for such work, moving forward.
Research directly targeting temperamental components is beginning to emerge
in the literature. Indirect support for this idea comes from evidence in
psychopharmacological research that has indicated that medications targeting
serotonin appear to reduce levels of N (e.g., Quilty, Meusel, & Bagby, 2008). In
addition, growing evidence suggests that personality traits, including N, show
response to treatment (Klein et al., 2011). Barlow and colleagues (2014) have
devoted considerable effort to developing a comprehensive treatment protocol
specifically aimed at transdiagnostic temperamental traits underlying
internalizing psychopathology. This cognitive–behavioral intervention was
designed to target negative emotional experiences and the individual’s responses
to such experiences, and early evidence on efficacy has been promising.
Individuals with comorbid mental and physical health problems show more
severe variants of health concerns, increased demand for health services, and
more severe prognosis (Baune, Adrian, & Jacobi, 2007; Druss et al., 2009;
McCaffery et al., 2006). Furthermore, research suggests that interventions
targeting one internalizing disorder frequently yield improvements in other
comorbid internalizing disorders that were not actually targeted in treatment
(Barlow et al., 2014). This overlapping treatment response implicates the
potential role of shared underlying features such as high N, and suggests that
interventions directly targeting high N may yield cascading effects on any
number of related mental and physical conditions. Research reviewed here
positions N as a particularly likely candidate for understanding the mechanisms
underlying mental and physical health problem cooccurrence, and thus
underscores the potential public health significance of trait N across these
conditions.
Summary
In sum, N is a higher-order personality trait that encompasses a tendency to
experience negative affect and emotions, including feelings of sadness, anxiety,
and anger. N is one of the most robust of higher-order personality traits,
emerging across different personality trait frameworks, as well as different
populations (e.g., based on age or culture). Although many conceptualizations of
N include antagonistic negative affect as well, this represents one of the most
conflicting aspects of N conceptualization across different trait models and in
different populations. N is associated with myriad adverse health outcomes,
including different types of psychopathology and physical health concerns. Low
N is also associated with positive outcomes such as quality of life.
As has been emphasized throughout this chapter, the personality trait of N has
enormous consequences for public health. The myriad mental and physical
health outcomes summarized here underscore its potential impact on individual
adaptation. Furthermore, economic costs associated with N are substantial,
exceeding even those for common mental disorders (Cuijpers et al., 2010). The
extent to which trait N will ultimately hold public health significance will
depend on the extent to which the literature summarized here translates into
more effective prevention and intervention efforts (Lahey, 2009).
In terms of future directions, one important next step for research in this area
is to move toward more sophisticated tests of causal mechanisms underlying
associations between N and mental and physical health outcomes. Evidence for
these associations is robust and numerous causal pathways have been articulated,
but are rarely tested in explicit and powerful research designs. In particular,
integrated models should be developed that directly account for physical and
mental health outcomes alongside one another to more fully explicate the
common associations among such outcomes and N (Lahey, 2009). For example,
one recent study examined the risk for depression and anxiety across multiple
adult cohorts with multiple predictors (e.g., trait N, physical health, cognitive
functioning, level of disability, body mass index, and socioeconomic status; Gale
et al., 2011). This examination demonstrated incremental unique associations
between N, various aspects of physical health, and depression/anxiety. Although
this study was not well situated to address causal mechanisms, it exemplifies the
type of integrative approach needed in order to better understand how N
associations with psychological and physical health manifest within individuals.
It is also anticipated that research will continue moving toward better evidence
for common underlying genetic and neurobiological factors that may partially
account for associations between N and these outcomes, as well as the extent to
which such factors are moderated by contextual features in influencing the
likelihood for adverse outcomes. Although such work is growing, research
explicitly integrating multiple levels of analysis will be needed to fully chart
causal pathways to pathological outcomes (Kendler, 2014).
References
Achenbach, T. M. (2006). As others see us: Clinical and research implications of cross-
informant correlations for psychopathology. Current Directions in Psychology Science,
15, 94–98.
Achenbach, T. M., McConaughy, S. H., & Howell, C. T. (1987). Child/adolescent
behavioral and emotional problems: Implications of cross-informant correlations for
situational specificity. Psychological Bulletin, 101, 213–232.
Allen, B., & Lauterbach, D. (2007). Personality characteristics of adult survivors of
childhood trauma. Journal of Traumatic Stress, 20, 587–595.
disorders (5th ed.). Arlington, VA: American Psychiatric Association.
Armon, G., Shirom, A., & Melamed, S. (2012). The big five personality factors as
predictors of changes across time in burnout and its facets. Journal of Personality, 80,
403–427.
Arrindell, W. A., Heesink, J., & Feij, J. A. (1999). The satisfaction with life scale
(SWLS): Appraisal with 1700 healthy young adults in the Netherlands. Personality and
Barger, S. D., & Sydeman, S. J. (2005). Does generalized anxiety disorder predict
coronary heart disease risk factors independently of major depressive disorder? Journal
of Affective Disorders, 88, 87–91.
Barlow, D. H., Sauer-Zavala, S., Carl, J. R., Bullis, J. R., & Ellard, K. K. (2014). The
nature, diagnosis, and treatment of neuroticism: Back to the future. Clinical
Barrocas, A. L., & Hankin, B. L. (2011). Developmental pathways to depressive
symptoms in adolescence: A multi-wave prospective study of negative emotionality,
stressors, and anxiety. Journal of Abnormal Child Psychology, 39, 489–500.
Bates, J. E., Pettit, G. S., Dodge, K. A., & Ridge, B. (1998). Interaction of temperamental
resistance to control and restrictive parenting in the development of externalizing
behavior. Developmental Psychology, 13, 183–214.
Baune, B. T., Adrian, I., & Jacobi, F. (2007). Medical disorders affect health outcome and
general functioning depending on comorbid major depression in the general population.
Journal of Psychosomatic Research, 62, 109–118.
Bolger, N. (1990). Coping as a personality process: A prospective study. Journal of
Bolger, N., & Zuckerman, A. (1995). A framework for studying personality in the stress
process. Journal of Personality and Social Psychology, 69, 890–902.
Borella, P., Bargellini, A., Rovesti, S., Pinelli, M., Vivoli, R., Vivoli, G., & Solfrini, V.
(1999). Emotional stability, anxiety, and natural killer activity under examination
stress. Psychoneuroendocrinology, 24, 613–627.
Bouhuys, A. L., Flentge, F., Oldehinkel, A. J., & van den Berg, M. D. (2004). Potential
psychosocial mechanisms linking depression to immune function in elderly subjects.
Psychiatry Research, 127, 237–245.
Boyce, C. J., Wood, A. M., & Powdthavee, N. (2013). Is personality fixed? Personality
changes as much as “Variable” economic factors more strongly predict changes to life
satisfaction. Social Indicators Research, 111, 287–305.
Bredemeier, K., Berenbaum, H., Most, S. B., & Simons, D. J. (2011). Links between
neuroticism, emotional distress, and disengaging attention: Evidence from a single-
target RSVP task. Oxford, England: Psychology Press.
Breslau, N., Novak, S. P., & Kessler, R. C. (2004). Psychiatric disorders and stages of
smoking. Biological Psychiatry, 55, 69–76.
Brickman, A. L., Yount, S. E., Blaney, N. T., Rothberg, S. T., & De-Nour, A. K. (1996).
Personality traits and long-term health. Psychosomatics, 37, 459–468.
Brown, T. A. (2007). Temporal course and structural relationships among dimensions of
temperament and DSM-IV anxiety and mood disorder constructs. Journal of Abnormal
Brown, T. A., & Barlow, D. H. (2009). A proposal for a dimensional classification system
based on the shared features of the DSM-IV anxiety and mood disorders: Implications
for assessment and treatment. Psychological Assessment, 21, 256–271.
Brown, S. M., & Hariri, A. R. (2006). Neuroimaging studies of serotonin gene
polymorphisms: Exploring the interplay of genes, brain, and behavior. Cognitive,
Affective and Behavioral Neuroscience, 6, 44–52.
Buske-Kirschbaum, A., Geiben, A., & Hellhammer, D. (2001). Psychobiological aspects
of atopic dermatitis: An overview. Psychotherapy and Psychosomatics, 70, 6–16.
Canli, T. (2004). Functional brain mapping of extraversion and neuroticism: Learning
from individual differences in emotion processing. Journal of Personality, 72, 1105–
1132.
Capitanio, J. P., Mendoza, S. P., & Cole, S. W. (2011). Nervous temperament in infant
monkeys is associated with reduced sensitivity of leukocytes to cortisol’s influence on
trafficking. Brain, Behavior, and Immunity, 25, 151–159.
Carey, G., & DiLalla, D. L. (1994). Personality and psychopathology: Genetic
perspectives. Journal of Abnormal Psychology, 103, 32–43.
Caspi, A. (2000). The child is father of the man: Personality continuities from childhood
to adulthood. Journal of Personality and Social Psychology, 78, 158–172.
Caspi, A., Harrington, H., Milne, B., Amell, J. W., Theodore, R. F., & Moffitt, T. E.
(2003). Children’s behavioral styles at age 3 are linked to their adult personality traits
at age 26. Journal of Personality, 71, 495–514.
Caspi, A., Houts, R. M., Belsky, D. W., Goldman-Mellor, S. J., Harrington, H., Israel, S.,
… Moffitt, T. E. (2014). The p factor: One general psychopathology factor in the
structure of psychiatric disorders? Clinical Psychological Science, 2, 119–137.
Caspi, A., & Moffitt, T. (2006). Gene-environment interactions in psychiatry: Joining
forces with neuroscience. Nature Reviews Neuroscience, 7, 583–590.
Caspi, A., & Silva, P. A. (1995). Temperamental qualities at age three predict personality
traits in young adulthood: Longitudinal evidence from a birth cohort. Child
Development, 66, 486–498.
Chaturvedi, S. K. (1986). Chronic idiopathic pain disorder. Journal of Psychosomatic
Research, 30, 199–203.
Christensen, A. J., Ehlers, S. L., Wiebe, J. S., Moran, P. J., Raichle, K., Ferneyhough, K.,
& Lawton, W. J. (2002). Patient personality and mortality: A 4-year prospective
examination of chronic renal insufficiency. Health Psychology: Official Journal of the
Division of Health Psychology, American Psychological Association, 21, 315–320.
Clark, L. A., & Watson, D. (1991). Tripartite model of anxiety and depression:
Psychometric evidence and taxonomic implications. Journal of Abnormal Psychology,
100, 316.
Cohen, J. (1992). A power primer. Psychological bulletin, 112(1), 155.
Contrada, R. J., Cather, C., & O’Leary, A. (1999). Personality and health: Dispositions
and processes in disease susceptibility and adaptation to illness. In L. A. Pervin & O. P.
John (Eds.), Handbook of personality (Vol. 2, pp. 576–604). New York, NY: Guilford
Press.
Cooper, M. L., Agocha, V. B., & Sheldon, M. S. (2000). A motivational perspective on
risky behaviors: The role of personality and affect regulatory processes. Journal of
Costa, P. T., & McCrae, R. R. (1987). Neuroticism, somatic complaints, and disease: is
the bark worse than the bite? Journal of Personality, 55(2), 299–316.
Costa, P. T., & McCrae, R. R. (1992). Four ways five factors are basic. Personality and
Cuijpers, P., Smit, F., Penninx, B. W. J. H., de Graaf, R., ten Have, M., & Beekman, A.
T. F. (2010). Economic costs of neuroticism: A population based study. Archives of
General Psychiatry, 67, 1086–1093.
Currie, S. R., & Wang, J. (2005). More data on major depression as an antecedent risk
factor for first onset of chronic back pain. Psychological Medicine, 35, 1275–1282.
Daruna, J. H. (1996). Neuroticism predicts normal variability in the number of circulating
leucocytes. Personality and Individual Differences, 20, 103–108.
Davies, P. T., & Windle, M. (2001). Interparental discord and adolescent adjustment
trajectories: The potentiating and protective role of intrapersonal attributes. Child
De Bolle, M., De Clercq, B., Van Leeuwen, K., Decuyper, M., Rosseel, Y., & De Fruyt,
F. (2009). Personality and psychopathology in Flemish referred children: Five
perspectives of continuity. Child Psychiatry and Human Development, 40, 269–285.
De Fruyt, F., Bartels, M., Van Leeuwen, K. G., De Clercq, B., Decuyper, M., &
Mervielde, I. (2006). Five types of personality continuity in childhood and adolescence.
de Moor, M., Derringer, J., Amin, N., Willemsen, G., Hottenga, J., Distel, M., …
Madden, P. (2012). Meta-analysis of genome-wide association studies for personality.
Molecular Psychiatry, 17, 337–349.
De Moor, M. H. M., van den Berg, S. M., Verweij, K. J. H., Krueger, R. F., Luciano, M.,
Vasquez, A. A. … Boomsma, D. I. The influence of common genetic variants on
neuroticism and their overlap with major depression: Evidence from the genetics of
personality consortium. Under review.
De Pauw, S. S., & Mervielde, I. (2010). Temperament, personality and developmental
traits. Child Psychiatry and Human Development, 41, 313–329.
De Pauw, S. S. W., Mervielde, I., & van Leeuwen, K. G. (2009). How are traits related to
problem behavior in preschoolers? Similarities and contrasts between temperament and
personality. Journal of Abnormal Child Psychology, 37, 309–325.
De Young, C. G., Quilty, L. C., & Peterson, J. B. (2007). Between facets and domains: 10
aspects of the Big Five. Journal of Personality and Social Psychology, 93, 880.
Di Simplicio, M., Costoloni, G., Western, D., Hanson, B., Taggart, P., & Harmer, C. J.
(2012). Decreased heart rate variability during emotion regulation in subjects at risk for
psychopathology. Psychological Medicine, 42, 1775–1779.
Donnellan, M. B., Conger, R. D., & Bryant, C. M. (2004). The Big Five and enduring
marriages. Journal of Research in Personality, 38, 481–504.
Drossman, D., Leserman, J., Li, Z., Keefe, F., Hu, Y., & Toomey, T. (2000). Effects of
coping on health outcome among women with gastrointestinal disorders.
Psychosomatic Medicine, 62, 309–317.
Druss, B., Hwang, I., Petukhova, M., Sampson, N., Wang, P., & Kessler, R. (2009).
Impairment in role functioning in mental and chronic medical disorders in the United
States: Results from the national comorbidity survey replication. Molecular Psychiatry,
14, 728–737.
Durbin, C. E. (2010). Modeling temperamental risk for depression using developmentally
sensitive laboratory paradigms. Child Development Perspectives, 4, 168–173.
Eaton, N., Krueger, R., & Oltmanns, T. (2011). Aging and the structure and long-term
stability of the internalizing spectrum of personality and psychopathology. Psychology
and Aging, 26, 987–993.
Eaves, L. J., Eysenck, H. J., & Martin, N. G. (1989). Genes, culture, and personality: An
empirical approach. San Diego, CA: Academic Press.
Ehring, T., Ehlers, A., & Glucksman, E. (2006). Contribution of cognitive factors to the
prediction of post-traumatic stress disorder, phobia and depression after motor vehicle
accidents. Behaviour Research and Therapy, 44, 1699–1716.
Espejo, E. P., Hammen, C., & Brennan, P. A. (2012). Elevated appraisals of the negative
impact of naturally occurring life events: A risk factor for depressive and anxiety
disorders. Journal of Abnormal Child Psychology, 40, 303–315.
Fanous, A., Gardner, C. O., Prescott, C. A., Cancro, R., & Kendler, K. S. (2002).
Neuroticism, major depression and gender: A population-based twin study.
Psychological Medicine, 32, 719–728.
Fanous, A. H., Neale, M. C., Aggen, S. H., & Kendler, K. S. (2007). A longitudinal study
of personality and major depression in a population-based sample of male twins.
Fergusson, D. M., Woodward, L. J., & Horwood, L. J. (2000). Risk factors and life
processes associated with the onset of suicidal behavior during adolescence and early
adulthood. Psychological Medicine, 30, 23–39.
Finch, J. F., & Graziano, W. G. (2001). Predicting depression from temperament,
personality, and patterns of social relations. Journal of Personality, 69, 27–55.
Friedman, H. S. (2000). Long-term relations of personality and health: Dynamisms,
mechanisms, tropisms. Journal of Personality, 68, 1089–1107.
Fullerton, J. (2006). New approaches to the genetic analysis of neuroticism and anxiety.
Behavior Genetics, 36, 147–161.
Fullerton, J., Neale, M. C., Fiddy, S., Mott, R., Allison, D. B., Flint, J., … Goodwin, G.
(2003). Linkage analysis of extremely discordant and concordant sibling pairs
identifies quantitative-trait loci that influence variation in the human personality trait
neuroticism. The American Journal of Human Genetics, 72, 879–890.
Futterman, A. D., Kemeny, M. E., Shapiro, D., & Fahey, J. L. (1994). Immunological and
physiological changes associated with induced positive and negative mood.
Gale, C. R., Hardy, R., Craig, L., Deary, I. J., Sayer, A. A., Cooper, C., … HALCyon
Study Team. (2011). Factors associated with symptoms of anxiety and depression in
five cohorts of community-based older people: The HALCyon (healthy ageing across
the life course) programme. Psychological Medicine, 41, 2057–2073.
Gattis, K. S., Berns, S., Simpson, L. E., & Christensen, A. (2004). Birds of a feather or
strange birds? Ties among personality dimensions, similarity, and marital quality.
Journal of Family Psychology, 18, 564–574.
Gillespie, N. A., Evans, D. E., Wright, M. M., & Martin, N. G. (2004). Genetic simplex
modeling of Eysenck’s dimensions of personality in a sample of young Australian
twins. Twin Research and Human Genetics, 7, 637–648.
Gleason, M. E. J., Powers, A. D., & Oltmanns, T. F. (2012). The enduring impact of
borderline personality pathology: Risk for threatening life events in later middle-age.
Journal of Abnormal Psychology, 121, 447–457.
Goldberg, L. R. (1993). The structure of phenotypic personality traits. The American
Psychologist, 48, 26–34.
Golimbet, V. E., Alfimova, M. V., Manandyan, K. K., Mitushina, N. G., Abramova, L. I.,
Kaleda, V. G., … Trubnikov, V. I. (2002). 5HTR2A gene polymorphism and
personality traits in patients with major psychoses. European Psychiatry, 17, 24–28.
Goubert, L., Crombez, G., & Van Damme, S. (2004). The role of neuroticism, pain
catastrophizing and pain-related fear in vigilance to pain: A structural equations
approach. Pain, 107, 234–241.
Gunthert, K. C., Cohen, L. H., & Armeli, S. (1999). The role of neuroticism in daily
stress and coping. Journal of Personality and Social Psychology, 77, 1087–1100.
Haas, B. W., Omura, K., Constable, R. T., & Canli, T. (2007). Emotional conflict and
neuroticism: Personality-dependent activation in the amygdala and subgenual anterior
cingulate. Behavioral Neuroscience, 121, 249–256.
Hampson, S. E., & Goldberg, L. R. (2006). A first large-cohort study of personality-trait
stability over the 40 years between elementary school and midlife. Journal of
Hankin, B. L., Fraley, R. C., & Abela, J. R. Z. (2005). Daily depression and cognitions
about stress: Evidence for a traitlike depressogenic cognitive style and the prediction of
depressive symptoms in a prospective daily diary study. Journal of Personality and
Hariri, A. R., Mattay, V. S., Tessitore, A., Kolachana, B., Fera, F., Goldman, D., …
Weinberger, D. R. (2002). Serotonin transporter genetic variation and the response of
the human amygdala. Science, 297, 400–403.
Hemström, Ö. (1996). Is marriage dissolution linked to differences in mortality risks for
men and women? Journal of Marriage and Family, 58, 366–378.
Hennig, J., Pössel, P., & Netter, P. (1996). Sensitivity to disgust as an indicator of
neuroticism: A psychobiological approach. Personality and Individual Differences, 20,
589–596.
Hettema, J. M., An, S. S., Neale, M. C., Bukszar, J., van den Oord, E. J. C. G., Kendler,
K. S., & Chen, X. (2006). Association between glutamic acid decarboxylase genes and
anxiety disorders, major depression, and neuroticism. Molecular Psychiatry, 11, 752–
762.
Hettema, J. M., Neale, M. C., Myers, J. M., Prescott, C. A., & Kendler, K. S. (2006). A
population-based twin study of the relationship between neuroticism and internalizing
disorders. The American Journal of Psychiatry, 163, 857–864.
Hirsh, J. B., & Inzlicht, M. (2008). The devil you know: Neuroticism predicts neural
response to uncertainty. Psychological Science, 19, 962–967.
Holt-Lunstad, J., Smith, T. W., & Uchino, B. N. (2008). Can hostility interfere with the
health benefits of giving and receiving social support? the impact of cynical hostility on
cardiovascular reactivity during social support interactions among friends. Annals of
Behavioral Medicine: A Publication of the Society of Behavioral Medicine, 35, 319–
330.
Hoyle, R. H., Fejfar, M. C., & Miller, J. D. (2000). Personality and sexual risk taking: A
quantitative review. Boston, MA and Oxford, England: Blackwell.
Huovinen, E., Kaprio, J., & Koskenvuo, M. (2001). Asthma in relation to personality
traits, life satisfaction, and stress: A prospective study among 11 000 adults. Allergy,
56, 971–977.
Hutchinson, J. G., & Ruiz, J. M. (2011). Neuroticism and cardiovascular response in
women: Evidence of effects on blood pressure recovery. Journal of Personality, 79,
277–302.
Hutchinson, J. G., & Williams, P. G. (2007). Neuroticism, daily hassles, and depressive
symptoms: An examination of moderating and mediating effects. Personality and
Hyde, L. W., Gorka, A., Manuck, S. B., & Hariri, A. R. (2011). Perceived social support
moderates the link between threat-related amygdala reactivity and trait anxiety.
Neuropsychologia, 49, 651–656.
Ikeda, A., Iso, H., Toyoshima, H., Fujino, Y., Mizoue, T., Yoshimura, T., … JACC Study
Group. (2007). Marital status and mortality among Japanese men and women: The
Japan collaborative cohort study. BMC Public Health, 7, 73–79.
Insel, T., Cuthbert, B., Garvey, M., Heinssen, R., Pine, D. S., Quinn, K., … Wang, P.
(2010). Research domain criteria (RDoC): Toward a new classification framework for
research on mental disorders. American Journal of Psychiatry, 167, 748–751.
Jabbi, M., Korf, J., Kema, I. P., Hartman, C., van der Pompe, G., Minderaa, R. B., et al.
(2007). Convergent genetic modulation of the endocrine stress response involves
polymorphic variations of 5-HTT, COMT, and MAOA. Molecular Psychiatry, 12,
483–490.
Jacobs, N., Kenis, G., Peeters, F., Derom, C., Vlietinck, R., & van Os, J. (2006). Stress-
related negative affectivity and genetically altered serotonin transporter function:
Evidence of synergism in shaping risk of depression. Archives of General Psychiatry,
63, 989–996.
Johnson, W. (2007). Genetic and environmental influences on behavior: Capturing all the
interplay. Psychological Review, 114, 423–440.
Karney, B. R., & Bradbury, T. N. (1997). Neuroticism, marital interaction, and the
trajectory of marital satisfaction. Journal of Personality and Social Psychology, 72,
1075–1092.
Kelly, E. L., & Conley, J. J. (1987). Personality and compatibility: A prospective analysis
of marital stability and marital satisfaction. Journal of Personality and Social
Kendler, K. S. (2014). The structure of psychiatric science. American Journal of
Kendler, K. S., Gardner, C. O., & Prescott, C. A. (2002). Toward a comprehensive
developmental model for major depression in women. The American Journal of
Psychiatry, 159, 1133–1145.
Kendler, K. S., Gardner, C. O., & Prescott, C. A. (2006). Toward a comprehensive
developmental model for major depression in men. The American Journal of
Kendler, K. S., Gatz, M., Gardner, C. O., & Pedersen, N. L. (2006). Personality and major
depression: A Swedish longitudinal, population-based twin study. Archives of General
Psychiatry, 63, 1113–1120.
Kendler, K. S., Kuhn, J., & Prescott, C. A. (2004). The interrelationship of neuroticism,
sex, and stressful life events in the prediction of episodes of major depression. The
American Journal of Psychiatry, 161, 631–636.
Kendler, K. S., Neale, M. C., Kessler, R. C., Heath, A. C., & Eaves, L. J. (1993). A
longitudinal twin study of personality and major depression in women. Archives of
General Psychiatry, 50, 853–862.
Kessler, R. C., Chiu, W. T., Demler, O., Merikangas, K. R., & Walters, E. E. (2005).
Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the national
comorbidity survey replication. Archives of General Psychiatry, 62, 617–627.
Khan, A. A., Jacobson, K. C., Gardner, C. O., Prescott, C. A., & Kendler, K. S. (2005).
Personality and comorbidity of common psychiatric disorders. The British Journal of
Klein, D. N., Kotov, R., & Bufferd, S. J. (2011). Personality and depression: Explanatory
models and review of the evidence. Annual Review of Clinical Psychology, 7, 269–295.
Kotov, R., Gamez, W., Schmidt, F., & Watson, D. (2010). Linking “big” personality traits
to anxiety, depressive, and substance use disorders: A meta-analysis. Psychological
Bulletin, 136, 768–821.
64, 241–256.
Lahey, B. B., Applegate, B., Chronis, A. M., Jones, H. A., Williams, S. H., Loney, J., et
al. (2008). Psychometric characteristics of a measure of emotional dispositions
developed to test a developmental propensity model of conduct disorder. Journal of
Clinical Child and Adolescent Psychology, 37, 794–807.
Lahey, B. B., Applegate, B., Hakes, J. K., Zald, D. H., Hariri, A. R., & Rathouz, P. J.
(2012). Is there a general factor of prevalent psychopathology during adulthood?
Journal of Abnormal Psychology, 121, 971.
Lahey, B. B., Rathouz, P. J., Applegate, B., Tackett, J. L., & Waldman, I. D. (2010).
Psychometrics of a self-report version of the child and adolescent dispositions scale.
Journal of Clinical Child and Adolescent Psychology, 39, 351–361.
Lake, R. I. E., Eaves, L. J., Maes, H. H. M., Heath, A. C., & Martin, N. G. (2000). Further
evidence against the environmental transmission of individual differences in
neuroticism from a collaborative study of 45,850 twins and relatives on two continents.
Lang, U. E., Hellweg, R., Kalus, P., Bajbouj, M., Lenzen, K. P., Sander, T., … Gallinat, J.
(2005). Association of a functional BDNF polymorphism and anxiety-related
personality traits. Psychopharmacology, 180, 95–99.
Larkins, J. M., & Sher, K. J. (2006). Family history of alcoholism and the stability of
personality in young adulthood. Psychology of Addictive Behaviors, 20, 471–477.
Larsen, R. J., & Ketelaar, T. (1991). Personality and susceptibility to positive and
negative emotional states. Journal of Personality and Social Psychology, 61, 132–140.
LeBlanc, J., Ducharme, M. B., & Thompson, M. (2004). Study on the correlation of the
autonomic nervous system responses to a stressor of high discomfort with personality
traits. Physiology and Behavior, 82, 647–652.
Lee, C., & Gramotnev, H. (2007). Life transitions and mental health in a national cohort
of young Australian women. Developmental Psychology, 43, 877–888.
Lee, S., Cho, E., Grodstein, F., Kawachi, I., Hu, F. B., & Colditz, G. A. (2005). Effects of
marital transitions on changes in dietary and other health behaviours in US women.
International Journal of Epidemiology, 34, 69–78.
Leonardo, E. D., & Hen, R. (2006). Genetics of affective and anxiety disorders. Annual
Lesch, K. P., & Canli, T. (2006). 5-HT1A receptor and anxiety-related traits:
Pharmacology, genetics, and imaging. In T. Canli (Ed.), Biology of personality and
individual differences (pp. 273–294). New York, NY: Guilford Press.
Littlefield, A. K., Agrawal, A., Ellingson, J. M., Kristjansson, S., Madden, P. A. F.,
Bucholz, K. K., … Sher, K. J. (2011). Does variance in drinking motives explain the
genetic overlap between personality and alcohol use disorder symptoms? A twin study
of young women. Alcoholism: Clinical and Experimental Research, 35, 2242–2250.
Littlefield, A. K., & Sher, K. J. (2010). The multiple, distinct ways that personality
contributes to alcohol use disorders. Social and Personality Psychology Compass, 4,
767–782.
Lotrich, F. E., & Pollock, B. G. (2004). Meta-analysis of serotonin transporter
polymorphisms and affective disorders. Psychiatric Genetics, 14, 121–129.
Lynn, M., & Steel, P. (2006). National differences in subjective well-being: The
interactive effects of extraversion and neuroticism. Journal of Happiness Studies, 7,
155–165.
Magee, C. A., Heaven, P. C. L., & Miller, L. M. (2013). Personality change predicts self-
reported mental and physical health. Journal of Personality, 81, 324–334.
Magnus, K., Diener, E., Fujita, F., & Pavot, W. (1993). Extraversion and neuroticism as
predictors of objective life events: A longitudinal analysis. Journal of Personality and
Maier, S. F., & Watkins, L. R. (1998). Cytokines for psychologists: Implications of
bidirectional immune-to-brain communication for understanding behavior, mood, and
cognition. Psychological Review, 105, 83–107.
Malouff, J. M., Thorsteinsson, E. B., & Schutte, N. S. (2005). The relationship between
the five-factor model of personality and symptoms of clinical disorders: A meta-
analysis. Journal of Psychopathology and Behavioral Assessment, 27, 101–114.
Malouff, J. M., Thorsteinsson, E. B., & Schutte, N. S. (2006). The five-factor model of
personality and smoking: A meta-analysis. Journal of Drug Education, 36, 47–58.
McCaffery, J. M., Frasure-Smith, N., Dubé, M., Théroux, P., Rouleau, G. A., Duan, Q., &
Lespérance, F. (2006). Common genetic vulnerability to depressive symptoms and
coronary artery disease: A review and development of candidate genes related to
inflammation and serotonin. Psychosomatic Medicine, 68, 187–200.
McCann, S. J. H. (2011). Emotional health and the big five personality factors at the
American state level. Journal of Happiness Studies, 12, 547–560.
McCleery, J. M., & Goodwin, G. M. (2001). High and low neuroticism predict different
cortisol responses to the combined dexamethasone-CRH test. Biological Psychiatry,
49, 410–415.
McCrae, R. R., & Costa, J. P. T. (1988). Recalled parent-child relations and adult
personality. Journal of Personality, 56, 417–434.
Mikolajewski, A. J., Allan, N. P., Hart, S. A., Lonigan, C. J., & Taylor, J. (2013).
Negative affect shares genetic and environmental influences with symptoms of
childhood internalizing and externalizing disorders. New York, NY: Springer.
Morissette, S. B., Tull, M. T., Gulliver, S. B., Kamholz, B. W., & Zimering, R. T. (2007).
Anxiety, anxiety disorders, tobacco use, and nicotine: A critical review of
interrelationships. Psychological Bulletin, 133, 245–272.
Mroczek, D. K., & Spiro, A. (2003). Modeling intraindividual change in personality
traits: Findings from the Normative Aging Study. The Journals of Gerontology Series
B: Psychological Sciences and Social Sciences, 58, P153–P165.
Mullins-Sweatt, S. N., & Widiger, T. A. (2010). Personality-related problems in living:
An empirical approach. Personality Disorders: Theory, Research, and Treatment, 1,
230–238.
Munafò, M. R., Clark, T., & Flint, J. (2005). Promise and pitfalls in the meta-analysis of
genetic association studies: A response to Sen and Schinka. Molecular Psychiatry, 10,
895–897.
Munafò, M. R., Clark, T., Roberts, K. H., & Johnstone, E. C. (2006). Neuroticism
mediates the association of the serotonin transporter gene with lifetime major
depression. Neuropsychobiology, 53, 1–8.
Murberg, T. A. (2004). Long-term effect of social relationships on mortality in patients
with congestive heart failure. International Journal of Psychiatry in Medicine, 34, 207–
217.
Murray, G., Allen, N. B., Trinder, J., & Burgess, H. (2002). Is weakened circadian
rhythmicity a characteristic of neuroticism? Journal of Affective Disorders, 72, 281–
289.
Muth, E. R., Koch, K. L., & Stern, R. M. (2000). Significance of autonomic nervous
system activity in functional dyspepsia. Digestive Diseases and Sciences, 45, 854–863.
Nakaya, N., Hansen, P. E., Schapiro, I. R., Eplov, L. F., Saito-Nakaya, K., Uchitomi, Y.,
& Johansen, C. (2006). Personality traits and cancer survival: A Danish cohort study.
British Journal of Cancer, 95, 146–152.
Nash, M. W., Gray, J. A., Ball, D., Asherson, P., Mann, A., Goldberg, D., … Prince, M.
(2004). Genome-wide linkage analysis of a composite index of neuroticism and mood-
related scales in extreme selected sibships. Human Molecular Genetics, 13, 2173–
2182.
Nater, U. M., Hoppmann, C., & Klumb, P. L. (2010). Neuroticism and conscientiousness
are associated with cortisol diurnal profiles in adults—Role of positive and negative
affect. Psychoneuroendocrinology, 35, 1573–1577.
Nave, C. S., Sherman, R. A., Funder, D. C., Hampson, S. E., & Goldberg, L. R. (2010).
On the contextual independence of personality teachers’ assessments predict directly
observed behavior after four decades. Social Psychological and Personality Science, 1,
327–334.
Norris, C., Larsen, J., & Cacioppo, J. (2007). Neuroticism is associated with larger and
more prolonged electrodermal responses to emotionally evocative pictures.
Psychophysiology, 44, 823–826.
Olson, S. L., Bates, J. E., Sandy, J. M., & Schilling, E. M. (2002). Early developmental
precursors of impulsive and inattentive behavior: From infancy to middle childhood.
Journal of Child Psychology and Psychiatry, 43, 435–447.
Omura, K., Constable, R. T., & Canli, T. (2005). Amygdala gray matter concentration is
associated with extraversion and neuroticism. Neuroreport, 16, 1905–1908.
Ormel, J., Bastiaansen, A., Riese, H., Bos, E. H., Servaas, M., Ellenbogen, M., …
Aleman, A. (2013). The biological and psychological basis of neuroticism: Current
status and future directions. Neuroscience and Biobehavioral Reviews, 37, 59–72.
Ormel, J., Jeronimus, B. F., Kotov, R., Riese, H., Bos, E. H., Hankin, B., … Oldehinkel,
A. J. (2013). Neuroticism and common mental disorders: Meaning and utility of a
complex relationship. Clinical Psychology Review, 33, 686–697.
Oswald, L., Zandi, P., Nestadt, G., Potash, J., Kalaydjian, A., & Wand, G. (2006).
Relationship between cortisol responses to stress and personality.
Neuropsychopharmacology, 31, 1583–1591.
Overbeek, G., Vollerbergh, W., de Graaf, R., Scholte, R., de Kemp, R., & Engels, R.
(2006). Longitudinal associations of marital quality and marital dissolution with the
incidence of DSM-III-R disorders. Journal of Family Psychology, 20, 284–291.
Pace, T. W. W., Mletzko, T. C., Alagbe, O., Musselman, D. L., Nemeroff, C. B., Miller,
A. H., & Heim, C. M. (2006). Increased stress-induced inflammatory responses in male
patients with major depression and increased early life stress. The American Journal of
Psychiatry, 163, 1630–1633.
Parslow, R. A., Jorm, A. F., & Christensen, H. (2006). Associations of pre-trauma
attributes and trauma exposure with screening positive for PTSD: Analysis of a
community-based study of 2085 young adults. New York, NY: Cambridge University
Press.
Passamonti, L., Cerasa, A., Gioia, M. C., Magariello, A., Muglia, M., Quattrone, A., &
Fera, F. (2008). Genetically dependent modulation of serotonergic inactivation in the
human prefrontal cortex. Neuroimage, 40, 1264–1273.
Penninx, B. W. J. H., Beekman, A. T. F., Honig, A., Deeg, D. J. H., Schoevers, R. A., van
Eijk, J. T. M., & van Tilburg, W. (2001). Depression and cardiac mortality: Results
from a community-based longitudinal study. Archives of General Psychiatry, 58, 221–
227.
Perreira, K. M., & Sloan, F. A. (2001). Life events and alcohol consumption among
mature adults: A longitudinal analysis. Journal of Studies on Alcohol, 62, 501–508.
Phillips, A. C., Carroll, D., Burns, V. E., & Drayson, M. (2005). Neuroticism, cortisol
reactivity, and antibody response to vaccination. Psychophysiology, 42, 232–238.
Portella, M. J., Harmer, C. J., Flint, J., Cowen, P., & Goodwin, G. M. (2005). Enhanced
early morning salivary cortisol in neuroticism. The American Journal of Psychiatry,
162, 807–809.
Powers, A. D., & Oltmanns, T. F. (2013). Personality pathology as a risk factor for
negative health perception. Journal of Personality Disorders, 27, 359–370.
Quilty, L. C., Meusel, L. A., & Bagby, R. M. (2008). Neuroticism as a mediator of
treatment response to SSRIs in major depressive disorder. Journal of Affective
Disorders, 111, 67–73.
Reti, I. M., Samuels, J. F., Eaton, W. W., Bienvenu, O. J., Costa, J., Paul, T., & Nestadt,
G. (2002). Influences of parenting on normal personality traits. Psychiatry Research,
111, 55–64.
Rettew, D. C., Vink, J. M., Willemsen, G., Doyle, A., Hudziak, J. J., & Boomsma, D. I.
(2006). The genetic architecture of neuroticism in 3301 Dutch adolescent twins as a
function of age and sex: A study from the Dutch Twin Register. Twin Research and
Human Genetics, 9, 24–29.
Riese, H., Rosmalen, J. G. M., Ormel, J., Van Roon, A. M., Oldehinkel, A. J., & Rijsdijk,
F. V. (2007). The genetic relationship between neuroticism and autonomic function in
female twins. Psychological Medicine, 37, 257–267.
Rietschel, M., Jamra, R. A., Schumacher, J., Höfels, S., Kumsta, R., Entringer, S., …
Bürger, C. (2008). G72 and its association with major depression and neuroticism in
large population-based groups from Germany. The American Journal of Psychiatry,
165, 753–762.
Rijsdijk, F. V., & Sham, P. C. (2002). Analytic approaches to twin data using structural
equation models. Briefings in Bioinformatics, 3, 119–133.
Roberts, B. W., Walton, K. E., & Viechtbauer, W. (2006). Patterns of mean-level change
in personality traits across the life course: A meta-analysis of longitudinal studies.
Robinson, G. E. (2004). Genomics. Beyond nature and nurture. Science (New York, N.Y.),
304, 397–399.
Robles, T. F., Glaser, R., & Kiecolt-Glaser, J. K. (2005). Out of balance: A new look at
chronic stress, depression, and immunity. Current Directions in Psychological Science,
14, 111–115.
Rogge, R. D., Bradbury, T. N., Hahlweg, K., Engl, J., & Thurmaier, F. (2006). Predicting
marital distress and dissolution: Refining the two-factor hypothesis. Journal of Family
Rothbart, M. K., Ahadi, S. A., Hershey, K. L., & Fisher, P. (2001). Investigations of
temperament at three to seven years: The Children’s Behavior Questionnaire. Child
Rothbart, M. K., & Bates, J. E. (2006). Temperament. In N. Eisenberg, W. Damon, & R.
M. Lerner (Eds.), Handbook of child psychology: Vol. 3, Social, emotional, and
personality development (6th ed., pp. 99–166). Hoboken, NJ: John Wiley.
Roy, A. (2002). Childhood trauma and neuroticism as an adult: Possible implication for
the development of the common psychiatric disorders and suicidal behaviour.
Russo, J., Katon, W., Lin, E., Von Korff, M., Bush, T., Simon, G., & Walker, E. (1997).
Neuroticism and extraversion as predictors of health outcomes in depressed primary
care patients. Psychosomatics, 38, 339–348.
Rutter, M. (2008). Biological implications of gene–environment interaction. Journal of
Abnormal Child Psychology, 36, 969–975.
Sareen, J., Cox, B. J., Clara, I., & Asmundson, G. J. G. (2005). The relationship between
anxiety disorders and physical disorders in the U. S. National Comorbidity Survey.
Depression and Anxiety, 21, 193–202.
Saulsman, L. M., & Page, A. C. (2004). The five-factor model and personality disorder
empirical literature: A meta-analytic review. Clinical Psychology Review, 23, 1055–
1085.
Schinka, J. A. (2005). Measurement scale does moderate the association between the
serotonin transporter gene and trait anxiety: Comments on Munafo et al. Molecular
Schinka, J. A., Busch, R. M., & Robichaux-Keene, N. (2004). A meta-analysis of the
association between the serotonin transporter gene polymorphism (5-HTTLPR) and
trait anxiety. Molecular Psychiatry, 9, 197–202.
Schneiderman, N., Ironson, G., & Siegel, S. D. (2005). Stress and health: Psychological,
behavioral, and biological determinants. Annual Review of Clinical Psychology, 1,
607–628.
Schommer, N. C., Kudielka, B. M., Hellhammer, D. H., & Kirschbaum, C. (1999). No
evidence for a close relationship between personality traits and circadian cortisol
rhythm or a single cortisol stress response. Psychological Reports, 84, 840–842.
Sen, S., Burmeister, M., & Ghosh, D. (2004). Meta-analysis of the association between a
serotonin transporter promoter polymorphism (5-HTTLPR) and anxiety-related
personality traits. Hoboken, NJ: Wiley Subscription Services. A Wiley Company.
Sen, S., Burmeister, M., & Ghosh, D. (2005). 5-HTTLPR and anxiety-related personality
traits meta-analysis revisited: Response to Munafo and colleagues. Molecular
Sen, S., Nesse, R. M., Stoltenberg, S. F., Li, S., Gleiberman, L., Chakravarti, A., …
Burmeister, M. (2003). A BDNF coding variant is associated with the NEO personality
inventory domain neuroticism, a risk factor for depression. Neuropsychopharmacology,
28, 397–401.
Shi, J., Badner, J. A., Gershon, E. S., & Liu, C. (2008). Allelic association of G72/G30
with schizophrenia and bipolar disorder: A comprehensive meta-analysis.
Schizophrenia Research, 98, 89–97.
Shifman, S., van den Oord, E. J. C. G., Kendler, K. S., Chen, X., Boomsma, D. I.,
Middeldorp, C. M., … Neale, M. C. (2008). A whole genome association study of
neuroticism using DNA pooling. Molecular Psychiatry, 13, 302–312.
Shiner, R. L. (2010). Mapping the landscape of personality in childhood and adolescence.
Social and Personality Psychology Compass, 4, 1084–1097.
Shiner, R., & Tackett, J. L. (2014). Personality and personality disorders. In E. Mash &
R. Barkley (Eds.), Child psychopathology (3rd ed.). New York, NY: Guilford Press.
Shipley, B., Weiss, A., Der, G., Taylor, M., & Deary, I. (2007). Neuroticism,
extraversion, and mortality in the UK health and lifestyle survey: A 21-year
prospective cohort study. Psychosomatic Medicine, 69, 923–931.
Silberg, J., Rutter, M., Neale, M., & Eaves, L. (2001). Genetic moderation of
environmental risk for depression and anxiety in adolescent girls. The British Journal
of Psychiatry, 179, 116–121.
Simonsick, E., Wallace, R., Blazer, D., & Berkman, L. (1995). Depressive
symptomatology and hypertension-associated morbidity and mortality in older adults.
Smith, T. W., & MacKenzie, J. (2006). Personality and risk of physical illness. Annual
Review of Clinical Psychology, 2, 435–467.
Soto, C. J., John, O. P., Gosling, S. D., & Potter, J. (2011). Age differences in personality
traits from 10 to 65: Big Five domains and facets in a large cross-sectional sample.
Spiller, R. C. (2007). Role of infection in irritable bowel syndrome. Journal of
Gastroenterology, 42, S41–S47.
Starr, L. R., Hammen, C., Brennan, P. A., & Najman, J. M. (2012). Serotonin transporter
gene as a predictor of stress generation in depression. Journal of Abnormal Psychology,
121, 810–818.
Steel, P., Schmidt, J., & Shultz, J. (2008). Refining the relationship between personality
and subjective well-being. Psychological Bulletin, 134, 138–161.
Stein, D. J., & Stein, M. B. (2008). Social anxiety disorder. The Lancet, 371, 1115–1125.
Suls, J., & Bunde, J. (2005). Anger, anxiety, and depression as risk factors for
cardiovascular disease: The problems and implications of overlapping affective
dispositions. Psychological Bulletin, 131, 260–300.
Suls, J., & Martin, R. (2005). The daily life of the garden-variety neurotic: Reactivity,
stressor exposure, mood spillover, and maladaptive coping. Journal of Personality, 73,
1–25.
in children and adolescents. Clinical Psychology Review, 26, 584–599.
Tackett, J. L. (2011). Parent informants for child personality: Agreement, discrepancies,
and clinical utility. Journal of Personality Assessment, 93, 539–544.
Tackett, J. L., Herzhoff, K., Reardon, K., Smack, A., & Kushner, S. K. (2013). The
relevance of informant discrepancies for the assessment of adolescent personality
pathology. Clinical Psychology: Science and Practice, 20, 378–392.
Tackett, J. L., Krueger, R. F., Iacono, W. G., & McGue, M. (2008). Personality in middle
childhood: A hierarchical structure and longitudinal connections with personality in
late adolescence. Journal of Research in Personality, 42, 1456–1462.
Tackett, J. L., Kushner, S. C., Mervielde, I., & De Fruyt, F. (2013). Delineating
personality traits in childhood and adolescence: Associations across measures,
temperament, and behavioral problems. Assessment, 20, 738–751.
Tackett, J. L., Slobodskaya, H., Mar, R. A., Deal, J., Halverson, C. F., Jr., Baker, S. R., …
Besevegis, E. (2012). The hierarchical structure of childhood personality in five
countries: Continuity from early childhood to early adolescence. Journal of
Tellegen, A., & Waller, N. G. (2008). Exploring personality through test construction:
Development of the Multidimensional Personality Questionnaire. The SAGE Handbook
of Personality Theory and Assessment, 2, 261–292.
Terracciano, A., Costa, J., & Paul T. (2004). Smoking and the five-factor model of
personality. Addiction (Abingdon, England), 99, 472–481.
Trobst, K. K., Wiggins, J. S., Costa, J. P. T., Herbst, J. H., McCrae, R. R., & Masters, H.
L. (2000). Personality psychology and problem behaviors: HIV risk and the five-factor
model. Journal of Personality, 68, 1233–1252.
Tucker, J. S., Kressin, N. R., Spiro, A., & Ruscio, J. (1998). Intrapersonal characteristics
and the timing of divorce: A prospective investigation. Journal of Social and Personal
Relationships, 15, 211–225.
Turkheimer, E. (2000). Three laws of behavior genetics and what they mean. Current
Directions in Psychological Science, 9, 160–164.
Uchino, B. N. (2006). Social support and health: A review of physiological processes
potentially underlying links to disease outcomes. Journal of Behavioral Medicine, 29,
377–387.
Uher, R., & McGuffin, P. (2010). The moderation by the serotonin transporter gene of
environmental adversity in the etiology of depression: 2009 update. Molecular
Van den Berg, S. M., de Moor, M. H. M., McGue, M., Pettersson, E., Terracciano, A.,
Verweij, K. J. H., … Boomsma, D. I. (2014). Harmonization of neuroticism and
extraversion phenotypes across inventories and cohorts in the genetics of personality
consortium: An application of item response theory. Behavior Genetics, 44, 295–313.
van den Oord, E. J. C. G., Kuo, P., Hartmann, A. M., Webb, B. T., Moller, H., Hettema, J.
M., … Rujescu, D. (2008). Genomewide association analysis followed by a replication
study implicates a novel candidate gene for neuroticism. Archives of General
Psychiatry, 65, 1062–1071.
van Ockenburg, S. L., de Jonge, P., van der Harst, P., Ormel, J., & Rosmalen, J. G. M.
(2014). Does neuroticism make you old? Prospective associations between neuroticism
and leukocyte telomere length. Psychological Medicine, 44, 723.
van Os, J., & Jones, P. B. (2001). Neuroticism as a risk factor for schizophrenia.
van Os, J., Park, S. B. G., & Jones, P. B. (2001). Neuroticism, life events and mental
health: Evidence for person–environment correlation. The British Journal of
Psychiatry, 178, S72–S77.
Vassend, O., Røysamb, E., & Nielsen, C. S. (2012). Neuroticism and self-reported
somatic health: A twin study. Psychology and Health, 27, 1–12.
Vazire, S. (2010). Who knows what about a person? The self–other knowledge
asymmetry (SOKA) model. Journal of Personality and Social Psychology, 98, 281.
Viken, R. J., Rose, R. J., Kaprio, J., & Koskenvuo, M. (1994). A developmental genetic
analysis of adult personality: Extraversion and neuroticism from 18 to 59 years of age.
Vogeltanz, N. D., & Hecker, J. E. (1999). The roles of neuroticism and
controllability/predictability in physiological response to aversive stimuli. Personality
and Individual Differences, 27, 599–612.
Watkins, L. L., Blumenthal, J. A., Davidson, J. R. T., Babyak, M. A., McCants, C. B., &
Sketch, M. H. Jr. (2006). Phobic anxiety, depression, and risk of ventricular
arrhythmias in patients with coronary heart disease. Psychosomatic Medicine, 68, 651–
656.
Weiss, A., Bates, T. C., & Luciano, M. (2008). Happiness is a personal(ity) thing: The
genetics of personality and well-being in a representative sample. Psychological
Science, 19, 205–210.
Weiss, A., & Costa, P. T. Jr. (2005). Domain and facet personality predictors of all-cause
mortality among Medicare patients aged 65 to 100. Psychosomatic Medicine, 67, 724–
733.
Widiger, T. A. (2009). Neuroticism. In M. Leary & R. Hoyle (Eds.), Handbook of
individual differences in social behavior (pp. 129–146). New York, NY: Guilford
Press.
Widiger, T. A., & Smith, G. T. (2008). Personality and psychopathology. In O. P. John,
R. W. Robins, & L. A. Pervin (Eds.), Handbook of personality: Theory and research
(3rd ed., pp. 743–769). New York, NY: Guilford Press.
disorder: Shifting to a dimensional model. The American Psychologist, 62, 71–83.
Willemsen, G., & Boomsma, D. I. (2007). Religious upbringing and neuroticism in Dutch
twin families. Twin Research and Human Genetics: The Official Journal of the
International Society for Twin Studies, 10, 327–333.
Willis-Owen, S. A. G., Fullerton, J., Surtees, P. G., Wainwright, N. W. J., Miller, S., &
Flint, J. (2005). The Val66Met coding variant of the brain-derived neurotrophic factor
(BDNF) gene does not contribute toward variation in the personality trait neuroticism.
Biological Psychiatry, 58, 738–742.
Wilson, R., Krueger, K., Gu, L., Bienias, J., De Leon, C., & Evans, D. (2005).
Neuroticism, extraversion, and mortality in a defined population of older persons.
Wilson, R. S., Mendes de Leon, C. F., Bienias, J. L., Evans, D. A., & Bennett, D. A.
(2004). Personality and mortality in old age. Journals of Gerontology Series B:
Psychological Sciences and Social Sciences, 59, P110–P116.
Yeo, A., Bahari, B., Crocker, N., Rallan, R., Varsani, S., Montgomery, D., … Taylor, I.
(2004). Association between a functional polymorphism in the serotonin transporter
gene and diarrhoea predominant irritable bowel syndrome in women. Gut, 53, 1452–
1458.
Zautra, A., Affleck, G., Tennen, H., Reich, J., & Davis, M. (2005). Dynamic approaches
to emotions and stress in everyday life: Bolger and Zuckerman reloaded with positive
as well as negative affects. Journal of Personality, 73, 1511–1538.
Zobel, A., Barkow, K., Schulze-Rauschenbach, S., von Widdern, O., Metten, M., Pfeiffer,
U., … Maier, W. (2004). High neuroticism and depressive temperament are associated
with dysfunctional regulation of the hypothalamic-pituitary-adrenocortical system in
healthy volunteers. Acta Psychiatrica Scandinavica, 109, 392–399.
Extraversion
Joshua Wilt and William Revelle
Abstract
This chapter provides a review of extraversion, defined as a dimension of
personality reflecting individual differences in the tendencies to experience and
exhibit positive affect, assertive behavior, decisive thinking, and desires for social
attention. Extraversion is one of five basic tendencies in the Five Factor Model
(FFM) of personality. In the FFM, basic tendencies are conceptualized as
including the following characteristics. They are organized hierarchically, based
in biology, develop over time according to intrinsic maturation principles, are
manifested in characteristic adaptations (i.e., are expressed in affective,
behavioral, and cognitive tendencies), influence one’s objective biography, are
reflected in the self-concept, and have both adaptive and maladaptive variants.
This chapter is organized around the theory and research on extraversion relevant
to each of the aforementioned characteristics.
Key Words: extraversion, surgency, gregariousness, friendliness, assertiveness,
leadership, sociability
Personality trait dimensions are abstractions used to describe and explain

consistency and coherence in affect, behavior, cognition, and desire—the
“ABCDs” of personality (Ortony, Norman, & Revelle, 2005; Revelle, 2008)—
over time and space. Introversion–extraversion (referred to from here on as
extraversion) is a higher order dimension of personality reflecting tendencies to
experience and exhibit positive affect, assertive behavior, decisive thinking, and
desires for social attention (Wilt & Revelle, 2009). More extraverted individuals
are characterized by energy, dominance, spontaneity, and sociability, whereas
more introverted individuals tend to be described as more lethargic, inhibited,
reflective, and quiet.
Generally speaking, it is important to study extraversion due to its emergence
as one of the basic and fundamental dimensions in almost all current theories
and taxonomies of normal personality traits (Ashton & Lee, 2007; Costa &
McCrae, 1992; Eysenck & Himmelweit, 1947; Goldberg, 1990; Hogan, 1982;
Norman, 1963), its role in contributing to effective functioning and well-being in
a number of different domains (Lucas & Fujita, 2000; Magee, Heaven, & Miller,
2013; Ozer & Benet-Martinez, 2006; Roberts, Kuncel, Shiner, Caspi, &
Goldberg, 2007), and its relations to various forms of psychopathology
(Krueger, Caspi, Moffitt, Silva, & McGee, 1996; Widiger, 2005). This chapter
focuses in particular on extraversion from the framework of the Five Factor
Model (FFM; Costa & McCrae, 1992; McCrae & Costa, 2008) of personality.
Although it is recognized that extraversion in the FFM is similar to the
extraversion constructs in other theories of personality in terms of its conceptual
and operational natures (McCrae & Costa, 2008), and indeed all studies of
extraversion are likely tapping into common features of the trait, it is worth
noting that the FFM provides a unique perspective from which to view and
organize current theory and research on extraversion. FFM categorizes
Extraversion as one of the five basic tendencies (along with Neuroticism,
Openness, Agreeableness, and Conscientiousness). In the FFM, basic tendencies
are conceptualized as being organized hierarchically, biologically based,
developing over time according to intrinsic maturation principles, being
manifested in characteristic adaptations (i.e., are expressed in affective,
behavioral, and cognitive tendencies), influencing our objective biography,
being reflected in the self-concept, and possibly having both adaptive and
maladaptive variants. This chapter is organized around the research relevant to
each of the aforementioned characteristics as they apply to extraversion with the
aim of providing an overview of what is known about this important trait.
Extraversion as a Basic Tendency

Extraversion Is in All Prominent Models
C. G. Jung (Jung, 1921/1971) first introduced the term extraversion,
describing more extraverted individuals as being more focused on the outer
world, in contrast to more introverted individuals who were thought to be
focused more on their own inner mentality. Extraversion for Jung was engaging
with the world, whereas introversion was being drawn inward into thought.
Although Jung originated the name, Gerard Heymans and Wilhelm Wundt
perhaps did more to establish the empirical basis for studying extraversion.
Heymans and Wiersma (1909), using early techniques that were crude cousins of
factor analysis, identified extraversion along a continuum of “strong” and
“weak” functioning (Eysenck, 1992). Reanalysis of the original data using factor
analysis has confirmed the presence of a factor similar to extraversion, ranging
from energetic to submissive (Van der Werff, 1985). Wundt (1897) reorganized
the temperaments of Hippocrates and Galen into two dimensions, changeability
and excitability; the choleric and sanguine temperaments were thought to be
more changeable, whereas the melancholic and phlegmatic temperaments were
conceptualized as being less changeable. The changeability dimension was later
conceptualized as extraversion by Hans Eysenck (Eysenck, 1981; Eysenck &
Himmelweit, 1947).
Perhaps nobody has done as much for extraversion as Eysenck, and he
demonstrated the importance of extraversion as a fundamental dimension of
personality in a series of experimental and taxometric studies (Eysenck, 1952;
Eysenck & Himmelweit, 1947). He was one of the first to attempt a thorough
description and measurement of extraversion that evolved with the development
of several inventories—the Maudsley Personality Questionnaire (MPQ;
Eysenck, 1959), the Eysenck Personality Inventory (EPI; Eysenck & Eysenck,
1964), the Eysenck Personality Questionnaire (EPQ; Eysenck & Eysenck, 1975),
and the Eysenck Personality Profile (EPP; Eysenck & Wilson, 1991)—the
content of which included items mainly assessing sociability and impulsivity in
varying proportions depending on the inventory (Rocklin & Revelle, 1981).
Another early measure that also went through several versions (Guilford &
Guilford, 1934) and that deserves mention is the Guilford–Zimmerman
Temperament Survey (GZTS; Guilford & Zimmerman, 1949). The GZTS
included a dimension defined at one end by the tendency for quiet reflection and
at the other end by impulsivity (named introversion–extraversion), and it
included yet another dimension that contained sociability content similar to
Eysenck’s extraversion. The differences between Eysenck’s and Guilford’s
conceptualizations of extraversion led to an influential debate about the
appropriate content of extraversion (Eysenck, 1977; Guilford, 1975, 1977).
Extraversion consistently emerged from early lexical analyses aimed at
determining the fundamental dimensions of personality (see also the chapter by
De Raad and Mlačić). Allport and Odbert’s (1936) list of trait words extracted
from an unabridged dictionary formed the basis for Raymond Cattell’s
Herculean efforts to catalogue and organize the trait domain (e.g., Cattell,
1943a,b, 1947). Over a number of years, Cattell narrowed Allport and Odbert’s
(1936) list of trait adjectives to 171 paragraph descriptors, then 35 paragraph
descriptors, and finally through factor analysis to 12 factors and four additional
scales that in turn were measured by the 16PF inventory of primary personality
factors (Cattell, 1947). In the 16PF, a higher order factor of extraversion
encompasses five of the primary factors that together contain content reflecting
impulsivity, sociability, and ascendance (Cattell, 1957). The work of Fiske
(1949) and Tupes and Christal (1961) examined the structure of peer ratings
based on the paragraph descriptors of Cattell and consistently found five factors,
one of which was labeled surgency or extraversion. This work laid the
foundation for Warren Norman’s (1963) seminal factor analysis on what he
considered to be the best marker scales from Tupes and Christal (1961) that
revealed what are now known as the Big Five factors of personality (Goldberg,
1990).
Each of the Big Five is conceptualized as a broad factor subsuming a number
of narrower traits. The first factor in the Big Five, Surgency (also called
extraversion), consists of more specific traits such as talkative, energetic,
assertive, and adventurous. Lewis Goldberg (1993) and John Digman (1990)
have perhaps conducted the most rigorous and influential research attesting to
the validity of the Big Five structure, which was replicated in the languages of
many different cultures (Goldberg, 1990, 1992). A wide range of inventories has
been developed over the past 20 years to assess the Big Five (e.g., Goldberg,
1992; Hofstee, de Raad, & Goldberg, 1992; John, Donahue, & Kentle, 1991;
Rammstedt & John, 2007; see also the chapter by Simms, Williams, and
Simms). The most extensive assessment of the Big Five is the Abridged Big Five
Circumplex (AB5C; Hofstee et al., 1992); the adjectives included in this
inventory have high loadings on two factors1 (each adjective has a primary
loading on one factor and a secondary loading on the second one) such that pairs
of the Big Five dimensions have a circumplex structure. Facets including items
with their primary loadings in the extraversion domain in this inventory are
gregariousness, friendliness, assertiveness, poise, leadership, provocativeness,
self-disclosure, talkativeness, and sociability.
The FFM (McCrae & Costa, 2008) identifies personality dimensions similar
to the Big Five and has also been replicated across many cultures. Although
often used synonymously with the Big Five, the FFM was derived from factor
analysis of questionnaires rather than adjectives. The most comprehensive
instrument used to assess the traits in the FFM, the NEO Personality Inventory-
Revised (NEO PI-R; Costa & McCrae, 1992), assumes a hierarchical structure
with each higher order factor seen as the aggregate of six lower order facets.
Extraversion’s lower order facets in this inventory are warmth, gregariousness,
assertiveness, activity, excitement seeking, and positive emotion. DeYoung,
Quilty, and Peterson (2007) recently contributed an important addition to the
assessment of Big Five/FFM traits by developing the Big Five Aspects Scales
(BFAS), an inventory that was empirically derived from the NEO PI-R and an
open-source measure of the AB5C included in the International Personality Item
Pool (IPIP) (Goldberg, 1999; Goldberg et al., 2006). In the BFAS, extraversion
is represented by the lower order aspects of enthusiasm and assertiveness.
Various other models of traits identify extraversion as a basic dimension of
personality. Also relying on factor analysis of adjectives from the dictionary,
Tellegen (1985) developed a seven-factor taxonomy including five factors that
resemble the Big Five and two additional factors of positive and negative
evaluation. Tellegen’s (1982) inventory, the Minnesota Personality
Questionnaire, operationalizes extraversion hierarchically as well, with its lower
order facets termed well-being, social potency, social closeness, and
achievement. Hogan’s (1982) Socioanalytic Theory includes a higher order
factor similar to extraversion that consists of the facets sociability and ambition,
and the HEXACO model of personality (Ashton & Lee, 2001, 2007; Lee &
Ashton, 2004) represents extraversion with four lower order facets termed
expressiveness, liveliness, sociability, and social boldness.
Defining Extraversion’s Lower Order Structure

The idea from FFM that basic tendencies are hierarchically organized is borne
out in the previous description of theories and inventories including
extraversion, as most of the inventories include lower order facets that together
comprise the higher order trait of extraversion. There is clearly quite a bit of
overlap across inventories in terms of what content is included in the
extraversion domain, although there are also differences in which content is
emphasized as well as the overall breadth of coverage. The difficulty of defining
facets in a nonarbitrary way (Costa & McCrae, 1998) perhaps contributes to
some of the inconsistency in the operationalization of extraversion across
inventories. This presents a challenge for moving toward a better understanding
of extraversion because a detailed and precise knowledge of lower order
constructs is necessary in order to achieve a greater comprehension of the broad
trait (McCrae & Costa, 1992).
Notwithstanding the inherent difficulties in generating a comprehensive list of
lower order constructs within a trait domain, the NEO PI-R (Costa & McCrae,
1992) may be viewed as a prototypical example of a traditional hierarchical
representation of facet structure. The NEO PI-R was rationally derived through
extensive literature reviews, theory building, and intuition, and it shows good
convergent and discriminant validity (McCrae & Costa, 1992). The NEO PI-R
(and many other inventories designed to assess the hierarchical structure of traits
described earlier) assumes a simple structure, in which lower order facets are
thought to associate with only one higher order trait and are empirically
associated with each other only through the shared variance of the latent, higher
order trait that they have in common. In the case of the NEO PI-R extraversion,
this implies that the facets of warmth, gregariousness, assertiveness, activity,
excitement seeking, and positive emotion are all thought to be connected through
some common process.
Advocates of the FFM have proposed that the common process is the
disposition to engage in social behavior (McCrae & Costa, 1997), whereas other
theorists have posited that the core of extraversion is positive emotionality
(Tellegen, 1985; Watson & Clark, 1997) or the propensity to seek social
attention (Ashton, Lee, & Paunonen, 2002). However, as Hofstee, de Raad, and
Goldberg (1992) showed in their development of the AB5C, personality
inventories do not typically show simple structure empirically, as lower order
constructs across different traits tend to be associated with one another even
when controlling for the variance of higher order factors (see also the chapter by
Wright). Additionally, the use of simple structure as applied to personality data
leads to imprecise and inaccurate solutions (Pettersson & Turkheimer, 2014).
These findings raise questions about the traditional way of thinking about the
hierarchical structure of extraversion (and traits in general), and so alternative
ways of representing traits at different levels of breadth deserve consideration.
Eysenck (1970) proposed an intriguing multilevel representation of
extraversion that consisted of four distinct levels of abstraction. Specific
responses are the most narrow level in this schematic and comprise individual
behaviors such as flirting on a single occasion. One level above are habitual
responses such as behaving in a lively manner at various parties on recurring
occasions. At the next level are facet-level constructs, such as gregariousness,
and finally at the highest level of the hierarchy is the broad trait of extraversion.
This hierarchy is unique in at least two respects. First, constructs at lower levels
are specified as residing within only one higher order level. Second, Eysenck did
not specify links between constructs included at the same level of a hierarchy
nor between constructs across levels. This type of hierarchy may be a useful way
of understanding how extraversion manifests at different levels of specificity;
however, little research has examined whether this elegant hypothetical structure
can be operationalized reliably.
Another unique perspective on higher order traits is provided by the
increasingly popular network perspective (Borsboom & Cramer, 2013; Cramer
et al., 2012; Goekoop, Goekoop, & Scholte, 2012; van Os, Lataster, Delespaul,
Wichers, & Myin-Germeys, 2014). The network perspective proposes that
higher order traits such as extraversion emerge from the interactions between
lower order affective, behavioral, and cognitive (ABC) states. In contrast to the
latent variable view of traditional trait hierarchies, the lower order constructs are
thought to be correlated due to their influences on each other over time rather
than sharing a common, higher order factor. This view therefore shifts the focus
of organization away from the trait level to the trait’s more narrow aspects. As
applied to extraversion, the network approach, in stark contrast to Eysenck’s
hierarchy, posits strong associations between specific ABC constructs within the
domain of extraversion. For an empirical example of how NEO PI-R
extraversion can be visualized from the network perspective, see Cramer et al.
(2012). This approach is appealing due to its ability to accommodate the
complexity of associations between lower order levels of personality, but issues
regarding how best to model network variables and the relegation of higher order
traits such as extraversion to summary variables with no causal or organizing
influence demand attention.
In addition to issues pertaining to how best to organize the traits residing in
the extraversion domain, there is also debate about just which traits should be
considered a part of extraversion. Perhaps the most longstanding and lively
debate has centered around whether constructs similar to impulsivity should be
thought of as lower order features of extraversion (Revelle, 1997). Eysenck
included impulsivity content in his original inventories, yet he appeared to be
ambivalent about whether impulsivity should be thought of as a central feature.
Analyses of the structure of the EPI and the EPQ showed that the greatest
difference between the inventories was that the EPI contained a substantial
amount of both sociability and impulsivity, whereas the EPQ contained much
more sociability than impulsivity (Rocklin & Revelle, 1981). Jeffrey Gray’s
original version of Reinforcement Sensitivity Theory (Gray, 1970, 1981, 1982)
conceptualized Eysenck’s extraversion as impulsivity minus anxiety;
specifically, Gray believed that impulsivity graphically rotated 45 degrees in
conceptual space from extraversion.
Zuckerman (1991) likewise included a construct similar to impulsivity in his
general theory of personality, identifying a factor of sensation seeking that
reflects a lack of planning, impulsive decision making, and taking risks for the
sake of novelty. In yet another model of personality to grapple with issues about
distinguishing among the aforementioned constructs, Cloninger, Svrakic, and
Przybeck (1993) considered impulsivity to be part of a factor labeled novelty
seeking that also contained approach behavior, high responsivity to reward, and
a quickness to lose one’s temper. Although still far from settling this debate,
studies including a wide array of scales covering the terrain of extraversion,
impulsivity, and sensation seeking have been making progress toward
delineating the structure of these constructs. Factor analyses of the NEO PI-R
and various impulsivity and sensation-seeking scales showed that some forms of
impulsivity were more similar to NEO PI-R conscientiousness, whereas
sensation seeking emerged as more highly associated with NEO PI-R
extraversion (Whiteside & Lynam, 2001). Additionally, a recent study (Quilty,
DeYoung, Oakman, & Bagby, 2014) used confirmatory factor analysis to show
that sensation seeking is related to but not subsumed by extraversion’s aspects of
assertiveness and enthusiasm.
The ABCD Approach as an Organizing Framework

The disagreements about extraversion’s content should not be discouraging to
those hoping for rapprochement regarding the scope of the extraversion domain;
rather, it is just this type of healthy scientific debate that produces advancement
in knowledge. Steps toward a synthesis of current ideas will likely arise out of an
organizing framework that is capable of sorting out the differences among
prominent theories’ conceptualizations of extraversion. The aforementioned
affect, behavior, cognition, and desire (ABCD) approach to personality might be
one way of integrating different theories and operationalizations of traits (Ortony
et al., 2005; Revelle, Wilt, & Condon, 2011; Wilt, Oehlberg, & Revelle, 2011).
Putting this approach to the test, Rauthmann and Will (2011) showed that
recurrent themes in the scientific literature pertaining to the trait of
Machiavellianism can be organized coherently into ABCD aspects. To
understand this approach, it is useful to first define the ABCD domains.
Despite its ostensibly intuitive nature, there have been inconsistencies in the
way the term behavior has been applied in the field of personality. Adopted
herein is the definition of behavior offered by (1) Furr (2009)—“behaviour may
be defined as verbal utterances (excluding verbal reports in psychological
assessment contexts) or movements that are potentially available to careful
observers using normal sensory processes” (p. 372)—and added to it (2) the
observation of Ortony et al. (2005) that behavior encompasses physical actions
that may not be observable through normal sensory processes (e.g., contractions
of the gut). Behavior is how the mental processes of affect, cognition, and
motivation manifest themselves and become tangible and concrete (Shweder,
1999). Descriptions of affect tend to converge on the definition of affect as a
higher order category subsuming valenced condition such as moods, emotions,
feelings, feeling-like states, and preferences (Ortony et al., 2005; Pytlik Zillig,
Hemenover, & Dienstbier, 2002; Scherer, 1995). Cognition, or cognitive
activity, is also thought of as a higher order category and encompasses mental
contents and processes (Gruszka, Matthews, & Szymura, 2010). Attention,
memory, knowledge, problem solving, beliefs, appraisals, interpretations,
representations, and expectations are all included in the domain of cognition
(Cervone, 2004; Ellsworth, 1994; Ortony et al., 2005; Revelle, 1995). The
domain of desire refers to people’s (conscious or unconscious) goals, needs,
wants, and wishes (Winter, John, Stewart, Klohnen, & Duncan, 1998). What
binds these constructs in common is that they represent states that people would
like to bring about or to prevent (Chulef, Read, & Walsh, 2001).2 By relying on
basic dimensions of phenomenological experience, the ABCD approach defines,
clarifies, and explicates the meaning of traits in terms of psychological content.
Thus, it can bring a more logical and meaningful organization to the
characteristics that together characterize extraversion.
Across the inventories summarized previously, extraversion is defined by
themes such as enthusiasm, assertiveness, sociability, dominance, agency,
gregariousness, and warmth. Although these terms together do well to describe
the landscape of extraversion, they provide little insight into the dynamic ABCD
processes that differentiate individuals residing at different levels of the
extraversion continuum. In contrast, take, for example, the results of an initial
effort to delineate extraversion by its ABCD components (Wilt, 2014).
Extraversion items from the International Personality Item Pool (Goldberg,
1999; Goldberg et al., 2006) versions of the NEO PI-R and AB5C that were
rated by experts as containing relatively high amounts of A, B, C, and D content
were aggregated into scales reflecting the ABCD components of extraversion
(see Table 4.1). These items together paint a coherent psychological portrait of
the extraversion continuum as comprising positive affects (especially around
people), gregarious and talkative behaviors, spontaneous and decisive
cognitions, and desires for attention and influence.
Table 4.1. Extraversion Items and ABCD Content

AB5C Primary NEO Domain A B C D
Affect items
Love surprise parties. E 90 2 3 5
Love excitement. E 77 10 7 7
Feel comfortable around people. E E 76 4 13 8
Have a lot of fun. E E 74 10 10 6
Express childlike joy. E E 72 26 2 1
Dislike neighbors living too close. E- 72 9 4 15
Often feel uncomfortable around others. E- E- 70 10 13 8
Behavior items
Make a lot of noise. E 6 83 6 6
Speak loudly. E 7 82 5 7
Start conversations. E 7 81 5 8
Speak softly. E- 10 79 3 8
Am the first to act. E 7 79 4 10
Don’t talk a lot. E- 8 78 7 7
Never stop talking. E 7 78 7 8
Cognition items
Come up with a solution right away. E 6 27 59 10
Know no limits. E 13 23 48 16
Know how to captivate people. E 13 29 45 13
Can take strong measures. E 11 35 43 12
Know what I want. E 9 1 36 54
React quickly. E 23 39 34 3
Let things proceed at their own pace. E- 15 37 33 15
Desire items
Seek to influence others. E 6 15 21 58
Seek adventure. E 11 23 9 57
Demand to be the center of interest. E 14 26 5 55
Know what I want. E 9 1 36 54
Seek quiet. E- E- 9 35 11 45
Try to lead others. E E 5 43 8 44
Can easily push myself forward. E 14 26 20 40
Note: The second and third columns show trait domains for each item for the IPIP-AB5C and IPIP-NEO PI-
R. Trait domain indicators with a “-” sign next to them signify that the item was reverse-scored with respect
to extraversion. Numbers indicate the average percentage of ABCD content for each item calculated across
independent raters.
The ABCDs are just one possibility for organizing the facet-level constructs
of extraversion in meaningful ways. Another conceptualization about the nature
of traits that shows promise is the division of traits into the situations in which
trait-relevant behaviors are carried out and the explanations for those behaviors
(Yang et al., 2014). For example, the situation of meeting new people at a party
might elicit conversation for the more extraverted individual, because she or he
believes that will facilitate social connections. For individuals who are more
introverted, a party might send them in search of a quiet spot alone because they
are overwhelmed by the pressure to interact socially. Read et al. (2010) provide
an excellent review and simulation study showing how these scenarios may play
out in dynamic fashion. Future research may seek points of contact and
departure between this approach, the ABCD approach, and other intriguing
explanatory models of traits (e.g., DeYoung, 2015; Fleeson, 2012; Read et al.,
2010), with the overarching and related aims of refining the conceptual
definition of extraversion and devising more accurate assessment techniques for
all levels of the extraversion hierarchy.
Evolution, Genetics, and Biology

Interest in the physiological bases of human personality dates back at least
2,500 years to the linking of bodily humors to the four temperaments described
by Hippocrates and Galen: blood for sanguine, yellow bile for choleric, black
bile for melancholic, and phlegm for phlegmatic (Stelmack & Stalikas, 1991).
Current theories of the evolutionary, genetic, and neurophysiological
underpinnings of personality, however, differ dramatically from their origins.
The fundamental notion that any logical explanation of traits needs to be
consistent with basic biology though remains as true now as it did then. The
FFM (McCrae & Costa, 2008) proposes that the five basic tendencies all have a
strong biological foundation. We now review research on the topic of the
evolutionary, genetic, and biological basis of extraversion.
Evolutionary Perspectives
The idea that traits evolved as strategies to meet adaptive challenges in the
social environment is a popular notion among evolutionary psychologists (e.g.,
Buss, 2009; Denissen & Penke, 2008; Nettle, 2006). Genetic polymorphisms that
relate to variations in traits, such as extraversion, can be maintained by natural
selection in a number of ways (Buss, 1991, 2009; Nettle, 2006). Selection
pressures vary over time or geographic location; as such, different phenotypes
may become more or less adaptive. Natural selection can also maintain variation
in traits in the case of frequency dependent selection, in which the fitness of a
phenotype depends on its frequency relative to other phenotypes in a given
population. As applied to extraversion, the evolutionary result of any or all of
these circumstances would be between-person variations in genes that bias
individuals toward developing more extraverted or introverted phenotypes.
Studies showing that extraversion is moderately heritable, h2 = 0.45–0.50,
with little if any shared environmental influence (Bouchard & Loehlin, 2001),
support the idea that extraversion has a substantial genetic basis (see also the
chapter by South). Establishing heritability is the first step in uncovering specific
genetic pathways, with optimistic theorists positing that extraversion may
eventually be linked to specific polymorphic genes (Munafò, 2009; Penke,
Denissen, & Miller, 2007). There has been some progress on this front, as
extraversion has been associated with several genetic polymorphisms (Canli,
2006; Ebstein, Benjamin, & Belmaker, 2003; Luo, Kranzler, Zuo, Wang, &
Gelernter, 2007).
Compelling evidence for the genetic basis of extraversion also comes from
studies of nonhuman animals. If extraversion was simply a byproduct of human
culture, traits similar to extraversion would not be expected to be found in other
species. However, Gosling and John (1999) synthesized research on personality
factors in nonhuman animals and found that factor labels that reside in the
domain of extraversion were nearly ubiquitous across species. For example,
individual differences in pigs and rhesus monkeys can be described by
sociability, dogs and cats by energy, and octopi by approach-avoidance
tendencies. In a vivid example, it was noted that more introverted octopi tend to
stay in their dens and hide themselves by changing color and releasing ink.
The variation in extraversion across a multitude of species raises questions
about how different levels of extraversion contribute to fitness. Nettle (2005,
2006) has proposed that there are fitness trade-offs at the poles of the
extraversion continuum. A potential fitness benefit of higher extraversion may
be the enhanced ability to form and sustain interpersonal relationships (Ashton
& Lee, 2007; Nettle, 2005). Indeed, extraversion promotes social status and
more extraverted individuals may enjoy the benefits of greater social influence
and dominance (Anderson, John, Keltner, & Kring, 2001). Importantly, for
arguments about whether the social benefits of extraversion actually increase
fitness, extraversion is related to having more sexual partners (Nettle, 2005). So,
why don’t we live in a world of all extraverts? For one, more extraverted
individuals may expose themselves to more safety risks as indicated by being
hospitalized more for injuries (Nettle, 2005), perhaps due in part to engaging in
increased antagonistic competition (Schaller & Murray, 2008). Introversion is
also a protective factor against exposure to infectious illness, and thus
geographic regions with high infectious disease prevalence may select for
genetic polymorphisms that bias individuals toward introversion (Nettle, 2005;
Schaller & Murray, 2008).
An example of how different levels of extraversion may be more or less
adaptive depending on the environment comes from a study conducted by
Camperio Ciani, Capiluppi, Veronese, and Sartori (2007) that assessed the
personality of people living on the mainland in Italy and on the small islands off
the mainland. It was found that compared to people living on the mainland and
recent immigrants to the islands, individuals from families that inhabited the
islands for 20 generations or more were less extraverted. Furthermore, emigrants
from the islands were more extraverted than islanders who never emigrated.
Camperio Ciani et al. (2007) proposed that selective emigration from the islands
based on genetic differences is the underlying cause for these population
differences in extraversion. More generally, genetically driven selective
emigration might be one plausible contributor to differences in extraversion (or
any personality trait) across regions within the same country (Rentfrow et al.,
2013) as well as across countries (McCrae & Terracciano, 2005).
In contrast to the view of traits as evolutionarily adaptive, Tooby and
Cosmides (1990) described an alternative model in which individuals engage in
facultative calibration of their traits to personal and environmental cues over the
course of development. That is, given a certain set of environmental conditions
or physical characteristics, individuals will differ in their behavioral strategies
based on which strategies are most adaptive for those specific circumstances. An
example of how this might play out comes from a study showing that physical
attractiveness and strength explained a large portion of the variance in
extraversion scores (Lukaszewski & Roney, 2011). Lukaszewski and Roney
reasoned that because reproductive success rates for extraverted behavioral
strategies are likely to depend in part on these physical qualities, stronger and
more attractive individuals tend to favor extraversion so as to increase their
likelihood of obtaining mates.
Have different levels of extraversion contributed to adaptive fitness across
phylogenetic history, or has extraversion simply calibrated over the course of
ontogeny? Questions such as these are likely to stir controversy, but they are also
likely to keep areas of study related to evolutionary personality psychology
moving forward rapidly. After overcoming early criticisms that evolutionary
topics were not amenable to empirical tests, researchers have found creative
ways to operationalize hypotheses based on evolutionary theories. Future
research will require even greater innovation, but it will be critical to meet the
many challenges that lie in wait if personality is to realize the aspiration of
having evolutionary theory as its meta-theoretical anchor (Ashton & Lee, 2001;
Buss, 1995; McAdams & Pals, 2006).
Brain Systems Underlying Variation in Extraversion

Evolutionary and genetic influences represent the most distal steps in tracing
the biological underpinnings of extraversion. Genes do not act directly on
behavior; rather, their effects are mediated by brain structure and function.
Research investigating how individual differences in brain systems may lead to
variations in extraversion has a rich history, starting with two titans (Eysenck
and Gray) in the history of biological theories for extraversion (see also the
chapter by Allen and DeYoung).
Eysenck and Gray

The now famous debate between Hans Eysenck and Jeffrey Gray marks the
beginning of contemporary theories about the neurobiological basis of
extraversion (Matthews & Gilliland, 1999). Eysenck laid the groundwork for
biological theorizing with his arousal hypothesis of extraversion (Eysenck,
1967). The basis of Eysenck’s theory was the idea that a person’s extraversion
was dependent upon their threshold for arousal in the ascending reticular
activating system, or ARAS (the ARAS is a feedback loop connecting the cortex
to the reticular activating system). Eysenck chose the ARAS because of its
known roles in attention and learning (Eysenck, 1973), two processes that he had
long believed were integral sources of individual differences in extraversion
(Eysenck, 1957). Eysenck posited that extraversion was related to higher
thresholds for arousal and thus to lower levels of cortical arousal at baseline.
Based on Wundt’s notion that people try to maintain moderate arousal (Wundt,
1897), Eysenck believed that this low arousal at baseline could explain the
relations between extraversion and the pursuit of stimulating activities such as
stimulant drugs (cigarettes), sexual activities, and social interaction. However,
problematic for Eysenck’s theory are studies showing that resting brain activity
rarely differs as a function of extraversion (Stelmack, 1990, 1997), as well as
electroencephalography (EEG) and functional magnetic resonance imaging
(fMRI) studies finding that the association between extraversion and cortical
arousal is sometimes positive and sometimes negative (Matthews & Gilliland,
1999; Zuckerman, 2005). Additionally, arousal-based frameworks are limited in
capturing key components of extraversion concerning reward processing,
incentive motivation, and behavioral approach (Depue & Collins, 1999; Gray,
1981; Smillie, Pickering, & Jackson, 2006).
Jeffrey Gray proposed an alternative causal theory of extraversion that was
well-suited to explain extraversion’s relations to approach processes, termed
Reinforcement Sensitivity Theory (RST; Gray, 1970, 1981, 1982). The original
formulation of RST, which was based on animal research, postulated the
existence of three separate neural systems underlying behavior: (1) the
Behavioral Approach System (BAS), (2) the Behavioral Inhibition System
(BIS), and (3) the Fight–Flight System (FFS). The primary emphasis was on the
effects of the BIS and BAS. Sensitivity of the BAS was thought to underlie trait
impulsivity, and sensitivity of the BIS was thought to underlie trait anxiety.
These traits were conceptualized as primary traits that together could explain
Eysenck’s extraversion. Specifically, Eysenck’s extraversion was thought to be
impulsivity minus anxiety.
Gray’s theory has undergone drastic revisions that are beyond the scope of
this chapter (Corr, 2008; Gray & McNaughton, 2000; Smillie, 2015; Smillie et
al., 2006),3 but it is worth mentioning that the BAS is thought to mediate
reactions to all appetitive stimuli and to produce various characteristics
associated with extraversion: the emotion of anticipatory pleasure and the pursuit
of rewarding, impulsive, and risky behaviors (Corr & Cooper, 2015). Indeed,
although Gray described only one system for approach, he did not rule out the
possibility that approach processes are multidimensional. From an evolutionary
standpoint, the diversification of approach systems would be consistent with an
evolutionary arms race in which predators must evolve multiple strategies for
catching their prey (Dawkins & Krebs, 1979). The penalty for a failed approach
is not as severe as for failed avoidance according to the so-called “life-dinner”
principle; as the predator loses a meal, the prey would lose its life (Dawkins &
Krebs, 1979). Psychometric assessments of the BAS (Carver & White, 1994;
Corr, 2008) reflect the complexity of approach, with scales assessing varied
components such as interest in rewards, emotional reactivity to rewards,
persistence in obtaining rewards, pleasure-seeking behavior, and impulsive
obtainment of an incentive.
Given their conceptual similarities, it is surprising that more research has not
explored the associations between BAS-related characteristics and extraversion.
In the first study examining the relationships between Carver and White’s (1994)
BIS/BAS scales and extraversion, Smits and Boeck (2006) found that the overall
BAS scale and all three of the subscales (drive, fun-seeking, and reward-
reactivity) were positively associated with extraversion. Likewise, Keiser and
Ross (2011) found a positive relationship between Carver and White’s (1994)
total BAS scale and extraversion. In the only study to examine links between
Carver and White’s (1994) BAS scales and the NEO PI-R, Segarra, Poy, López,
and Moltó (2014) showed that BAS fun-seeking was uniquely related to the
facets of warmth, gregariousness, activity, excitement-seeking, and positive
emotions; reward responsivity was related to warmth, activity, excitement-
seeking, and positive emotions; and drive was related only to assertiveness.
Further specification of the associations between approach-related phenomena
and extraversion has the potential to better situate revised-RST (Gray &
McNaughton, 2000) as a viable biological foundation for the FFM of
extraversion (Wilt & Revelle, 2009).
Reward Processing and Dopaminergic Functioning

Gray’s efforts were just the starting points in relating extraversion to brain
mechanisms engaged in reward processing. Depue (1995) proposed a
neurological Behavioral Facilitation System (BFS) as the causal basis for agentic
components of extraversion (Depue, 1995; Depue & Collins, 1999).4 The
neuroanatomical correlate of the BFS is hypothesized as the mesocorticolimbic
dopaminergic system, which is integral in desire and reward and is thought to
facilitate behavioral approach by increasing the salience of positive stimuli.
Depue’s model of behavioral facilitation is a threshold model in that dopamine
must reach a certain level for approach behavior to be elicited. Thus, approach
behavior is thought to depend on tonic level of dopamine as well as on phasic
level (Depue, 1995). A growing body of evidence directly implicates
dopaminergic function in extraversion (e.g., Depue & Collins, 1999; Wacker,
Chavanon, & Stemmler, 2006). DeYoung (2010) and Smillie (2008) reviewed
seminal work that has linked extraversion to genetic variations in dopamine
function and reward-seeking behavior (Reuter, 2007), size of brain structures
involved in reward processing (Omura, Constable, & Canli, 2005; Rauch et al.,
2005), brain activity in response to rewarding stimuli (Canli, 2004; Rauch et al.,
2005), and responses to psychotropic drugs that influence the functioning of
dopamine (Rammsayer, 1998; Wacker et al., 2006).
Animal studies have also generated evidence in support of dopamine’s
involvement in reward processing. Drugs that increase dopamine (dopamine
agonists), such as amphetamines, have been shown to increase the degree to
which rats pursue rewards (Wyvell & Berridge, 2000). In contrast, drugs that
block dopamine (dopamine antagonists) decrease reward-seeking behaviors
(Wise, 2004). Mice bred without the ability to synthesize dopamine show
deficits in reward-seeking behavior; however, if dopamine production is restored
in the dorsal striatum of those mice via gene therapy, they exhibit increases in
goal pursuit (Robinson, Sotak, During, & Palmiter, 2006).
It is notable that the studies on animals have concerned the role of dopamine
in the pursuit of reward without mentioning consumption of rewards. Indeed,
this is consistent with descriptions of separate reward systems for mediating
appetitive, incentive-seeking behaviors (“wanting”), as opposed to
consummatory behaviors (“liking”) (Berridge, 2007, 2012). The idea that
dopaminergic functioning underlies reward-seeking behaviors rather than
reward-liking behaviors emerges from this distinction. Taking another step
forward in this line of reasoning leads to the hypothesis that extraversion, due to
its association with dopaminergic functioning, should in turn be associated
specifically with features of reward-seeking behavior instead of reward-liking
behavior. That is, more extraverted people should respond with energized and
excited affects when in appetitive situations; conversely, extraversion should not
be related to increases in pleasantness when simply enjoying a reward. These
two predictions have been supported across a series of recent studies (Smillie,
Cooper, Wilt, & Revelle, 2012; Smillie, Geaney, Wilt, Cooper, & Revelle,
2013). More extraverted people consistently responded with higher levels of
energy and vigor to situations meant to elicit reward-pursuit behavior (e.g.,
imagining buying a lottery ticket and winning); however, extraversion did not
relate to an experience of pleasantness when people were presented with merely
pleasant scenarios that lacked a reward-pursuit component (e.g., imagining lying
on a beach on a pleasant day). These findings led to the specification and
narrowing of the affect-reactivity model of extraversion (Larsen & Ketelaar,
1991; Strelau, 1987), which originally stated that extraversion should relate to
reacting more strongly to all forms of positive stimuli. These studies also
illustrate how biologically informed theories can generate fruitful predictions at
the behavioral level.
Development
Evidence indicating that extraversion has a strong biological component
indicates that early forms of what will later be called extraversion should appear
when people are relatively young. Indeed, according to the FFM, genetic and
biological factors influence the development of extraversion across the lifespan
(McCrae et al., 2000) and are much more important than social/environmental
factors for shaping the trajectory of trait development in general [but see
Bleidorn, Kandler, Riemann, Angleitner, and Spinath (2012) and Wood and
Roberts (2006) for opposing viewpoints]. To begin tracking the development of
extraversion, we first examined first its temperamental origins (see also the
chapter by De Pauw).
Childhood Temperament
In the study of children, temperament refers to individual differences in
reactivity and self-control that arise from a constitutional basis (Durbin, Klein,
Hayden, Buckley, & Moerk, 2005; Rothbart, 1981). Observational studies of
infants in the laboratory show that temperamental precursors of extraversion
appear as early as 3 months, and by 6 months the familiar smiling, laughing, and
approach behaviors of extraversion are readily apparent (Rothbart, Derryberry,
& Hershey, 2000). In preschool-aged children, observational studies (Wilson,
Schalet, Hicks, & Zucker, 2013) yielded a dimension termed “anxious
introversion,” which reflects differences on a dimension characterized at one
pole by shyness and inhibition and at the other by liveliness and activity. A
factor reflecting extraversion emerges in studies of parent-reported temperament
in childhood as well (Rothbart & Bates, 1998); this factor includes activity level,
sociability, and enjoyment. In a testament to the prominence of extraversion,
parent reports identify an extraversion factor in youth as young as 3 years and up
to age 20 years (Soto & John, 2014).
Throughout childhood, features related to extraversion appear to be important
in determining how children interact with their peers. From ages 5 to 12 years,
children who are more sociable and less withdrawn are more popular and are
less likely to experience rejection (Newcomb, Bukowski, & Pattee, 1993). More
extraverted children and adolescents also tend to enjoy higher degrees of peer
support (Asendorpf & van Aken, 2003). Although this seems to be good news
for extraverted youth and their parents, it has been noted that findings such as
these highlight the importance of attending more closely to the social needs of
more introverted individuals during their formative years (Cain, 2013).
Adolescence Through Adulthood
Questions regarding how extraversion changes from adolescence through
adulthood have received a considerable amount of attention. During late
adolescence (around ages 16–20 years), extraversion increases slightly (Bleidorn
et al., 2013; Lüdtke, Roberts, Trautwein, & Nagy, 2011). Obtaining a job during
this time, however, is related to decreases in extraversion (Bleidorn et al., 2013),
perhaps suggesting that entering roles in which responsibility is valued is
conducive to introversion among adolescents. Extraversion continues to increase
during the years spent at university, at least on average (Vaidya, Gray, Haig, &
Watson, 2002). Analyses looking at change in extraversion at the level of the
individual rather than group-level change show that whereas some individuals
increase in extraversion during college (about 17%), most stay the same (80%),
and a small minority (3%) of people show decreases in extraversion (Vaidya et
al., 2002).
After emerging from adolescence and entering adulthood, extraversion
exhibits high differential stability, or rank-order stability (Lucas & Donnellan,
2011; Specht, Egloff, & Schmukle, 2011). This means that a person’s level of
extraversion will remain relatively stable in relation to the extraversion levels of
others. That is, on average, more extraverted younger adults tend to be more
extraverted older adults. Differential stability tends to be highest among middle-
aged individuals (around ages 40 to 60 years), with lower levels found in
younger and older people.
Although the rank ordering of individuals with regard to extraversion remains
relatively stable throughout adulthood, there are still interesting patterns of
change in extraversion throughout the lifespan. In a national sample of over
10,000 American adults, cross-sectional analyses showed a linear decrease in
extraversion between individuals in their thirties and in those in their eighties
(Costa et al., 1986). Cross-cultural studies of individuals between college age
and middle age have corroborated the finding that over time, extraversion
decreases slightly and in a linear fashion (McCrae et al., 1999; McCrae &
Terracciano, 2005).
A more nuanced story emerges when considering change in extraversion
among different birth cohorts and when examining change at the facet level. In a
study of three birth cohorts of men (1897–1919, 1920–1929, and 1930–1945)
over the span of 12 years (people in the study were initially ages 43 to 91 years),
Mroczek and Spiro (2003) found that the overall trajectory of extraversion by
age showed the same small linear decrease reported in previous studies. Yet the
two younger cohorts showed slight increases in extraversion, whereas the oldest
cohort showed a slight decrease.
Extraversion’s facets have distinct patterns of age-related change. Roberts,
Walton, and Viechtbauer (2006) summarized the results of 113 longitudinal
studies involving over 50,000 people and concluded that social dominance (i.e.,
independence, dominance) increases from adolescence to the thirties and then
levels out through the fifties, whereas social vitality (i.e., sociability, positive
affect) increases from adolescence to young adulthood, stays stable throughout
the fifties, and then declines slightly in old age. These findings, in conjunction
with the analyses of individual-level change in extraversion during college
(Vaidya et al., 2002), emphasize the importance of carefully investigating what
at first may appear to be relatively straightforward findings about the
development of extraversion.
Are Societies as a Whole Becoming More Extraverted?

At least in select western societies, the answer to this question is a resounding
yes. From the late 1960s to early 1990s, cross-temporal meta-analyses done on
59 studies involving over 16,000 American college students revealed that for
both men and women, extraversion has increased by nearly one standard
deviation (Twenge, 2001); however, this finding should be treated with caution
as it may be an artifact of using different scales to measure extraversion at
different times. In a study of nearly 9,000 college students in The Netherlands,
mean extraversion scores show a clear, positive trend from 1982 to 2007 (Smits,
Dolan, Vorst, Wicherts, & Timmerman, 2011). Various reasons have been put
forward for the increased prevalence of extraversion in western cultures, such as
more opportunity for contact with a wider variety of people, less strict parenting
styles, and an increase in service jobs in relation to industry (Smits et al., 2011).
It has also been suggested that western societies increasingly value outgoing and
assertive traits over more reflective and quiet characteristics (Cain, 2013). As it
seems as if these trends will continue for the foreseeable future, so may the
societal shift upward in extraversion.
Extraversion Characteristic Adaptations

The FFM proposes that basic tendencies such as extraversion should be
expressed in characteristic adaptations (McCrae, 2009). It is worth noting again,
as discussed in previous chapters, that characteristic adaptations are not simply
observable trait content. Rather, they are conceptualized as the manifest
affective, behavioral, and cognitive products of the interactions between
biologically based traits and the environment (McCrae & Costa, 2008). As noted
earlier in this chapter, desire or motivation can be added to the aforementioned
affective, behavioral, and cognitive domains, and in so doing a comprehensive
“ABCD” description of the psychological terrain of traits as they interface with
different social environments over time and space can be formed. Thus, the
issues addressed in the sections on characteristic adaptations concern whether
and how extraversion is related to different ABCDs.
The most general answer to the question of whether extraversion is related to
ABCDs in daily life can be found by examining whether the dispositional trait
of extraversion is related to the personality state of extraversion. Personality
states are short-term, concrete and contextualized patterns of ABCDs compared
to the more stable and decontextualized ABCD components of personality traits
(Bleidorn, 2009; Fleeson, 2001). Personality states may be described in the same
way as personality traits, and so extraversion states broadly encompass short-
term manifestations of vitality, assertiveness, spontaneity, and desires for social
attention. Experience-sampling studies suggest that trait measures of
extraversion indeed correlate highly with aggregate mean levels of extraversion
states (Ching et al., 2014; Fleeson & Gallagher, 2009; Heller, Komar, & Lee,
2007; Schutte, Malouff, Segrera, Wolf, & Rodgers, 2003; Wilt, Noftle, Spain, &
Fleeson, 2011), to single extraversion states, and also to the median, mode,
maximum, and minimum of the distribution of extraversion states (Fleeson &
Gallagher, 2009). These findings might be taken to indicate that those scoring
highly in extraversion are always found in extraverted states (and more
introverted individuals are always found in introverted states), but in fact this is
far from true. In actuality, people display a wide variability of extraversion states
throughout the course of their lives; sometimes highly extraverted people act
very introverted and vice versa (Fleeson, 2001). This makes good sense when
extraversion states are thought of as characteristic adaptations reflecting the
output of dispositional extraversion in combination with environmental variables
—different situations facilitate extraversion to varying degrees—even the most
extraverted individuals might remain quiet at church, whereas the most
introverted individuals will likely have a laugh (or two) at a lively party.
We next turn to a review of the relationship between extraversion and each
individual ABCD domain of characteristic adaptations. As there are literally
thousands of studies that could be categorized as addressing the ABCDs
associated with extraversion, the following sections cannot even come close to
providing an exhaustive summary of this research. Rather, they are necessarily a
focused attempt to highlight findings in each area that are potentially important,
and in sum represent a selective overview of how extraversion is manifested in
ABCDs.
Affect
One of the best-known findings in all of personality is the robust relationship
between extraversion and positive affect. Trait extraversion is related to trait
levels of positive affect (Lucas & Baird, 2004; Lucas & Fujita, 2000; Watson &
Clark, 1992), aggregated ratings of momentary positive affect (Ching et al.,
2014; Flory, Manuck, Matthews, & Muldoon, 2004; Spain, Eaton, & Funder,
2000; Wilt, Noftle, et al., 2011), and even to single ratings of current positive
affect (Lucas & Baird, 2004; Uziel, 2006). Trait extraversion appears to be
specifically more strongly related to activated positive affect—feeling happy and
energetic, as opposed to deactivated positive affect—feeling relaxed or at ease
(Smillie, DeYoung, & Hall, 2014). A growing number of studies have also
shown that being in extraverted states over the course of daily life is conducive
to experiencing higher levels of state positive affect (Ching et al., 2014; Heller et
al., 2007; Lischetzke, Pfeifer, Crayen, & Eid, 2012; Wilt, Noftle, et al., 2011).
Moreover, experiments in which participants were instructed to act extraverted
or introverted revealed a causal effect of extraversion states on positive affect,
even for introverts (Fleeson, Malanos, & Achille, 2002; McNiel & Fleeson,
2006; McNiel, Lowman, & Fleeson, 2010). The experience of positive feelings
is no doubt a core characteristic of both trait and state extraversion (Watson &
Clark, 1997). These findings raise the more fundamental question of why
extraversion is related to positive affect.
A number of explanations have been put forward for the association between
trait levels of extraversion and positive affect. The original affect-reactivity
hypothesis (Gross, Sutton, & Ketelaar, 1998) posited that extraverts, due to their
more reactive reward processing system, should exhibit stronger positive
reactions in all forms of positive situations. This hypothesis received mixed
support across a number of studies (Lucas & Baird, 2004). Studies that assessed
positive affect using terms reflecting energy and arousal found support for the
affective-reactivity hypothesis, but those that favored affective terms reflecting
pleasantness and happiness did not (Smillie et al., 2012). This led to a
specification of the affective–reactivity hypothesis (discussed previously)
indicating that extraverted people should exhibit stronger activated positive
reactions in rewarding situations, which has been replicated consistently in
experiments (Smillie et al., 2012, 2013) and which has received initial support in
natural environments (Oerlemans & Bakker, 2014). Yet as these results concern
only activated positive feelings, they fail to explain why extraversion is then
related to pleasantness and happiness.
Another explanation for the extraversion-positive affect association that has
been put forward is the social activity hypothesis (Watson, 1988; Watson, Clark,
McIntyre, & Hamaker, 1992), which states that extraversion is related to positive
affect due to greater participation in social activities. Although sensible, this
hypothesis has achieved only weak and inconsistent support across a number of
studies (Argyle & Lu, 1990; Diener, Sandvik, Pavot, & Fujita, 1992; Lucas &
Diener, 2001; Lucas, Le, & Dyrenforth, 2008; Oerlemans & Bakker, 2014;
Srivastava, Angelo, & Vallereux, 2008). A revision of the social activity
hypothesis, that extraversion is related to positive affect due to the quality (rather
than the quantity) of social experiences, has received some initial support
(Smillie, Wilt, Kabbani, Garratt, & Revelle, 2015), but awaits further replication.
Additionally, studies have identified specific mediators of the extraversion-
positive affect association, such as mood regulation abilities (Lischetzke & Eid,
2006), resilience (Lü, Wang, Liu, & Zhang, 2014), and perceived uniqueness
(Koydemir, Şimşek, & Demir, 2014). Further theoretical advances are necessary
to integrate these seemingly disparate findings into a coherent conceptual
framework.
A further explanation for the extraversion–happiness association is that trait
extraversion increases the likelihood of being in extraverted states (Fleeson &
Gallagher, 2009) that lead directly to more positive affect states (Fleeson et al.,
2002). The accumulation of positive states might therefore lead individuals
higher in extraversion to report higher levels of positive affect in general (Wilt,
Noftle, et al., 2011). Aspects of this hypothesis have been supported in multiple
experience sampling studies (Wilt, Noftle, et al., 2011) and even across multiple
cultures, including the United States, Venezuela, the Philippines, China, and
Japan (Ching et al., 2014). If the association between trait extraversion and trait
positive affect can be explained by the association between state extraversion
and state positive affect (i.e., it is what extraverts do that leads to higher levels of
positive affect), then explaining the state-level association between extraversion
and positive affect becomes necessary to understanding the trait-level
association. Little research has explored the mechanisms connecting state
extraversion to state positive affect (but see Lischetzke et al., 2012 for evidence
that state extraversion is related to state positive affect through intentional mood
regulation), but a recent article (Smillie, 2013) nicely summarized a number of
potential explanations. Briefly, Smillie (2013) reviewed research suggesting that
state extraversion may be associated with state positive affect through increased
reward-processing states, social reinforcement, the social desirability of
extraverted behavior, the perception that extraversion states are effective at
producing progress toward goals, the physical actions involved in extraversion
states, and the psychological significance of bodily states associated with
extraversion. It is clear that the study of extraversion and positive affect has been
enormously fruitful, and it is not difficult to predict that this topic will continue
to stimulate innovative investigations for a long time to come.
Behavior
Evidence for the role of extraversion in behavior comes from a variety of
different methodologies. Investigations relying on self-report show that
extraversion associates with the content of behavior as well as specific
behaviors. More extraverted individuals describe their behaviors as bold,
socially adept, and secure (Funder, Furr, & Colvin, 2000), and they report
consuming more alcohol, going to more parties, dating more people, and
exercising more often (Paunonen, 2003). These studies suggest that extraversion
may be highly relevant to a wide spectrum of interpersonal behaviors. Findings
from a recent study (DeYoung, Weisberg, Quilty, & Peterson, 2013) supported
this idea by showing that the aspects of extraversion are uniquely associated with
the dimensions of the interpersonal circumplex (Wiggins, 1996): assertiveness
was related to the interpersonal dimension of dominance-submissiveness,
whereas enthusiasm was related to the dimension running from gregarious to
aloof.
The social nature of extraversion may act as a cue allowing people to
accurately assess others’ levels of extraversion. Acquaintances, experimenters,
and confederates are able to correctly identify more extraverted people after
observing a number of short tasks involving social activities (Borkenau, Mauer,
Riemann, Spinath, & Angleitner, 2004). Perhaps one characteristic that signifies
extraversion is a greater use of gestures. In an experiment that involved
describing the meaning of words to another person, more extraverted people
tended to accompany their speech with physical movements meant to convey
meaning (Hostetter & Potthoff, 2012). Another feature that seems to be
emblematic of extraversion is simply the propensity to talk more frequently.
Judges listening to recordings of random samples of activity throughout the
course of people’s daily lives rated those who were more talkative as more
extraverted (Mehl, Gosling, & Pennebaker, 2006). This turned out to be a good
heuristic, as coding the recordings revealed that self-described extraverts did
indeed spend more time talking to people and more time with others in general
(Mehl et al., 2006).
Differences in the communication styles depending on extraversion extend
from talking and gesturing to writing and electronic communication. When
asked to write in a stream of consciousness mode about their feelings related to
being in college, more extraverted university students include more positive
emotion words as well as more socially relevant words—such as references to
communicating or being with other people—in their written descriptions
(Pennebaker & King, 1999). Similarly, the online blogs of more extraverted
people contain more positive emotions and social references to friends, family,
and sexual behaviors (Yarkoni, 2010). Breaking down the relation between
extraversion and blog content by NEO PI-R facets showed that friendliness,
gregariousness, and cheerfulness accounted for these findings rather than the
facets of excitement seeking, assertiveness, or activity level (activity level was,
however, related to more achievement-related references). The text messages
that extraverted college students send surprisingly do not contain more positive
words but, similar to the blogs of more extraverted people, they do include more
social and sexual references (Holtgraves, 2011). Extraversion is also related to
more total time spent texting (Butt & Phillips, 2008).
It should come as no surprise, given the foregoing discussion, that more
extraverted individuals report a higher quantity of social participation when
asked to recall their daily activities (Srivastava et al., 2008). There are also
differences in the quality of the social participation of extraverts. In a laboratory
study of dynamic social interactions (Eaton & Funder, 2003), it was found that
not only did extraverts behave in more social ways, but they also influenced the
emotions, behaviors, and cognitive interpersonal judgments of their conversation
partners to create a more positive social environment. Recent work suggests that
extraversion is so ingrained with positive social interactions that more
extraverted people automatically and implicitly associate people with rewards
(Wilkowski & Ferguson, 2014). Extraverts seem to reap the benefits of their
social adroitness, as they exhibit levels of social well-being higher than
introverts (Hill, Turiano, Mroczek, & Roberts, 2012; Smillie et al., 2015; Wilt,
Cox, & McAdams, 2010).
It is obvious that extraversion is related to sociability, but this does not mean
that introverts do not value social interactions nor that introverted behavior is
inherently asocial. Introverts actually talk as much as extraverts in one-on-one
situations, but, as group size increases, more extraverted individuals spend a
disproportionately large amount of time talking (Antill, 1974). More introverted
individuals might also value quality rather than quantity when it comes to
socializing, preferring a few good friends to a large number of acquaintances
(Cain, 2013). A mixture of extraverted and introverted behaviors might be more
valuable in the literal, monetary sense when it comes to sales. In a study of
outbound call representatives, it was found that ambiverted individuals, those
toward the middle of the distribution of extraversion scores, generated the most
sales revenue (Grant, 2013). As Grant (2013) suggests, perhaps listening has
been underrated as a social skill.
Cognition
As people navigate their daily lives, they encounter a range of environments
that might present positive and/or negative consequences. Broadly speaking,
people are motivated to engage with positive stimuli and avoid negative stimuli,
but many situations are ambiguous with regard to their objective valence. For
example, is a job interview objectively rewarding due to the chance to have
intellectually stimulating conversations, or is it punishing because of the
potential for being negatively evaluated by a possible employer? The answer, of
course, is that a job interview, like so many other complex social environments,
contains a mixture of positive and negative (and neutral) elements. Individual
differences further complicate the landscape, as what some see as positive or
neutral, say, public speaking, might be viewed as an extremely negative situation
by others. Individual differences in how people perceive and categorize their
environments (i.e., individual differences in cognition) will to a large extent
determine their engagement with the world.
The section outlining the associations between extraversion and positive affect
suggested that extraversion should relate to judging the environment more
positively. This notion is borne out in a number of studies. Extraversion is
associated with judging neutral events more positively (Uziel, 2006) and with
recognizing positive stimuli more quickly after an initial positive prime
(Robinson, 2007). Extraverts judge positively valenced words (“hug” and
“smile”) as more similar than negatively valenced words (“grief” and “death”)
and as more similar than words that are related by semantic quality (“smile and
face”) (Rogers & Revelle, 1998; Weiler, 1992). The section describing
extraversion’s association with social behaviors hints at the idea that
extraversion should associate with more favorable cognitions regarding social
situations. Indeed, extraversion’s association with more positive and less
negative beliefs about interacting with others in extraverted ways (Zelenski et
al., 2013) perhaps explains why introverts do not engage in high levels of
extraverted behavior even though they experience the positive affective benefits
of acting extraverted (e.g., Fleeson et al., 2002; Wilt, Noftle, et al., 2011).
Moving past the general idea that extraversion relates to seeing the world in a
rose-colored tinge are studies of information processing tasks that vary as a
function of extraversion. There is considerable evidence that extraversion is
associated with superior recall on traditional, verbal short-term memory tasks
(M. Eysenck, 1981; Matthews, 1992) and with poorer vigilance (Beauducel,
Brocke, & Leue, 2006; Koelega, 1992). An excellent integrative review
(Matthews, Deary, & Whiteman, 2003) concluded that extraverts show
advantages with regard to dividing attention between tasks, resisting distractions,
and short-term memory. Introverts, in contrast, are better at sustained attention
tasks, solving complex problems, and long-term recall. These findings, taken
together, suggest that extraversion may relate to excelling in complex
environments where a variety of stimuli are competing for attention, whereas
introversion might be better suited to quiet tasks requiring persistence. It is
perhaps due to these differences in cognition that dynamic social environments
seem to be the extravert’s natural habitat.
Desire
People with different levels of extraversion pursue and relate to their goals in
different ways. Echoing previous discussions about the integral relationship
between extraversion and reward pursuit, extraversion is associated with
attaching more importance to goals, more intense goal pursuit, greater optimism
about achieving goals, and higher expectations for happiness when goals are
achieved (Romero, Villar, Luengo, & Gómez-Fraguela, 2009). These findings
add to the already large amount of evidence reviewed linking extraversion to
heightened engagement with rewarding stimuli. It is therefore clear that
extraversion is associated with approach motivation (Elliot & Thrash, 2002;
Heller et al., 2007), an energizing drive that directs behavior toward rewards
(Elliot, 2006). The following discussion focuses on the specific rewards that
extraverts desire to attain.
Extraversion relates to higher general motivation for social contact, intimacy,
and interdependence, as well as to drives for power, status, and positive affect
(Emmons, 1986; King, 1995; King & Broyles, 1997; Olson & Weber, 2004).
These findings suggest that extraversion is associated with the broad motivations
for affiliation and agency (Depue & Morrone-Strupinsky, 2005), or getting along
and getting ahead (Hogan, 1982). These motives permeate the lives of
extraverts. With regard to getting along, extraversion is related to the pursuit of
communal life goals and careers in the social domain; with regard to getting
ahead, extraverts desire lives in which they accomplish more goals related to
personal agency, in domains such as economics, aesthetics, politics, and
hedonism (Bleidorn et al., 2010; Larson, Rottinghaus, & Borgen, 2002; Roberts
& Robins, 2000, 2004). Extraverted states may also facilitate goals related to
getting along and getting ahead. People with higher levels of affiliation and
achievement goals enact more extraverted states over time (Bleidorn, 2009).
Additionally, the short-term goals of being more sociable, enthusiastic, and
assertive are associated with state extraversion (McCabe & Fleeson, 2012).
Objective Biography
The characteristic adaptations just described are psychological structures and
patterns that bridge the gap between basic traits and objective biography—a
person’s factual life story. Objective biography brings personality traits back to
the person by describing the real successes, struggles, failures, and redemptions
that people experience as they navigate their lives.
A person’s level of extraversion in late adolescence is an important
determinant of subsequent life events in the near future. Magnus, Diener, Fujita,
and Pavot (1993) determined that extraversion in a sample of college
undergraduates prospectively predicted the occurrence of objective, positive life
events over the course of 4 years. In this study, the composite of positive life
events included 20 events that received high ratings on the dimensions of
objectivity and positivity. This list included seminal events in the domains of
relationships, education, career, and leisure (e.g., getting engaged, getting
married, getting into graduate school, receiving a promotion or raise, beginning a
hobby). Extraversion was unrelated to the occurrence of objective, negative life
events (e.g., divorce, death of a loved one, getting fired, being the victim of a
crime). Vaidya et al. (2002) used similar lists of positive/negative life events and
found that in a sample of undergraduates, not only was extraversion related to a
higher occurrence of positive events over 2.5 years, but the occurrence of
positive events over that time was also related to increases in extraversion. In a
study tracking German high school students for 4 years, Lüdtke et al. (2011)
pinpointed specific, positive events that were most highly related to
extraversion; this list included getting promoted, starting a job, going abroad,
and starting a relationship.
In adulthood, extraversion remains a robust predictor of social outcomes.
People with higher levels of extraversion have a greater number of social
relationships and greater social support (Berkman, Glass, Brissette, & Seeman,
2000). As noted in the section on evolutionary costs and benefits, extraversion is
associated with having more sexual partners (Nettle, 2005). Extraversion is
related to greater marriage satisfaction with (Watson et al., 2004) but also to
higher rates of infidelity (Nettle, 2005). Extraversion is also particularly
powerful in predicting occupational outcomes. Extraverted individuals are more
satisfied with their jobs (Thoresen, Kaplan, Barsky, Warren, & de Chermont,
2003) and show higher levels of job performance (Sackett & Walmsley, 2014);
adolescent ratings of extraversion predict higher income and job status 46 years
later, even after controlling for cognitive ability (Judge, Higgins, Thoresen, &
Barrick, 1999). All of these findings attest to the conclusion that extraversion is
highly relevant to the lives that people lead (Ozer & Benet-Martinez, 2006;
Roberts et al., 2007), and thus people with different levels of extraversion have
very different objective biographies.
Self-Concept and Identity

Self-concept refers to how people think about, perceive, and evaluate
themselves (Baumeister, 1999). It includes memories and evaluations of past,
present, and future selves (Markus & Nurius, 1986). In the FFM, self-concept is
defined as being consistent with traits, and is also influenced by characteristic
adaptations and objective biography (McCrae & Costa, 2008). Said differently,
when answering the question “Who am I?,” people draw on information from
their basic tendencies, their characteristic ABCD patterns, and actual events and
circumstances in their lives. Indeed, extraversion is pertinent to how we define
ourselves.
In support of the claim that self-concept is consistent with our traits,
introversion is related to describing the “true” or “authentic” self as more
introverted, whereas people who are more extraverted endorse an extraverted
true-self concept (Fleeson & Wilt, 2010). Extraversion is also relevant to
evaluative aspects of the self-concept. Higher levels of extraversion are
associated with higher self-esteem (Aluja, Rolland, García, & Rossier, 2007;
Watson, Suls, & Haig, 2002), which refers to a global evaluation of general
worth as a person (Crocker & Luhtanen, 2003). This finding is sensible given the
generally positively valenced characteristic adaptations and objective
biographies of more extraverted people. Positive affect and social support have
been identified as mediators of the association between extraversion and self-
esteem (Swickert, Hittner, Kitos, & Cox-Fuenzalida, 2004).
Self-concept includes a person’s identity, which is a sense of coherence and
unity based on self-defined investments in life choices (Erikson, 1963). An
increasingly influential theory conceptualizes identity as internalized life stories
that together comprise what has been termed narrative identity (McAdams,
1993; Singer, 2004; Thorne & Nam, 2009). Life stories are self-authored and
psychologically constructed integrations of the remembered past, experienced
present, and imagined future that represent one way in which people potentially
instill their lives with unity, meaning, and purpose. Life stories are constructed
within the narrative mode of human cognition (Bruner, 2004), and thus narrative
terms (e.g., imagery, plot, theme, scene, setting, conflict, character, ending) are
thought to be the best way to describe and characterize life stories (McAdams,
2008).
Few studies have examined the links between extraversion and narrative
identity, but the results from those that have done so suggest that they are indeed
linked. Research has shown that extraversion relates to the structure and contents
of life stories as well as the ways that the different scenes in the life story are
classified (McAdams et al., 2004; McLean & Fournier, 2008; Raggatt, 2006;
Thomsen, Olesen, Schnieber, & Tonnesvang, 2013). More extraverted
individuals tell stories with a high degree of positive emotions, see events as
having more positive connections to the self, and organize scenes from their
stories by characteristics such as status, optimism, sociability, and activity.
People with higher levels of extraversion are more concerned with interpersonal
trust when describing life story scenes from childhood and adolescence, and they
are concerned with giving back to society, or generativity, in scenes from their
adulthood. The social nature of extraversion is apparent in the way that they
communicate their life stories. More extraverted people share important
memories with others more often and report a high degree of comfort in doing so
(McLean & Pasupathi, 2006).
Extraversion and Psychopathology
Thus far, extraversion has been discussed in the context of “normal”
personality functioning as opposed to “abnormal” or clinical manifestations of
the trait. Yet Eysenck had very early recognized the importance of extraversion
in psychiatric disorders (e.g., Eysenck & Himmelweit, 1947) and continued to
emphasize the application of normal personality traits to psychopathology
(Eysenck, 1957). In the FFM, basic tendencies are also conceptualized as having
both adaptive and maladaptive variants (McCrae & Costa, 2008). Provided next
is a brief summary of how both low and high levels of extraversion relate to
psychopathology.
A variety of studies show that categorically defined personality disorders
(PDs) may be conceptualized as maladaptive or “extreme” variants of traits
included in models of normal personality (e.g., Bagby, 2013; Gore & Widiger,
2013; Samuel, Carroll, Rounsaville, & Ball, 2013; Sellbom, Anderson, & Bagby,
2013; Van den Akker et al., 2013; see also the chapter by Widiger, Gore, Crego,
Rojas, and Oltmanns). Extreme introversion is defined by characteristics such as
social withdrawal, social detachment, intimacy avoidance, restricted affectivity,
and anhedonia (Gore & Widiger, 2013; Skodol et al., 2011; Watson, Stasik, Ro,
& Clark, 2013), all of which may be relevant to maladaptive personality
functioning. Indeed, Skodol et al. (2011) theorized that these features may
represent core components of the schizoid, schizotypal, and avoidant PDs. Low
levels of extraversion and related traits reflecting low levels of social connection
have additionally been related to problems with anxiety and depression (Jylhä &
Isometsä, 2006; Krueger et al., 1996; Trull & Sher, 1994; Watson, Gamez, &
Simms, 2005). However, extremely high extraversion poses risks for personality
pathology as well, as people falling at this end of the continuum are more likely
to be sexually promiscuous, emotionally intrusive, and engage in excessive self-
disclosure and thrill-seeking behaviors (McCrae, Löckenhoff, & Costa, 2005).
People with high levels of extraversion are also more likely to have difficulties
with substance abuse (Atherton, Robins, Rentfrow, & Lamb, 2014; Krueger et
al., 1996), possibly due to their elevated reward-seeking tendencies.
Conclusions
Several years ago, we (Wilt & Revelle, 2009) highlighted three areas of
research on extraversion about which we were particularly enthusiastic: the role
of extraversion in ongoing functioning, the integration of psychological and
biological theories of extraversion, and the use of public domain personality
assessment to study the structure of extraversion and its predictive validity in
important domains. We were optimistic at the time, but we did not anticipate just
how quickly progress would occur in these and many other areas, as reviewed in
this chapter. The rapid accumulation of research on an already expansive topic
makes even more important the existence of an overarching theoretical
framework. The FFM provides a comprehensive and parsimonious
organizational architecture by which to classify and group the myriad findings
emerging from this exciting field. We are confident that in the next decade and
beyond we will see many more such advances.
Notes
1. There were few adjectives that had substantial loadings on up to three factors.
2. Desire is chosen over the term “motivation” due to desire’s more specific connotation
of referring to what people want, as compared with motivation’s more general
connotation of referring to the factors that energize, direct, and select behavior
(Atkinson & Raynor, 1978; Heckhausen, 1991; Humphreys & Revelle, 1984).
Whereas the factors that guide behavior may include affect, cognition, desire, and
even behavior itself, desire links more naturally to goals, wants, and wishes.
3. The main change to the theory is that the system formerly referred to as the FFS (now
FFFS—“Fight, Flight, Freeze System”) has been given a greater role, mediating
responses to all aversive stimuli and generating the fear response.
4. Depue has also proposed that “affiliative extraversion,” encompassing warmth and
social closeness, may be related to opiate functioning (Depue & Morrone-Strupinsky,
2005)
References
Allport, G. W., & Odbert, H. S. (1936). Trait names: A psycholexical study.
Psychological Monographs, 47(211), 171.
Aluja, A., Rolland, J.-P., García, L. F., & Rossier, J. (2007). Dimensionality of the
Rosenberg self-esteem scale and its relationships with the Three- and the Five-Factor
personality models. Journal of Personality Assessment, 88(2), 246–249.
Anderson, C., John, O. P., Keltner, D., & Kring, A. M. (2001). Who attains social status?
effects of personality and physical attractiveness in social groups. Journal of
Personality and Social Psychology, 81(1), 116–132.
Antill, J. K. (1974). The validity and predictive power of introversion-extraversion for
quantitative aspects of conversational patterns. Dissertation Abstracts International,
35(1-B), 532.
Argyle, M., & Lu, L. (1990). The happiness of extraverts. Personality and Individual
Differences, 11(10), 1011–1017.
Asendorpf, J. B., & van Aken, M. A. G. (2003). Personality-relationship transaction in
adolescence: Core versus surface personality characteristics. Journal of Personality,
71(4), 629–666.
Ashton, M. C., & Lee, K. (2001). A theoretical basis for the major dimensions of
personality. European Journal of Personality, 15(5), 327–353.
Ashton, M. C., & Lee, K. (2007). Empirical, theoretical, and practical advantages of the
HEXACO model of personality structure. Personality and Social Psychology Review,
11(2), 150–166.
Ashton, M. C., Lee, K., & Paunonen, S. V. (2002). What is the central feature of
extraversion?: Social attention versus reward sensitivity. Journal of Personality and
Social Psychology, 83(1), 245–251.
Atherton, O. E., Robins, R. W., Rentfrow, P. J., & Lamb, M. E. (2014). Personality
correlates of risky health outcomes: Findings from a large internet study. Journal of
Atkinson, J. W., & Raynor, J. O. (1978). Personality, motivation, and achievement.
Washington, DC: Hemisphere.
Bagby, R. M. (2013). Introduction to special issue on the personality inventory for DSM-
5 (PID-5). Assessment, 20(3), 267–268.
Baumeister, R. F. (1999). The self in social psychology. New York, NY: Psychology
Press.
Beauducel, A., Brocke, B., & Leue, A. (2006). Energetical bases of extraversion: Effort,
arousal, EEG, and performance. International Journal of Psychophysiology, 62(2),
212–223.
Berkman, L. F., Glass, T., Brissette, I., & Seeman, T. E. (2000). From social integration
to health: Durkheim in the new millennium. Social Science & Medicine, 51(6), 843–
857.
Berridge, K. C. (2007). The debate over dopamine’s role in reward: The case for
incentive salience. Psychopharmacology, 191(3), 391–431.
Berridge, K. C. (2012). From prediction error to incentive salience: Mesolimbic
computation of reward motivation. European Journal of Neuroscience, 35(7), 1124–
1143.
Bleidorn, W. (2009). Linking personality states, current social roles and major life goals.
European Journal of Personality, 23(6), 509–530.
Bleidorn, W., Kandler, C., Hülsheger, U. R., Riemann, R., Angleitner, A., & Spinath, F.
M. (2010). Nature and nurture of the interplay between personality traits and major life
goals. Journal of Personality and Social Psychology, 99(2), 366–379.
Bleidorn, W., Kandler, C., Riemann, R., Angleitner, A., & Spinath, F. M. (2012). Genetic
and environmental influences on personality profile stability: Unraveling the
normativeness problem. Journal of Personality, 80(4), 1029–1060.
Bleidorn, W., Klimstra, T. A., Denissen, J. J., Rentfrow, P. J., Potter, J., & Gosling, S. D.
(2013). Personality maturation around the world: A cross-cultural examination of
social-investment theory. Psychological Science, 24(12), 2530–2540.
Borkenau, P., Mauer, N., Riemann, R., Spinath, F. M., & Angleitner, A. (2004). Thin
slices of behavior as cues of personality and intelligence. Journal of Personality and
Borsboom, D., & Cramer, A. O. (2013). Network analysis: An integrative approach to the
structure of psychopathology. Annual Review of Clinical Psychology, 9, 91–121.
Bouchard, T. J., & Loehlin, J. C. (2001). Genes, evolution, and personality. Behavior
Genetics, 31(3), 243–273.
Bruner, J. (2004). Life as narrative. Social Research, 71(3), 691–710.
Buss, D. M. (1991). Evolutionary personality psychology. Annual Review of Psychology,
42, 459–491.
Buss, D. M. (1995). Evolutionary psychology: A new paradigm for psychological
science. Psychological Inquiry, 6(1), 1–30.
Buss, D. M. (2009). How can evolutionary psychology successfully explain personality
and individual differences? Perspectives on Psychological Science, 4(4), 359–366.
Butt, S., & Phillips, J. G. (2008). Personality and self reported mobile phone use.
Computers in Human Behavior, 24(2), 346–360.
Cain, S. (2013). Quiet: The power of introverts in a world that can’t stop talking. New
York, NY: Broadway Books.
Camperio Ciani, A. S., Capiluppi, C., Veronese, A., & Sartori, G. (2007). The adaptive
value of personality differences revealed by small island population dynamics.
European Journal of Personality, 21(1), 3–22. doi:10.1002/per.595.
Canli, T. (2004). Functional brain mapping of extraversion and neuroticism: Learning
from individual differences in emotion processing. Journal of Personality, 72(6),
1105–1132.
Canli, T. (2006). Genomic imaging of extraversion (pp. 93–115). New York, NY: The
Guilford Press.
Carver, C. S., & White, T. L. (1994). Behavioral inhibition, behavioral activation, and
affective responses to impending reward and punishment: The BIS/BAS scales.
Cattell, R. (1947). Confirmation and clarification of primary personality factors.
Psychometrika, 12(3), 197–220.
Cattell, R. B. (1943a). The description of personality: Basic traits resolved into clusters.
The Journal of Abnormal and Social Psychology, 38(4), 476–506.
Cattell, R. B. (1943b). The description of personality. I. Foundations of trait
measurement. Psychological Review, 50(6), 559–594.
Cattell, R. B. (1957). Personality and motivation structure and measurement. Oxford,
England: World Book.
Cervone, D. (2004). The architecture of personality. Psychological Review, 111(1), 183–
204.
Ching, C. M., Church, A. T., Katigbak, M. S., Reyes, J. A. S., Tanaka-Matsumi, J.,
Takaoka, S., … Rincon, B. C. (2014). The manifestation of traits in everyday behavior
and affect: A five-culture study. Journal of Research in Personality, 48, 1–16.
Chulef, A. S., Read, S. J., & Walsh, D. A. (2001). A hierarchical taxonomy of human
goals. Motivation and Emotion, 25(3), 191–232.
Cloninger, C., Svrakic, D. M., & Przybeck, T. R. (1993). A psychobiological model of
temperament and character. Archives of General Psychiatry, 50(12), 975–990.
Corr, P. J. (2008). The reinforcement sensitivity theory of personality. Cambridge,
England: Cambridge University Press.
Corr, P. J., & Cooper, A. (2015). The Corr-Cooper reinforcement sensitivity theory
personality questionnaire (RST-PQ): Development and validation. Unpublished
manuscript, City University, London, UK.
Costa, J., McCrae, R. R., Zonderman, A. B., Barbano, H. E., Lebowitz, B., & Larson, C.
M. (1986). Cross-sectional studies of personality in a national sample: II. Stability in
neuroticism, extraversion, and openness. Psychology and Aging, 1(2), 144–149.
doi:10.1037/0882–7974.1.2.144.
Costa, P. T., & McCrae, R. R. (1992). Four ways five factors are basic. Personality and
Individual Differences, 13(6), 653–665.
Costa, P. T., & McCrae, R. R. (1998). Six approaches to the explication of facet-level
traits: Examples from conscientiousness. European Journal of Personality, 12(2), 117–
134.
Cramer, A. O., Sluis, S., Noordhof, A., Wichers, M., Geschwind, N., Aggen, S. H., …
Borsboom, D. (2012). Dimensions of normal personality as networks in search of
equilibrium: You can’t like parties if you don’t like people. European Journal of
Personality, 26(4), 414–431.
Crocker, J., & Luhtanen, R. K. (2003). Level of self-esteem and contingencies of self-
worth: Unique effects on academic, social, and financial problems in college students.
Personality and Social Psychology Bulletin, 29(6), 701–712.
Dawkins, R., & Krebs, J. R. (1979). Arms races between and within species. Proceedings
of the Royal Society of London. Series B, Biological Sciences, 205(1161), 489–511.
Denissen, J. J., & Penke, L. (2008). Motivational individual reaction norms underlying
the Five-Factor Model of personality: First steps towards a theory-based conceptual
framework. Journal of Research in Personality, 42(5), 1285–1302.
Depue, R. A. (1995). Neurobiological factors in personality and depression. European
Journal of Personality, 9(5), 413–439.
Depue, R. A., & Collins, P. F. (1999). Neurobiology of the structure of personality:
Dopamine, facilitation of incentive motivation, and extraversion. Behavioral and Brain
Sciences, 22(3), 491–569.
Depue, R. A., & Morrone-Strupinsky, J. V. (2005). A neurobehavioral model of
affiliative bonding: Implications for conceptualizing a human trait of affiliation.
Behavioral and Brain Sciences, 28(3), 313–395.
DeYoung, C. G. (2010). Personality neuroscience and the biology of traits. Social and
DeYoung, C. G. (2015). Cybernetic Big Five theory. Journal of Research in Personality,
56, 33–58.
aspects of the Big Five. Journal of Personality and Social Psychology, 93(5), 880–896.
DeYoung, C. G., Weisberg, Y. J., Quilty, L. C., & Peterson, J. B. (2013). Unifying the
aspects of the Big Five, the interpersonal circumplex, and trait affiliation. Journal of
Personality, 81(5), 465–475.
Diener, E., Sandvik, E., Pavot, W., & Fujita, F. (1992). Extraversion and subjective well-
being in a U.S. national probability sample. Journal of Research in Personality, 26(3),
205–215.
Digman, J. M. (1990). Personality structure: Emergence of the Five-Factor Model.
Annual Review of Psychology, 41, 417–440.
Durbin, C., Klein, D. N., Hayden, E. P., Buckley, M. E., & Moerk, K. C. (2005).
Temperamental emotionality in preschoolers and parental mood disorders. Journal of
Abnormal Psychology, 114(1), 28–37.
Eaton, L. G., & Funder, D. C. (2003). The creation and consequences of the social world:
An interactional analysis of extraversion. European Journal of Personality, 17(5), 375–
395.
Ebstein, R. P., Benjamin, J., & Belmaker, R. H. (2003). Behavioral genetics, genomics,
and personality (pp. 365–388). Washington, DC: American Psychological Association.
Elliot, A. J. (2006). The hierarchical model of approach-avoidance motivation.
Motivation and Emotion, 30(2), 111–116.
Elliot, A. J., & Thrash, T. M. (2002). Approach-avoidance motivation in personality:
Approach-avoidance temperaments and goals. Journal of Personality and Social
Ellsworth, P. C. (1994). William James and emotion: Is a century of fame worth a century
of misunderstanding? Psychological Review, 101(2), 222–229.
Emmons, R. A. (1986). Personal strivings: An approach to personality and subjective
well-being. Journal of Personality and Social Psychology, 51(5), 1058–1068.
Erikson, E. H. (1963). Childhood and society (2nd ed.). Oxford, England: Norton & Co.
Eysenck, H. J. (1952). The scientific study of personality. London, England: Routledge &
Kegan Paul.
Eysenck, H. J. (1957). The dynamics of anxiety and hysteria; an experimental application
of modern learning theory to psychiatry. Oxford, England: Praeger.
Eysenck, H. J. (1959). The “Maudsley Personality Inventory” as determinant of neurotic
tendency and extraversion. Zeitschrift fur Experimentelle und Angewandte
Psychologie, 6 Apr–Jun, 167–190.
Eysenck, H. J. (1967). The biological basis of personality. Springfield, NJ: Thomas.
Eysenck, H. J. (1970). The structure of human personality. London, England: Methuen.
Eysenck, H. J. (1973). Eysenck on extraversion. New York, NY: Wiley.
Eysenck, H. J. (1977). Personality and factor analysis: A reply to Guilford. Psychological
Bulletin, 84(3), 405–411.
Eysenck, H. J. (1981). General features of the model (pp. 1–37). Berlin, Germany:
Springer-Verlag.
Eysenck, H. J. (1992). A hundred years of personality research, from Heymans to modern
times. Houten, The Netherlands: Bohn.
Eysenck, H. J., & Eysenck, S. B. G. (1964). Eysenck Personality Inventory. San Diego,
CA: Educational and Industrial Testing Service.
Eysenck, H. J., & Himmelweit, H. T. (1947). Dimensions of personality; a record of
research carried out in collaboration with H.T. Himmelweit [and others]. London,
England: Kegan Paul, Trench.
Eysenck, H. J., & Wilson, G. D. (1991). The Eysenck Personality Profiler (version 6).
Worthing, England: Psi-Press.
Eysenck, M. (1981). Learning, memory and personality (pp. 169–209). New York, NY:
Springer.
Eysenck, S. B., & Eysenck, H. J. (1975). Manual of the Eysenck Personality
Questionnaire. London, England: Houer & Stoughton.
Fiske, D. W. (1949). Consistency of the factorial structures of personality ratings from
different sources. Journal of Abnormal and Social Psychology, 44, 329–344.
Fleeson, W. (2001). Toward a structure- and process-integrated view of personality:
Traits as density distributions of states. Journal of Personality and Social Psychology,
80(6), 1011–1027.
Fleeson, W. (2012). Perspectives on the person: Rapid growth and opportunities for
integration. In K. Deaux & M. Snyder (Eds.), The Oxford handbook of personality and
social psychology (p. 33). New York, NY: Oxford University Press.
Fleeson, W., & Gallagher, P. (2009). The implications of Big Five standing for the
distribution of trait manifestation in behavior: Fifteen experience-sampling studies and
a meta-analysis. Journal of Personality and Social Psychology, 97(6), 1097–1114.
Fleeson, W., Malanos, A. B., & Achille, N. M. (2002). An intraindividual process
approach to the relationship between extraversion and positive affect: Is acting
extraverted as “good” as being extraverted? Journal of Personality and Social
Psychology, 83(6), 1409–1422.
Fleeson, W., & Wilt, J. (2010). The relevance of Big Five trait content in behavior to
subjective authenticity: Do high levels of within-person behavioral variability
undermine or enable authenticity achievement? Journal of Personality, 78(4), 1353–
1382.
Flory, J. D., Manuck, S. B., Matthews, K. A., & Muldoon, M. F. (2004). Serotonergic
function in the central nervous system is associated with daily ratings of positive mood.
Psychiatry Research, 129(1), 11–19.
Funder, D. C., Furr, R., & Colvin, C. (2000). The Riverside Behavioral Q-sort: A tool for
the description of social behavior. Journal of Personality, 68(3), 451–489.
Furr, R. M. (2009). Personality psychology as a truly behavioural science. European
Goekoop, R., Goekoop, J. G., & Scholte, H. S. (2012). The network structure of human
personality according to the NEO-PI-R: Matching network community structure to
factor structure. PLoS One, 7(12), e51558.
Goldberg, L. R. (1990). An alternative “description of personality”: The Big-Five factor
structure. Journal of Personality and Social Psychology, 59(6), 1216–1229.
Goldberg, L. R. (1992). The development of markers for the Big-Five factor structure.
Psychological Assessment, 4(1), 26–42.
Goldberg, L. R. (1993). The structure of phenotypic personality traits. American
Psychologist, 48(1), 26–34.
measuring the lower-level facets of several five-factor models (Vol. 75, pp. 7–28).
Tilburg, The Netherlands: Tilburg University Press.
Goldberg, L. R., Johnson, J. A., Eber, H. W., Hogan, R., Ashton, M. C., Cloninger, C. R.,
& Gough, H. G. (2006). The international personality item pool and the future of
public-domain personality measures. Journal of Research in Personality, 40(1), 84–96.
Gore, W. L., & Widiger, T. A. (2013). The DSM-5 dimensional trait model and Five-
Factor Models of general personality. Journal of Abnormal Psychology, 122(3), 816.
Gosling, S. D., & John, O. P. (1999). Personality dimensions in nonhuman animals: A
cross-species review. Current Directions in Psychological Science (Wiley-Blackwell),
8(3), 69–75.
Grant, A. M. (2013). Rethinking the extraverted sales ideal: The ambivert advantage.
Psychological Science, 24(6), 1024–1030.
Gray, J. A. (1970). The psychophysiological basis of introversion-extraversion.
Behaviour Research and Therapy, 8(8), 249–266.
Gray, J. A. (1981). A critique of Eysenck’s theory of personality (pp. 246–277). Berlin,
Germany: Springer-Verlag.
Gray, J. A. (1982). Neuropsychological theory of anxiety: An investigation of the septal-
hippocampal system. Cambridge, England: Cambridge University Press.
Gray, J. A., & McNaughton, N. (2000). The neuropsychology of anxiety: An enquiry into
the functions of the septo-hippocampal system (2nd ed.). Oxford, England: Oxford
University Press.
Gross, J. J., Sutton, S. K., & Ketelaar, T. (1998). Relations between affect and
personality: Support for the affect-level and affective reactivity views. Personality and
Social Psychology Bulletin, 24(3), 279–288.
Gruszka, A., Matthews, G., & Szymura, B. (2010). Handbook of individual differences in
cognition: Attention, memory, and executive control. New York, NY: Springer-Verlag.
Guilford, J. P. (1975). Factors and factors of personality. Psychological Bulletin, 82(5),
802–814.
Guilford, J. P. (1977). Will the real factor of extraversion-introversion please stand up? A
reply to Eysenck. Psychological Bulletin, 84(3), 412–416.
Guilford, J. P., & Guilford, R. B. (1934). An analysis of the factors in a typical test of
introversion-extroversion. Journal of Abnormal and Social Psychology, 28(4), 377–
399.
Guilford, J. P., & Zimmerman, W. S. (1949). The Guilford-Zimmerman Temperament
Survey. Oxford, England: Sheridan Supply.
Heckhausen, H. (1991). Motivation and action. New York, NY: Springer-Verlag.
Heller, D., Komar, J., & Lee, W. B. (2007). The dynamics of personality states, goals,
and well-being. Personality and Social Psychology Bulletin, 33(6), 898–910.
Heymans, G., & Wiersma, E. (1909). Beitrage zur apeziellen psychologie auf grund einer
masse-nuntersuchung [contributions to differential psychology based on a large scale
investigation]. Zeitschrift fur Psychologic, 51, 1–72.
Hill, P. L., Turiano, N. A., Mroczek, D. K., & Roberts, B. W. (2012). Examining
concurrent and longitudinal relations between personality traits and social well-being in
adulthood. Social Psychological and Personality Science, 3(6), 698–705.
Hofstee, W. K., de Raad, B., & Goldberg, L. R. (1992). Integration of the Big Five and
circumplex approaches to trait structure. Journal of Personality and Social Psychology,
63(1), 146–163.
Hogan, R. (1982). A socioanalytic theory of personality, vol. 30 of Nebraska Symposium
on Motivation. Lincoln, NE: University of Nebraska Press.
Holtgraves, T. (2011). Text messaging, personality, and the social context. Journal of
Research in Personality, 45(1), 92–99.
Hostetter, A. B., & Potthoff, A. L. (2012). Effects of personality and social situation on
representational gesture production. Gesture, 12(1), 62–83.
Humphreys, M. S., & Revelle, W. (1984). Personality, motivation, and performance: A
theory of the relationship between individual differences and information processing.
Psychological Review, 91(2), 153–184.
John, O. P., Donahue, E. M., & Kentle, R. L. (1991). The Big Five Inventory-Versions 4a
and 54. Berkeley, CA: University of California, Berkeley, Institute of Personality and
Social Research.
Judge, T. A., Higgins, C. A., Thoresen, C. J., & Barrick, M. R. (1999). The Big Five
personality traits, general mental ability, and career success across the life span.
Personnel Psychology, 52(3), 621–652.
Jung, C. G. (1921/1971). Psychological types: Collected Works (Vol. 6). Princeton, NJ:
Princeton University Press.
Jylhä, P., & Isometsä, E. (2006). The relationship of neuroticism and extraversion to
symptoms of anxiety and depression in the general population. Depression & Anxiety,
23(5), 281–289.
Keiser, H. N., & Ross, S. R. (2011). Carver and Whites BIS/BAS scales and domains and
facets of the Five Factor Model of personality. Personality and Individual Differences,
51(1), 39–44.
King, L. A. (1995). Wishes, motives, goals, and personal memories: Relations of
measures of human motivation. Journal of Personality, 63(4), 985–1007.
King, L. A., & Broyles, S. J. (1997). Wishes, gender, personality, and well-being. Journal
of Personality, 65(1), 49–76.
Koelega, H. S. (1992). Extraversion and vigilance performance: 30 years of
inconsistencies. Psychological Bulletin, 112(2), 239–258.
Koydemir, S., Şimşek, O. F., & Demir, M. (2014). Pathways from personality to
happiness: Sense of uniqueness as a mediator. Journal of Humanistic Psychology,
54(3), 314–335.
Krueger, R. F., Caspi, A., Moffitt, T. E., Silva, P. A., & McGee, R. (1996). Personality
traits are differentially linked to mental disorders: A multitrait-multidiagnosis study of
an adolescent birth cohort. Journal of Abnormal Psychology, 105(3), 299–312.
Larsen, R. J., & Ketelaar, T. (1991). Personality and susceptibility to positive and
negative emotional states. Journal of Personality and Social Psychology, 61(1), 132–
140.
Larson, L. M., Rottinghaus, P. J., & Borgen, F. H. (2002). Meta-analyses of Big Six
interests and Big Five personality factors. Journal of Vocational Behavior, 61(2), 217–
239.
Lee, K., & Ashton, M. C. (2004). Psychometric properties of the HEXACO personality
inventory. Multivariate Behavioral Research, 39(2), 329–358.
Lischetzke, T., & Eid, M. (2006). Why extraverts are happier than introverts: The role of
mood regulation. Journal of Personality, 74(4), 1127–1162.
Lischetzke, T., Pfeifer, H., Crayen, C., & Eid, M. (2012). Motivation to regulate mood as
a mediator between state extraversion and pleasant and unpleasant mood. Journal of
Lü, W., Wang, Z., Liu, Y., & Zhang, H. (2014). Resilience as a mediator between
extraversion, neuroticism and happiness, PA and NA. Personality and Individual
Lucas, R. E., & Baird, B. M. (2004). Extraversion and emotional reactivity. Journal of
Lucas, R. E., & Diener, E. (2001). Understanding extraverts’ enjoyment of social
situations: The importance of pleasantness. Journal of Personality and Social
Psychology, 81(2), 343–356.
Lucas, R. E., & Donnellan, M. B. (2011). Personality development across the life span:
Longitudinal analyses with a national sample from Germany. Journal of Personality
and Social Psychology, 101(4), 847–861.
Lucas, R. E., & Fujita, F. (2000). Factors influencing the relation between extraversion
and pleasant affect. Journal of Personality and Social Psychology, 79(6), 1039–1056.
Lucas, R. E., Le, K., & Dyrenforth, P. S. (2008). Explaining the extraversion/positive
affect relation: Sociability cannot account for extraverts’ greater happiness. Journal of
Personality, 76(3), 385–414.
Lüdtke, O., Roberts, B. W., Trautwein, U., & Nagy, G. (2011). A random walk down
university avenue: Life paths, life events, and personality trait change at the transition
to university life. Journal of Personality and Social Psychology, 101(3), 620–637.
Lukaszewski, A. W., & Roney, J. R. (2011). The origins of extraversion: Joint effects of
facultative calibration and genetic polymorphism. Personality and Social Psychology
Bulletin, 37(3), 409–421.
Luo, X., Kranzler, H. R., Zuo, L., Wang, S., & Gelernter, J. (2007). Personality traits of
agreeableness and extraversion are associated with ADH4 variation. Biological
Psychiatry, 61(5), 599–608.
Magee, C. A., Heaven, P. C. L., & Miller, L. M. (2013). Personality change predicts self-
reported mental and physical health. Journal of Personality, 81(3), 324–334.
Markus, H., & Nurius, P. (1986). Possible selves. American Psychologist, 41(9), 954.
Matthews, G. (1992). Extraversion, Vol. 3: State and trait (pp. 95–126). San Diego, CA:
Academic Press.
Matthews, G., Deary, I. J., & Whiteman, M. C. (2003). Personality traits. Cambridge,
England: Cambridge University Press.
Matthews, G., & Gilliland, K. (1999). The personality theories of H. J. Eysenck and J. A.
Gray: A comparative review. Personality and Individual Differences, 26(4), 583–626.
McAdams, D. P. (1993). The stories we live by: Personal myths and the making of the
self. New York, NY: William Morrow & Co.
McAdams, D. P. (2008). Personal narratives and the life story (3rd ed., pp. 242–264).
McAdams, D. P., Anyidoho, N. A., Brown, C., Huang, Y. T., Kaplan, B., & Machado, M.
A. (2004). Traits and stories: Links between dispositional and narrative features of
personality. Journal of Personality, 72(4), 761–784.
McAdams, D. P., & Pals, J. L. (2006). A new Big Five: Fundamental principles for an
integrative science of personality. American Psychologist, 61(3), 204–217.
McCabe, K. O., & Fleeson, W. (2012). What is extraversion for? Integrating trait and
motivational perspectives and identifying the purpose of extraversion. Psychological
Science, 23(12), 1498–1505.
McCrae, R. R. (2009). The Five-Factor Model of personality traits: Consensus and
controversy (Chap. 9, pp. 148–161). Cambridge, England: Cambridge University Press.
McCrae, R. R., & Costa, P. T. (1992). Discriminant validity of NEO-PIR facet scales.
Educational and Psychological Measurement, 52(1), 229–237.
McCrae, R. R., & Costa, P. T. (1997). Personality trait structure as a human universal.
American Psychologist, 52(5), 509–516.
McCrae, R. R., & Costa, P. T. (2008). The five-factor theory of personality (3rd ed., pp.
McCrae, R. R., Costa, P. T., de Lima, M. P., Simões, A., Ostendorf, F., Angleitner, A., …
Piedmont, R. L. (1999). Age differences in personality across the adult life span:
Parallels in five cultures. Developmental Psychology, 35(2), 466–477.
McCrae, R. R., Costa, P. T., Jr., Ostendorf, F., Angleitner, A., Hrebícková, M., Avia, M.
D., … Woodfield, R. (2000). Nature over nurture: Temperament, personality, and life
span development. Journal of Personality and Social Psychology, 78(1), 173.
McCrae, R. R., Löckenhoff, C. E., & Costa, J. (2005). A step toward DSM-V:
Cataloguing personality-related problems in living. European Journal of Personality,
19(4), 269–286. doi:10.1002/per.564.
McCrae, R. R., & Terracciano, A. (2005). Universal features of personality traits from the
observer’s perspective: Data from 50 cultures. Journal of Personality and Social
Psychology, 88(3), 547–561.
McLean, K. C., & Fournier, M. A. (2008). The content and processes of autobiographical
reasoning in narrative identity. Journal of Research in Personality, 42(3), 527–545.
McLean, K. C., & Pasupathi, M. (2006). Collaborative narration of the past and
extraversion. Journal of Research in Personality, 40(6), 1219–1231.
McNiel, J., & Fleeson, W. (2006). The causal effects of extraversion on positive affect
and neuroticism on negative affect: Manipulating state extraversion and state
neuroticism in an experimental approach. Journal of Research in Personality, 40(5),
529–550.
McNiel, J. M., Lowman, J. C., & Fleeson, W. (2010). The effect of state extraversion on
four types of affect. European Journal of Personality, 24(1), 18–35.
Mehl, M. R., Gosling, S. D., & Pennebaker, J. W. (2006). Personality in its natural
habitat: Manifestations and implicit folk theories of personality in daily life. Journal of
Mroczek, D. K., & Spiro, I. (2003). Modeling intraindividual change in personality traits:
Findings from the normative aging study. The Journals of Gerontology: Series B:
Psychological Sciences and Social Sciences, 58B(3), P153–P165.
Munafò, M. R. (2009). Behavioural genetics: From variance to DNA (pp. 287–304). New
York, NY: Cambridge University Press.
Nettle, D. (2005). An evolutionary approach to the extraversion continuum. Evolution
and Human Behavior, 26(4), 363–373.
Nettle, D. (2006). The evolution of personality variation in humans and other animals.
Newcomb, A. F., Bukowski, W. M., & Pattee, L. (1993). Children’s peer relations.
Psychological Bulletin, 113(1), 99–128.
Replicated factors structure in peer nomination personality ratings. Journal of
Abnormal and Social Psychology, 66(6), 574–583.
Oerlemans, W. G., & Bakker, A. B. (2014). Why extraverts are happier: A day
reconstruction study. Journal of Research in Personality, 50, 11–22.
Olson, K. R., & Weber, D. A. (2004). Relations between Big Five traits and fundamental
motives. Psychological Reports, 95(3), 795–802.
Omura, K., Constable, R., & Canli, T. (2005). Amygdala gray matter concentration is
associated with extraversion and neuroticism. Neuroreport: For Rapid Communication
of Neuroscience Research, 16(17), 1905–1908.
Ortony, A., Norman, D. A., & Revelle, W. (2005). Effective functioning: A three level
model of affect, motivation, cognition, and behavior (pp. 173–202). New York, NY:
Oxford University Press.
Paunonen, S. V. (2003). Big Five factors of personality and replicated predictions of
behavior. Journal of Personality and Social Psychology, 84(2), 411–422.
Penke, L., Denissen, J. J. A., & Miller, G. F. (2007). The evolutionary genetics of
Pennebaker, J. W., & King, L. A. (1999). Linguistic styles: Language use as an individual
difference. Journal of Personality and Social Psychology, 77(6), 1296–1312.
Pettersson, E., & Turkheimer, E. (2014). Self-reported personality pathology has complex
structure and imposing simple structure degrades test information. Multivariate
Behavioral Research, 49(4), 372–389.
when we assess a Big 5 trait? A content analysis of the affective, behavioral and
cognitive processes represented in the Big 5 personality inventories. Personality and
Quilty, L. C., DeYoung, C. G., Oakman, J. M., & Bagby, R. M. (2014). Extraversion and
behavioral activation: Integrating the components of approach. Journal of Personality
Assessment, 96(1), 87–94.
Raggatt, P. (2006). Putting the Five-Factor Model into context: Evidence linking Big Five
traits to narrative identity. Journal of Personality, 74(5), 1321–1347.
Rammsayer, T. H. (1998). Extraversion and dopamine: Individual differences in response
to changes in dopaminergic activity as a possible biological basis of extraversion.
European Psychologist, 3(1), 37–50.
Rammstedt, B., & John, O. P. (2007). Measuring personality in one minute or less: A 10-
item short version of the Big Five inventory in English and German. Journal of
Rauch, S. L., Milad, M. R., Orr, S. P., Quinn, B. T., Fischl, B., & Pitman, R. K. (2005).
Orbitofrontal thickness, retention of fear extinction, and extraversion. Neuroreport:
For Rapid Communication of Neuroscience Research, 16(17), 1909–1912.
Rauthmann, J. F., & Will, T. (2011). Proposing a multidimensional Machiavellianism
conceptualization. Social Behavior & Personality: An International Journal, 39(3),
391–403.
Read, S. J., Monroe, B. M., Brownstein, A. L., Yang, Y., Chopra, G., & Miller, L. C.
(2010). A neural network model of the structure and dynamics of human personality.
Rentfrow, P. J., Gosling, S. D., Jokela, M., Stillwell, D. J., Kosinski, M., & Potter, J.
(2013). Divided we stand: Three psychological regions of the United States and their
political, economic, social, and health correlates. Journal of Personality and Social
Psychology, 105(6), 996–1012.
Reuter, M. (2007). Neuroimaging and genetics. Cambridge, England: Cambridge
University Press.
Revelle, W. (1995). Personality processes. Annual Review of Psychology, 46, 295–328.
Revelle, W. (1997). Extraversion and impulsivity: The lost dimension? In H. Nyborg
(Ed.), The scientific study of human nature: Tribute to Hans J. Eysenck at eighty (pp.
189–212). Amsterdam, The Netherlands: Pergamon/Elsevier Science Inc.
Revelle, W. (2008). The contribution of reinforcement sensitivity theory to personality
theory (pp. 508–527). Cambridge, England: Cambridge University Press.
Revelle, W., Wilt, J., & Condon, D. M. (2011). Individual differences and differential
psychology: A brief history and prospect (Chap. 1, 2nd ed., pp. 3–38). London,
England: Wiley-Blackwell.
Roberts, B. W., & Robins, R. W. (2000). Broad dispositions, broad aspirations: The
intersection of personality traits and major life goals. Personality and Social
Psychology Bulletin, 26(10), 1284–1296.
Roberts, B. W., & Robins, R. W. (2004). Person-environment fit and its implications for
personality development: A longitudinal study. Journal of Personality, 72(1), 89–110.
Robinson, M. D. (2007). Personality, affective processing, and self-regulation: Toward
process-based views of extraversion, neuroticism, and agreeableness. Social and
Robinson, S., Sotak, B. N., During, M. J., & Palmiter, R. D. (2006). Local dopamine
production in the dorsal striatum restores goal-directed behavior in dopamine-deficient
mice. Behavioral Neuroscience, 120(1), 196–200.
Rocklin, T., & Revelle, W. (1981). The measurement of extraversion: A comparison of
the Eysenck Personality Inventory and the Eysenck Personality Questionnaire. British
Journal of Social Psychology, 20(4), 279–284.
Rogers, G. M., & Revelle, W. (1998). Personality, mood, and the evaluation of affective
and neutral word pairs. Journal of Personality and Social Psychology, 74(6), 1592–
1605.
Romero, E., Villar, P., Luengo, M. A., & Gómez-Fraguela, J. A. (2009). Traits, personal
strivings and well-being. Journal of Research in Personality, 43(4), 535–546.
Rothbart, M., & Bates, J. (1998). Temperament (Vol. III, pp. 105–176). New York, NY:
Wiley.
Rothbart, M. K. (1981). Measurement of temperament in infancy. Child Development,
52(2), 569–578.
Rothbart, M. K., Derryberry, D., & Hershey, K. (2000). Stability of temperament in
childhood: Laboratory infant assessment to parent report at seven years (Chap. 4, pp.
85–120). New York, NY: Routledge.
Sackett, P. R., & Walmsley, P. T. (2014). Which personality attributes are most important
in the workplace? Perspectives on Psychological Science, 9(5), 538–551.
Samuel, D. B., Carroll, K. M., Rounsaville, B. J., & Ball, S. A. (2013). Personality
disorders as maladaptive, extreme variants of normal personality: Borderline
personality disorder and neuroticism in a substance using sample. Journal of
Personality Disorders, 27(5), 625–635.
Schaller, M., & Murray, D. R. (2008). Pathogens, personality, and culture: Disease
prevalence predicts worldwide variability in sociosexuality, extraversion, and openness
to experience. Journal of Personality and Social Psychology, 95(1), 212–221.
doi:10.1037/0022–3514.95.1.212.
Scherer, K. R. (1995). Plato’s legacy: Relationships between cognition, emotion, and
motivation. Geneva Studies in Emotion and Communication, 9(1), 1–7.
Schutte, N. S., Malouff, J. M., Segrera, E., Wolf, A., & Rodgers, L. (2003). States
reflecting the Big Five dimensions. Personality and Individual Differences, 34(4), 591–
603.
Segarra, P., Poy, R., López, R., & Moltó, J. (2014). Characterizing Carver and White’s
BIS/BAS subscales using the Five Factor Model of personality. Personality and
Sellbom, M., Anderson, J. L., & Bagby, R. M. (2013). Assessing DSM-5 Section III
personality traits and disorders with the MMPI-2-RF. Assessment, 20(6), 709–722.
Shweder, R. A. (1999). Why cultural psychology? Ethos, 27(1), 62–73.
Singer, J. A. (2004). Narrative identity and meaning making across the adult lifespan: An
introduction. Journal of Personality, 72(3), 437–459.
Skodol, A. E., Clark, L. A., Bender, D. S., Krueger, R. F., Morey, L. C., Verheul, R., …
Oldham, J. M. (2011). Proposed changes in personality and personality disorder
assessment and diagnosis for DSM-5 Part I: Description and rationale. Personality
Smillie, L. (2008). What is reinforcement sensitivity theory. European Journal of
Personality, 22(5), 359–384.
Smillie, L. D. (2013). Why does it feel good to act like an extravert? Social and
Smillie, L. D. (2015). What is reinforcement sensitivity? Neuroscience paradigms for
approach-avoidance processes in personality.
Smillie, L. D., Cooper, A. J., Wilt, J., & Revelle, W. (2012). Do extraverts get more bang
for the buck? Refining the affective-reactivity hypothesis of extraversion. Journal of
Personality and Social Psychology, 103(2), 306.
Smillie, L. D., DeYoung, C. G., & Hall, P. J. (2014). Clarifying the relation between
extraversion and positive affect. Journal of Personality, 83(5), 564–574.
Smillie, L. D., Geaney, J. T., Wilt, J., Cooper, A. J., & Revelle, W. (2013). Aspects of
extraversion are unrelated to pleasant affective-reactivity: Further examination of the
affective-reactivity hypothesis. Journal of Research in Personality, 47(5), 580–587.
Smillie, L. D., Pickering, A. D., & Jackson, C. J. (2006). The new reinforcement
sensitivity theory: Implications for personality measurement. Personality and Social
Psychology Review, 10(4), 320–335.
Smillie, L. D., Wilt, J., Kabbani, R., Garratt, C., & Revelle, W. (2015). Quality of social
experience explains the relation between extraversion and positive affect. Emotion,
15(3), 339–349.
Smits, D. J. M., & Boeck, P. D. (2006). From BIS/BAS to the Big Five. European
Smits, I. A. M., Dolan, C. V., Vorst, H. C. M., Wicherts, J. M., & Timmerman, M. E.
(2011). Cohort differences in Big Five personality factors over a period of 25 years.
Journal of Personality and Social Psychology, 100(6), 1124–1138.
Soto, C. J., & John, O. P. (2014). Traits in transition: The structure of parent-reported
personality traits from early childhood to early adulthood. Journal of Personality,
82(3), 182–199.
Spain, J. S., Eaton, L. G., & Funder, D. C. (2000). Perspectives on personality: The
relative accuracy of self versus others for the prediction of emotion and behavior.
Specht, J., Egloff, B., & Schmukle, S. C. (2011). Stability and change of personality
across the life course: The impact of age and major life events on mean-level and rank-
order stability of the Big Five. Journal of Personality and Social Psychology, 101(4),
862–882.
Srivastava, S., Angelo, K. M., & Vallereux, S. R. (2008). Extraversion and positive
affect: A day reconstruction study of person-environment transactions. Journal of
Stelmack, R. M. (1990). Biological bases of extraversion: Psychophysiological evidence.
Stelmack, R. M. (1997). Toward a paradigm in personality: Comment on Eysenck’s
(1997) view. Journal of Personality and Social Psychology, 73(6), 1238–1241.
Stelmack, R. M., & Stalikas, A. (1991). Galen and the humour theory of temperament.
Personality and Individual Differences, 12(3), 255–263.
Strelau, J. (1987). Emotion as a key concept in temperament research. Journal of
Swickert, R., Hittner, J. B., Kitos, N., & Cox-Fuenzalida, L.-E. (2004). Direct or indirect,
that is the question: A re-evaluation of extraversion’s influence on self-esteem.
Tellegen, A. (1982). Brief manual for the differential personality questionnaire.
Unpublished manuscript, University of Minnesota, Minneapolis.
Tellegen, A. (1985). Structures of mood and personality and their relevance to assessing
anxiety, with an emphasis on self-report (pp. 681–706). Hillsdale, NJ: Lawrence
Erlbaum Associates.
Thomsen, D. K., Olesen, M. H., Schnieber, A., & Tonnesvang, J. (2013). The emotional
content of life stories: Positivity bias and relation to personality. Cognition & Emotion,
28(2), 260–277.
Thoresen, C. J., Kaplan, S. A., Barsky, A. P., Warren, C. R., & de Chermont, K. (2003).
The affective underpinnings of job perceptions and attitudes: A meta-analytic review
and integration. Psychological Bulletin, 129(6), 914–945. doi:10.1037/0033–
2909.129.6.914.
Thorne, A., & Nam, V. (2009). The storied construction of personality (Chap. 28, pp.
491–505). Cambridge, England: Cambridge University Press.
Tooby, J., & Cosmides, L. (1990). On the universality of human nature and the
uniqueness of the individual: The role of genetics and adaptation. Journal of
Personality, 58(1), 17–67.
Trull, T. J., & Sher, K. J. (1994). Relationship between the Five-Factor Model of
personality and Axis I disorders in a nonclinical sample. Journal of Abnormal
Psychology, 103(2), 350–360.
Tupes, E. C., & Christal, R. E. (1961). Recurrent personality factors based on trait ratings.
Technical Report 61-97, USAF ASD Technical Report. Reprinted in Journal of
Personality, 1992, 60, 225–251.
Twenge, J. M. (2001). Birth cohort changes in extraversion: A cross-temporal meta-
analysis, 1966–1993. Personality and Individual Differences, 30(5), 735–748.
Uziel, L. (2006). The extraverted and the neurotic glasses are of different colors.
Vaidya, J. G., Gray, E. K., Haig, J., & Watson, D. (2002). On the temporal stability of
personality: Evidence for differential stability and the role of life experiences. Journal
of Personality and Social Psychology, 83(6), 1469–1484.
Van den Akker, A. L., Prinzie, P., Deković, M., De Haan, A. D., Asscher, J. J., &
Widiger, T. (2013). The development of personality extremity from childhood to
adolescence: Relations to internalizing and externalizing problems. Journal of
Personality and Social Psychology, 105(6), 1038.
Van der Werff, J. (1985). Heymans’ temperamental dimensions in personality research.
Journal of Research in Personality, 19(3), 279–287.
van Os, J., Lataster, T., Delespaul, P., Wichers, M., & Myin-Germeys, I. (2014).
Evidence that a psychopathology interactome has diagnostic value, predicting clinical
needs: An experience sampling study. PloS One, 9(1), e86652.
Wacker, J., Chavanon, M.-L., & Stemmler, G. (2006). Investigating the dopaminergic
basis of extraversion in humans: A multilevel approach. Journal of Personality and
Watson, D. (1988). Intraindividual and interindividual analyses of positive and negative
affect: Their relation to health complaints, perceived stress, and daily activities.
Watson, D., & Clark, L. A. (1992). On traits and temperament: General and specific
factors of emotional experience and their relation to the Five-Factor Model. Journal of
Personality, 60(2), 441–476.
Watson, D., & Clark, L. A. (1997). Extraversion and its positive emotional core (pp. 767–
793). San Diego, CA: Academic Press.
Watson, D., Clark, L. A., McIntyre, C. W., & Hamaker, S. (1992). Affect, personality,
and social activity. Journal of Personality and Social Psychology, 63(6), 1011–1025.
Watson, D., Gamez, W., & Simms, L. J. (2005). Basic dimensions of temperament and
their relation to anxiety and depression: A symptom-based perspective. Journal of
Watson, D., Klohnen, E. C., Casillas, A., Nus Simms, E., Haig, J., & Berry, D. S. (2004).
Match makers and deal breakers: Analyses of assortative mating in newlywed couples.
Watson, D., Stasik, S. M., Ro, E., & Clark, L. A. (2013). Integrating normal and
pathological personality: Relating the DSM-5 trait-dimensional model to general traits
of personality. Assessment, 20(3), 312–326.
Watson, D., Suls, J., & Haig, J. (2002). Global self-esteem in relation to structural models
of personality and affectivity. Journal of Personality and Social Psychology, 83(1),
185–197.
Weiler, M. A. (1992). Sensitivity to affectively valenced stimuli. Ph.D. thesis.
Whiteside, S. P., & Lynam, D. R. (2001). The Five Factor Model and impulsivity: Using
a structural model of personality to understand impulsivity. Personality and Individual
Differences, 30(4), 669–689.
Widiger, T. A. (2005). Five Factor Model of personality disorder: Integrating science and
practice. Journal of Research in Personality, 39(1), 67–83.
Wiggins, J. S. (1996). An informal history of the interpersonal circumplex tradition.
Journal of Personality Assessment, 66(2), 217–233.
Wilkowski, B. M., & Ferguson, E. L. (2014). Just loving these people: Extraverts
implicitly associate people with reward. Journal of Research in Personality, 54, 93–
102.
Wilson, S., Schalet, B. D., Hicks, B. M., & Zucker, R. A. (2013). Identifying early
childhood personality dimensions using the California Child Q-Set and prospective
associations with behavioral and psychosocial development. Journal of Research in
Personality, 47(4), 339–350.
Wilt, J. (2014). A New Form and Function for Personality (Doctoral dissertation).
Retrieved from Proquest. (http://search.proquest.com/docview/1558901195)
Wilt, J., Cox, K. S., & McAdams, D. P. (2010). The Eriksonian life story: Developmental
scripts and psychosocial adaptation. Journal of Adult Development, 17(3), 156–161.
Wilt, J., Noftle, E. E., Spain, J. S., & Fleeson, W. (2011). The dynamic role of personality
states in mediating the extraversion–positive affect relationship. Journal of Personality,
80(5), 1205–1236.
Wilt, J., Oehlberg, K., & Revelle, W. (2011). Anxiety in personality. Personality and
Wilt, J., & Revelle, W. (2009). Extraversion (pp. 27–45). New York, NY: Guilford Press.
Winter, D. G., John, O. P., Stewart, A. J., Klohnen, E. C., & Duncan, L. E. (1998). Traits
and motives: Toward an integration of two traditions in personality research.
Wise, R. A. (2004). Dopamine, learning and motivation. Nature Reviews Neuroscience,
5(6), 483–494.
Wood, D., & Roberts, B. W. (2006). The effect of age and role information on
expectations for Big Five personality traits. Personality and Social Psychology
Bulletin, 32(11), 1482–1496.
Wundt, W. (1897). Outlines of psychology. Oxford, England: Engelmann.
Wyvell, C. L., & Berridge, K. C. (2000). Intra-accumbens amphetamine increases the
conditioned incentive salience of sucrose reward: Enhancement of reward “wanting”
without enhanced “liking” or response reinforcement. The Journal of Neuroscience,
20(21), 8122–8130.
Yang, Y., Read, S. J., Denson, T. F., Xu, Y., Zhang, J., & Pedersen, W. C. (2014). The
key ingredients of personality traits situations, behaviors, and explanations. Personality
and Social Psychology Bulletin, 40(1), 79–91.
Yarkoni, T. (2010). Personality in 100,000 words: A large-scale analysis of personality
and word use among bloggers. Journal of Research in Personality, 44(3), 363–373.
Zelenski, J. M., Whelan, D. C., Nealis, L. J., Besner, C. M., Santoro, M. S., & Wynn, J.
E. (2013). Personality and affective forecasting: Trait introverts underpredict the
hedonic benefits of acting extraverted. Journal of Personality and Social Psychology,
104(6), 1092–1108.
Zuckerman, M. (1991). Psychobiology of personality. Cambridge, England: Cambridge
University Press.
Zuckerman, M. (2005). Psychobiology of personality (2nd ed., rev. & updated). New
Openness
Angelina R. Sutin
Abstract
Despite its early struggles to survive, openness is now recognized as a personality
trait with far-reaching consequences. This chapter is an overview of how
individual differences in cognitive flexibility, sensitivity to aesthetics, depth of
feeling, and preference for novelty contribute to important domains of
functioning. Briefly reviewed will be conceptualizations of openness, some
measurement considerations, and where it fits within the nomological net of
related constructs. The chapter is then devoted to the nature and consequences of
openness, arranged from the biological to the societal. Research on the biological
roots of openness and its developmental trajectory from early childhood through
old age are then covered. Also considered is how openness contributes to nearly
every aspect of functioning, including health and well-being, employment, person
presentation and perception, marriage and family, and its geographic implications.
Key Words: openness, intellect, unconventionality, imagination, creativity, Five
Factor Model
Openness is perhaps the most resilient trait within the Five Factor Model
(FFM) of personality. In contrast to traits such as neuroticism, which are well
represented in most models of personality, openness has traditionally struggled
to be recognized, first as replicable and then as meaningful. Indeed, there was no
early consensus on this trait, and this disagreement contributed to the ambiguity
of openness and the question of its relevance. In the early days of the FFM,
when openness was retained as a meaningful trait, it was variously labeled
intellect, culture, imagination, and unconventionality. The diversity of labels
underscores both the heterogeneous nature of this trait and the need for a
unifying theoretical framework. Openness did survive, and the triumph of this
struggle indicates that individual differences in imagination, sensitivity to
aesthetics, depth of feeling, preference for novelty, cognitive flexibility, and
social and political values are universal and consequential.
The initial research on openness focused on its definition, measurement, and
links with related constructs, such as creativity. It soon became clear, however,
that the consequences of openness stretched far beyond intellectual and artistic
pursuits (e.g., McCrae, 1996). This chapter is primarily devoted to highlighting
the significance of openness to a range of important life outcomes. Because
others have conceptualized openness at length (e.g., DeYoung, Grazioplene, &
Peterson, 2012; McCrae & Costa, 1997), how openness has been defined, some
measurement considerations, and its placement within the nomological net of
related constructs will be considered only briefly. The bulk of the chapter will be
devoted to its nature and consequences, arranged from the biological to the
societal. Research on the biological roots of openness and its developmental
trajectory from early childhood through old age will first be described. Then the
importance of openness for nearly every aspect of functioning, including health
and well-being, employment, how it is perceived by others, how it shapes
personal relationships, and its geographic implications will be reviewed. This
chapter is not meant to be exhaustive but rather is a survey of some of the vast
literature that illustrates the extent of openness’s reach.
Openness: An Overview
Definition
The most cited definition of openness comes from McCrae and Costa (1997),
who argued that “Openness must be viewed in both structural and motivational
terms. Openness is seen in the breadth, depth, and permeability of
consciousness, and in the recurrent need to enlarge and examine experience” (p.
826). The breadth of openness refers to the wide range of interests that is so
characteristic of open people. Open people like to try new things and go to new
places. Their hobbies and interests are many and varied. Closed individuals, in
contrast, tend to be more set in their ways and prefer familiarity to learning new
things. The depth of openness refers to the density of associations that open
individuals hold between ideas. That is, ideas and concepts are associated in the
brain; open people easily make more remote and creative connections between
ideas, whereas individuals lower in openness tend to make more literal and
concrete connections. This ability is reflected in the correlates of openness, such
as better divergent thinking skills. Finally, permeability refers to malleability of
mental boundaries. This malleability is seen in everything from nuanced and
well-differentiated political beliefs to synesthesia (e.g., seeing music). And from
a motivational perspective, open individuals are inherently curious and have a
real need for variety and novelty and actively seek out such experiences.
Openness, perhaps more than any other of the FFM traits, has inspired debate
and controversy over how it should be defined. Like all FFM traits, openness
emerged and converged across the lexical tradition and from factor analysis of
psychological questionnaires. These two traditions historically emphasized
different aspects of openness (Connelly, Ones, & Chernyshenko, 2014). McCrae
and Costa (1997) derived their conception of openness from psychological
questionnaires, and thus focus relatively more on the experiential aspects of
openness. The lexical approach, in contrast, gives relatively more weight to the
intellectual aspects of this trait. When measured with adjectives, openness is
defined by words such as intelligent, clever, intellectual, thoughtful, and ignorant
(reverse scored), which clearly reflect intellectual engagement (Goldberg, 1990).
And, indeed, there are moderate correlations between aspects of cognitive
functioning and openness (DeYoung, Quilty, Peterson, & Gray, 2014). These
connections have led some to argue that intelligence is best construed as a facet
of openness (DeYoung, Grazioplene, & Peterson, 2012). Others contend that
openness and intelligence are best conceptualized as separate but related
individual differences (Connelly, Ones, & Chernyshenko, 2014; McCrae &
Greenberg, 2015).
Despite the debate, there is more consistency than differences in the way that
openness is typically conceptualized. One commonality across approaches is the
recognition that openness is a heterogeneous trait, with many separate tendencies
sharing an overall openness core. As such, it is helpful to distinguish more
specific components, or facets, of openness. As operationalized by the NEO
Personality Inventory-Revised (NEO-PI-R; Costa & McCrae, 1992), openness is
defined by a vivid imagination (O1: Fantasy), an appreciation of art and beauty
(O2: Aesthetics), depth of emotions (O3: Feelings), an eagerness to try new
things (O4: Actions), intellectual curiosity (O5: Ideas), and being liberal (O6:
Values). Other approaches to the facets have revealed a slightly different
structure. Connelly and colleagues (Connelly, Ones, Davies, & Birkland, 2014),
for example, used a critical incidents methodology to identify four distinct facets
—aestheticism, openness to sensations, nontraditionalism, and introspection—as
well as a number of “openness compounds” or facets with moderate correlations
with other FFM traits (e.g., thrill-seeking from extraversion, tolerance from
agreeableness). An exploratory factor analysis of 36 openness-related scales
revealed six facets—intellectual efficiency, ingenuity, curiosity, aesthetics,
tolerance, and depth—that replicated across samples and had good cross-cultural
equivalence (Woo et al., 2014). There may also be an intermediate level between
the broad openness domain and the more specific facets, with facets related to
cognitive flexibility grouped together under intellect and the more experiential
facets grouped together under openness (DeYoung, Quilty, & Peterson, 2007;
Woo et al., 2014). Despite disagreement about the number and exact content of
each facet, there is considerable overlap among the facets derived from these
different measurement approaches.
The debate over how openness should be conceptualized is not limited to
models of normal personality. The most recent revision of the Diagnostic and
Statistical Manual of Mental Disorders (i.e., the DSM-5) of the American
Psychiatric Association (APA, 2013) rekindled discussion about the utility of a
dimensional approach to personality disorders and where to place openness
within a model of abnormal personality. It has long been argued that
maladaptive personality disorders can be interpreted as variants of normal
personality functioning (see also the chapter by Widiger, Gore, Crego, Rojas,
and Oltmanns). Although the categorical approach to personality disorders was
retained in the current DSM revision, the APA did include a hybrid model with a
dimensional assessment in Section III for emerging models and measures. From
this dimensional perspective, personality disorders are viewed as maladaptive or
extreme variants of normal personality structure. Four of the proposed
dimensions map almost directly to the FFM factors: Negative affectivity
resembles neuroticism, detachment resembles (low) extraversion, antagonism
resembles (low) agreeableness, and disinhibition resembles (low)
conscientiousness.
A fifth (and sometimes sixth) factor typically emerges in factor analyses of
scales that measure maladaptive personality functioning. This factor has been
variously labeled oddity (Watson, Clark, & Chmielewski, 2008), peculiarity
(Tackett, Silberschmidt, Krueger, & Sponheim, 2009), and psychoticism (APA,
2013). Although the exact content of the factor varies depending on what scales
are analyzed, this fifth dimension, labeled psychoticism in the DSM-5, typically
reflects cognitive and perceptional dysregulation. Given that openness does not
correlate strongly with DSM personality disorders, psychoticism is thought by
some to reflect yet another dimensional trait that is more specific to abnormal,
than normal, personality.
Accumulating evidence, however, suggests that psychoticism belongs on the
same factor as openness. For example, a factor analysis of scales with items that
assess unconventionality, eccentricity, and peculiarity, which are thought to be
the maladaptive variants of openness, revealed that these scales load more
strongly on a factor with openness than on any of the other factors (Gore &
Widiger, 2013). Gore and Widiger (2013) point out that one reason openness
may be correlated only weakly with personality disorders is that items on
standard personality instruments, such as the NEO-PI-R, reflect the adaptive
aspects of high openness, not the potentially pathological aspects of this trait.
And indeed, when items on the NEO are changed to reflect maladaptive variants
of openness (e.g., “I have an excessive imagination” versus “I have an active
imagination”), openness is correlated with the DSM personality disorders
(Haigler & Widiger, 2001). And using scales that included items that tapped into
both positive and negative aspects of openness, Piedmont and colleagues
(Piedmont, Sherman, & Sherman, 2012; Piedmont, Sherman, Sherman, Dy-
Liacco, & Williams, 2009) identified two maladaptive variants of both low and
high openness (i.e., superficial, rigid and odd/eccentric, excessively unrestricted,
respectively). Thus, the association between openness and personality disorders
may be more limited by measurement issues than by conceptual ones.
Measurement Considerations
There are many scales that measure the FFM personality traits, including
openness. A thorough discussion of these scales is beyond the scope of this
chapter (see also the chapter by Simms, Williams, and Simms), but there are
some measurement considerations to keep in mind. As described above, there
are many facets of openness, and different personality measures emphasize
different components of this trait. Care should be taken when selecting a
measure to ensure that the scale includes the components of openness most
relevant to the research question. Likewise, because all openness scales are not
equivalent, care needs to be taken when evaluating the correlates of openness.
Different scales may produce different correlates, depending on which aspects of
openness are captured on each of the scales.
Several questionnaire scales, such as the full NEO PI-R and the shorter NEO-
FFI, offer a comprehensive assessment of openness and the more specific facets.
Despite this good domain coverage, there are also limitations to using these
measures. Of all the traits, the items that measure openness tend to be the most
complex and require a higher level of literacy to comprehend. Indeed, the items
that measure openness on the NEO require a 2.5 grade level higher literacy than
items that measure the other four factors, and literacy is a strong predictor of
data quality (Sutin, Costa, Evans, & Zonderman, 2013). In addition, the content
of the items may not be relevant for all demographic groups (Salva, Davey,
Costa, & Whitfield, 2007). Openness is the most complex of the five traits to
assess, and the expression of openness may be more context dependent than
items that measure the other traits. Despite these limitations, there still tends to
be adequate convergence with the normative structure across diverse groups
(McCrae & Terracciano, 2005a).
Nomological Net
It is helpful to place openness within the nomological net of related constructs
(McCrae, 1996; McCrae & Sutin, 2009). Openness shares conceptual and
empirical links with a number of constructs, including authoritarianism, need for
cognition, need for closure, emotional intelligence, and traits from other
conceptually similar models of personality.
Openness is strongly inversely related to authoritarianism. Individuals who
score low on openness tend to be conventional and conservative, and their
political attitudes tend to align with their psychological orientation. And indeed,
there is a strong negative association between openness and measures of “right
wing authoritarianism” (r = –.50, p < .001). This negative association tends to be
stronger for the more cognitive components of openness, rather than its more
experiential components (Onraet, Van Hiel, Roets, & Cornelis, 2011).
Openness is also strongly related to constructs commonly used in social
psychology, such as the need for cognition and the need for closure. The need
for cognition is defined as the extent to which people engage in and enjoy
activities that require cognitive effort (Petty, Briñol, Loersch, & McCaslin,
2009). Individuals high in the need for cognition seek out cognitively
challenging activities, whereas those low in this trait prefer activities that require
less thought. Given that open individuals and those high in the need for
cognition share an interest in intellectual engagement, there tends to be a
moderate correlation between the two (r = .41, p < .01; Fleischhauer, Enge,
Brocke, Ullrich, & Strobel, 2010). This correlation is higher than the correlation
between the need for cognition and any of the other four traits (r ≤ |.26|).
Individuals high in openness tend to score lower on the need for closure, defined
asthe desire for definite and final answers (Webster & Kruglanski, 1994).
One defining feature of openness is the experience of more differentiated
emotional states. This greater intrapersonal sensitivity to emotions may extend to
identifying emotions in other people. And, in fact, openness is associated with
better emotion recognition (Terracciano, Merritt, Zonderman, & Evans, 2003).
Open individuals are also more aware of their emotional states, express those
feelings to others, and are better able to regulate emotions in themselves and
others, all key components of “emotional intelligence” (Downey, Lee, & Stough,
2011).
A number of other models of personality have traits conceptually similar to
openness. For example, openness is strongly associated with the intuition type,
as defined by the Myers-Briggs Type Inventory (Costa & McCrae, 1992): Open
individuals prefer abstract and theoretical information (intuition), whereas closed
individuals prefer concrete and tangible information (sensing). The Experience
Seeking subscale of the Sensation Seeking scale is strongly related to openness
(Zuckerman, Kuhlman, Joireman, Teta, & Kraft, 1993). Openness is also related
to several scales from the Temperament and Character Inventory, including Self-
Transcendence, (low) Harm Avoidance, and Novelty Seeking (de Fruyt, van De
Wiele, & van Heeringen, 2000).
These constructs should not be construed as equivalent to openness. They are
all only moderately correlated with openness at best and each has a specific
content that gives it unique correlates. When considered together, however, the
core of each of these constructs would be defined primarily by an openness
factor (McCrae, 1996).
Openness: From Biology to Society

Biological Basis
Genetics. Estimates derived from twin studies routinely suggest that
approximately half of the variance in openness can be attributed to genetic
factors (e.g., Kendler & Myers, 2010; Riemann, Angleitner, & Strelau, 1997;
Vernon, Villani, Vickers, & Harris, 2008). Of the five traits, openness tends to
have the highest estimate of additive genetic effects and the lowest estimate of
nonshared environmental effects (see also the chapter by Jarnecke and South).
Similar to the other traits, there appears to be minimal shared environmental
associations. The comparatively larger additive genetic association with
openness may be due, in part, to the shared genetic association between
openness and general cognitive ability (Wainwright, Wright, Luciano, Geffen, &
Martin, 2008).
Findings from behavioral genetics inspired the search for the actual genes
associated with personality. There are two approaches to this search: candidate
genes and genome-wide association studies (GWAS). Within the candidate gene
approach, specific single nucleotide polymorphisms (SNPs) are hypothesized a
priori to be related to the trait. The candidate genes are usually selected based on
knowledge of the known biological action of the gene and its theoretical
relevance to the trait in question. Much of the research using this approach has
focused on neuroticism, but some work has addressed whether openness is
associated with genes within the dopaminergic system that have been implicated
in exploratory behavior. In a sample of children and adults, the associations
between openness and the dopamine D4 receptor gene (DRD4) and the catechol-
O-methyltransferase gene (COMT) were qualified by gene–gene interactions and
interactions with both demographic and environmental factors (DeYoung et al.,
2011). These interactions suggest that there is no straightforward association
between these SNPs and openness.
In contrast to the candidate gene approach, the GWAS approach is
atheoretical and completely data driven. GWAS scan the entire genome for
common genetic variation associated with the trait of interest. In general,
findings from GWAS have not supported the candidate genes hypothesized to be
associated with personality. In the first GWAS of openness, none of the top
SNPs passed the commonly accepted threshold for significance (p < 10–8), and
neither DRD4 nor COMT was among the top hits (Terracciano, Sanna, et al.,
2010). The largest GWAS meta-analysis to date (>17,000 participants) also did
not support the hypothesized candidate genes (de Moor et al., 2012). This study
did find an association between openness and two SNPs located downstream
from RASA1, a gene that is expressed weakly in the brain. Although these
associations passed the genome-wide significance threshold in the discovery
sample, neither SNP was significant in the replication sample (de Moor et al.,
2012). A recent GWAS also failed to replicate either the candidate genes or the
top hits for openness in previous GWAS (Bae et al., 2013). As such, consistent
genetic associations have yet to emerge for this trait.
The evidence for the molecular genetic basis of openness from large meta-
analyses, such as the one from de Moor and colleagues (2012), has been
disappointing. The scarcity of findings points to the critical need for larger
samples and the necessity of replication. It is possible that there are complex
interactions between individual SNPs that contribute to openness, and that these
interactions may differ across different demographic groups. At the same time,
interactions are difficult to find and difficult to replicate; care must be taken
before firm conclusions can be drawn. New technologies may be more
promising. For example, as the cost of sequencing the entire genome decreases,
more researchers will be able to take advantage of it. By increasing the coverage
of the genome, sequencing expands the number of variants that can be tested,
especially rare variants.
Neuroscience. The search for the neurobiological underpinnings of openness
to experience is more complex than the search for its genetic roots. Part of this
complexity stems from the number of ways to conceptualize neurological
correlates; the correlates of openness have been examined in the context of the
volume of brain structures, resting-state blood flow, and functional connectivity,
among others (see also the chapter by Allen and DeYoung).
One of the defining features of openness is cognitive flexibility, or the ability
to store and manipulate information with ease. Such flexibility requires good
working memory, attention, and set shifting ability, all of which have well-
established neural correlates. Some have used these correlates together with a
conceptual understanding of openness to make theory-driven hypotheses about
how openness should be associated with brain structure and function. DeYoung
and colleagues (2010), for example, hypothesized that openness would be
positively related to structural variation in the prefrontal cortex. Others have
noted the inconsistencies in the literature on personality and brain structure and
have thus taken a more exploratory approach (Bjørnebekk et al., 2013; Liu et al.,
2013). In each of these studies, using slightly different techniques, no correlation
between openness and structure emerged (Bjørnebekk et al., 2013; DeYoung et
al., 2010; Liu et al., 2013). Kapogiannis and colleagues (Kapogiannis, Sutin,
Davatzikos, Costa, & Resnick, 2013) also took an exploratory approach and
found that openness was associated with gray matter volume of several brain
structures, including areas in the prefrontal cortex implicated in cognitive
control. Given the inconsistencies in the literature, however, caution should be
taken when evaluating these results.
The literature on openness and both the resting-state correlates and functional
connectivity is likewise mixed. Open individuals tend to have greater resting-
state activity in regions associated with cognitive flexibility, monitoring
processes, and reward and emotional processing including the prefrontal cortex,
the orbitofrontal cortex, and the cingulate cortex (Sutin, Beason-Held, Resnick,
& Costa, 2009). Novelty seeking, a trait somewhat related to openness, also
correlates positively with some of these regions when measurements are taken at
rest (e.g., Nakao et al., 2013; Sugiura et al., 2000; Youn et al., 2002). Not all,
however, find these relations (e.g., Kunisato et al., 2011). Adelstein and
colleagues (2011) hypothesized that openness would be associated with greater
connectivity in regions implicated in cognitive flexibility, and, indeed, they
found that open individuals had greater functional connectivity in the
dorsolateral prefrontal cortex. Xu and Potenza (2012) also found that openness
was associated with better white matter integrity in this area. Others, however,
have either found associations in different regions (e.g., Jung, Grazioplene,
Caprihan, Chavez, & Haier, 2010) or none at all (e.g., Bjørnebekk et al., 2013).
The search for the neural underpinnings of openness has been somewhat
hampered by methodological limitations. First, the sample sizes tend to be
modest. Compared to genetic studies that typically include tens of thousands of
participants, neuroimaging studies use a few hundred participants at best. The
modest sample sizes are not surprising, given the time and expense of
neuroimaging. At the same time, openness is complex, and the neurobiological
correlates are likely to be small; large samples are thus needed to reliably detect
significant associations. Second, different measures of openness tend to be used
across different studies. Even well-validated scales may not have consistent
correlates because different scales emphasize different aspects of openness. A
facet-level approach to the neural underpinnings may reveal more consistent
associations. Finally, neuroimaging studies tend to use samples that vary widely
in age. Although not a limitation, this difference may make it difficult to
compare findings across studies, given the significant changes in brain structure
and function that come with aging.
Biomarkers. In addition to the genetic and neurological correlates, there are
numerous biomarkers of biological relevance to openness. Although some
biomarkers have already been found to be unrelated to openness, such as brain-
derived neurotropic factor (BDNF; Terracciano et al., 2010) and telomere length
(Sadahiro et al., 2015), there are other promising biomarkers that have yet to be
tested in relation to openness. Elevated homocysteine is a risk factor for
dementia (Seshadri et al., 2002), and gut microbiota have been linked to
psychological functioning, as well as physical health (Cryan & Dinan, 2012);
both are plausible links to openness. A better understanding of the physiological
underpinnings may provide valuable insight into the nature of openness.
DEVELOPMENTAL TRAJECTORY
There are lay theories about how openness changes across the lifespan,
theories that are shared by people all over the world. When participants from
over 20 cultures were asked to rate the level of openness of the average
adolescent, the average adult, and the average older person, the perceived
trajectory was a clear decline across the lifespan: Adolescents were perceived to
be the most imaginative, intellectually curious, and liberal, whereas older adults
were perceived to be the most practical, the most closed-minded, and the most
conservative, with middle-aged adults falling in between (Chan et al., 2012).
Although these perceptions map to actual age differences in openness, such
stereotypes of openness also exaggerate age differences. For example,
adolescents are perceived to be much higher on fantasy, actions, and values than
they actually are, whereas older adults are perceived to be lower on these traits
than they actually score (Chan et al., 2012).
The exact trajectory of openness across the lifespan, however, is debatable,
starting with the emergence of openness in childhood. Traditional models of
temperament do not include a childhood antecedent of openness, and early
research suggested that openness might not emerge as a recognizable trait until
adulthood (Eder, 1990). Recent advancements, including improvements in
measurement, have challenged this assertion (see also the chapter by De Pauw).
Openness is present and measurable even in early childhood. Interestingly
though, openness may be more closely related to conscientiousness than
extraversion in childhood (Soto & John, 2014), a pattern that replicates across
cultures (Tackett et al., 2012). In early childhood (~ages 3–5), openness is also
defined in part by motor activity and energy level, as well as by the more
expected characteristics of imagination and creativity. The shared association
with activity level suggests that openness is rooted in physical as well as
cognitive exploration (Soto & John, 2014). In middle childhood, openness is
clearly associated with both conscientiousness and extraversion, with intellect-
related items more strongly related to conscientiousness, and imagination-related
items more strongly related to extraversion (Herzhoff & Tackett, 2012).
Adolescence is a particularly important and complicated developmental
period. There is a consensus that adolescents should score high in openness
(Chan et al., 2012). Indeed, the very definition of adolescence is equivalent to
openness: It is a time for children to explore and try new things. As such,
openness should increase across adolescence, but the evidence is equivocal. The
meta-analysis of Roberts et al. (2006) pointed to a slight, but not significant,
increase in openness across adolescence (ages 10–18 years). Using cross-
sectional, longitudinal, and cross-cultural data, McCrae and colleagues (2002)
did document an increase in openness between the ages of 12 and 18 years.
Using a comprehensive measure of openness that included experiential items in
addition to intellectual interests, adolescents increased in “their appreciation of
aesthetics, tolerance of alternative value systems, and sensitivity to
moods/emotions” (p. 1465). This increase in openness may reflect the growth in
cognitive capacity that is one of the defining characteristics of adolescence
(McCrae et al., 2002). Other studies, however, document a decline. In a short-
term longitudinal study using parent reports of personality, a domain labeled
imagination, which included facets of curiosity and creativity, decreased
between childhood and early adolescence (De Fruyt et al., 2006). And from the
largest cross-sectional study to date, the age difference in openness was not in
the expected direction: 10-year-old participants scored higher in openness than
20-year-old participants (Soto, John, Gosling, & Potter, 2011).
The transition between adolescence and young adulthood is more clearly
marked by an increase in openness. American college students, for example,
increase in openness across their time in college (Robins, Fraley, Roberts, &
Trzesniewski, 2001), as do Italian (Vecchione, Alessandri, Barbaranelli, &
Caprara, 2012) and German (Lüdtke, Roberts, Trautwein, & Nagy, 2011)
students. Given that education may increase openness, the increase in this trait
among students might be more a function of going to school than a function of
development. To address this question, Lüdtke and colleagues (2011) compared
the trajectories of emerging adults who did vocational training with those who
entered a university. Although the university students scored higher in openness
than the students in vocational training, the trajectory of the groups was
identical. This increase in openness in young adulthood may thus be more of an
intrinsic developmental process than a consequence of higher education.
Across adulthood, the trajectory of openness parallels lay perceptions of
change, but the exact shape of the decline is less clear. Terracciano and
colleagues (Terracciano, McCrae, Brant, & Costa, 2005) used hierarchical linear
modeling to model up to 11 assessments of openness per participant. At the
domain level, openness clearly declined between the ages of 30 and 90 years.
The trajectory of some of the facets, however, diverged from this overall pattern.
In particular, general tendencies toward an interest in art and beauty
(Aesthetics), an active imagination (Fantasy), and interests in intellectual
pursuits (Ideas) remained consistent across adulthood, whereas the general
tendencies toward being open-minded (Values), interested in trying new things
(Actions), and depth of emotion (Feelings) declined considerably from young
adulthood to old age. This pattern suggests that openness measures that have
more items related to values, actions, and feelings should show a greater decline
than measures weighted toward aesthetics, fantasy, and ideas.
A slightly different trajectory emerges from large-scale national samples that
now have personality measured twice, approximately 4 years apart. National
samples from the United States (Stephan, Sutin, & Terracciano, 2015), Germany
(Lucas & Donnellan, 2011), and Australia (Wortman, Lucas, & Donnellan,
2012) all show that openness declines over the 4-year follow-up period. The
cross-sequential nature of these studies also allows for modeling the trend across
adulthood. In this case, openness is relatively flat up until about age 60 years,
with accelerated decline from age 60 years into old age (Lucas & Donnellan,
2011; Wortman et al., 2012).
Longitudinal studies thus indicate that, in general, people become more
conservative with age. Studies of cross-sectional age differences, however,
challenge these findings. With a sample of over 1 million participants drawn
from internet users, Soto and colleagues (2011) found that 60-year-old
participants scored higher on openness than 20-year-old participants. A second
internet sample found curvilinear age differences, such that middle-aged
participants scored higher on openness, with adolescent and older participants
scoring lower; this pattern was similar across three different measures of
openness (Lehmann, Denissen, Allemand, & Penke, 2013). This different pattern
across cross-sectional internet studies versus longitudinal studies may be due in
large part to selection bias. That is, when deciding whether to participate in an
internet study, individuals who are more intellectually curious and introspective
may be particularly excited to fill out a questionnaire to learn about their
personality. This effect may be even stronger at older ages. Selection bias is also
likely to have an effect in other types of studies. That is, open individuals are
more likely to volunteer to participate in research. From this perspective,
longitudinal studies are particularly critical to track development; participants
must be followed longitudinally to see actual age-related decline. There are
limits to longitudinal studies as well, as open individuals are more likely to
remain in a study than more closed individuals (Jerant, Chapman, Duberstein, &
Franks, 2009). These trends need to be kept in mind in both designing studies
and interpreting results. A more concerted effort to provide incentives for closed
individuals to participate and remain in research studies would help to reduce
these biases.
Finally, developmental time is typically measured as time since birth (i.e.,
chronological age), but there are alternative measures. Subjective age refers to
how old (or young) someone feels relative to his or her chronological age. Open
individuals tend to feel younger than their chronological age, an association that
grows stronger with chronological age (Stephan, Demulier, & Terracciano,
2012). Stephan and colleagues argue that “from middle-to-old age, open
individuals’ tendency to search for a variety of new ideas, values, and
experiences could lead them to have interest and activities more indicative of
younger people than of people their own age, leading them to feel younger than
their actual age” (p. 878). A younger subjective age may also have implications
for the developmental trajectory of openness: Instead of showing the typical age-
related decline, people who feel younger than their age increase in their tendency
to be open to new experiences (Stephan et al., 2014).
PHYSICAL, MENTAL, AND COGNITIVE HEALTH

Physical health. Research on the personality predictors of longevity has
focused primarily on neuroticism and conscientiousness. Openness, in fact, is
more often than not unrelated to mortality risk. A study of over 76,000 adults
with a mean follow-up period of almost 6 years, for example, found that open
adults did not die any sooner or later than more closed individuals (Jokela et al.,
2013). The association between openness and longevity, however, may be more
nuanced than simply domain-level associations. In one study, individuals who
had a high preference for novelty (i.e., openness to actions) had a reduced risk of
all-cause mortality and individuals who appreciated art and beauty (i.e.,
openness to aesthetics) had a reduced risk of cardiac death, whereas domain-
level openness was unrelated to either all-cause or cardiac mortality (Jonassaint
et al., 2007). Likewise, in another sample, creativity, but not intellect, was
associated with a reduced mortality risk, whereas there was no association at the
domain level (Turiano, Spiro, & Mroczek, 2012).
This same pattern holds for intermediate markers of health: Facet-level
associations tend to be more consistent than domain-level correlates. For
example, there is no clear association between domain-level openness and
systemic inflammation, measured with either interleukin-6 or C-reactive protein.
Although some studies find that domain-level openness is associated with lower
inflammation (Armon, Melamed, Shirom, Berliner, & Shapira, 2013; Chapman
et al., 2011), others have not (Sutin et al., 2010; Turiano, Mroczek, Moynihan, &
Chapman, 2013). In addition, when associations are found, they typically hold
for some populations but not others (e.g., Jonassaint et al., 2010) or have
complex relations (e.g., Mõttus, Luciano, et al., 2013). The aesthetics facet,
however, has been associated consistently with lower inflammation (e.g., lower
interleukin-6 and C-reactive protein (e.g., Chapman et al., 2011; Sutin et al.,
2010), even when openness at the domain level is unrelated to inflammation.
Individuals who are moved by works of art tend to have more favorable
inflammatory profiles.
Openness may be unrelated to risk of disease (Sutin, Zonderman, Ferrucci, &
Terracciano, 2013), but when individuals do get sick, those with experientially
rich lives may have slower disease progression. Among HIV+ adults, those with
higher openness, particularly those who are open to aesthetics and ideas, have
slower rates of increase in viral load and slower rates of decrease in CD4 cells
over 4 years (Ironson, O’Cleirigh, Weiss, Schneiderman, & Costa, 2008).
Among the elderly, domain-level openness has been linked to a reduced risk of
walking limitations over a 6-year follow-up period (Tolea et al., 2012). And in a
community sample, over a 10-year follow-up, open individuals were less likely
to develop incident coronary heart disease than more conservative individuals
(H. B. Lee et al., 2014).
There are at least two pathways through which openness may contribute to
slower disease progression. First, openness may help buffer against the negative
effects of stressful situations. For example, Williams and colleagues (Williams,
Rau, Cribbet, & Gunn, 2009) found that during a laboratory stressor, more open
individuals showed greater stress regulation at the physiological level (e.g., less
blood pressure reactivity) as well as the emotional level (e.g., greater increases
in positive affect) level. Second, open individuals may be more open to try
experimental treatment options or alternative therapies (Honda & Jacobson,
2005).
In contrast to traits such as neuroticism and conscientiousness, behavioral
factors do not seem to contribute to the association between openness and health.
That is, openness tends to be unrelated to many health-risk behaviors, such as
smoking (Terracciano & Costa, 2004). Open individuals are also not more or
less likely to exercise than closed individuals, a pattern found in both younger
(Rhodes & Smith, 2006) and older (Stephan, Boiché, Canada, & Terracciano,
2014) adults. Open and closed individuals, however, may have different
motivations for engaging in physical activity. Open individuals tend to engage in
physical activity to manage stress and because they enjoy it, whereas open and
closed individuals are equally likely to engage in physical activity for health,
appearance, maintaining a healthy weight, or social reasons (Courneya &
Hellsten, 1998).
Openness does emerge as a consistent predictor of at least two health-related
behaviors: marijuana use and diet. From early studies (Eisenman, Grossman, &
Goldstein, 1980) to more recent ones (Fridberg, Vollmer, O’Donnell, & Skosnik,
2011), current and frequent marijuana users consistently score higher in
openness. This association may be due, in part, to the shared association between
openness and intelligence. For example, more intelligent 5-year-old children are
more likely to have ever tried marijuana by age 16 years and are more likely to
report use within the past 12 months at age 30 years (White & Batty, 2012).
Daly (2013) argued that openness may be the key to the association between
intelligence and cannabis use. Specifically, he found that openness explained the
marijuana-intelligence link, and he concluded that open individuals may “seek
out illicit substances and to select into cognitively stimulating environments that
improve neuropsychological functioning” (p. E979). Although openness is
unrelated to harder drug use in some samples (e.g., Terracciano, Löckenhoff,
Crum, Bienvenu, & Costa, 2008), some studies find that openness is protective
against current opiate/cocaine use (Kornør & Nordvik, 2007; Sutin, Evans, &
Zonderman, 2013). With any study of drug use, however, it is unclear whether
scores on openness are truly predictive of current use or whether the drug has an
effect on the individual’s level of openness.
Of the five traits, openness has the most consistent associations with healthier
eating patterns. In a large study of Estonians who ranged in age from 18 to 89
years, open individuals were more likely to report eating a “health aware” diet, a
diet characterized by a high consumption of cereal and dairy products, fish,
vegetables, and fruits. In addition, participants who scored lower in openness
were more likely to consume a “traditional” diet, a diet characterized by greater
consumption of potatoes, meat, meat products, and bread (Mõttus et al., 2012).
In a large sample of older Scottish adults, open individuals were more likely to
consume a Mediterranean (defined as the consumption of vegetables, fish,
poultry, pasta, oil, and beans) and health aware (defined as the consumption of
fruit and less consumption of meat, eggs, and alcohol) diet, whereas more
conservative individuals were more likely to consume a convenience diet
(defined as tinned vegetables, beans, meat or chicken pies, pastries and sausage,
mashed potatoes) and sweets (Mõttus, McNeill, et al., 2013). And in a large
sample of older Finnish adults, openness was associated with consuming more
fruits and vegetables and fewer sweets and soft drinks, but was unrelated to the
consumption of cereals, dairy, fats, meat, or fish (Tiainen et al., 2013).
This evidence thus suggests that open individuals eat healthier diets than more
closed individuals. This healthy eating pattern, however, does not translate into
healthier body weight. In American (Chapman, Fiscella, Duberstein, Coletta, &
Kawachi, 2009; Sutin, Ferrucci, Zonderman, & Terracciano, 2011), British
(Cheng & Furnham, 2013; Mõttus, McNeill, et al., 2013), Italian (Terracciano et
al., 2009), and Israeli (Armon, Melamed, Shirom, Shapira, & Berliner, 2013)
samples, openness is unrelated to body mass index (BMI) and risk of obesity
(i.e., BMI ≥ 30). A meta-analysis of nearly 80,000 participants found that
openness was associated with a lower risk of obesity, but the effect disappeared
after controlling for education (Jokela et al., 2013). This pattern suggests that
any protective effect of openness on BMI likely functions through higher
education. Still, some have found that openness is associated with lower waist
circumference (van Reedt Dortland, Giltay, van Veen, Zitman, & Penninx,
2012), and there is also evidence that the association between openness and
anthropomorphic measures may be complex. For example, in a study of midlife
adults, open men had lower BMI than more traditional men, an association that
decreased across middle adulthood, whereas openness was unrelated to BMI for
women across this time period (Brummett et al., 2006). In the context of the
association with a healthier diet, it may be that open individuals consume better
foods, but still consume a similar number of calories as more closed individuals.
Mental health. As described above, there is great interest in integrating
normal and pathological personality functioning for all five traits, and there has
been much discussion about the conceptual and empirical overlap between
openness and psychoticism (as defined by the DSM-5). In addition to personality
disorders, openness may play a role in other aspects of mental health. Rather
than being a strong component of mental disorders (such as neuroticism),
openness may help to differentiate among similar but distinct disorders. In
clinical samples, openness is the defining trait difference in distinguishing
between bipolar disorder (BD) and major depressive disorder (MDD). That is,
the personality profiles of BD and MDD are identical except for openness:
Individuals with BD and MDD score higher in neuroticism and lower in
extraversion, agreeableness, and conscientiousness, but BD is characterized by
high openness, whereas openness is unrelated to MDD (Barnett et al., 2011). As
such, high openness seems to be a risk factor for mania or hypomania versus
other disorders with an affective component. Although openness predicts the
type of depressive disorder, it is generally unrelated to manic symptomatology
(Quilty, Pelletier, DeYoung, & Bagby, 2013) and does not seem to share genetic
roots with BD (Middeldorp et al., 2011).
Just as openness appears to differentiate BD from MDD, it also may
differentiate positive from negative schizotypy. Symptoms of positive
schizotypy generally refer to beliefs in things that are implausible and perceptual
and bodily distortions that are schizophrenic-like, whereas symptoms of negative
schizotypy generally refer to deficits in sensory and aesthetic pleasure. Cluster
analyses of schizotypal symptoms show that positive schizotypy is characterized
by higher openness, whereas low openness is characteristic of negative
schizotypy (e.g., Kwapil, Barrantes-Vidal, & Silvia, 2008). In fact, there is about
a one standard deviation difference in openness between the positive and
negative schizotypy clusters (Barrantes-Vidal, Lewandowski, & Kwapil, 2010).
This difference may be one reason that openness is not consistently associated
with schizotypy, unless the positive and negative aspects of it are distinguished.
In addition to psychiatric disorders, personality traits contribute to more
general evaluations of subjective well-being, such as life satisfaction and
happiness. The role of openness in these evaluations is more limited than that of
the other traits. Open individuals experience a broader range and depth of
emotions but that does not translate into feeling happier or sadder, on average
(Lamers, Westerhof, Kovács, & Bohlmeijer, 2012). Although openness tends to
be unrelated to measures of happiness and life satisfaction, open individuals tend
to have greater psychological well-being. That is, individuals who are
intellectually curious and have experientially rich lives report more autonomy, a
greater sense of mastery, and more personal growth (Cox, Wilt, Olson, &
McAdams, 2010; Lamers et al., 2012). Openness is also the trait that has the
strongest association with societal engagement. Open individuals report making
more positive contributions to society and the world (Cox et al., 2010). This
association may be due, in part, to the greater generativity among open
individuals.
Cognitive health. Open individuals are perceived to be verbally fluent,
humorous, and expressive (Sneed, McCrae, & Funder, 1998). This shared
perception may be due to their fluency with language, a fluency that is apparent
across the lifespan. Open adolescents, for example, score higher on the verbal
section of aptitude tests (Noftle & Robins, 2007), open young adults perform
well on vocabulary and comprehension tasks (Bates & Shieles, 2003), open
older adults score higher on verbal measures of cognition, such as analogies and
synonyms (Sharp, Reynolds, Pedersen, & Gatz, 2010), and across the lifespan,
open individuals perform better on verbal fluency tasks (Sutin, Terracciano, et
al., 2011). Open individuals may perform better on verbal tasks both because of
their reading abilities (Ritchie, Luciano, Hansell, Wright, & Bates, 2013) and
because they tend to have better executive functioning (Ayotte, Potter, Williams,
Steffens, & Bosworth, 2009), which contributes to the cognitive flexibility
needed to perform well on verbal fluency tasks. In addition, open people tend to
seek out new experiences, which may naturally lead to more opportunities to
learn new things.
With the increasing prevalence of dementia worldwide, identifying factors
that help maintain cognitive function in older adulthood is critical. It is often
asserted that openness should be protective against Alzheimer’s disease and
other forms of dementia. After all, open individuals tend to engage in the
behaviors that form the core of the cognitive reserve/resilience hypothesis. Yet
openness rarely emerges as a significant predictor of Alzheimer’s disease. In
fact, it took a meta-analysis of over 5,000 participants for the protective effect of
openness to emerge (Terracciano et al., 2014). Even then, neuroticism and
conscientiousness were stronger predictors of Alzheimer’s disease than
openness.
Openness is likewise unrelated to cognitive decline prior to dementia onset. It
appears that going in to old age, open individuals have higher cognitive
functioning but decline at the same rate as less open individuals. In a sample of
older adults, for example, open participants performed well on tasks that
measured reading, memory, and nonverbal reasoning; these effects persisted
even after controlling for childhood intelligence. Over the follow-up period,
however, openness was unrelated to change in these cognitive dimensions
(Hogan, Staff, Bunting, Deary, & Whalley, 2012). Similarly, in another sample
of older adults, open individuals performed better on measures of five different
aspects of cognition (verbal, spatial, memory, speed, and general intelligence),
but these individuals declined at the same rate as more closed individuals (Sharp
et al., 2010).
Such findings are surprising, given the strong associations between openness
and intelligence and cognitive functioning in young adulthood. Part of the
explanation may lie in the shared association among cognition, openness, and
education. Education is among the most protective behavioral factors against
dementia and cognitive decline (Barnes & Yaffe, 2011), and open individuals
are more likely to seek out opportunities for higher education. Studies of
personality and cognition in older adulthood typically control for the effect of
education, which may absorb the effect of openness along with it. There are, of
course, plenty of open people who do not have the opportunity to continue in
school and vice versa. For people with less education, openness may be more
protective of cognitive function, whereas this protective effect may diminish
with more education (Franchow, Suchy, Thorgusen, & Williams, 2013).
It is also surprising given that open individuals engage in cognitively
stimulating behaviors that are thought to help preserve cognition. In two large
samples, for example, openness was strongly associated with “developmental”
activities, including reading, writing letters, stories, or journal entries, using
technology, and attending educational lectures or courses (Stephan et al., 2014).
At the same time, however, closed individuals are more likely to participate in
religious activities (e.g., Jopp & Hertzog, 2010), and this social engagement may
also be protective against cognitive decline in old age (Fratiglioni, Paillard-Borg,
& Winblad, 2004). As such, the tendencies and behavioral patterns associated
with both ends of the openness continuum may be protective, albeit for different
reasons.
Employment
For most adults, employment is of great economic and psychological
importance. In addition to the economic necessity, working can be a significant
source of stress and/or personal fulfillment. Given the economic and
psychological significance of employment to both the individual and society,
there is tremendous interest in the role of personality traits in the employment
experience (see also the chapter by Siebert and DeGeest). The classic 10-year
longitudinal study from Kohn and Schooler (1982) was among the first to show
that personality plays a role in work experiences. Specifically, men who scored
higher on ideational flexibility (i.e., openness) were employed in more
substantively complex jobs with less supervision, fewer working hours, and a
greater likelihood of a self-directed position at the 10-year follow-up. The
association between openness and work experiences was not unidirectional. Men
employed in jobs that required more complex work that had more time pressures
increased in ideational flexibility over 10 years, whereas men who were
supervised closely, worked in manual labor, or routinization work decreased.
Clausen and Gilens (1990) subsequently found a similar pattern among women.
These early studies set the stage for more systematic investigations of the role of
personality, particularly openness, in employment.
From looking for a job to retirement, openness is a key personality contributor
to this process. Open individuals, for example, seek out and prefer jobs that are
intrinsically motivating, such as jobs that have autonomy, variety, and an
opportunity for growth (Bipp, 2010). These preferences fit well with their
general tendencies for intellectual curiosity and preference for novelty. In
contrast to these intrinsic motivations, openness is unrelated to extrinsic
motivations to work: Open and closed individuals are both equally likely to
value job security, working hours, and relationships at work (Bipp, 2010).
Once in a job, the characteristics of the working environment vary greatly
across occupations. That is, some jobs are highly demanding, some jobs allow
for a great degree of autonomy and decision-making capabilities, and other jobs
are more restrictive. A longitudinal study that spanned approximately 10 years
partially replicated Kohn and Schooler’s (1982) classic findings with well-
validated measures of both personality and job characteristics (Sutin & Costa,
2010). Open individuals tended to be employed in positions with a great deal of
decision-making latitude, and this latitude increased over the 10 years between
assessments. Openness was unrelated to having either psychologically or
physically demanding jobs. Surprisingly, these characteristics of the working
environment were unrelated to changes in openness. Most individuals spend a
great deal of their waking hours working, and thus it is surprising that the day-
to-day working environment would have little effect on personality.
All jobs involve some evaluation of performance. In the classic meta-analysis
by Barrick and Mount (1991), openness was unrelated to job performance. The
meta-analytic correlation was essentially zero, which indicated that open people
did not perform much better or much worse than closed people. A number of
more recent studies have noted that looking only at domain-level openness
misses the complexity of the relation between openness and job performance.
And, indeed, facet-level analyses indicate a more nuanced story. Ziegler et al.
(2014) had supervisors rate their apprentices during their first year of training on
a number of work-related characteristics (e.g., acquired skills and knowledge,
teamwork). Across a range of professions, domain-level openness was unrelated
to training success, but openness to ideas and openness to fantasy both shared
significant correlations with supervisor ratings—in opposite directions. In
situations that require a great deal of flexibility and learning, individuals who are
intellectually curious may perform well, presumably because they use their
intellectual engagement to learn how to do the job effectively. In contrast,
individuals high in fantasy are rated worse, perhaps because their tendency to
daydream distracts them from learning how to do the job.
There are a number of ways to define a successful career, including the
objective prestige of the job, the amount of money earned, and the individual’s
own subjective evaluation. Open individuals tend to be employed in objectively
higher prestige jobs (defined as a combination of the education required for the
job and the average income of the profession, derived from the census, not the
individual’s actual education or income) but are not more likely to earn more
money than individuals lower in openness (Sutin, Costa, Miech, & Eaton, 2009).
The association between openness and the more subjective evaluation of success
is complex. Although some have found that openness is unrelated to job
satisfaction (Sutin et al., 2009), others have noted that this association varies by
the type of job and that collapsing across professions may obscure the relation
between openness and satisfaction (Lounsbury et al., 2003). For example, open
individuals are more satisfied with jobs in consulting, whereas closed individuals
are more satisfied in manufacturing. Open individuals also tend to report overall
satisfaction with their career choice, even if they are not necessarily satisfied
with their current job (Lounsbury et al., 2003); they also tend to draw more
meaning from their work (Bipp, 2010) than more conservative individuals.
In many jobs, it is essential to be able to work effectively in a team. Teams
can be more creative than individuals working alone. It turns out that the
personality composition of the team is important for the team’s creative output.
Similar to individual-level openness, teams high in openness tend to develop
more creative solutions to problems. But it is not just the average level of
openness that matters; the variability in openness within the team matters too.
That is, teams made up of individuals both high and low in openness perform
better than teams made up of individuals either all high or all low in openness
(Schilpzand, Herold, & Shalley, 2011). Groups may need some open team
members to generate unusual and different ideas, and other team members to
ground those ideas in reality. Openness, and the variability of openness within a
group, is unrelated to other group metrics, such as conflict or satisfaction with
the team (Tekleab & Quigley, 2014).
Finally, some people will need to switch jobs over the course of their working
lives and most people will retire. Openness is associated with these employment
transitions. For example, following unemployment, individuals who are more
open are reemployed more quickly (Kanfer, Wanberg, & Kantrowitz, 2001). As
retirement nears, open individuals endorse more aspirational reasons for
retirement (Robinson, Demetre, & Corney, 2010). For example, they report that
they have become more interested in their lifestyle than in how much money
they make and that they want more time to pursue new opportunities, such as
self-employment. Interestingly, openness is unrelated to either positive or
negative experiences at work that contribute to the decision to retire, and it is
unrelated to life satisfaction in retirement (Robinson et al., 2010). That is, both
open and closed individuals are equally likely to be satisfied with their
retirement.
Person Presentation and Perception

Open individuals are creative, curious, and crave variety; individuals lower on
openness tend to prefer the conventional and more conservative. These
characteristics of open and closed individuals are expressed across a wide
variety of circumstances and mediums. From physical living spaces to massively
multiplayer online role-play games, individuals express their personality in
numerous ways. Observers often pick up on these behavioral manifestations of
openness and can accurately judge openness based on these cues. At the same
time, lay perceivers have their own ideas about what is indicative of openness,
and these ideas are not always diagnostic.
In their physical spaces, such as offices and bedrooms, consistent with their
intellectual interests, open individuals tend to own and display a wider variety of
books and magazines (Gosling, Ko, Mannarelli, & Morris, 2002). Open
individuals also decorate these spaces in ways that are considered unique and
unusual. It is not just their physical space that is described as unusual. As judged
from photographs, open individuals themselves tend to look more distinctive and
more messy, and they also tend to look away from the camera. More
conservative individuals, in contrast, look healthier, more ordinary, and neater
(Naumann, Vazire, Rentfrow, & Gosling, 2009). Observers do not always pick
up on these cues. Whereas a distinctive appearance is diagnostic of openness and
is perceived as such by observers, cues such as smiling and looking energetic are
taken as signs of openness, when they are actually unrelated (Naumann et al.,
2009). Likewise, people judge cheerful and colorful offices as occupied by
someone high in openness, when in fact these cues are unrelated to openness
(Gosling et al., 2002). Laypersons confuse interpersonal openness with openness
to experience (Sneed et al., 1998) and thus may misattribute these cues to
openness.
For some cues there is strong consensus, but little accuracy. In one study of
World of Warcraft (WoW) usernames, people with and without experience with
the game rated the personality of the usernames. There was consensus among
both users and nonusers on all five of the traits, including openness. Despite the
consensus among observers, there was no correlation with actual scores on
openness—raters agreed with each other on what cues should be indicative of
openness, but there was little accuracy with actual cues (Graham & Gosling,
2012). Open players tended to have fantasy- and sci-fi-related names, whereas
more closed individuals used names that were funny, emotional, or self-
deprecating. Observers thought that closed individuals would use more
gendered-related names and more aggressive names, but these cues were
unrelated to the individual’s openness.
The rise of social media has led to new ways of examining how personality
leaves its residue in everyday life. In particular, the words that people use when
writing blogs, status updates, and tweets should reflect elements of their
characteristic ways of thinking, feeling, and behaving. Given that openness is
associated with verbal fluency, open individuals may be more expressive in their
writing. In an analysis of blog posts, openness correlated significantly with 393
words—over three times as many words as the trait with the next highest number
of significant correlations (agreeableness at 110 words). Many of the words
correlated with openness reflected the very nature of this trait. That is, open
individuals value artistic pursuits and use words such as poet, art, films, and
literature in their blogs (Yarkoni, 2010).
Open language analysis is one new and innovative way to characterize the
language of personality without being confined to predetermined word sets. This
approach makes use of “Big Data” to visualize the words and phrases most
commonly associated with a trait of interest. One benefit of this approach is that
it is purely data driven and thus has the power to uncover associations with a
wide variety of written expressions. For example, in an analysis of Facebook
updates from almost 70,000 participants, high openness was associated with
words that reflect artistic and intellectual interests such as music, writing,
universe, and book, and, more than any other trait, low openness was associated
with misspellings and the use of shorthand, such as “wat,” “ur,” and “2day.”
This difference reflects the difference in intellectual and cultural sophistication
between individuals high and low in openness; it could also be due to differences
in education. The language choices of open individuals also reflect their
intellectual and sensory pursuits, with words related to causation (makes, origin,
rationale, used, why), insight (accept, become, believe, know, recall), and
sensory processes (delicious, feel, flavor, sour, press). More closed individuals,
in contrast, make more references to their family (Kern et al., 2013).
This approach has likewise been used to test how accurately personality can
be judged from online language use. In an analysis of approximately 75,000
Facebook users, of the five traits, openness was the trait most accurately
predicted from language use (Schwartz et al., 2013). Consistent with other
analyses of language, words related to aesthetic interests and misspelled
words/contractions were the words that best discriminated openness. A similar
analysis of what people “Like” on Facebook also revealed that openness is the
trait best predicted by what users endorse. For example, knowing that someone
likes The Colbert Report is indicative of openness, whereas fewer open people
liked Mitt Romney (Kosinski, Stillwell, & Graepel, 2013). Individuals high (or
low) on openness tend to inhabit similar social spaces and interests and this is
reflected in what they like. As such, inferences about someone’s personality,
particularly their openness, can be accurately derived from their activity on
social media. That is, status updates, “likes,” and tweets reveal psychologically
meaningful information about the individual who posted it.
Finally, in any interpersonal context, individuals strive to present themselves
in a certain way, which may or may not reflect who they actually are. These
biases may be particularly likely to be present on social media, as people have a
lot of control over how they present themselves. And yet personality still comes
through accurately. In fact, how people present themselves on social networking
sites is more strongly associated with their actual personality than their ideal
personality; this effect is strongest for openness (Back et al., 2010).
Relationships, Marriage, and Family

In any relationship, the dynamics of the interaction are shaped, in part, by the
personalities of the individuals involved. Although true for any dyadic
interaction, most evidence comes from research on romantic relationships and
married couples. At each stage, from whom to date to whether to get married to
parenting, openness shapes choices, interactions, and consequences.
When asked what they are looking for in a partner, single individuals say they
want someone who resembles them on openness; by comparison, agreeableness
and extraversion are a distant second and third, respectively (Figueredo, Sefcek,
& Jones, 2006). When actually evaluating potential mates, there are sex
differences in the types of traits that people value. Asendorpf and colleagues
(Asendorpf, Penke, & Back, 2011) used a speed-dating paradigm to examine
what characteristics contributed to who the speed-daters would choose for actual
dates. Of the five personality traits, women tended to choose men who scored
the highest on openness, presumably because openness is a marker of future
resource potential. Men, in contrast, did not evaluate women based on any FFM
trait. Men’s own openness, however, may shape what they look for in a mate.
For example, when evaluating potential romantic partners, open men care more
about perceived intelligence than the facial attractiveness of a potential mate.
Women’s own openness, however, is unrelated to how she evaluates a potential
mate (A. J. Lee, Dubbs, von Hippel, Brooks, & Zietsch, 2014).
Psychodynamic approaches to mate selection traditionally indicate that
individuals are motivated (consciously or not) to select mates who are similar to
their parents. McCrae and colleagues (McCrae, Willemsen, & Boomsma, 2012)
tested this hypothesis for personality traits. Specifically, they examined whether
individuals selected partners with personality traits similar to that of their
parents. The strongest selection effect was for openness: Women tended to
choose husbands who resembled both their mother’s and father’s level of
openness. A similar, but nonsignificant, pattern emerged for men. There was no
evidence of any selection effects for the other four traits.
Whether people select mates with similar traits and characteristics is of
considerable interest to behavioral geneticists, who typically assume no
assortative mating in calculating heritability estimates. That is, behavioral
geneticists assume that an open man would be just as likely to marry a closed
woman as an open woman. For most of the FFM personality traits, this
assumption seems to be true; assortative mating correlations tend to be lower for
personality traits than for other personal characteristics (Watson et al., 2004).
The one exception is openness.
McCrae et al. (2008) analyzed trait similarity using both self-reports and
spouse ratings of personality in married couples across three cultures. The
correlations between spouses for each of the five traits were generally modest,
with openness showing the strongest spousal similarity (mean r for openness
across the three cultures = .22). Facet-level analyses again revealed a more
nuanced story. Openness to values showed the most similarity across the
different cultures: Liberals seek out other liberals, whereas conservatives seek
out other conservatives. Part of this pairing is likely due to convenience; these
two types of people inhabit very different social worlds. Different ideologies
may also be a source of argument and conflict within the relationship that would
lead to early dissolution. Although lower in magnitude, individuals also married
partners similar to themselves in openness to aesthetics. While getting to know
each other during the early stages of dating, couples may engage in shared
interests, such as going to art museums or the symphony. The relationship may
not last long if one partner adores the arts and the other one is bored stiff. In
contrast to values and aesthetics, there was no evidence for similarity for fantasy
or feelings. Across the three cultures, individuals with an active imagination and
depth of feeling were just as likely to be paired with someone with similar or
opposite tendencies. Perhaps the internal nature of these experiences matters less
when selecting a mate than the other tendencies associated with openness.
Couples may resemble each other in openness (or any other trait) because
partners select each other based on the trait (i.e., selection) or because partners
became more (or less) similar in openness over the course of their relationship
(i.e., convergence). The trait similarity on openness among married couples
comes more from initial choice than convergence over time (Watson, Beer, &
McDade-Montez, 2014): People with the same values and intellectual pursuits
seek each other out rather than mold each other into their likenesses over time.
With many large-scale longitudinal studies now including personality
assessments, it is possible to take a more sophisticated approach to couple
similarity. For example, Rammstedt, Spinath, Richter, and Schupp (2013) not
only examined whether couples’ personality traits became more congruent over
time but also examined how trait congruence changed before couples dissolve.
For couples that remained in the relationship over the 4-year follow-up, the
correlations were similar across both measurement occasions: Partners’ scores
on openness were moderately correlated and did not increase or decrease over
time. A number of couples did separate and/or divorce across the follow-up
period. At the first assessment, couples who were to separate already showed
lower convergence than couple who were to stay together, and, at the follow-up,
their openness scores were actually negatively correlated. Openness was the only
trait that showed a significant change in similarity across the follow-up period.
Although marriage is a normative and expected event, and there is often
considerable pressure to get married, some choose to remain single. Those who
remain single tend to score higher on openness and related constructs (Johnson,
McGue, Krueger, & Bouchard, 2004; Jokela, Alvergne, Pollet, & Lummaa,
2011). It may be that open individuals have other types of relationships and
activities that are just as fulfilling as marriage, and, without a strong internal
need to conform to the expectations of society, pursue these interests instead of
potential mates. When open individuals do get married, they tend to marry at an
older age compared to more closed individuals, a pattern that holds for both men
and women (Jokela et al., 2011). Open individuals are also less likely to have
children, and if they do, they do so at an older age. This effect, however, might
be confounded by education.
The association between personality and marriage is not immune to societal
pressure, and the personality correlates of marriage and divorce, especially for
openness, may be partly driven by the social landscape. Replicating the study of
Jokela et al. (2011), Lundberg (2012) found that open individuals were much
less likely to be married by age 35 years. This association, however, held only
for more recent cohorts. In the older cohorts, openness was strongly related to
divorce, whereas openness was virtually unrelated to divorce in the younger
cohorts. With changing social norms, there is less pressure for people to marry
young. Without this societal expectation, open people may wait until they find a
partner who is truly a good match or they may forego marriage altogether.
In addition to the length of the relationship, how happy individuals are within
the relationship is another measure of the success of the partnership. Even
though people tend to seek out and couple with individuals who have similar
levels of openness, openness is the trait that is the least associated with
relationship satisfaction. That is, open and closed individuals are equally likely
to report being satisfied or dissatisfied with their romantic partner (Malouff,
Thorsteinsson, Schutte, Bhullar, & Rooke, 2010). In addition, the degree of
similarity in openness between members of the couple is likewise unrelated to
relationship satisfaction (Dyrenforth, Kashy, Donnellan, & Lucas, 2010; Furler,
Gomez, & Grob, 2013). It may, however, be associated with more specific
aspects of relationship functioning, such as sexual satisfaction. Specifically,
open wives tend to have more sexually satisfying marriages, as rated by both the
wife and the husband (Donnellan, Conger, & Bryant, 2004). Donnellan and
colleagues speculated that openness is related to sexual satisfaction because the
motivation to seek out new and varied experiences may also extend to sexual
experiences. Wives who are more open to trying new things may translate into
greater sexual satisfaction for both partners.
Finally, openness not only shapes marriage and childbearing patterns but also
how individuals parent their children. That is, the different philosophies that
characterize low and high openness—obedience and deference to authority
versus open-mindedness and tolerance—also characterize the parenting styles of
those low and high in openness. For example, closed parents expect obedience
and limit their children’s autonomy, whereas open parents are more likely to
encourage their children to voice their opinions (Metsäpelto & Pulkkinen, 2003).
Open mothers tend to have more knowledge about parenting, feel more
competent and invested in their parenting role, and tend to provide more
stimulating environments for their children (Bornstein, Hahn, & Haynes, 2011).
The consequence of these different parenting styles may be evident in their
children’s behavior: Open parents are less likely to report child misbehavior as a
major daily stressor (Lee-Baggley, Preece, & DeLongis, 2005). Rather than
having well-behaved children, however, open parents may just be more tolerant
of child misbehavior.
Geographic Openness
We typically think about personality as a characteristic of individuals, but it
can also be a characteristic of places. That is, depending on the psychological
make-up of its inhabitants, different regions may have different aggregate
psychological profiles. Large-scale cross-cultural studies suggest that mean
levels of personality traits differ systematically across nations. McCrae and
colleagues (McCrae, 2002; McCrae, Terracciano, 2005b) used both self-reported
personality (36 cultures) and observer ratings (51 cultures) to derive mean
aggregate personality scores for each culture. Convergent culture-level
correlations across the two datasets were significant for four of the five factors
and for all but four of the facets. In particular, there was a correlation of .50 for
total openness with the facets ranging from .44 for openness to actions to .75 for
openness to values. Thus, different samples that used different methods
generally converged in describing the citizens of some cultures as more open
than others. Individual differences within culture, however, were generally much
greater than the differences between cultures.
Which cultures are most open? Of the 28 cultures with both self-reports and
observer ratings, French-speaking Switzerland, Serbia, Austria, Germany, and
German-speaking Switzerland had the highest openness, with T-scores ranging
from 53 to 59; Croatia, Spain, Hong Kong, Malaysia, and India scored the
lowest, with T-scores ranging from 46 to 49. Although it is not clear why
openness scores are so much higher in Serbia than Croatia, the other findings
make some sense: Modern, progressive, well-educated countries are higher in
openness than are more traditional cultures. The United States was about average
on aggregate openness.
Although aggregate openness for the United States is about average, there is
considerable variability in openness across different regions within the United
States. Indeed, there is a shared perception that the West Coast is more open than
the Deep South (Rogers & Wood, 2010). These regional stereotypes do seem to
accurately reflect regional variability in personality across the country. Studies
over the past 40 years have shown remarkable consistency in regional variations,
especially for openness: The Northeast and Western parts of the country are
higher in openness than the Deep South (reviewed in Rentfrow, 2010).
Rentfrow and colleagues (Rentfrow et al., 2013) used five samples to
distinguish psychological regions across the United States. Cluster analyses of
these samples revealed three distinct regions, which they labeled “Friendly and
Conventional,” “Relaxed and Creative,” and “Temperamental and Uninhibited.”
The first two clusters were particularly defined by their standing on openness:
The Friendly and Conventional cluster was characterized by low openness,
whereas high openness was one of the defining traits of the Relaxed and
Creative cluster. Not surprisingly, the first cluster mapped to the north central
Great Plains and the South and the second cluster mapped on to states in the
West and along the Eastern Seaboard.
Aggregate openness scores can be treated similarly to individual-level
openness and be correlated with outcomes of interests. In general, the correlates
of aggregate openness tend to parallel those of individual-level openness.
Cultures with more open-minded individuals, for example, tend to be less
conservative and more likely to have democratic regimes, even after controlling
for gross domestic product (Terracciano & Chan, 2013). Within the United
States, aggregate openness is a strong predictor of voting patterns: Open states
tend to vote Democrat, whereas closed states tend to vote Republican. Political
and social attitudes tend to follow this pattern as well, with open states more
likely to endorse liberal values, such as legalization of marijuana and same-sex
marriage and place less value on organized religion (Rentfrow, Gosling, &
Potter, 2008). Care should be taken, however, when interpreting correlations
across different levels of analysis, as aggregate-level correlates may not
generalize to the individual level and vice versa (McCrae & Terracciano, 2008;
Rentfrow, 2010). Nonetheless, aggregate openness reveals meaningful
psychological characteristics and correlates of geographic regions.
Conclusions
This chapter was meant to provide a brief overview of the varied ways in
which openness matters. And, indeed, the broad range of correlates highlights
the reach of this trait: The consequences of openness range from the biological
to the societal. Once thought to be relevant primarily for creativity and
intelligence, openness contributes to functioning across a surprisingly wide array
of important life domains.
The research on openness has come of age. Early research on openness was
devoted to the reliability and validity of this trait and to establishing its rightful
place within the FFM. After openness was recognized as a meaningful trait,
research on openness then moved to documenting how critical this trait is to a
wide variety of domains. The next generation of openness research will need to
advance our understanding of its biological basis, better elucidate its relation to
psychopathology, and uncover the processes and mechanisms that link openness
to important life outcomes. Openness no longer needs to struggle to survive; that
it is thriving is a testament to the enduring power and significance of this trait.
References
Adelstein, J. S., Shehzad, Z., Mennes, M., DeYoung, C. G., Zuo, X. N., Kelly, C., …
Milham, M. P. (2011). Personality is reflected in the brain’s intrinsic functional
architecture. PLoS One, 6, e27633.
American Psychiatric Association (2013). Diagnostic and statistical manual of mental
disorders (5th ed.). Arlington, VA: American Psychiatric Publishing.
Armon, G., Melamed, S., Shirom, A., Berliner, S., & Shapira, I. (2013). The associations
of the Five Factor Model of personality with inflammatory biomarkers: A four-year
prospective study. Personality and Individual Differences, 54, 750–755.
Armon, G., Melamed, S., Shirom, A., Shapira, I., & Berliner, S. (2013). Personality traits
and body weight measures: Concurrent and across-time associations. European Journal
of Personality, 27, 398–402.
Asendorpf, J. B., Penke, L., & Back, M. D. (2011). From dating to mating and relating:
Predictors of initial and long-term outcomes of speed-dating in a community sample.
Ayotte, B. J., Potter, G. G., Williams, H. T., Steffens, D. C., & Bosworth, H. B. (2009).
The moderating role of personality factors in the relationship between depression and
neuropsychological functioning among older adults. International Journal of Geriatric
Psychiatry, 24, 1010–1019.
Back, M. D., Stopfer, J. M., Vazire, S., Gaddis, S., Schmukle, S. C., Egloff, B., &
Gosling, S. D. (2010). Facebook profiles reflect actual personality, not self-
idealization. Psychological Science, 21, 372–374.
Bae, H. T., Sebastiani, P., Sun, J. X., Andersen, S. L., Daw, W., Terracciano, A., … Perls,
T. T. (2013). Genome-wide association study of personality traits in the Long Life
Family Study. Frontiers in Genetics, 4, 65.
Barnes, D. E., & Yaffe, K. (2011). The projected effect of risk factor reduction on
Alzheimer’s disease prevalence. Lancet Neurology, 10, 819–828.
Barnett, J. H., Huang, J., Perlis, R. H., Young, M. M., Rosenbaum, J. F., Nierenberg, A.
A., … Smoller, J. W. (2011). Personality and bipolar disorder: Dissecting state and trait
associations between mood and personality. Psychological Medicine, 41, 1593–1604.
Barrantes-Vidal, N., Lewandowski, K. E., & Kwapil, T. R. (2010). Psychopathology,
social adjustment and personality correlates of schizotypy clusters in a large nonclinical
sample. Schizophrenic Research, 122, 219–225.
Barrick, M. R., & Mount, M. K. (1991). The Big 5 personality dimensions and job-
performance: A meta-analysis. Personnel Psychology, 44, 1–26.
Bates, T. C., & Shieles, A. (2003). Crystallized intelligence as a product of speed and
drive for experience: The relationship of inspection time and openness to g and Gc.
Intelligence, 31, 275–287.
Bipp, T. (2010). What do people want from their jobs? The Big Five, core self-
evaluations and work motivation. International Journal of Selection and Assessment,
18, 28–39.
Bjørnebekk, A., Fjell, A. M., Walhovd, K. B., Grydeland, H., Torgersen, S., & Westlye,
L. T. (2013). Neuronal correlates of the five factor model (FFM) of human personality:
Multimodal imaging in a large healthy sample. Neuroimage, 65, 194–208.
Bornstein, M. H., Hahn, C. S., & Haynes, O. M. (2011). Maternal personality, parenting
cognitions, and parenting practices. Developmental Psychology, 47, 658–675.
Brummett, B. H., Babyak, M. A., Williams, R. B., Barefoot, J. C., Costa, P. T., & Siegler,
I. C. (2006). NEO personality domains and gender predict levels and trends in body
mass index over 14 years during midlife. Journal of Research in Personality, 40, 222–
236.
Chan, W., McCrae, R. R., De Fruyt, F., Jussim, L., Lockenhoff, C. E., De Bolle, M., …
Terracciano, A. (2012). Stereotypes of age differences in personality traits: Universal
and accurate? Journal of Personality and Social Psychology, 103, 1050–1066.
Chapman, B. P., Fiscella, K., Duberstein, P., Coletta, M., & Kawachi, I. (2009). Can the
influence of childhood socioeconomic status on men’s and women’s adult body mass
be explained by adult socioeconomic status or personality? Findings from a national
sample. Health Psychology, 28, 419–427.
Chapman, B. P., van Wijngaarden, E., Seplaki, C. L., Talbot, N., Duberstein, P., &
Moynihan, J. (2011). Openness and conscientiousness predict 34-SSweek patterns of
interleukin-6 in older persons. Brain, Behavior, and Immunity, 25, 667–673.
Cheng, H., & Furnham, A. (2013). Personality traits, education, physical exercise, and
childhood neurological function as independent predictors of adult obesity. PLoS One,
8, e79586.
Clausen, J. A., & Gilens, M. (1990). Personality and labor force participation across the
life course: A longitudinal study of women’s careers. Sociological Forum, 5, 595–618.
Connelly, B. S., Ones, D. S., & Chernyshenko, O. S. (2014). Introducing the special
section on openness to experience: Review of openness taxonomies, measurement, and
nomological net. Journal of Personality Assessment, 96, 1–16.
Connelly, B. S., Ones, D. S., Davies, S. E., & Birkland, A. (2014). Opening up openness:
A theoretical sort following critical incidents methodology and a meta-analytic
investigation of the trait family measures. Journal of Personality Assessment, 96, 17–
28.
Costa, P. T., Jr., & McCrae, R. R. (1992). Revised NEO Personality Inventory (NEO-PI-
R) and the NEO Five-Factor Inventory (NEO-FFI) professional manual. Odessa, FL:
Psychological Assessment Resources.
Courneya, K. S., and Hellsten, L. A. M. (1998). Personality correlates of exercise
behavior, motives, barriers and preferences: An application of the five-factor model.
Personality and Individual Differences, 24, 625–633.
Cox, K. S., Wilt, J., Olson, B., & McAdams, D. P. (2010). Generativity, the big five, and
psychosocial adaptation in midlife adults. Journal of Personality, 78, 1185–1208.
Cryan, J. F., & Dinan, T. G. (2012). Mind-altering microorganisms: The impact of the
guy microbiota on brain and behavior. Nature Reviews Neuroscience, 13, 701–712.
Daly, M. (2013). Personality may explain the association between cannabis use and
neuropsychological impairment. Proceedings of the National Academy of Sciences of
the United States of America, 110, E979.
de Fruyt, F., Bartels, M., Van Leeuwen, K. G., De Clercq, B., Decuyper, M., &
de Fruyt, F., van De Wiele, L., & van Heeringen, C. (2000). Cloninger’s psychobiological
model of temperament and character and the five-factor model of personality.
de Moor, M. H., Costa, P. T., Terracciano, A., Krueger, R. F., de Geus, E. J., Toshiko, T.,
… Boomsma, D. I. (2012). Meta-analysis of genome-wide association studies for
personality. Molecular Psychiatry, 17, 337–349.
DeYoung, C. G., Cicchetti, D., Rogosch, F. A., Gray, J. R., Eastman, M., & Grigorenko,
E. L. (2011). Sources of cognitive exploration: Genetic variation in the prefrontal
dopamine system predicts openness/intellect. Journal of Research in Personality, 45,
364–371.
DeYoung, C. G., Grazioplene, R. G., & Peterson, J. B. (2012). From madness to genius:
The Openness/Intellect trait domain as a paradoxical simplex. Journal of Research in
(2010). Testing predictions from personality neuroscience. Brain structure and the Big
Five. Psychological Science, 21, 820–828.
DeYoung, C. G., Quilty, L. C., Peterson, J. B., & Gray, J. R. (2014). Openness to
experience, intellect, and cognitive ability. Journal of Personality Assessment, 96, 46–
52.
Donnellan, M. B., Conger, R. D., & Bryant, C. M. (2004). The Big Five and enduring
marriages. Journal of Research in Personality, 38, 481–504.
Downey, L. A., Lee, B., & Stough, C. (2011). Recruitment consultant revenue:
Relationships with IQ, personality, and emotional intelligence. International Journal of
Selection and Assessment, 19, 280–286.
Dyrenforth, P. S., Kashy, D. A., Donnellan, M. B., & Lucas, R. E. (2010). Predicting
relationship and life satisfaction from personality in nationally representative samples
from three countries: The relative importance of actor, partner, and similarity effects.
Eder, R. A. (1990). Uncovering young children’s psychological selves: Individual and
developmental differences. Child Development, 61, 849–863.
Eisenman, R., Grossman, J. C., & Goldstein, R. (1980). Undergraduate marijuana use as
related to internal sensation novelty seeking and openness to experience. Journal of
Figueredo, A. J., Sefcek, J. A., & Jones, D. N. (2006). The ideal romantic partner
personality. Personality and Individual Differences, 41, 431–441.
Fleischhauer, M., Enge, S., Brocke, B., Ullrich, J., & Strobel, A. (2010). Same or
different? Clarifying the relationship of need for cognition to personality and
intelligence. Personality and Social Psychology Bulletin, 36, 82–96.
Franchow, E. I., Suchy, Y., Thorgusen, S. R., & Williams, P. (2013). More than
education: Openness to experience contributes to cognitive reserve in older adulthood.
Journal of Aging Science, 1.
Fratiglioni, L., Paillard-Borg, S., & Winblad, B. (2004). An active and socially integrated
lifestyle in late life might protect against dementia. Lancet Neurology, 3, 343–353.
Fridberg, D. J., Vollmer, J. M., O’Donnell, B. F., & Skosnik, P (2011). Cannabis users
differ from non-users on measures of personality and schizotypy. Psychiatry Research,
186, 46–52.
Furler, K., Gomez, V., & Grob, A. (2013). Personality similarity and life satisfaction in
couples. Journal of Research in Personality, 47, 369–375.
Goldberg, L. R. (1990). An alternative “description of personality”: The big-five factor
Gore, W. L., & Widiger, T. A. (2013). The DSM-5 dimensional trait model and five-
factor models of general personality. Journal of Abnormal Psychology, 122, 816–821.
Gosling, S. D., Ko, S., Mannarelli, T., & Morris, M. E. (2002). A room with a cue:
Personality judgments based on offices and bedrooms. Journal of Personality and
Graham, L. T., & Gosling, S. D. (2012). Impressions of World of Warcraft players’
personalities based on their usernames: Interobserver consensus but no accuracy.
Haigler, E. D., & Widiger, T. A. (2001). Experimental manipulation of NEO-PI-R items.
Journal of Personality Assessment, 77, 339–358.
Herzhoff, K., & Tackett, J. L. (2012). Establishing construct validity for Openness-to-
Experience in middle childhood: Contributions from personality and temperament.
Hogan, M. J., Staff, R. T., Bunting, B. P., Deary, I. J., & Whalley, L. J. (2012). Openness
to experience and activity engagement facilitate the maintenance of verbal ability in
older adults. Psychology and Aging, 27, 849–854.
Honda, K., & Jacobson, J. S. (2005). Use of complementary and alternative medicine
among United States adults: The influences of personality, coping strategies, and social
support. Preventive Medicine, 40, 46–53.
Ironson, G. H., O’Cleirigh, C., Weiss, A., Schneiderman, N., & Costa, P. T. (2008).
Personality and HIV disease progression: Role of NEO-PI-R openness, extraversion,
and profiles of engagement. Psychosomatic Medicine, 70, 245–253.
Jerant, A., Chapman, B. P., Duberstein, P., & Franks, P. (2009). Is personality a key
predictor of missing study data? An analysis from a randomized controlled trial. Annals
of Family Medicine, 7, 148–156.
Johnson, W., McGue, M., Krueger, R. F., & Bouchard, T. J., Jr. (2004). Marriage and
personality: A genetic analysis. Journal of Personality and Social Psychology, 86,
285–294.
Jokela, M., Alvergne, A., Pollet, T. V., & Lummaa, V. (2011). Reproductive behavior and
personality traits of the Five Factor Model. European Journal of Personality, 25, 487–
500.
Jokela, M., Batty, G. D., Nyberg, S. T., Virtanen, M., Nabi, H., Singh-Manoux, A., &
Kivimäki, M. (2013). Personality and all-cause mortality: Individual-participant meta-
analysis of 3,947 deaths in 76,150 adults. American Journal of Epidemiology, 178,
667–675.
Jokela, M., Hintsanen, M., Hakulinen, C., Batty, G. D., Nabi, H., Singh-Manoux, A., &
Kivimäki, M. (2013). Association of personality with the development and persistence
of obesity: A meta-analysis based on individual-participant data. Obesity Reviews, 14,
315–323.
Jonassaint, C. R., Boyle, S. H., Kuhn, C. M., Siegler, I. C., Copeland, W. E., & Williams,
R. (2010). Personality and inflammation: The protective effect of openness to
experience. Ethnicity and Disease, 20, 11–14.
Jonassaint, C. R., Boyle, S. H., Williams, R. B., Mark, D. B., Siegler, I. C., & Barefoot, J.
C. (2007). Facets of openness predict mortality in patients with cardiac disease.
Jopp, D. S., & Hertzog, C. (2010). Assessing adult leisure activities: An extension of a
self-report activity questionnaire. Psychological Assessment, 22, 108–120.
Jung, R. E., Grazioplene, R., Caprihan, A., Chavez, R. S., & Haier, R. J. (2010). White
matter integrity, creativity, and psychopathology: Disentangling constructs with
diffusion tensor imaging. PLoS One, 5, e9818.
Kanfer, R., Wanberg, C. R., & Kantrowitz, T. M. (2001). Job search and employment: A
personality-motivational analysis and meta-analytic review. Journal of Applied
Kapogiannis, D., Sutin, A., Davatzikos, C., Costa, P., Jr., & Resnick, S. (2013). The five
factors of personality and regional cortical variability in the Baltimore longitudinal
study of aging. Human Brain Mapping, 34, 2829–2880.
Kendler, K. S., & Myers, J. (2010). The genetic and environmental relationship between
major depression and the five-factor model of personality Psychological Medicine, 40,
801–806.
Kern, M. L., Eichstaedt, J. C., Schwartz, H. A., Dziurzynski, L., Ungar, L. H., Stillwell,
D. J., … Seligman, M. E. (2013). The online social self: An open vocabulary approach
to personality. Assessment, 21, 158–169.
Kohn, M. L., & Schooler, C. (1982). Job conditions and personality: A longitudinal
assessment of their reciprocal effects. American Journal of Sociology, 87, 1257–1286.
Kornør, H., & Nordvik, H. (2007). Five-factor model personality traits in opioid
dependence. BMC Psychiatry, 7, 37.
Kosinski, M., Stillwell, D., & Graepel, T. (2013). Private traits and attributes are
predictable from digital records of human behavior. Proceedings of the National
Academy of Sciences of the United States of America, 110, 5802–5805.
Kunisato, Y., Okamoto, Y., Okada, G., Aoyama, S., Nishiyama, Y., Onoda, K., &
Yamawaki, S. (2011). Personality traits and the amplitude of spontaneous low-
frequency oscillations during resting state. Neuroscience Letters, 492, 109–113.
Kwapil, T. R., Barrantes-Vidal, N., & Silvia, P. J. (2008). The dimensional structure of
the Wisconsin Schizotypy Scales: Factor identification and construct validity.
Schizophrenia Bulletin, 34, 444–457.
Lamers, S. M. A., Westerhof, G. J., Kovács, V., & Bohlmeijer, E. T. (2012). Differential
relationships in the association of the Big Five personality traits with positive mental
health and psychopathology. Journal of Research in Personality, 46, 517–524.
Lee, A. J., Dubbs, S. L., von Hippel, W., Brooks, R. C., & Zietsch, B. P. (2014). A
multivariate approach to human mate preferences. Evolution and Human Behavior, 35,
193–203.
Lee, H. B., Offidani, E., Ziegelstein, R. C., Bienvenu, O. J., Samuels, J., Eaton, W. W., &
Nestadt, G. (2014). Five-Factor Model personality traits as predictors of incident
coronary heart disease in the community: A 10.5-year cohort study based on the
Baltimore Epidemiologic Catchment Area Follow-Up Study. Psychosomatics, 55, 352–
361.
Lee-Baggley, D. L., Preece, M., & DeLongis, A. (2005). Coping with interpersonal stress:
Role of Big Five Traits. Journal of Personality, 73, 1141–1180.
Lehmann, R., Denissen, J. J., Allemand, M., & Penke, L. (2013). Age and gender
differences in motivational manifestations of the Big Five from age 16 to 60.
Developmental Psychology, 49, 365–383.
Liu, W. Y., Weber, B., Reuter, M., Markett, S., Chu, W. C., & Montag, C. (2013). The
Big Five of personality and structural imaging revisited: A VBM—DARTEL study.
Neuroreport, 24, 375–380.
Lounsbury, J. W., Loveland, J. M., Sundstrom, E. D., Gibson, L. W., Drost, A. W., &
Hamrick, F. L. (2003). An investigation of personality traits in relation to career
success. Journal of Career Assessment, 11, 287–307.
to university life. Journal of Personality and Social Psychology, 101, 620–637.
Lundberg, S. (2012). Personality and marital surplus. IZA Journal of Labor Economics, 1,
3.
Malouff, J. M., Thorsteinsson, E. B., Schutte, N. S., Bhullar, N., & Rooke, S. E. (2010).
The Five-Factor Model of personality and relationship satisfaction of intimate partners:
A meta-analysis. Journal of Research in Personality, 44, 124–127.
McCrae, R. R. (1996). Social consequences of experiential openness. Psychological
Bulletin, 120, 323–337.
McCrae, R. R. (2002). NEO-PI-R data from 36 cultures: Further intercultural
comparisons. In R. R. McCrae & J. Allik (Eds.), The Five-Factor Model of personality
across cultures (pp. 101–125). New York, NY: Kluwer Academic/Plenum Publishers.
McCrae, R. R., & Costa, P. T., Jr. (1997). Conceptions and correlates of Openness to
Experience. In R. Hogan, J. A. Johnson, & S. R. Briggs (Eds.), Handbook of
personality psychology (pp. 825–847). Orlando, FL: Academic Press.
McCrae, R. R., Costa, P. T., Jr., Terracciano, A., Parker, W. D., Mills, C. J., De Fruyt, F.,
& Mervielde, I. (2002). Personality trait development from age 12 to age 18:
Longitudinal, cross-sectional, and cross-cultural analyses. Journal of Personality and
McCrae, R. R., & Greenberg, D. M. (2015). Openness to experience. In D. K. Simonton
(Ed.), Handbook of genius. West Sussex, England: Wiley-Blackwell.
McCrae, R. R., Martin, T. A., Hrebícková, M., Urbánek, T., Boomsma, D. I., Willemsen,
G., & Costa, P. T. (2008). Personality trait similarity between spouses in four cultures.
McCrae, R. R., & Sutin, A. R. (2009). Openness to Experience. In M. R. Leary and R. H.
Hoyle (Eds.), Handbook of Individual Differences in Social Behavior (pp. 257–273).
New York: Guilford.
McCrae, R. R., & Terracciano, A. (2005a). Personality profiles of cultures: Aggregate
personality traits. Journal of Personality and Social Psychology, 89, 407–425.
McCrae, R. R., Terracciano, A. (2005b). Universal features of personality traits from the
McCrae, R. R., & Terracciano, A. (2008). The Five-Factor Model and its correlates in
individuals and cultures. In F. R. J. v. d. Vijver, D. A. v. Hemert, & Y. H. Poortinga
(Eds.), Multilevel analysis of individuals and cultures (pp. 249–283). New York, NY:
Taylor & Francis.
McCrae, R. R., Willemsen, G., & Boomsma, D. I. (2012). Are parental personality traits a
basis for mate selection? Journal of Research in Personality, 46, 455–457.
Metsäpelto, R. L., & Pulkkinen, L. (2003). Personality traits and parenting: Neuroticism,
extraversion, and openness to experience as discriminative factors. European Journal
Middeldorp, C. M., de Moor, M. H., McGrath, L. M., Gordon, S. D., Blackwood, D. H.,
Costa, P. T., … Boomsma, D. I. (2011). The genetic association between personality
and major depression or bipolar disorder. A polygenic score analysis using genome-
wide association data. Translational Psychiatry, 1, e50.
Mõttus, R., Luciano, M., Starr, J. M., Pollard, M. C., & Deary, I. J. (2013). Personality
traits and inflammation in men and women in their early 70s: The Lothian Birth Cohort
1936 study of healthy aging. Psychosomatic Medicine, 75, 11–19.
Mõttus, R., McNeill, G., Jia, X., Craig, L. C., Starr, J. M., & Deary, I. J. (2013). The
associations between personality, diet and body mass index in older people. Health
Mõttus, R., Realo, A., Allik, J., Deary, I. J., Esko, T., & Metspalu, A. (2012). Personality
traits and eating habits in a large sample of Estonians. Health Psychology, 31, 806–814.
Nakao, T., Matsumoto, T., Shimizu, D., Morita, M., Yoshimura, S., Northoff, G., …
Yamawaki, S. (2013). Resting state low-frequency fluctuations in prefrontal cortex
reflect degrees of harm avoidance and novelty seeking: An exploratory NIRS study.
Frontiers in Systems Neuroscience, 7, 115.
Naumann, L. P., Vazire, S., Rentfrow, P. J., & Gosling, S. D. (2009). Personality
judgments based on physical appearance. Personality and Social Psychology Bulletin,
35, 1661–1671.
Five correlates of GPA and SAT scores. Journal of Personality and Social Psychology,
93, 116–130.
Onraet, E., Van Hiel, A., Roets, A., & Cornelis, I. (2011). The closed mind: “Experience”
and “cognition” aspects of openness to experience and need for closure as
psychological bases for right-wing attitudes. European Journal of Personality, 25,
184–197.
Petty, R., Briñol, P., Loersch, C., & McCaslin, M. J. (2009). The need for cognition. In
M. R. Leary & R. H. Hoyle (Eds.), Handbook of individual differences in social
behavior. New York, NY: Guilford Press.
Piedmont, R. L., Sherman, M. F., & Sherman, N. C. (2012). Maladaptively high and low
openness: The case for experiential permeability. Journal of Personality, 80, 1641–
1668.
Piedmont, R. L., Sherman, M. F., Sherman, N. C., Dy-Liacco, G. S., & Williams, J. E.
(2009). Using the five-factor model to identify a new personality disorder domain: The
case for experiential permeability. Journal of Personality and Social Psychology, 96,
1245–1258.
Quilty, L. C., Pelletier, M., DeYoung, C. G., & Michael Bagby, R. (2013). Hierarchical
personality traits and the distinction between unipolar and bipolar disorders. Journal of
Rammstedt, B., Spinath, F. M., Richter, D., & Schupp, J. (2013). Partnership longevity
and personality congruence in couples. Personality and Individual Differences, 54,
832–835.
Rentfrow, P. J. (2010). Statewide differences in personality: Toward a psychological
geography of the United States. American Psychologist, 65, 548–558.
Psychology, 105, 996–1012.
Rentfrow, P. J., Gosling, S. D., & Potter, J. (2008). A theory of the emergence,
persistence, and expression of geographic variation in psychological characteristics.
Perspectives on Psychological Science, 3, 339–369.
Rhodes, R. E., & Smith, N. E. I. (2006). Personality correlates of physical activity: A
review and meta-analysis. British Journal of Sports Medicine, 40, 958–965.
Riemann, R., Angleitner, A., & Strelau, J. (1997). Genetic and environmental influences
on personality: A study of twins reared together using the self- and peer report NEO-
FFI scales. Journal of Personality, 65, 449–475.
Ritchie, S. J., Luciano, M., Hansell, N. K., Wright, M. J., & Bates, T. C. (2013). The
relationship of reading ability to creativity: Positive, not negative associations.
Learning and Individual Differences, 26, 171–176.
Robins, R. W., Fraley, R. C., Roberts, B. W., & Trzesniewski, K. H. (2001). A
longitudinal study of personality change in young adulthood. Journal of Personality,
69, 617–640.
Robinson, O. C., Demetre, J. D., & Corney, R. (2010). Personality and retirement:
Exploring the links between the Big Five personality traits, reasons for retirement and
the experience of being retired. Personality and Individual Differences, 48, 792–797.
Rogers, K. H., & Wood, D. (2010). Accuracy of United States regional personality
stereotypes. Journal of Research in Personality, 44, 704–713.
Sadahiro, R., Suzuki, A., Enokido, M., Matsumoto, Y., Shibuya, N., Kamata, M., …
Otani, K. (2015). Relationship between leukocyte telomere length and personality traits
in healthy subjects. European Psychiatry, 30, 291–295.
Salva, J., Davey, A., Costa, P. T., & Whitfield, K. E. (2007). Replicating the NEO-PI-R
factor structure in African-American older adults. Personality and Individual
Schilpzand, M. C., Herold, D. M., & Shalley, C. E. (2011). Members’ Openness to
Experience and teams’ creative performance. Small Group Research, 42, 55–76.
Schwartz, H. A., Eichstaedt, J. C., Kern, M. L., Dziurzynski, L., Ramones, S. M., …
Unger, L. (2013). Personality, gender, and age in the language of social media: The
open-vocabulary approach. PLoS One, 8, e73791.
Seshadri, S., Beiser, A., Selhub, J., Jacques, P. F., Rosenberg, I. H., … Wolf, P. A.
(2002). Plasma homocysteine as a risk factor for dementia and Alzheimer’s Disease.
New England Journal of Medicine, 346, 476–483.
Sharp, E. S., Reynolds, C. A., Pedersen, N. L., & Gatz, M. (2010). Cognitive engagement
and cognitive aging: Is openness protective? Psychology and Aging, 25, 60–73.
Sneed, C. D., McCrae, R. R., & Funder, D. C. (1998). Lay conceptions of the five-factor
model and its indicators. Personality and Social Psychology Bulletin, 24, 115–126.
personality traits from early childhood to early adulthood. Journal of Personality, 82,
182–199.
Stephan, Y., Boiché, J., Canada, B., & Terracciano, A. (2014). Association of personality
with physical, social, and mental activities across the lifespan: Findings from US and
French samples. British Journal of Psychology, 105, 564–580.
Stephan, Y., Demulier, V., & Terracciano, A. (2012). Personality, self-rated health, and
subjective age in a life-span sample: The moderating role of chronological age.
Stephan, Y., Sutin, A. R., & Terracciano, A. (2015). Subjective age and personality
development: A 10-year study. Journal of Personality, 83, 142–154.
Sugiura, M., Kawashima, R., Nakagawa, M., Okada, K., Sato, T., Goto, R., … Fukuda, H.
(2000). Correlation between human personality and neural activity in cerebral cortex.
Neuroimage, 11, 541–546.
in the resting-state neural correlates of Openness to Experience among older adults.
Cerebral Cortex, 19, 2797–2802.
Sutin, A. R., & Costa, P. T., Jr. (2010). Reciprocal influences of personality and job
characteristics across middle adulthood. Journal of Personality, 78, 257–288.
Sutin, A. R., Costa, P. T., Evans, M. K., & Zonderman, A. B. (2013). Personality
assessment in a diverse urban sample. Psychological Assessment, 25, 1007–1012.
Sutin, A. R., Costa, P. T., Jr., Miech, R., & Eaton, W. W. (2009). Personality and career
success: Concurrent and longitudinal relations. European Journal of Personality, 23,
71–84.
Sutin, A. R., Evans, M. K., & Zonderman, A. B. (2013). Personality traits and illicit
substances: The moderating role of poverty. Drug and Alcohol Dependence, 131, 247–
251.
Sutin, A. R., Ferrucci, F., Zonderman, A. B., & Terracciano, A. (2011). Personality and
obesity across the adult lifespan. Journal of Personality and Social Psychology, 101,
579–592.
Sutin, A. R., Terracciano, A., Deiana, B., Naitza, S., Ferrucci, L., Uda, M., … Costa, P.
T., Jr. (2010). High neuroticism and low conscientiousness are associated with
interleukin-6. Psychological Medicine, 40, 1485–1493.
Sutin, A. R., Terracciano, A., Kitner-Triolo, M. H., Uda, M., Schlessinger, D., &
Zonderman, A. B. (2011). Personality traits prospectively predict verbal fluency in a
lifespan sample. Psychology and Aging, 26, 994–999.
Sutin, A. R., Zonderman, A. B., Ferrucci, L., & Terracciano, A. (2013). Personality traits
and chronic disease: Implications for adult personality development. Journals of
Gerontology Psychological Sciences, 68, 912–920.
Tackett, J. L., Silberschmidt, A. L., Krueger, R. F., & Sponheim, S. R. (2009). A
dimensional model of personality disorder: Incorporating DSM Cluster A
characteristics. Personality Disorders: Theory, Research, and Treatment, 117, 454–
459.
Tackett, J. L., Slobodskaya, H. R., Mar, R. A., Deal, J., Halverson, C. F., Baker, S. R., …
Besevegis, E. (2012). The hierarchical structure of childhood personality in five
Tekleab, A. G., & Quigley, N. R. (2014). Team deep-level diversity, relationship conflict,
and team members’ affective reactions: A cross-level investigation. Journal of Business
Research, 67, 394–402.
Terracciano, A., & Chan, W. (2013). Personality traits, national character stereotypes, and
climate-economic conditions. Behavioral Brain Sciences, 36, 501–502.
Terracciano, A., & Costa, P. T., Jr. (2004). Smoking and the Five-Factor Model of
personality. Addiction, 99, 472–481.
Terracciano, A., Löckenhoff, C. E., Crum, R. M., Bienvenu, O. J., & Costa, P. T., Jr.
(2008). Five-factor model personality profiles of drug users. BMC Psychiatry, 8.
Terracciano, A., Martin, B., Ansari, D., Tanaka, T., Ferrucci, L., Maudsley, S., … Costa,
P. T. (2010). Plasma BDNF concentration, Val66Met genetic variant and depression-
related personality traits. Genes Brain and Behavior, 9, 512–518.
modeling analyses of the NEO-PI-R scales in the Baltimore Longitudinal Study of
Aging. Psychology and Aging, 20, 493–506.
Terracciano, A., Merritt, M., Zonderman, A. B., & Evans, M. K. (2003). Personality traits
and sex differences in emotion recognition among African Americans and Caucasians.
Annals of the New York Academy of Sciences, 1000, 309–312.
Terracciano, A., Sanna, S., Uda, M., Deiana, B., Usala, G., Busonero, F., … Costa, P. T.,
Jr. (2010). Genome-wide association scan for five major dimensions of personality.
Molecular Psychiatry, 15, 647–656.
Terracciano, A., Sutin, A. R., An, Y., O’Brien, R. J., Ferrucci, L., Zonderman, A. B., &
Resnick, S. M. (2014). Personality and risk of Alzheimer’s disease: New data and
meta-analysis. Alzheimer’s and Dementia, 10, 179–186.
Terracciano, A., Sutin, A. R., McCrae, R. R., Deiana, B., Ferrucci, L., Schlessinger, D.,
… Costa, P. T., Jr. (2009). Facets of personality linked to underweight and overweight.
Tiainen, A. M., Männistö, S., Lahti, M., Blomstedt, P. A., Lahti, J., Perälä, M. M., …
Eriksson, J. G. (2013). Personality and dietary intake—findings in the Helsinki birth
cohort study. PLoS One, 8, e68284.
Tolea, M. I., Ferrucci, L., Costa, P. T., Faulkner, K., Rosano, C., Satterfield, S., …
Health, Aging, and Body Composition Study. (2012). Personality and reduced
incidence of walking limitation in late life: Findings from the Health, Aging, and Body
Composition Study. Journals of Gerontology Psychological Sciences, 67, 712–719.
Turiano, N. A., Mroczek, D. K., Moynihan, J., & Chapman, B. P. (2013). Big 5
personality traits and interleukin-6: Evidence for “healthy Neuroticism” in a US
population sample. Brain Behavior & Immunity, 28, 83–89.
Turiano, N. A., Spiro, A., & Mroczek, D. K. (2012). Openness to experience and
mortality in men: Analysis of trait and facets. Journal of Aging Health, 24, 654–672.
van Reedt Dortland, A. K., Giltay, E. J., van Veen, T., Zitman, F. G., & Penninx, B. W.
(2012). Personality traits and childhood trauma as correlates of metabolic risk factors:
The Netherlands Study of Depression and Anxiety (NESDA). Progress in
Neuropsychopharmacology and Biological Psychiatry, 36, 85–91.
Vecchione, M., Alessandri, G., Barbaranelli, C., & Caprara, G. (2012). Gender
differences in the Big Five personality development: A longitudinal investigation from
late adolescence to emerging adulthood. Personality and Individual Differences, 53,
740–746.
Vernon, P. A., Villani, V. A., Vickers, L. C., & Harris, J. A. (2008). A behavioral genetic
investigation of the Dark Triad and the Big 5. Personality and Individual Differences,
44, 445–452.
Wainwright, M. A., Wright, M. J., Luciano, M., Geffen, G. M., & Martin, N. G. (2008).
Genetic covariation among facets of openness to experience and general cognitive
ability. Twin Research Human Genetics, 11, 275–286.
Watson, D., Beer, A., & McDade-Montez, E. (2014). The role of active assortment in
spousal similarity. Journal of Personality, 82, 116–129.
Watson, D., Clark, L. A., & Chmielewski, M. (2008). Structures of personality and their
relevance to psychopathology: II. Further articulation of a comprehensive unified trait
structure. Journal of Personality, 76, 1545–1586.
Watson, D., Klohnen, E. C., Casillas, A., Simms, E. N., Haig, J., & Berry, D. S. (2004).
Webster, D. M., & Kruglanski, A. W. (1994). Individual differences in need for cognitive
closure. Journal of Personality and Social Psychology, 67, 1049–1062.
White, J., & Batty, G. D. (2012). Intelligence across childhood in relation to illegal drug
use in adulthood: 1970 British Cohort Study. Journal of Epidemiology and Community
Health, 66, 767–774.
Williams, P. G., Rau, H. K., Cribbet, M. R., & Gunn, H. E. (2009). Openness to
Experience and stress regulation. Journal of Research in Personality, 43, 777–784.
Woo, S. E., Chernyshenko, O. S., Longley, A., Zhang, Z. X., Chiu, C. Y., & Stark, S. E.
(2014). Openness to Experience: Its lower level structure, measurement, and cross-
cultural equivalence. Journal of Personality Assessment, 96, 29–45.
Wortman, J., Lucas, R. E., & Donnellan, M. B. (2012). Stability and change in the Big
Five personality domains: Evidence from a longitudinal study of Australians.
Xu, J., & Potenza, M. N. (2012). White matter integrity and five-factor personality
measures in healthy adults. Neuroimage, 59, 800–807.
Yarkoni, T. (2010). Personality in 100,000 words: A large-scale analysis of personality
and word use among bloggers. Journal of Research in Personality, 44, 363–373.
Youn, T., Lyoo, I. K., Kim, J. K., Park, H. J., Ha, K. S., Lee, D. S., … Kwon, J. S. (2002).
Relationship between personality trait and regional cerebral glucose metabolism
assessed with positron emission tomography. Biological Psychology, 60, 109–120.
Ziegler, M., Bensch, D., MaaB, U., Schult, V., Vogel, M., & Bühner, M. (2014). Big Five
facets as predictor of job training performance: The role of specific job demands.
Zuckerman, M., Kuhlman, D. M., Joireman, J., Teta, P., & Kraft, M. (1993). A
comparison of three structural models for personality: The Big Three, the Big Five, and
the Alternative Five. Journal of Personality and Social Psychology, 65, 757–768.
Agreeableness and the Five Factor Model
William G. Graziano and Renée M. Tobin
Abstract
Agreeableness is a summary label for individual differences in the motivation to
maintain positive relations with others. Agreeableness is one of the major
dimensions in the Big Five structural model of personality. It is also a major
domain in the Five Factor Model of personality. This chapter provides an
overview of the considerable body of research concerning the conceptualization,
assessment, and etiology of Agreeableness with a focus on its six facets. It
concludes with a discussion of alternative theoretical explanations for
Agreeableness. In particular, an opponent process model that involves two
competing motive systems is applied to the processes underlying Agreeableness.
Key Words: Agreeableness, facets, motives, emotion, personality, opponent
process
Agreeableness is one of the major dimensions in the Big Five structural model
of personality. It is also a major domain in the Five Factor Model of personality.
In his book, The Big Five Personality Factors, Boele De Raad (2000) described
Agreeableness as the Big Five personality dimension “with the shortest history”
(p. 91), perhaps due to the shifting construct labels used for it. Labels may have
shifted but many writers (e.g., De Raad, 2000; Matthews & Deary, 1998;
Wiggins, 1991) noted that relative to the other dimensions of the Big Five,
Agreeableness is associated distinctively with differences in social behavior and
interpersonal relations. Substantively, in their comprehensive overview,
Graziano and Eisenberg (1997) proposed that the Big Five dimension of
Agreeableness could be defined in motivational terms. Specifically, they
suggested that Agreeableness was a summary label for individual differences in
the motivation to maintain positive relations with others. Less prominence was
given to differences in thought and social cognition. Subsequent reviews
supported the utility of this perspective (Graziano & Tobin, 2009, 2013), but
added more material on social cognition.
From the Five Factor Model camp, McCrae and Costa (2003) also described
Agreeableness in terms of motives and emotions (differences in selfless concern
for others), but in addition, in terms of thoughts and attitudes (trusting and
generous sentiments). McCrae and Costa (2003) further described Agreeableness
in terms of characteristics of persons who score low or high on the domain.
Persons low in Agreeableness are critical, skeptical, try to push limits, express
hostility directly, and show condescending behavior to others. Persons high in
Agreeableness are sympathetic, considerate, warm, compassionate, generous,
and arouse liking from others. From the perspective of both the Big Five and the
Five Factor Model approaches, Agreeableness explains individual differences in
certain forms of social behavior through links to psychological processes of
cognition and affect. It follows that a comprehensive understanding of this
domain will ultimately require connections beyond differential psychology
toward the fields of psychology dealing explicitly with social cognition and the
psychology of interpersonal relations.
The two approaches (i.e., Big Five, Five Factor Model) to personality
structure and function are not isomorphic (e.g., De Raad, 2000; John & Robins,
1993; McCrae & Costa, 2003). The Five Factor Model is a specialized, theory-
based version of the empirical Big Five. At the operational level, however,
measures of Agreeableness derived from the five factor approach converge
consistently with corresponding measures of Agreeableness derived from the
Five Factor Model, except for some nuances involving the internal structure of
measures and external correlates (e.g., Costa & McCrae, 1995; Gow, Whiteman,
Pattie, & Deary, 2005; Rojas & Widiger, 2014; Samuel, Mullins-Sweatt, &
Widiger, 2013). Of particular note is the Five Factor Model facet substructure
operating beneath the superordinate supertrait domain of Agreeableness. A facet
indicates a local, subordinate organization of characteristics, and facets “reflect
specific sides or aspects of the broader domain” (McCrae & Costa, 2003, p. 47;
see also Axelrod, Widiger, Trull, & Corbitt, 1997).
The original measures of the Five Factor Model [NEO and NEO Personality
Inventory (NEO PI)] did not assess separate facets of Agreeableness, or its
theoretical cousin, Conscientiousness. The revised NEO PI-R (Costa & McCrae,
1992; Costa, McCrae, & Dye, 1991) introduced six new facets for
Agreeableness: in alphabetical order they are labeled altruism, compliance,
modesty, straightforwardness, tender-mindedness, and trust. Costa and McCrae
(1995) also noted that the NEO PI-R was different from the Big Five in being
constructed top-down (theory-based item construction) as opposed to the
bottom-up empirical approach of the Big Five, which helps to explain why the
five mega-trait dimensions are referred to as “domains,” not factors in this
model. The conventional assumption is that these facets all load only on
Agreeableness, the appropriate superordinate domain inside five factor space,
and form a positive manifold (Costa & McCrae, 1995; Samuel et al., 2013).
Piedmont and Weinstein (1993) were among the first to present evidence to
support these assumptions. One small fly in the ointment involved “warmth” as a
facet of Extraversion. It also had a significant (double) loading on the
Agreeableness factor. Costa and McCrae (1995) reported a similar secondary
loading of the NEO PI-R warmth facet of Extraversion on two different Big Five
domain-level measures of Agreeableness (see Costa & McCrae, 1995, Table 3,
p. 35). In the interest of giving the domain of Extraversion an affective and
interpersonal aspect, Costa and McCrae (1995) preferred a factor-analytic
rotation that assigned warmth to Extraversion rather than to Agreeableness (see
Costa & McCrae, 1995, p. 37). Based on other empirical research on emotions
this decision likely contributes to an underestimate of the role of Agreeableness
in emotional experience and expression (e.g., Tobin, Graziano, Vanman, &
Tassinary, 2000; Tobin & Graziano, 2011).
Another assumption, perhaps less conventional, is that the facets are co-
equals. That is, they all have comparable breadth, that no one of them is more
basic than another, and that they operate at the same or similar levels of
abstraction above the observable data, but beneath the superordinate domain of
Agreeableness (but see DeYoung, Quilty, & Peterson, 2007 for another view).
As noted previously, Costa and McCrae (1995) did not elaborate precisely on
what theoretical top-down basis they chose these substantive content areas, or
why they settled on six, as opposed to two, eight, or ten facets. They stated that
“there is nothing magical about the number six” (Costa & McCrae, 1995, p. 26).
The choices appear to have been driven largely by empirical considerations and
by the prospects for finding stable structure with six variables (Costa & McCrae,
1995, p. 27) rather than by conceptual/rational considerations.
Despite claims for a top-down approach, Costa et al. (1991) devote less time
to the explication of the substantive content of Agreeableness than do
supposedly atheoretical Big Five theorists such as Jerry Wiggins or Robert
Hogan. Wiggins (1991) framed the substantive content of Agreeableness in
terms of social exchanges, or underlying social motives such as agency and
communion. Hogan (1982) framed his conceptualization of the Agreeableness
domain in terms of the demands of social living, focusing on likeability and
social reputation. In contrast, Costa et al. (1991) devote less than half a page to
the substantive theoretical content for Agreeableness facets, and much of this
was measurement oriented. More details were offered later (Costa & McCrae,
1995). The considerations listed were (1) that more than six facet scales would
“lead to intellectual overload” (p. 27), (2) it is generally difficult to replicate
factors with fewer than five or six salient variables per factor, and (3) the
discovered factor pattern was similar in men versus women, younger versus
older participants, and white versus nonwhite respondents.
McCrae and Costa (2003) reported that within each Five Factor Model
supertrait domain, no one yet had a coherent theoretical rationale for identifying
the number or nature of these facet subdomains. Intuitively, however, it makes
sense to recognize local organization, subtleties, and nuances within the larger
supertrait domain. Beneath this intuition are a host of complex conceptual issues
(e.g., Costa & McCrae, 1995), some of which will be raised later. Are domains
and facets in the Five Factor Model both hypothetical constructs, or are facets
merely intervening variables (e.g., Grimm & Widaman, 2012, p. 622)? Do facets
have both predictive and explanatory roles, or are they merely “local
predictors”? Other subdomain structures were proposed based on factor analytic
methods by Soto and John (2012; 3: trustfulness, compassion, and humility) and
by DeYoung et al. (2007; 2: compassion and “politeness”).
Despite these differences, it could be argued that the empirical literature on
the Five Factor Model approach to Agreeableness prior to 1992 is functionally
the same as that of the Big Five approach to personality. After that date, the
facets became widely available, allowing for the possibility of refinements, or
even differences to emerge between the two approaches to Agreeableness. In
previous work, we reviewed comprehensively the literature on Agreeableness
making no sharp distinction between the two approaches (Graziano & Tobin,
2009, 2013), so we will not attempt to replow soil already tilled. Instead, here we
will concentrate on developing a coherent explanatory account of the supertrait
of Agreeableness, including its facets, within the framework of the Five Factor
Model. Special attention is devoted to studies published after 1992 that examine
Agreeableness facets empirically with a focus on “construct explication”
(Shadish, Cook, & Campbell, 2002). We consider issues of application (e.g.,
career counseling, workplace) and clinical utility only as they pertain to
construct elaboration.
In terms of advanced organizers, first we briefly discuss the emergence of the
empirical concept (Feigl, 1970; Salmon, 1984) of Agreeableness as a single,
coherent construct, and its subsequent elaboration within the Five Factor Model.
To what distinctive thoughts, feelings, and behaviors does Agreeableness refer?
What are the empirical correlates of Agreeableness that are especially
informative about its status within the Five Factor Model? Can the single domain
of Agreeableness (or its set of corresponding facets) be extracted from the five
factor system without distorting its properties? Conceptual issues arise in
extracting one construct from a system such as the Five Factor Model, and a
brief rationale is presented for doing so.
Second, we discuss critically the strategies used to specify the construct
underlying the domain of Agreeableness. Up to the introduction of facet
structure, the dominant strategy is based on “bottom-up” psychometric and
empirical principles of construct elaboration. This dominant strategy created a
large collection of empirical findings, but these have been difficult to organize
into coherent patterns.
Third, we consider major theoretical explanations for Agreeableness,
concentrating on those that point toward an underlying process or set of affective
and cognitive processes. These are empathy, the temperamental variable of
effortful control, and processes underlying social accommodation. Finally,
within the framework of Five Factor Theory (McCrae & Costa, 2003), we
propose processes that could provide the basic tendency underlying the entire
Agreeableness domain, including the facets.
Agreeableness: Emergence of an Empirical Concept

Many years before scientific psychology appeared, Aristotle commented on
the role of the moral virtue of Agreeableness in his Nicomachean Ethics. He
believed it was a characteristic that could be cultivated and used in the service of
group living and civic participation. Other writers commented on Agreeableness
in emotional life (e.g., Ricord, 1840). In modern scientific psychology research,
Agreeableness has an unusual history relative to other recognized dimensions of
personality. Unlike the two supertraits of Neuroticism and Extraversion,
Agreeableness did not initially receive attention because of deductive top-down
theorizing about its link to biology or especially to conspicuous processes such
as anxiety. It was not tied to distinctive psychological processes of brain activity
in the way that Neuroticism or Extraversion was. Instead, systematic
Agreeableness research was stimulated by observable regularities arising in
descriptions of others, and later in self-descriptions (Digman & Takemoto-
Chock, 1981; John & Robins, 1993).
In recent years, additional raw material has appeared to feed the understanding
of Agreeableness, both empirically and conceptually. A PsycINFO search of the
keyword “Agreeableness” identified 2,872 peer-reviewed journal articles written
in English from 1860 to February 7, 2015. More than 97% of these articles were
written after 1992. Within that set, more than 220 of these articles were
published in the year 2014 alone, suggesting a growing interest in the construct.
In a related search, we found 135 articles with “Agreeableness” in the title
between 1900 and 2014. For purposes of comparison, the corresponding
numbers for titles during this same time frame was 1,600 for Neuroticism, 1,398
for Extraversion, 363 for Conscientiousness, and 175 for Openness.
As a domain, Agreeableness has an impeccable empirical genealogy. It could
be argued that Agreeableness is a poster-child illustration of what logical
empiricists call an “empirical concept” (Feigl, 1970; Grimm & Widaman, 2012,
p. 622; Salmon, 1984). Agreeableness had its origins in modern scientific
psychology in an effort to build a bottom-up taxonomy of English language
personality-related words (Graziano & Eisenberg, 1997; John & Robins, 1993;
McCrae & John, 1992; see also the chapter by Costa and McCrae). In the United
States, Allport and Odbert (1936) extracted 17,953 personality-relevant terms
from the 1925 edition of Webster’s New International Dictionary. Words were
sorted into four lists, or “columns.” The first sorting divided “censorial” or
evaluative words (Column 3) from “neutral” words (Column 1). Words in
Column 1 were the “real” (quotes from Allport & Odbert, 1936) traits of
personality. Column 3 was the longest of the lists, and contained words such as
“amiable,” “agreeable,” and “appealing.” Allport and Odbert (1936) suggested
that words in Column 3 “should be avoided by psychologists unless they are
prepared to deal with the subject of social judgment” (p. vii).
The Allport–Odbert list (and taxonomy) was a major source of terms for many
(but not all) linguistic analyses of personality. When Cattell (1957) used the
technique of factor analysis to identify additional order beneath the relatively
crude categorization, he limited himself to Column 1 (see also the chapter by
Wright). Given this state of affairs, the prior probability seemed low that factor
analysis would uncover a major dimension devoted primarily to “social
evaluation.” Despite this strategic elimination, when Digman and Takemoto-
Chock (1981) reanalyzed data from six large studies, they labeled the first factor
to emerge as “friendly compliance vs. hostile non-compliance.”
At the least, bottom-up statistical patterns—now in many natural languages
(e.g., Kohnstamm, Halverson, Mervielde, & Havill, 1998)—provide good reason
to think diverse cultures develop words that describe differences in the
motivation for having cooperative, harmonious relationships with other people.
As Aristotle anticipated, Agreeableness is a property associated with efficient
group living across most cultural groups studied so far. What is less clear is how
to label the underlying construct. Construct labeling is not a trivial matter
because labels provide frames for research on construct elaboration. When a
person is described as “altruistic,” is the motivation underlying it inherently
prosocial (e.g., Graziano & Habashi, 2015) or is it compliant acquiescence to
social norms, conformity to social demands, or the actions of the less powerful
to find a place in the sun, so to speak? Digman and Takemoto-Chock (1981)
used the label “compliance” for the supertrait domain, but as we discuss later,
this word carries surplus meaning.
Until quite recently in modern psychology, Agreeableness was usually
described in behavioral terms, not in terms of underlying affective or cognitive
processes. That is, Agreeableness was described as differences in behavior
characterizing people who score high versus people who score low on
Agreeableness measures. From extreme group (high versus low) behavioral
differences, researchers can and do make inferences about underlying processes,
but rarely test them directly. For example, Agreeableness is presumed to be
related to the cognitive and affective processes of prejudice because persons who
score high on Agreeableness are more positive about most social groups, and are
less negative about outgroup members and traditional targets of prejudice, than
are persons who score low on Agreeableness (Graziano & Bruce, 2008;
Graziano, Bruce, Sheese, & Tobin, 2007; Graziano & Habashi, 2010). A similar
logic was used to connect Agreeableness to empathy and its components
(Graziano, Habashi, Sheese, & Tobin, 2007). Agreeableness has also been linked
to other overt behaviors, such as constructive conflict resolution (Graziano,
Jensen-Campbell, & Hair, 1996; Jensen-Campbell, Gleason, Adams, &
Malcolm, 2003; Jensen-Campbell & Graziano, 2001) and intragroup cooperation
(Graziano, Hair, & Finch, 1997; see Graziano & Tobin, 2009, for a review).
Perhaps because of its empirical origins, some findings for Agreeableness
emerged quasiaccidentally. Perhaps a better term is “incidentally.” That is, in
studies focused on specific behaviors (e.g., bullying) or on life outcomes (e.g.,
longitudinal stability of behavior patterns or life outcomes), some findings for
Agreeableness and its facets emerged. The focus in these studies was on the
behavioral phenomenon, per se, and not on Agreeableness and its facets. In some
cases, the patterns made sense, whereas in other cases they are best described as
intriguing curiosities begging for more focused follow-up. For example, why are
first-born children lower in Agreeableness than their siblings (Michalski &
Shackelford, 2002)? Why do college women high in Agreeableness weigh less
than their less agreeable peers, yet the former show less antifat prejudice than the
latter (Graziano, Bruce, et al., 2007)? Why do children lower in Agreeableness
like to play chess more than their higher-agreeable peers (Bilalic, McLeod, &
Gobet, 2007)? In one large longitudinal study (Costa, Herbst, McCrae, &
Siegler, 2000) two Agreeableness facets (trust and modesty) showed small 6-
year declines, whereas three other facets of the same domain of Agreeableness
(straightforwardness, altruism, and compliance) increased slightly over the same
interval. All changes were small, but why did this pattern occur? Are these
merely statistical noise or genuine effects? Each empirical snippet can be partly
explained retrospectively, but in the absence of theory, integrative coherence is
elusive.
Because of its behavioral and empirical origins, controversies appeared about
its sources within human psychology, its correlates, and even a suitable label for
this hypothetical construct. Alternative labels used to describe the dimension are
“friendly compliance versus hostile noncompliance,” “tender mindedness,”
“likeability,” “communion,” “benevolence,” and even “love versus hate.” What
we are describing here are alternative labels that are applied to the entire domain
of Agreeableness, not just one facet. The implication is that the label
characterizes the bulk of the variation associated with the domain. That is, if the
domain is labeled “friendly compliance,” then we might expect the entire
domain to be associated with a range of social influence phenomena such as
conformity, compliance, internalization, impression management, deference to
norms and conventions, and the like. If the domain is labeled “benevolence” or
“tender-mindedness,” however, then the expected pattern would be entirely
different. The domain should be associated with prosocial behavior, some forms
of which might even involve resisting and defying conformity pressure to help
victims. We consider this issue later in our discussion of “surface similarity” in
construct elaboration.
The verbal label of “friendly compliance” in particular has led to
misunderstandings about Agreeableness as a domain. For example, the term
“compliance” has a process-based meaning in social psychology that often
places it on a continuum of social influence with internalization and social
identification (e.g., Petty & Wegener, 1999). That variety of compliance is
considerably different in scope and meaning from the one used more casually in
personality to imply tendencies to follow rules and norms (Parks, 2011,
“Conformity,” pp. 530–531). “Friendly compliance” might imply a general
conforming personality, but the empirical evidence is thin, at best, for such a
connection. In fact, there is evidence to the contrary (Graziano & Tobin, 2002;
Xia & Habashi, 2015). For example, Xia and Habashi (2015) found that persons
high in Agreeableness were less persuaded by weak arguments, but more
persuaded by strong arguments, than were persons low in Agreeableness.
Overall, persons low in Agreeableness were less responsive to the nature of the
communications, suggesting that Agreeableness (at least at the domain level) is
more a matter of attention and responsiveness to other people than to a
generalized conforming disposition.
Repeated attempts to find a theory-based empirical link between
Agreeableness and compliance-related behavior has been met with stubborn
resistance in the form of null outcomes (see Kassner, 2011; but see Carlo, Okun,
Knight, & de Guzman, 2005, p. 1302; Erdheim, Wang, & Zichar, 2006; Habashi
& Wegener, 2008). Of course, there is a bias against publishing research
reporting null outcomes, but some work ran that null-bias gauntlet successfully
(e.g., Graziano, Jensen-Campbell, Shebilske, & Lundgren, 1993; Graziano &
Tobin, 2002). A possible outlier/exception is one experimental study from
France exploring obedience using the Milgram paradigm (Begue et al., 2015). It
did not focus on Agreeableness as a domain, or on Agreeableness facets, per se.
These authors found three significant correlations with obedience, operationally
defined as the level of shock participants delivered to the “learner.” They were
political orientation (r = .32), domain-level Conscientiousness (r = .34), and,
bringing up the rear, domain-level Agreeableness (r = .26) (for a different
perspective on higher-order personality factors and conformity, see DeYoung,
Peterson, & Higgins, 2002).
A similar construct-labeling problem arises with Agreeableness at the facet
level. For example, consider the facet label “altruism.” In the social psychology
literature on prosocial behavior, the term “altruism” has a restricted, process-
based meaning tied to the presumed preponderant intent of the actor (see Batson,
1991; Batson, Duncan, Ackerman, Buckley & Birch, 1981; Batson, O’Quin,
Fultz, Vanderplas, & Isen, 1983; Schroeder & Graziano, 2015, for a
comprehensive overview). Altruism is defined as a subset of the more
comprehensive construct of prosocial behaviors. Prosocial behaviors seem to be
much more common than altruistic behaviors because the former are presumed
to be motivated by self-interest, whereas altruistic behaviors are rare (perhaps
even nonexistent), presumably because they require ignoring self-interest.
Instead, they are motivated primarily by concern for the recipient. To label a
behavior or chronic disposition as altruistic implies that the labeler can isolate
the motives underlying the action in question. The NEO PI-R altruism facet
items, however, ask questions such as “Make people feel welcome,” “Love to
help others,” “Have a good word for everyone,” and “Am concerned about
others.” Persons who endorse these items may indeed be chronically prosocial,
but those items do not permit a clean diagnostic assessment of underlying
egoistic or altruistic motives. Furthermore, researchers have constructed
taxonomies for forms of prosocial behavior (e.g., emergency helping versus
sustained volunteerism). These different forms of prosocial behavior probably
have different predictors for eliciting and maintaining those behaviors (e.g.,
Dovidio, Piliavin, Schroeder & Penner, 2006; Pearce & Amato, 1980).
The implication here is that Agreeableness and its facets probably do not
apply uniformly across the full range of the prosocial taxonomy, and may be
differentially responsive to different eliciting conditions. Emergency helping is
associated with impulsive, often thoughtless actions (Extraversion?), whereas
volunteering over months or years is associated with other-oriented aspects of
Agreeableness (e.g., Penner, Fritzsche, Craiger, & Freifeld, 1995). Note it is not
our intent to promote one subdiscipline’s definitions over the others. There are,
of course, many ways of conceptualizing altruism and prosocial behaviors other
than the narrow ones offered by social psychology (e.g., John, 1984; Sober &
Wilson, 1998). Nevertheless, the narrow definitions lend themselves to
falsifiable hypotheses and focused process-based explanations. Because
Agreeableness is associated with distinctive forms of social and interpersonal
behaviors, and because social psychology has a major stake in the area, its
theoretical distinctions are especially worthy of attention (for a particularly
insightful personality-oriented article on this altruism/prosocial issue, see John,
1984).
Strategies for Construct Specification and Elaboration

Two strategies have been used to elaborate and explicate psychological
constructs (Grimm & Widaman, 2012). Both have been used for the Big Five
construct of Agreeableness. One is to expand the nomological network,
demonstrating empirical links to variables external to Agreeableness. In this
strategy, researchers explore Agreeableness as a predictor or moderator of
variables having some intuitive connection to Agreeableness such as
interpersonal conflict, intergroup and intragroup cooperation, helping, prejudice,
and various psychopathologies. In some cases, the intuitive connections were
corroborated (e.g., helping), but in other cases they were not (e.g.,
responsiveness to social desirability pressures). A second strategy is to focus on
the internal structure of the Agreeableness measures to identify subordinate,
localized pockets of coherence. This second strategy gave rise to the
development of facets, described as specific sides or aspects of the five broader
domains. Presumably, as more refined, internally coherent measures are
developed, more precise predictions can be made and theoretical explanations
become more systematic (e.g., Kausel & Slaughter, 2011). McCrae and Costa
(2003) noted that researchers from many different personality research traditions
agree about the five broad domains of the Big Five; they added that “the finer
distinctions within domains, however, are more arbitrary” (p. 47).
Just how arbitrary are these facet distinctions? The NEO PI-R has no
monopoly on the identification of facets. In an effort to make available public-
access measures of individual differences in personality, Goldberg (1999)
reported work on the International Personality Item Pool (IPIP). One aspect of
this work was the construction of facet items for Agreeableness (see also
Goldberg et al., 2006). How well do the NEO PI-R Agreeableness facets match
the IPIP facets? Let us look specifically at Agreeableness from the perspective of
prosocial behavior, which seems to be a pervasive aspect of domain-level
Agreeableness. In the NEO PI-R, at least three of the six Agreeableness facets
deal with prosocial behavior in one form or another. We already discussed
problems with the facet labeled “altruism.” There is a separate Agreeableness
facet for “straightforwardness,” which seems to map onto “sympathy” in the
IPIP items (e.g., Goldberg, 1999), and “compliance,” which maps onto
“morality” in the IPIP items. Perhaps trust and modesty facets also are predictors
of aspects of prosocial behavior at the group level (e.g., Fisher, Bell, Dierdorff,
& Belohlav, 2012; John, 1984; Kausel & Slaughter, 2011; Paulhus & John,
1998). In the IPIP, trust is measured with items such as “Believe that others have
good intentions,” “Trust what people say,” “Believe that people are basically
moral,” and “Believe in human goodness.” Modesty is indexed with items such
as “Boast about my virtues.”
Both NEO PI-R and IPIP facets may indeed be measuring different faces of
the broader domain of Agreeableness, but do the IPIP facets refer to the same
variables or processes as the NEO PI-R facets? More importantly, even if they
do converge, how many different facets of the larger Agreeableness domain are
we seeing? The conceptual problems here go far beyond simple issues of facet
labeling. It is plausible that there are Agreeableness-related individual
differences in, say, prosocial predispositions, but the links among facets and
psychological processes are not well articulated, perhaps because the processes
underlying such differences are still very much in question (see Graziano &
Habashi, 2015, “The quest for the prosocial personality”)
To explore convergent and discriminant validity empirically, Costa and
McCrae (1995) reported Agreeableness facet correlations with other measures of
the Big Five and Five Factor Model, namely Goldberg’s (1990) Transparent
Trait Rating Form (TTRF) and Wiggins’ Revised Interpersonal Adjective Scales
—Big Five version (IASR-B5; Trapnell & Wiggins, 1990). The Agreeableness
facet that correlated the highest with both of these scales was altruism, with rs =
.75 and .68 for the Wiggins IASR-B5 scale and the Goldberg TTRF,
respectively. Correlations this large are what we expect to see for measures of
the same construct (Grimm & Widaman, 2012, p. 632). Costa and McCrae
(1995) concluded that the NEO PI-R, TTRF, and IASR-B5 covered the same
range of traits. The Hogan Personality Inventory (HPI; Hogan, 1986) offers
another version of the Five Factor Model. In the HPI the domain of
Agreeableness is represented at the domain level as Likeability. The NEO PI-R
converges less well with the HPI than with the Wiggins or Goldberg measures.
Nevertheless, the NEO PI-R facet that correlates highest with the Hogan
Likeability domain is altruism (r = .37, followed closely by trust at .35).
More recent research by Widiger and colleagues (e.g., Rojas & Widiger,
2014; Samuel et al., 2013) examined convergent and discriminant validity for
several combinations of personality measures, including a modified one-page
version of the Five Factor Model Rating Form, the Five Factor Form. Rojas and
Widiger (2014) reported strong convergence correlations for the Five Factor
Form Agreeableness facets and IPIP, especially for trust (.70), altruism (.52),
and tender-mindedness (.52). Samuel et al. (2013) used exploratory structural
modeling to examine links among the Five Factor Model Rating Form facets of
Agreeableness. All six facets of Agreeableness loaded on the Agreeableness
factor. The highest loading was for tender-mindedness (.59), followed by trust
(.49) and modesty (.44). Last in rank was straightforwardness (.26).
Empirical Basis for the Agreeableness Facets: The Extraction
Problem
A distinguished personality psychologist who will remain unnamed once
referred to the “bad old days of personality,” referring to previous years in which
personality was investigated one trait variable at a time. This approach
undermined efforts to find a coherent structure underlying personality. The Big
Five and Five Factor Model promised to show a path out of the wilderness, but
these efforts to create coherence and integration can come at a price.
A special set of problems arises in explaining Agreeableness within the Five
Factor Model. First, the major domains of personality, not to mention the
subordinate facet structure, are parts of an integrated Five Factor Model system.
Extracting one domain or facet from the full set for examination works directly
against one of the main strengths of the Five Factor Model, namely integration.
Extraction could be potentially misleading if the domains or facets form
distinctive combinations or configurations “horizontally” (e.g., De Raad &
Hofstee, 1993; Hofstee, Ostendorf, & Boomsma, 1998, as cited by De Raad,
2000). An individual scoring high on Agreeableness could be one of several
different nuanced “varieties” (e.g., Costa, Herbst, McCrae, Samuels, & Ozer,
2002; DeRaad, 2000; Paulhus & Jones, 2015). By collapsing across the other
variables, we run the risk of failing to “generalize across” cases (Shadish et al.,
2002) if they also score high on other domains as well.
De Raad (2000) provides an illustration using trait word descriptors (see also
the chapter by de Raad and Mlačić). For example, a person high in
Agreeableness could also score high on (1) Extraversion (“sympathetic,” “kind,”
“warm”) or (2) Conscientiousness (“helpful,” “cooperative,” “sympathetic”) or
(3) emotional stability (“trustful,” “pleasant,” “tolerant”) or (4) Openness
(“genial,” “tactful”). These individuals do share some common properties, but at
least intuitively, the individual persons seem noticeably different from each
other. The nuanced differences might be more pronounced at the low end of the
Agreeableness continuum (e.g., Dark Triad configurations; Jakobwitz & Egan,
2006; Paulhus, 2015; Stead & Fekken, 2014). If this is true for domain-to-
domain configurations, then its influence could be even greater at the facet-to-
facet level.
Table 6.1. Results of PsycINFO Searches for Agreeableness Facet Keywords

Conducted on 2/7/2015
Construct No Peer-Reviewed, Combined with Only Multiple Total
Limits English Language Agreeableness this Facets
Journal Facet
Agreeableness 3,922 2,866 – – – –
Trust 23,935 15,561 69 8 10 18
Straightforwardness 76 51 23 2 17 19
Altruism 5,499 3,330 44 1 13 14
Compliance 27,518 20,177 31 2 13 15
Modesty 736 443 16 1 10 11
Tender-Mindedness 69 53 15 3 8 11
Such nuances notwithstanding, empirical research on personality would be

severely handicapped if each study was required to investigate all five domains
and all of their respective facets in every data collection. After all, each domain
does refer to a distinctive, general form of thoughts, feelings, and behavior. For
the foreseeable future, if the goal is construct elaboration (e.g., Shadish et al.,
2002, pp. 66–68), then it makes sense to focus on domain-specific or facet-
specific processes, allowing room for a few brave pioneers to study
comprehensive, nuanced differences among configurations and varieties.
Here we discuss the Agreeableness facets extracted from their superordinate
structure. Earlier, we described the key assumption that Agreeableness facets
represent distinctive aspects or faces of persons underneath the larger umbrella
of the Agreeableness domain. Since the introduction of facets, some research
was conducted to examine systematically the relations among them and overt
behaviors. As shown in Table 6.1, there is variability in how many of the facets
have been linked empirically to outcomes. Table 6.1 presents the results of
PsycINFO searches focused specifically on the Agreeableness facets. In the first
column, we present the total number of articles identified in keyword searches
for each of the constructs when no limits were set on the search. This
information provides a baseline for level of interest in the variables as “stand
alone” topics. The second column presents the number of studies for each
construct when the searches were limited to peer-reviewed journal articles
published in the English language. The third column presents the results of a
search in which the facet keyword was combined with Agreeableness. This
restriction produced a dramatic drop in numbers. As discussed previously, a
likely explanation for these reductions may be largely accounted for by the
different verbal labels for the Agreeableness domain and facet labels in the
psychological literature.
We then focused on each of the articles that included the combination of the
facet label and Agreeableness. Results of this investigation appear in the
remaining columns. The fourth column provides the number of articles that
reported results for only that facet, but not for any of the other five
simultaneously. The fifth column reports the number of articles that reported
results for more than one facet, including the target facet. Thus, the Multiple
Facets column reflects redundant tallying of the studies. For example, the 10
articles reported for trust overlap with four articles with significant findings for
the compliance facet, including one that also reported a significant effect for
straightforwardness, one that also reported significant effects for altruism and
tender-mindedness, and one that reported significant effects for all but one of the
facets (i.e., significant effects for trust, compliance, altruism,
straightforwardness, and tender-mindedness, but not modesty). All of the studies
reporting significant results for more than one facet were counted in the
“Multiple Facets” total for each of the facets. Finally, the last column provides
the total number of articles reporting findings for that facet, either reported alone
or with other significant facet results. Overall, these searches produced only
modest evidence of linkages between the facets and reportable outcomes.
In examining the results, it appears that the bulk of the articles did not make a
priori, facet-specific predictions or try to test them. The most common case
involved studies that report administering the NEO PI-R and examining
inductively how each of the facets is related to the outcomes of interest, a
situation that increases the chances of Type I errors. Furthermore, the majority of
studies providing significant results for the Agreeableness facets also report
significant relations with multiple facets, including facets outside of the
Agreeableness domain. Examination of the articles also revealed that topics in
industrial/organizational and clinical psychology most commonly generated the
studies using correlational methods only. The studies frequently focused on links
to characteristics of employees and psychopathology. Here we highlight the
findings for each facet in alphabetical order.
Altruism
As noted previously, the psychological literature on altruism and prosocial
behavior is substantial (e.g., Schroeder & Graziano, 2015). In our literature
search, however, we found only one empirical study that explicitly made a priori
predictions about the altruism facet of the Five Factor Model. Van Iddekinge,
McFarland, and Raymark (2007) found that the altruism facet predicted
defensive impression management behaviors during a structured interview task.
The sample consisted of 132 university students who were randomly assigned to
one of two interview conditions (impression motivation versus control). Most
striking was that all other studies of altruism reported findings for other facets
simultaneously (discussed subsequently) or they examined links between
domain-level Agreeableness and altruism, not the facet of altruism and its links
to other variables.
Compliance
There is a substantial Five Factor Model-oriented, empirical literature on
compliance. This research is generally not focused, however, on the facet of the
same name. Only a few studies report findings exclusively for the Agreeableness
compliance facet. When they do, it is not because the researchers designed the
study to examine compliance specifically. For example, in a large longitudinal
examination of the data from participants in the East Baltimore Epidemiologic
Catchment Area study, Lockenhoff, Terracciano, Patriciu, Eaton, and Costa
(2009) found that the compliance facet was decreased by the experience of
traumatic events. It is important to note, however, that Lockenhoff and
colleagues examined all 30 of the NEO PI-R facets in the study, not just the
compliance facet. The focus here was on the omnibus influence of traumatic
experience, not on construct elaboration for Agreeableness. A second
longitudinal study examined compliance in inner city boys, but not as measured
by the NEO PI-R. Using factor analysis of California Q-set data (Caspi et al.,
1992), Kern et al. (2013) found that compliance was related to important
academic, employment, relationship, and legal outcomes. Another study by
MacLaren, Ellery, and Knoll (2015) examined the compliance “aspect”
(DeYoung et al., 2007), finding that it predicted problem gambling in a
community sample of electronic gambling machine users. Again, the De Young
et al. (2007) compliance aspect of personality is not the same as the compliance
facet from the NEO PI-R; it is made up of the trust, altruism, and tender-
mindedness facets. For De Young et al. (2007) the politeness aspect of
personality is made up of the compliance facet along with straightforwardness
and modesty.
Modesty
Much like the research on the Agreeableness altruism and compliance facets,
the literature solely examining the modesty facet is scant. Furnham, Moutafi, and
Chamorro-Premuzic (2005) found that Agreeableness, and the modesty facet in
particular, was related to self-estimated intelligence. That is, in a two-study set
of undergraduate student participants (N = 230), they found that modesty was
related to underestimating intelligence relative to intelligence testing scores on
the Wonderlic Personnel Test (Wonderlic, 1992). This finding is interesting, but
it is not contrary to uninformed intuition, nor did it have strong a priori
theoretical backing.
Straightforwardness
The straightforwardness facet lacks a strong empirical base. Before publishing
the study linking compliance to problem gambling, MacLaren, Best, Dixon, and
Harrigan (2011) reported significant inverse relations between the
straightforwardness facet and a questionnaire measure of problem gambling in
undergraduate students. It is possible that the difference in samples (i.e., frequent
electronic gamblers versus university students) accounts for the emergence of
one significant facet in one study and the emergence of another in the other
study. Additional explanations seem more likely. In both cases, all of the facets
were entered into the analyses and the significant facet from each was reported
without discernible connection to theory. This approach to the examination of
facets increases the chances for spurious associations. For example, Piedmont
and Weinstein (1994) included all of the available facets in analyses of the
supervisor ratings of employee performance. They provided little rationale for
examining the specific facet, but they found that low straightforwardness
predicted positive supervisor ratings.
Tender-Mindedness
Few studies of the tender-mindedness facet were uncovered in our searches.
Koelsch, Enge, and Jentschke (2012) predicted a significant relation between
tender-mindedness and cardiac signatures (i.e., electrocardiogram amplitudes),
but they found no evidence of a significant relation. Relative to most of the
research on the other facets, however, those examining tender-mindedness seem
more purposeful and focused. As noted previously, Graziano, Bruce, et al.
(2007) found a relation between domain-level Agreeableness and prejudices.
They had not examined Agreeableness facets. In two different studies, the team
of Akrami and Ekehammar (Akrami & Ekehammar, 2006; Ekehammar &
Akrami, 2007) explored the relations between facet-level personality and
prejudice. Akrami and Ekehammar (2006) found that among the significant
facets of Agreeableness that predicted right-wing authoritarianism, tender-
mindedness was the strongest. Going outside the Five Factor Model, they also
found the strongest (inverse) relation between social dominance orientation and
right-wing authoritarianism. They had not made any specific predictions about
these relations. A year later, this team published another article in Journal of
Personality showing that the domain of Agreeableness as well as the tender-
mindedness facet was a significant predictor of sexism (inversely; Ekehammar &
Akrami, 2007).
Trust
Based on sheer frequency of publications, trust seems to be a popular stand-
alone topic. Among the Agreeableness facets, trust has been examined to a
greater degree than any of the other facets. Several of these studies examine the
trust facet in the workplace. Kausel and Slaughter (2011) examined the fit
between job applicant personality and recruitment methods in predicting
organizational attractiveness, hypothesizing that scores on the trust facet would
moderate the relation between organizational trustworthiness and workplace
attractiveness. They found that trustworthiness made a significant difference, but
only for participants low in trust. Timmerman (2004) found that trust was the
only significant predictor of supervisor ratings of job performance for 203 call
center employees. It is interesting to note that 10 years prior to this report,
Piedmont and Weinstein (1994) reported a significant relation between an
Agreeableness facet and supervisor ratings of job performance, but it was a
significant link to low straightforwardness, not trust, as was found in this study.
Significant trust facet results have also been reported for dysfunctional
behavior and psychopathology. Christopher, Zabel, and Miller (2013) found that
trust was inversely related to hostile and benevolent sexism. Again, we see
researchers presenting significant relations between an Agreeableness facet, but
it is not the same as the one reported by others who found tender-mindedness to
be the best predictor of sexism (Ekehammar & Akrami, 2007). In their first
study, Ekehammar and Akrami report significant correlations between
generalized prejudice and all of the facets except compliance. In more targeted
regression analyses, they found that only tender-mindedness and a facet of
Extraversion added significantly to the prediction of generalized prejudice
(Study 1) and sexism (Studies 2 and 3). Most importantly, the researchers made
a priori hypotheses that tender-mindedness would be linked to prejudice.
Beyond sexism, trust has also been examined in a clinical context, revealing
inverse relations to other problem behaviors. For example, Hopwood et al.
(2007) found that the trust facet meaningfully distinguished between problem
alcohol drinkers and other patients in the Collaborative Longitudinal Personality
Disorders Study (Gunderson et al., 2000). As another example, Bienvenu et al.
(2004) found that only the trust facet was related to social phobia as assessed
using a semistructured interview instrument with a sample of 398 participants
from the Hopkins Epidemiology of Personality Disorders Study (Samuels et al.,
2002). Unlike many of the other significant findings for facets, this relation is in
the absence of a significant relation at the domain level. That is, Bienvenu and
colleagues (2004) found that individuals with social phobia tended to be average
in Agreeableness and found only a significant relation with the trust facet. On
that basis, Glinski and Page (2010) examined the trust facet from a somewhat
different clinical angle building on the research specifically linking trust to
social phobia. They posited that individuals who are high in social anxiety have
an increased fear of evaluation from others and are more likely to be low in
Agreeableness, primarily because of the decrease on the trust facet. Furthermore,
they ventured that improving anxious symptoms would be reflected in
improvements in self-ratings of personality. To test this hypothesis, they
examined the influence of group therapy for social anxiety on changes
(preintervention to postintervention) in facet and factor scores for 25 adults
completing an average of nine group therapy sessions. They found that
postintervention trust scores were higher than they were before the intervention,
suggesting malleability of this facet of Agreeableness. Although these results are
positive from a clinical perspective, it remains to be seen whether the results
reported from this study are replicated with a larger sample in the future.
Multiple Facets
In terms of multiple facet results, a growing literature, including meta-analytic
reviews, links the Agreeableness facets inversely to psychopathy and
antisocial/aggressive behavior (e.g., Axelrod et al., 1997; DeCuyper, De Pauw,
De Fruyt, De Bolle, & De Clercq, 2009; Jones, Miller, & Lynam, 2011; see also
the chapter by Widiger, Gore, Crego, Oltmanns, and Rojas) as early as
adolescence (Salekin, Debus, & Barker, 2010; see also the chapter by De Fruyt,
De Clercq, and De Bolle). Across these articles, each of the six facets of
Agreeableness has been linked to psychopathy and/or antisocial/aggressive
behavior in at least one study. Among the studies presenting more than one
significant relation with an Agreeableness facet, the most frequently reported
was the straightforwardness facet (DeCuyper et al., 2009; Jones et al., 2011).
Importantly, these articles also found significant relations for the
Agreeableness domain as well as its facets when analyses at the domain level
were reported. For example, Stead and Fekken (2014) found that Agreeableness
and its facets were related to the Dark Triad. Specifically, they found that
Agreeableness was inversely related to all three aspects of the Dark Triad. Factor
analyses revealed that Agreeableness and its facets were central to
Machiavellianism, but only modesty and straightforwardness loaded on
Narcissism and Psychopathy. That Agreeableness was also a significant
predictor in these studies draws into question the amount of benefit gained from
facet analysis, particularly in the absence of a priori hypotheses. Furthermore,
the spread of significant relations across all six facets of Agreeableness suggests
that the variance in psychopathy is distributed across multiple aspects of
Agreeableness.
These associations are also reflected in the significant inverse relations
between the Agreeableness domain and its facets and destructive behaviors, such
as acts of violence, delinquency, and risky sexual behavior. For example, Voller
and Long (2010) found that rape perpetrators had lower levels of Agreeableness,
tender-mindedness, and altruism relative to sexual assault perpetrators and
nonperpetrators. Similarly, Madsen, Parsons, and Grubin (2006) found that
convicted child molesters with personality disorders were lower in
Agreeableness, trust, straightforwardness, and compliance in a modest sample of
40 men. Heaven (1996) found significant inverse relations between
Agreeableness and interpersonal violence and vandalism/theft among 216
adolescents attending an Australian Catholic school. In their second study, they
found that the trust facet in particular predicted these acts in a sample of 90
undergraduate students. Exploring the relations between Agreeableness facets
and vengefulness, Bellah, Bellah, and Johnson (2003) found that low
straightforwardness accounted for the most variance (27%), followed by
modesty (8%) and tender-mindedness (6%). Finally, Miller et al. (2004) linked
risky sexual behavior in a sample of 481 participants (mean age 21 years) to low
Agreeableness at the domain level and to low trust and straightforwardness at the
facet level.
In addition to antisocial and aggressive behavior, other impairments in social–
emotional functioning have also been linked to multiple Agreeableness facets. In
their investigation of emotional functioning, Luminet, Bagby, Wagner, Taylor,
and Parker (1999) found no evidence that the Agreeableness domain was related
to alexithymia, but they found significant relations at the facet level for altruism,
modesty, and tender-mindedness. In terms of internalizing psychopathology,
Bienvenu et al. (2004) found that low trust was related to the experience of
social phobia and low trust and low compliance were related to experiences of
agoraphobia and panic disorder. Consistent with our earlier comments, the
authors identify one of the major concerns with examining the Five Factor
Model facets: They state that “Replication of our facet-level findings is needed,
as the likelihood of Type I errors (rejecting the null hypothesis when it is, in fact,
true) is high, given the large number of comparisons we made.”
Costa and McCrae (1995) note that the detailed information provided by a
facet analysis can either support or call into question the construct validity of
other scales. They reported Agreeableness facet correlations with empirically
based occupational scales from the Hogan Personality Inventory and California
Psychological Inventory (Gough, 1987). These measures include variables for
Service Orientation, Stress Tolerance, Reliability, Clerical Potential, Sales
Potential, and Managerial Potential. The strongest correlations between the
Agreeableness facets for Service Orientation were for compliance (.52), altruism
(.39), and straightforwardness (.35). For Reliability, the largest correlations were
for compliance (.53), straightforwardness (.52), and modesty (.29). The
Agreeableness facets have a special relation to Sales Potential: most of them are
negative: modesty (–.48), straightforwardness (–.36), and compliance (–.32).
Only one Agreeableness facet, modesty, correlated with Management (–.31).
The Agreeableness facets with the largest correlations to Work Orientation, a
measure of the so-called Protestant Work Ethic, were compliance (.34) and
altruism (.24). In another study of the workplace, Thomason, Weeks, Bernardin,
and Kane (2011) found that high levels on the altruism and tender-mindedness
facets of Agreeableness were most appealing to peers when evaluating
managerial potential, whereas high Conscientiousness and achievement striving
appealed to supervisors (see also the chapter by Siebert and DeGeest). These
results highlight the importance of interdependence when evaluating colleagues.
Beyond clinical settings and the workplace, multiple Agreeableness facets
have been connected to a wide range of behaviors, including compulsive buying
(Otero-Lopez & Pol, 2013), caregiver relationships (Riffin, Lockenhoff,
Pillemer, Friedman, & Costa, 2013), birth order (Jefferson, Herbst, & McCrae,
1998), executive functioning (Williams, Suchy, & Kraybill, 2010), therapeutic
orientation (Scandell, Wlazelek, & Scandell, 1997), and reactance (Seemann,
Buboltz, Thomas, Soper, & Wilkinson, 2005), and even the characteristics of
rock musicians (Gillespie & Myors, 2000). Overall, these studies do not provide
compelling a priori predictions, but they offer stimulating possibilities for future
research.
HEXACO
Ashton et al. (2004) have suggested that six, not just five, factors have
repeatedly emerged from both English and non-English lexical studies of
personality structure. As a consequence, Ashton and colleagues proposed a new
six-dimensional model of personality structure forming the acronym
“HEXACO”: Honesty–Humility, Emotionality, Extraversion, Agreeableness,
Conscientiousness, and Openness to Experience. Articles published on
HEXACO deal with both theory and measurement, but conceptually and
empirically the work has its origins in a bottom-up format. Specifically, the
Agreeableness and emotionality factors of HEXACO do not correspond
isomorphically to the Agreeableness and Neuroticism factors of the Big
Five/Five Factor Model (B5/FFM). The Agreeableness factor of six-factor
solutions observed in various languages is characterized by content such as
forgiveness, tolerance, and even-temper versus irritability, anger, and harshness.
It is their contention that B5/FFM Agreeableness is not characterized by content
related to even-temper versus anger, which instead chiefly defines B5/FFM
emotional stability (i.e., low Neuroticism). Moreover, unlike the B5/FFM
Neuroticism versus emotional stability factor, the emotionality factor observed
in the six-factor solutions of various languages is not defined by anger-related
content. The latter dimension, which they believe to be summarized more
accurately and less pejoratively by the name of emotionality, is instead
characterized by content such as anxiety, sentimentality, and vulnerability versus
independence, toughness, and fearlessness. Conversely, they purport that content
related to sentimentality is not a central feature of B5/FFM Neuroticism (i.e.,
low emotional stability) but rather is aligned mainly with B5/FFM
Agreeableness.
Lee and Ashton (2006) suggest that content of the HEXACO Agreeableness
and emotionality factors corresponds roughly to rotational variants of the
B5/FFM Agreeableness and emotional stability factors. Consistent with this
interpretation, HEXACO-PI Agreeableness correlated .72 with IPIP
pleasantness, a scale designed to measure the high Agreeableness/high emotional
stability axis of the Big Five space, and HEXACO-PI emotionality correlated .74
with IPIP imperturbability, a scale designed to measure the high emotional
stability/low Agreeableness axis of the Big Five space. Despite apparent
differences in orientation and focus, work on HEXACO could potentially
contribute to construct elaboration for the Five Factor Model domain of
Agreeableness and its facets. Here we briefly describe three recent programs of
HEXACO-based research relevant to construct elaboration of Agreeableness.
Using a self-report questionnaire format, Stürmer et al. (2013) found that traits
associated with “endeavor” (Extraversion, Openness, Conscientiousness) were
better predictors of xenophilia than traits associated with altruism and
cooperation-related traits (i.e., Honesty–Humility, Emotionality, and
Agreeableness). The HEXACO outcomes suggested that high levels of
Agreeableness are associated with increased concerns for becoming exploited by
others (Ashton & Lee, 2007, see especially p. 156). These researchers conclude
that high levels in traits predisposing individuals to be gentle and cooperative in
dealing with others on an interpersonal level are often insufficient to explain
how people will behave in encounters with members of culturally different or
unfamiliar outgroups.
Looking at other forms of cooperation in a laboratory setting, Hilbig, Zettler,
Leist, and Heydasch (2013) found that Honesty–Humility (but not
Agreeableness) predicted active cooperation (nonexploitation in the dictator
game), whereas Agreeableness (but not Honesty–Humility) was linked to
reactive cooperation (nonretaliation in the ultimatum game). Finally, examining
another aspect of interpersonal relations, namely adolescent bullying, Book,
Anthony, Volk, and Hosker (2012) found that bullying was negatively associated
with personality traits such as fairness and modesty (honesty–humility), but was
unrelated to traits such as forgiveness and tolerance (Agreeableness), based on
adolescent self-report (N = 310). Work of this sort contributes to construct
elaboration of two major domains of personality, but also contributes to a greater
understanding of the structure of complex interpersonal situations.
On the one hand, several aspects of the HEXACO-based work are noteworthy.
First, there are only a few studies, but several of those that are published are
based on a priori hypotheses. Second, the studies are diagnostic, either in terms
of discriminant validation (e.g., honesty–humility versus Agreeableness) or in
construct elaboration (e.g., Agreeableness and wariness about stranger
exploitation). On the other hand, the links to the Big Five and Five Factor Model
(measures and theory) are not direct. The honesty–humility domain seems to
parse Agreeableness-related behaviors in ways that make direct comparisons
with other approaches difficult.
Taking Stock of Empirical Research on Five Factor Model
Agreeableness Facets
Where does this leave us? We restate the claim we made previously, but here
it is based on frequency counts of published empirical articles open to public
scrutiny. There is far more published empirical research on the domain of
Agreeableness than on individual Agreeableness facets. Facets promised to give
us greater precision, not to mention enhanced predictive and explanatory power,
than we would get from an omnibus measure of the superordinate domain of
Agreeableness. To date, there is little evidence to support that promise.
Inferences from small literatures are risky. Often, the facet information is
collected as part of an investigation of some phenomena (e.g., traumatic
experience, gambling), without any a priori theoretical rationale for the inclusion
of the Agreeableness facet. Often, these studies provide little or no increment in
predicting external criteria over the comprehensive Agreeableness domain
measure. As a collective, the empirical studies of Agreeableness facets
contribute less than they could to construct elaboration of the superordinate
Agreeableness domain. It might be argued, however, that construct elaboration
was not the top priority for researchers in this area. Furthermore, there is
heterogeneity in the research literature. Some facets have been investigated more
extensively (e.g., trust) than others (e.g., modesty, tender-mindedness). Perhaps
the most prudent conclusion is that the promise of Five Factor Model-oriented
Agreeableness facets needs to be more tightly focused on construct elaboration
rather than on diverse correlates of Agreeableness (and other domain) facets. In
this sense, the potential value of Agreeableness facets still remains to be met.
It could be rebutted that we are quibbling about semantics and verbal labels
for facets. In some of the best studies, the correlations involving facets converge
reasonably with other measures, align with intuition, and make sense (e.g., Rojas
& Widiger, 2014). It other cases, however, the facets have correlates that seem
tangential to the label. For example, Costa and McCrae (1995) reported that the
largest correlation with the Agreeableness compliance facet in the California Q
Set is “behaves in a sympathetic manner.” It is difficult to understand how this is
a manifestation of compliance, but not altruism. In contrast, Costa and McCrae
(1995) reported, as an empirical finding, that the altruism facet does not correlate
with “behaves in a sympathetic manner.” At the same time, the largest single
correlation with the Agreeableness altruism facet in the California Q Set is
“basically distrustful” at r = –.34. We might think that this behavior best fits
under the Agreeableness trust facet. It does (r = –.54), but this Q Set item has
significant correlations with three different Agreeableness facets. Is this pattern
because all three facets fall under the same Agreeableness domain or because
there are serious problems with construct validity? A skeptic might wonder
about discriminant validity, or even whether the labels match the content
faithfully. At the least, the NEO PI-R Agreeableness facet labels can create
problems due to misinterpretation.
Shadish et al. (2002) describe how such scientific misinterpretations can
happen. They note that practicing scientists routinely make causal
generalizations in their research, and they almost never use formal probability
theory as the basis for generalization. Instead, they use five closely related
principles; these are (1) surface similarity, (2) ruling out irrelevancies, (3)
making discriminations, (4) interpolation and extrapolation, and (5) causal
explanation. The principle of surface similarity seems to be most relevant here.
Shadish et al. (2002) define surface similarity as the act of “assessing the
apparent similarities between study operations and the prototypical
characteristics of the target of generalization” (p. 357).
Part of the meaning of surface similarity is captured by the now-almost-
obsolete concept of face validity. Obsolete or not, it is widely used by practicing
researchers to make decisions about which predictors should be explored with
which outcomes. If the prototypical properties of an Agreeableness facet seem
similar, intuitively or linguistically, to the prototypical thought, feeling, or
behavior in question, then that facet becomes a predictor. That is, they choose
plausible-sounding labels that appear to be reasonable, with the scientist
accepting the label on faith. The prototypical features associated with the verbal
label “altruism” probably share more intuitive surface features with helping,
volunteering, and donating to charity than being trusting/distrustful.
Nevertheless, intuition-based heuristics that lead to surface similarity matching
can be false prophets. The prototypical features associated with the verbal label
“compliance” probably share more intuitive surface features with compliance,
conformity, and obedience than with a “sympathetic manner.” Begue et al.
(2015) thought so. As an empirical fact, the largest single correlation with the
compliance facet and the California Q Set was “behaves in a sympathetic
manner” (Costa & McCrae, 1995, Table 2, p. 32; see also Shadish et al., 2002,
pp. 357–361, for details on surface similarity as a pervasive influence on
research decisions).
The more general point is that labeling and measurement methods have
contributed to skepticism about Agreeableness in particular as a domain and
scientific construct. Lacking the kind of theoretical foundation associated with
domains such as Extraversion and Neuroticism, Agreeableness is vulnerable;
mere empirical links for Agreeableness seem ephemeral. Researchers often
frame questions in terms of labels and surface similarity. As Costa and McCrae
(1995) noted, perhaps Agreeableness absorbs so much rating variability because
it is assessed with such broad traits (e.g., “kind”) that necessarily covary with
many more lower-level trait words than do narrower traits for other domains. In
rebuttal, broadly assessed or not, Agreeableness as a domain predicts behaviors
external to the Agreeableness measure.
Agreeableness has been measured through observation by knowledgeable
informants such as spouses (Costa & McCrae, 1988), employment supervisors
(Hogan, Hogan, & Roberts, 1996), and teachers/childcare supervisors (e.g.,
Digman & Takemoto-Chock, 1981; Tobin & Graziano, 2011). Still, self-report
measures continue to be the most commonly used method (Costa & McCrae,
1988; Goldberg, 1992; Goldberg et al., 2006; John & Srivastava, 1999). Finch,
Panter, and Caskie (1999), Graziano and Tobin (2009), Hofstee (1994), and
Paulhus and John (1998) provide more thorough discussions of Agreeableness-
related issues in assessment.
Because self-report is the most common way of measuring Agreeableness, it
is reasonable to ask questions about self-favoring biases emerging from self-
report. Given that words indicative of high Agreeableness (e.g., friendly, likes to
cooperate with others) are considered more favorable than those indicating low
Agreeableness (e.g., cool, aloof, unkind), items assessing Agreeableness have
raised social desirability concerns. After all, one of the main facets of
Agreeableness is compliance, and through surface similarity, should we not
expect a link between Agreeableness and the compliance-related activity of
social desirability responding? Observations of these measures have generated
suspicions that the entire domain of Agreeableness may simply reflect
responsiveness to the direction of the prevailing social wind (e.g., Begue et al.,
2105; DeYoung et al., 2002).
If this proposition is true, then an individual’s standing on Agreeableness
could be altered by manipulating the social desirability of Agreeableness
experimentally. Both correlational and experimental data, however, do not
support this alternative explanation for Agreeableness findings. Finch et al.
(1999) conducted both joint and interbattery factor analyses of the Big Five, as
measured with NEO PI (Costa & McCrae, 1988) and Murray’s needs, as
measured with the Personality Research Form (Jackson, 1984). Both methods
produced a large factor for Agreeableness (and for each of the other Big Five
dimensions), but both analyses failed to find evidence that social desirability-
related needs such as abasement, defendance, succorance, or desirability loaded
on the Agreeableness factor.
Taking a more focused experimental approach, Graziano and Tobin (2002)
examined the relations among Big Five dimensions and social desirability in
three studies. In one study, they pretested participants for domain-level
Agreeableness and then randomly assigned them to one of three different social
desirability conditions, namely Good to Be Agreeable, Bad to Be Agreeable, or
No Information Control. Graziano and Tobin (2002) found that self-ratings of
Agreeableness for participants who were randomly assigned to the bad to be
agreeable condition remained the same or actually increased their self-reports of
Agreeableness. Furthermore, they found that other domains of the Big Five (i.e.,
Neuroticism and Conscientiousness) had stronger correlations with measures of
social desirability than did Agreeableness. It might be argued that these studies
used domain-level Agreeableness, not the compliance facet, as its predictor. The
domain-level assessment may have been too molar to capture the process; the
compliance facet may have given a more precise prediction. In rebuttal, the
domain-level assessment is not too molar to predict helping, cooperation, or
conflict tactics. If facets are assumed to operate at comparable levels of breadth
beneath their respective domains, then the “too molar criticism” is less plausible.
Taken together, these and other findings suggest that Agreeableness effects are
probably not primarily (or even preponderantly) artifacts of social desirability.
Instead, they are assessing an authentic personality domain linked to important
differences in interpersonal relations and social behavior.
Theoretical Explanations for Agreeableness

The process of searching for correlates and measurement artifacts could be
endless in the absence of guiding theory. Theories vary in size, scope, and
compatibility with other theories. First, we will briefly revisit the problem of
extracting any one domain or facet from a larger system of personality for
purposes of theoretical explanation. Second, we discuss attempts to link
Agreeableness and its facets to various basic processes such as empathy,
temperament, and differences in motivation to accommodate to the social
environment. Third, we will discuss a specific endogenous mechanism as a
prime candidate for the basic tendency underlying Agreeableness, as outlined in
Five Factor Theory.
Earlier, we provided a rationale for extracting domains and facets from the
Five Factor Model for focused explanation. We noted that extracting one domain
or facet from the full set for examination works directly against one of the main
strengths of the Five Factor Model, namely integration. Extraction could be
potentially misleading if the domains or facets form distinctive combinations or
configurations “horizontally” (e.g., De Raad & Hofstee, 1993; Hofstee et al.,
1998, as cited by De Raad, 2000). An individual scoring high on Agreeableness
could be one of several different nuanced “varieties” (e.g., Costa et al., 2002;
DeRaad, 2000; Paulhus & Jones, 2015). By collapsing across the other variables,
we run the risk of failing to “generalize across” cases (Shadish et al., 2002) if
they also score high on other domains as well.
Extraction comes with tradeoffs. On the benefits side of the ledger, extraction
allows researchers to focus on specific behaviors and processes. On the
disadvantage side, extraction could be misleading in the absence of the larger
system. For example, Ode, Robinson, and Wilkowski (2008) showed that at
higher levels of Agreeableness, the link between anger and Neuroticism was
considerably reduced. Similarly, Ode and Robinson (2008) found that
Agreeableness moderated the relation between Neuroticism and depressive
symptoms. As another example, Jensen-Campbell and Graziano (2005) found
that higher Conscientiousness partially compensated for low Agreeableness in
predicting cheating during a resistance to temptation task with adolescents. In
each of these cases the substantive concern was affect regulation. The
configuration of personality patterns (versus the examination of one personality
dimension at a time) is a front-line issue in personality theory and measurement,
generally following the tradition of the Abridged Big Five Circumplex (AB5C),
but it suggests avenues for refinement of associations among personality
domains and their shared relation with behavior (De Raad, 2000, p. 83; De Raad,
Hendriks, & Hofstee, 1994). This issue is especially relevant to the discussion of
our opponent process approach to Agreeableness (Graziano & Habashi, 2010;
Graziano & Tobin, 2009, 2013).
With these complexities noted, there is still the challenge of explicating each
domain and its facets. What do the diverse thoughts, feelings, and behaviors
associated with the Agreeableness domain have in common? By asking this
question, we can perhaps identify a variable or set of variables that could qualify
as the endogenous basic tendency underlying Agreeableness as described
(abstractly) in Five Factor Theory. Here we consider three candidates: (1)
dispositional empathy, (2) the temperamental process of effortful control, and (3)
processes associated with social accommodations. Finally, we will consider a
learning-oriented opponent-process mechanism compatible with Five Factor
Theory.
Agreeableness and Empathy

Agreeableness as a domain may not be highly related to the other four major
structural domains of personality, but it is probably related to other dispositions
(e.g., Finch et al., 1999). What processes might be candidates for the basic
tendency underlying Agreeableness? Intuitively, we might expect empathy to be
a basic tendency, or at least a component of the basic tendency, of
Agreeableness. Davis (1996, 2015) conceptualized dispositional empathy as a
psychological system that is linked to thoughts and feelings of connectedness to
other people. For Davis, empathy has a cognitive component (perspective
taking), a self-focused emotional component (personal distress), and a third
other-oriented emotional component (empathic concern; a fourth component,
“empathic fantasy,” is rarely studied). Given the nature of the Agreeableness
facets (e.g., altruism, tender-mindedness, and perhaps also trust and
compliance), empathy becomes a plausible candidate. Empirically, studies show
that Agreeableness is related to dispositional empathy. Empirical research
supports the claim that Agreeableness is related both to empathy and to overt
helping (e.g., Graziano, Habashi et al., 2007). Persons high in Agreeableness
report greater ease in seeing the world through others’ eyes (perspective taking)
and feeling the suffering of others (empathic concern), but not necessarily in
experiencing self-focused negative emotions (personal distress) when observing
victims in sorrow. Past research showed that these cognitive and emotional
processes are related to overt helping, so we might expect persons high in
Agreeableness to experience the kind of empathic processes that motivate
prosocial behavior (e.g., giving aid to others, even to strangers) more than their
peers. They do (Graziano & Habashi, 2015).
Nevertheless, Graziano and Habashi (2015) expressed reservations about
Agreeableness or empathy being the single basic process underlying altruism, or
even the less restrictive prosocial behavior. Despite their tendency to take the
perspective of a victim, and to experience other-oriented empathic concern,
persons high in Agreeableness will decline helping victims if they have a clear
opportunity to escape from the helping situation. If they cannot escape, they will
help. Persons high in Agreeableness may indeed be more helpful and empathic
in naturalistic settings than their peers, but the motivation underlying the helping
is probably egoistic: It is personally aversive to them to experience the suffering
of others. To eliminate their own unpleasant state, they help. Such helping
makes them more prosocial, but probably not more altruistic, than their peers.
Adding further to the reservation is the conceptualization of empathy as a
composite of two different processes. The self-focused process of personal
distress generally undermines prosocial behavior and helping, whereas the other-
focused process of empathic concern promotes it. Given this pattern, we might
expect the two dispositions to be negatively correlated, but they are correlated
positively in most published studies that report separate components (e.g.,
Graziano, Bruce, et al., 2007). Graziano and Habashi (2015) proposed that
personal distress and empathic concern must be configured in a different pattern
in persons high and low in Agreeableness. At the least, this conceptualization
suggests that some additional variable moderates the link between empathy and
Agreeableness as the process underlying prosocial behavior.
Agreeableness and Temperament

If there is a third-variable moderator in the empathy–Agreeableness system,
then what kind of variable is needed? One particularly promising line of theory
offers some clues to the nature of the basic tendency. It is provided by theory
and research on temperament and development. Agreeableness may be tied
distinctively to temperament-based systems of self-regulation, especially as they
apply to regulating frustrating emotions in social relations (Cumberland-Li,
Eisenberg, & Reiser, 2004; Eisenberg & Spinrad, 2004; Graziano & Tobin,
2013; Jensen-Campbell & Graziano, 2005; Jensen-Campbell & Malcolm, 2007;
Laursen, Pulkkinen, & Adams, 2002; Tobin & Graziano, 2011). Consistent with
these findings, Smits and Boeck (2006) found that Agreeableness had a positive
relation with the Behavioral Inhibition System (BIS) and a negative one with the
drive scale of the Behavioral Approach System (BAS).
Ahadi and Rothbart (1994) were the first to propose that an early appearing
temperamental variable, effortful control, provides a developmental substrate for
subsequent personality structure in children, adolescents, and adults. They
proposed that effortful control is part of a common developmental system
underlying two of the major dimensions in the Big Five, namely Agreeableness
and Conscientiousness. Rothbart and her colleagues (e.g., Evans & Rothbart,
2007; Rothbart & Bates, 2006; Rothbart & Posner, 1985) proposed that effortful
control modulates other temperament systems as the frontal cortex matures.
Effortful control is related to early-appearing differences in the ability to sustain
and shift attention and the ability to initiate and inhibit action voluntarily (e.g.,
Kochanska & Knaack, 2003; Kochanska, Murray, & Coy, 1997; MacDonald,
2008). Effortful control seems to be related to the ability to suppress a dominant
behavior to perform a subdominant response or even an opposing dominant
response, as is commonly the case for Agreeableness. Applying this to our
helping example, upon witnessing a victim in distress, people experience the
unpleasant self-focused emotion of personal distress. The dominant behavior
associated with this aversive state is to escape without helping. However, people
high in dispositional effortful control have the resources to suppress that
dominant response, allowing subdominant prosocial responses to come on line.
Graziano and Habashi (2015) as well expressed some reservations about this
explanation. First, the temperament of effortful control is assumed to lay the
foundation for Agreeableness and Conscientiousness. That conceptualization
implies an intimate, perhaps even part–whole link between effortful control and
Agreeableness. It is possible that one part of a system could act as a moderator
on another part, but where does effortful control end and Agreeableness begin?
Within the framework of Five Factor Theory, perhaps effortful control skills
provide the basic tendency that precedes the later developing characteristic
adaptation of Agreeableness.
Based on patterns of correlations using the Adult Temperament Questionnaire
(Thomas, Mittelman, Chess, Korn, & Cohen,1982), a few writers suggested that
effortful control (especially attentional control) may not be the developmental
precursor of Agreeableness, and therefore it is not a good candidate for the basic
tendency for Agreeableness. That is because temperamental effortful control is
more closely tied to Conscientiousness than to Agreeableness. That may or may
not be true because much of the evidence supporting this claim is open to
alternative interpretations, or at least is not differentially diagnostic. For
example, the original content of the Adult Temperament Questionnaire did not
include items dealing explicitly with the social and interpersonal expression of
effortful control. It is possible that effortful control appears to be more closely
correlated with Conscientiousness than Agreeableness in this case because the
measure itself does not assess social and interpersonal expressions of effortful
control (but see also Evans & Rothbart, 2007). Even then, some social and
interpersonal tasks (e.g., resistance to temptation) will require processes that
place more stress on Agreeableness than on Conscientiousness, whereas other
tasks (e.g., reactions to disappointing gifts) could reverse that pattern (see, for
example, Jensen-Campbell & Graziano, 2005; Tobin & Graziano, 2011). These
issues are worthy of further empirical investigation.
Consistent with this theoretical connection, Jensen-Campbell et al. (2002)
found that both Agreeableness and Conscientiousness were associated with the
Stroop Test and the Wisconsin Card Sorting Task, both of which are considered
traditional measures of self-regulation. Furthermore, Haas, Omura, Constable,
and Canli (2007) found that Agreeableness is related to activation of the right
lateral prefrontal cortex following exposure to negative emotional stimuli. These
results suggest that individuals high in Agreeableness automatically engage in
processes of emotion regulation when exposed to negative stimuli. In a
converging multimethod three-study set, Tobin et al. (2000) presented evidence
consistent with this connection: They found that individuals high in
Agreeableness reported experiencing stronger emotional reactions when they
were presented experimentally with negatively valenced stimuli. However, based
on both objective observations and on self-report, they also exerted greater
efforts to regulate these emotions than their peers. Similarly, Tobin and Graziano
(2011) found that Agreeableness predicted regulation of negative affect in
school-aged children using the disappointing gift paradigm (Cole, 1986; Saarni,
1984). Thus, the Agreeableness dimension has been associated with both
negative affect expression and regulation in more than one age range (i.e.,
school-aged children and university students).
Agreeableness and Social Accommodation

Perhaps the problem with empathy and effortful control explanations is that
they are too closely connected to narrow specific behaviors. Perhaps what is
needed is a more abstract variable that covers more diverse forms of
Agreeableness-related behaviors. Graziano and Tobin (2013) attempted to
expand the explanatory structure surround Agreeableness by moving past
behavioral regularities toward underlying processes. They proposed that
Agreeableness was closely linked to interpersonal accommodation, a process
central to social psychology. Here is a bridge between personality and
mainstream social psychology. Gordon Allport (1968) defined social psychology
in terms of the adjustments in thoughts, feelings, and behavior that individuals
make as a result of the actual, imagined, or implied presence of other people. In
keeping with this perspective, we proposed that Agreeableness refers to the
motivation to accommodate to other people with the goal of maintaining smooth
interpersonal relationships (Graziano & Eisenberg, 1997; Graziano & Habashi,
2010, 2015). More formally, Agreeableness is defined as a superordinate
summary term for a set of interrelated dispositions and characteristics manifested
as differences in being likable, pleasant, and harmonious in relations with others.
Agreeableness can be indexed through individual differences, but it is a larger
construct likely reflecting a set of underlying psychological processes.
Presumably, the Five Factor Model facet structure captures the key processes in
that set. Research shows that persons who are described by others as “trusting”
are also described as “kind” and “warm.” Taken together, this combination of
characteristics points toward a domain of behaviors involving flexibility and
tolerance in dealing with others. We now know that these tendencies are
relatively stable over time (Costa et al., 2000; De Fruyt et al., 2006; Hair &
Graziano, 2003; Laursen et al., 2002; Shiner, Masten, & Roberts, 2003).
One problem with this interpretation is that it is largely descriptive, and does
not identify the precise mechanism that generates the stability of differences in
accommodation tendencies over time. Furthermore, this approach implies that
persons high in Agreeableness will be more accommodating than their peers to
most people most of the time. There is evidence that persons high in
Agreeableness are selective and discriminating in their accommodation (e.g.,
Graziano, Bruce, et al., 2007; Graziano, Habashi, et al., 2007; White et al.,
2012), suggesting there is still much to be learned about these processes.
Agreeableness and Five Factor Theory

The most comprehensive theory covering Agreeableness is the Five Factor
Theory of Personality (McCrae & Costa, 2003; see also the chapter by Costa and
McCrae). First, we will briefly summarize key aspects of the theory as they
apply to the special case of Agreeableness and its facets. Next, we will discuss
the theory’s compatibility with lower level theories and with empirical
outcomes. McCrae and Costa (2003) note that the first and most central fact to
be explained in personality is the stability of personality traits over long periods
of time (e.g., Costa et al., 2000). Yet some other aspects of personality such as
habits and self-images do seem to change. For personality to be both stable and
also changeable, personality must consist of at least two different components.
In the Five Factor Theory, personality consists of three central components:
Basic Tendency, Characteristic Adaptation, and Self-Concept (in Five Factor
Theory, self-concepts are a form of characteristic adaptation, and will not be
discussed further). The key distinction is between a Basic Tendency and a
Characteristic Adaptation because Basic Tendencies are assumed to be stable
over long periods, whereas Characteristic Adaptations are open to change. Basic
Tendencies are the abstract capacities and tendencies of the individual, whereas
Characteristic Adaptations are the concrete, acquired structures that emerge as
the individual interacts with the environment.
In its original form, Five Factor Theory made the strong assumption that the
environment did not influence Basic Tendencies; they were endogenous,
biological structures. Five Factor Theory focused on personality traits, but other
kinds of predisposing tendencies are Basic Tendencies as well. Included in this
list are cognitive abilities, artistic talents, sexual orientation, and the entire
neurocognitive ensemble underlying psychological activities such as perceiving
and thinking. All learned skills are Characteristic Adaptations, including habits,
interests, attitudes and beliefs, and language. Some Characteristic Adaptation
skills and tendencies such as language or self-concepts are stable over time, but
generally Characteristic Adaptations are less stable over time than Basic
Tendencies. These changes over time could involve learning or the unfolding of
endogenous aspects of persons (e.g., maturation) (see also Costa et al., 2000).
In Five Factor Theory, presumably the process underlying the supertrait
domain of Agreeableness is a Basic Tendency. As such, the process is assumed
to be endogenous, and has its origins as an inherited biological system. Behavior
genetic twin studies suggest that Agreeableness shows broad heritability
comparable to (but possibly slightly lower than) the other Big Five domains,
depending on assumptions about nonadditive genetic effects and
shared/nonshared environments (e.g., Loehlin, 1992; but see as well Matthews &
Deary, 1998, pp. 116–117). Exactly what is being inherited?
From these simple assumptions, many important questions follow. First,
precisely what is the process that underlies the Basic Tendency of
Agreeableness? To qualify as a Basic Tendency, it must be an abstract tendency
that would give rise to Characteristic Adaptations. The large collection of
behavioral differences associated with Agreeableness cannot be the Basic
Tendency itself. They are instead acquired residues of interactions of
Characteristic Adaptations and the environment. If this is true, then personality
development is a process in which Basic Tendencies predispose individuals to
acquire certain kinds of skills and habits, depending on the availability of
environmental conditions relevant to the Basic Tendency. In the absence of those
environmental conditions, the typical Characteristic Adaptation will not develop,
nor be present in form typically found with persons of that Basic Tendency. In
this view, the Basic Tendency underlying Agreeableness provides a mechanism
for a dynamic “reaction range” (Dobzhansky, 1955; Griffiths & Tabery, 2007)
determining responsiveness to environmental variations. The Basic Tendency
would be a species-wide mechanism (Eastwick, 2009) with some small variation
in operating parameters among individuals, whereas the Characteristic
Adaptations associated with it would show considerably more variability due to
differential social learning experiences in the environment. If this is true, then
we have an explanation for why persons high in Agreeableness can appear as
different phenotypic varieties, as described previously. Thus, a key question is:
What are the fundamental, endogenous processes underlying the Basic
Tendencies that make up Agreeableness?
The next important conceptual question for Five Factor Theory is precisely
how to differentiate Basic Tendencies from Characteristic Adaptations. At
present, we must rely on intuitions. Perhaps we can perform a process similar to
“backward engineering.” In a study that may address the Characteristic
Adaptations associated with Agreeableness, Laursen et al. (2002) tracked 194
individuals longitudinally for 25 years collecting peer and teacher reports of self-
control, compliance, and aggression, with compliance at age 8 and again at age
33 years. They found distinct patterns in childhood and adulthood that varied by
level of Agreeableness. Specifically, fewer problems on noncompliance and
inattention were reported for children high in Agreeableness relative to those
low in Agreeableness. Boys high in Agreeableness also had fewer behavior
problems and better grades relative to boys low in Agreeableness. Laursen et al.
(2002) noted continuity in Agreeableness status, with more positive outcomes
associated with high Agreeableness. Specifically, adults high in Agreeableness
reported fewer arrests and problems with alcohol, less depression, and greater
career stability relative to their peers. Based on these results, Laursen et al.
(2002) concluded that Agreeableness is linked to processes of self-regulation
that appear stable from childhood through adulthood (see also De Fruyt et al.,
2006).
Agreeableness as a Complex of Motivational and Cognitive Processes

Graziano and Eisenberg (1997) first identified Agreeableness as a summary
term for individual differences in motives to maintain positive interpersonal
relations. More than a decade later, Graziano and Tobin (2009) reported reliable
behavioral differences tied to Agreeableness, implicating these motivational
processes, but they noted the need for additional research to reveal its underlying
processes and mechanisms. In the language of Five Factor Theory, we seemed to
know more about Characteristic Adaptations associated with Agreeableness than
its most central Basic Tendency. Graziano and Tobin (2009) made progress in
identifying these psychological processes by noting parallels in the way
Agreeableness related to the two outwardly opposite social behaviors of
prejudice and helping. The main concern was not to dissect two different social
behaviors—admitting these two classes of behavior qualify as important—but to
identify a process of mechanism underlying them and to explain its connection
to Agreeableness and social accommodation.
In keeping with everyday intuition, there is evidence from behavior genetics
that prosocial and antisocial systems may be different (e.g., Krueger, Hicks, &
McGue, 2001), but Graziano and Tobin (2009) noted that the commonalities
between behaviors of prejudice and helping are not merely coincidences:
Underlying both are accommodative processes with approach and avoidance
elements. It is plausible that a common regulatory system linking approach,
avoidance, and Agreeableness, may underlie both forms of behavior, and likely
others. In general, Agreeableness and social accommodation might be tied
closely to self-regulation, and to balancing approach and avoidance tendencies in
particular. Here we offer this claim: That system may provide the Basic
Tendency underlying the panoply of Agreeableness-related events.
CHARACTERISTIC ADAPTATIONS
How would such a system work? In overview, the system we describe is based
on an opponent process model of learning (Solomon, 1980; Solomon & Corbit,
1974; see also Eisenberger, Lieberman & Satpute, 2005 for a related two
component approach). The dual-process system has an endogenous mechanism
(corresponding to Five Factor Theory’s Basic Tendency) that responds to the
environment and accumulates “residues of experience” in patterned ways. The
operative mechanism within the system is endogenous, and in keeping with
McCrae and Costa (2003), the basic mechanism itself is not influenced by the
environment. Operationally, the system consists of two components that operate
in time and includes an initial emotional reaction to events or people (State A)
followed by a subsequent homeostatic “correction” (State B). It is the nature of
the system for State A to activate an opponent State B. State A and State B are
yoked endogenously through hard wiring. In terms of outcomes, State B
automatically reduces the effects of State A. With repeated exposure, the
opponent State B will be activated by State A with shorter and shorter latencies.
That is, when activated, State A is expressed at lower levels, whereas State B is
expressed at higher levels. The mechanism is endogenous because it is an
evolved homeostatic device that maintains the integrity of the organism when it
is faced with challenges.
Detailed aspects of the application of this model to Agreeableness are
provided elsewhere (i.e., Graziano & Tobin, 2009, 2013). Here, we present an
overview of the opponent approach, as shown in Figure 6.1. In keeping with
Five Factor Theory, the mechanism itself is endogenous. It is the single, simple
Basic Tendency underlying the entire domain of Agreeableness. The first
process activated is labeled State A. Its activation is automatic. In keeping with
Solomon’s model, it is an unconditioned response to the onset of an
environmental stimulus. It is active while the evocative stimulus is present and
ends when the stimulus is removed. State B is the second process activated as an
opponent. It is not activated as quickly, but it remains well after State A
terminates. States A and B are opponents, but State A is the first to come on line
in response to an environmental event. Thus, State A operates in almost pure
form (without an opponent) initially. As discussed in Graziano and Tobin
(2013), this model can be applied to helping behavior: If State A is Personal
Distress and State B is Empathic Concern, then the initial response to a victim
would be unopposed Personal Distress. If escape is possible in this interval, the
victim will not receive help. Using the same logic, first reactions to unusual
cases (e.g., victims of misfortune) as well as to members of outgroups would be
personal distress and avoidance. As time passes, these processes have the
opportunity to unfold: State B (Empathic Concern) can be activated, opposing
the processes of State A.
IMPLICATIONS FOR AGREEABLENESS

The opponent process approach provides a conceptualization of the nature of
Agreeableness and its links to social accommodation, but it leaves some
questions unanswered. First, what appropriate time window captures the
expression of social accommodation? It is plausible that if the window we
examine is too narrow, we would miss the dynamics underlying these processes.
When these systems operate simultaneously, as in the opponent process, one
system (State B) may diminish the influence of another (State A) once State B is
activated. When studying a helping incident, it is common to assume that a
single process is operating, but it is more likely that these components are at
work simultaneously.
A second question is, what is the role of individual differences such as
Agreeableness, empathic concern, and personal distress? One approach to
individual differences, whether at the domain or facet level, is to regard them as
proxies or markers of differences in cognitive or emotional processes. Then the
question shifts to identifying processes. Temperament researchers (e.g., Rueda,
Posner, & Rothbart, 2005) remind us that each individual is born with an
emotional core that interacts with the social learning environment over the life
course, leaving residues such as internal working models, social learning
histories, and individual differences in personality. Individuals learn about others
(including outgroup members) as they develop, but what exactly are they
learning (Biesanz, West, & Millevoi, 2007)? Evidence suggests that most
individuals are selective in attention, encoding, retrieval, and general
information processing.
As we noted previously, the Basic Tendencies of personality would condition
that selectivity. Personality development is a process in which Basic Tendencies
predispose individuals to acquire certain kinds of skills and habits, depending on
the availability of environmental conditions relevant to the Basic Tendency. In
the absence of those environmental conditions, the typical Characteristic
Adaptation will not develop, nor be present in a form typically found with
persons of that Basic Tendency. The Basic Tendency underlying Agreeableness
provides a mechanism for a dynamic “reaction range” (Dobzhansky, 1955;
Griffiths & Tabery, 2007) determining responsiveness to environmental
variations. In this view, persons high in Agreeableness are predisposed to learn
about others, to be flexible and accommodating in dealing with others, and to be
biased toward learning prosocial (but not necessarily altruistic) thoughts,
feelings, and behavior. In normative environments, person high in Agreeableness
will typically acquire Characteristic Adaptations of the sort described by Laursen
et al. (2002). These Characteristic Adaptations will be stable over time. From the
perspective of Five Factor Theory, the Characteristic Adaptations were induced
by the endogenous Basic Tendencies outlined in the opponent process model.
Figure 6.1. Opponent process model of motivation (adapted from Solomon & Corbit,
1974).
Theorists (Brown & Brown, 2006; Dijker & Koomen, 2007; Eastwick, 2009;
Jonas et al., 2014; Porges & Carter, 2012) posited that evolution left humans
with two powerful motive systems in Fight/Flight and Care. Dijker and Koomen
(2007) describe Fight/Flight as unconscious responses that are deeply rooted in
our evolutionary history to protect the organism. This system allows the
individual to escape from danger or fight if necessary. In contrast, the Care
system is tied to kin selection and attending to the needs of others. It has the
capacity to override the Fight/Flight system. These motives systems are
considered universal, but there are probably individual differences in the relative
strength of these motives. In terms of overt behaviors, observers might identify
these socially important behavioral differences as domain-level Neuroticism and
Agreeableness, respectively (Graziano & Habashi, 2010).
Common approaches to studying these processes are to build structural
models or collect data showing intercorrelations among variables such as Care,
Agreeableness, its six facets, and some other disposition such as empathy or self-
esteem. These methods would likely miss important aspects of the dynamic
processes, their major dispositional inputs, and the range of influence of the
individual difference under consideration. Nevertheless, from the perspective of
Five Factor Theory, repeated exposure to certain kinds of environmental events
could change the Characteristic Adaptation, but not the basic operation of the
inherited dispositions and motives. For example, recent experimental research
showed that ostracism and social rejection can induce lower Agreeableness, at
least temporarily (Hales, Kassner, Williams, & Graziano, under review).
Presumably, those experiences would reset the level of the Characteristic
Adaptation, but not alter the opponent process, the endogenous Basic Tendency
mechanism underlying Agreeableness. At this time, it is not clear if
Agreeableness facets are more responsive to environmental experiences than the
superordinate domain, or are differentially responsive among themselves to
environmental events. It is possible that facets differ in how much their
operation is determined by the endogenous Basic Tendency, or even whether
they are themselves Characteristic Adaptations derived from the endogenous
Basic Tendency.
Connecting the affective components of empathy and the personality domain
of Agreeableness to interpersonal behaviors and to more general self-regulatory
processes (Graziano & Tobin, 2009) within the framework of Five Factor
Theory is novel. Not surprisingly, several questions remain unanswered. Is
Agreeableness as a domain tied to the Care system only or to Fight/Flight as
well? Does the facet structure of the Five Factor Model give us some leverage in
understanding the different processes at work within the domain? Is the
Agreeableness domain and/or its facets tied to both personal distress and
empathic concern, to both prejudice and the suppression of prejudice, or to just
one of these elements in each pair? We believe that the opponent process
approach to Agreeableness allows us to conceptualize these processes in ways
that will help guide future research.
Graziano and Habashi (2010) offered a few tentative ideas. For example,
delayed helping appears to be worthy of special attention (see Penner et al.,
1995), given that rates of helping are affected by the time interval between the
request for help and the opportunity to provide it. If the opponent process system
operates as described here, then differences may be observed between helping
provided after a short delay relative to helping following an immediate request.
Delaying the request may allow the initial Fight/Flight reaction to come under
the control of the opponent Care system, in effect disinhibiting helping with
time. Undoubtedly, we would also see characteristic emotions, such as relief at
finally having an opportunity to provide assistance. Based on the previous
rationale, we would also expect persons high in Agreeableness to offer more
help, sooner and with less influence of delay, than persons low in Agreeableness.
Future research guided by this model is likely to help us understand the dynamic
processes underlying the Agreeableness domain and related interpersonal
processes.
It could be argued that it is misleading to describe Agreeableness primarily as
a domain-level individual difference. The measurement at the domain level is
too molar to be valuable for either prediction or explanation. Evidence cited in
this chapter as well as previous review chapters (e.g., Graziano & Tobin, 2013)
argues against that statement. Another critique involves ways that the domain-
level approach biased the basic definition of the domain. Inevitably, say critics,
persons low in Agreeableness are defined by default, as simply lacking qualities
that are possessed by persons high in Agreeableness. There is more to the
concept of “cold” beyond being merely the absence of “heat.” If persons high in
Agreeableness show empathic concern, show a willingness to accommodate to
the goals of others, and are motivated to minimize conflict with others, then do
persons low in Agreeableness simply lack these qualities? By implication, some
of the distinctiveness of persons low in Agreeableness might be lost. We offer
several comments in response.
First, all of the Big Five structural domains of personality, including
Agreeableness, are better represented conceptually as a set of continuous
variables than as categories (e.g., Finch et al., 1999). The major domains of
personality allow for the description of larger patterns and trends in thoughts,
feelings, and behaviors. The domain may be less well suited than the facets to
predicting specific unique, one-time behaviors such as preferring chess or
holding a specific Theory of Mind. Perhaps this is the place in which
Agreeableness facets could prove especially valuable in the future. Second, the
opponent process model described here suggests that Agreeableness is a proxy
for understanding a larger dynamic process of social accommodation. That there
are noticeable individual differences suggests that facet-level analyses may
prove valuable for construct elaboration. Presumably, the social accommodation
process itself operates in different situations for different individuals (e.g.,
Kelley, Holmes, Kerr, Reis, Rusbult & van Lange, 2003; Pursell, Laursen,
Rubin, Booth-LaForce, & Rose-Krasnor, 2008). We infer that variability in
Agreeableness means that both distal evolutionary processes and more proximal
social dynamics can find uses for individuals at most levels along a continuum
of social accommodation (e.g., Savani, Morris, Naidu, Kumar, & Berlia, 2011).
Third, there is much to be gained by comparing individuals who are located at
different places along a single continuum. These comparisons can implicate
processes underlying the dimension as a whole. As such, our opponent process
model is a product of these comparisons. Subsequent research may corroborate
or refute such inferences, but we are optimistic that these are worthwhile
pursuits.
None of these comments is directed against the measurement of facets
subordinate to the domain of Agreeableness. The promise of facets is greater
precision in prediction and explanation. So far, however, this enhanced precision
promise has not been met by the empirical literature. There are several pervasive
problems. Many researchers have not based their facet-based research on a priori
predictions. Instead, most NEO-based facet research is atheoretical and
inductive. In addition, Five Factor Theorists should address the issue of
comprehensiveness of coverage. Many Agreeableness-related forms of thoughts,
feelings, and behaviors are not well described by the current set of Five Factor
Model facets. The altruism and compliance facets in particular are problematic.
A second issue related to comprehensiveness is exhaustiveness. In particular, the
separate NEO PI-R Agreeableness facets are summed to capture
comprehensively the full range of thoughts, feelings, and behaviors associated
with the Agreeableness domain as a whole. There is a part–whole relation
between the facets and the domain score. That tactic implies that the current
facets cover all the relevant issues in the domain; adding other forms of behavior
beyond those in the six facets (e.g., somatic complaints) or even other facets will
not improve the predictive power of Agreeableness. We think that is premature
closure.
It is not clear that this second exhaustiveness criticism applies with
comparable force to the domain level, because the Agreeableness domain is
clearly a hypothetical construct, implicating something unseen beyond the
empirical relations (Grimm & Widaman, 2012, p. 623). Facets may be better
characterized in more local, empirical terms as shorthand verbal labels for
objective phenomena or results. These are “intervening variables”
(MacCorquodale & Meehl, 1948), which are local manifestations limited by the
empirical results. Whether this distinction is valid, these are issues that need
discussion within the framework of Five Factor Theory and the Five Factor
Model.
References
Ahadi, S. A., & Rothbart, M. K. (1994). Temperament, development, and the Big Five. In
C. F. Halverson, Jr., G. A. Kohnstamm, & R. P. Martin (Eds.), The developing
structure of temperament and personality from infancy to adulthood (pp. 189–207).
Hillsdale, NJ: Erlbaum.
Akrami, N., & Ekehammar, B. (2006). Right-Wing authoritarianism and social
dominance orientation: Their roots in Big-Five personality factors and facets. Journal
of Individual Differences, 27, 117–126.
Allport, G. W. (1968). The historical background of modern social psychology. In G.
Lindsey & E. Aronson (Eds.), The handbook of social psychology (2nd ed., Vol. 1, pp.
1–80). Reading, MA: Addison-Wesley.
Allport, G. W., & Odbert, H. S. (1936). Trait names: A psycho-lexical study.
Psychological Monographs, 47, (1, Whole No. 211).
11, 150–166.
Ashton, M. C., Lee, K., Perugini, M., Szarota, P., de Vries, R. E., DiBlas, L., … De Raad,
B. (2004). A six-factor structure of personality-descriptive adjectives: Solutions from
psycholexical studies in seven languages. Journal of Personality and Social
Axelrod, S. R., Widiger, T. A., Trull, T. J., & Corbitt, E. M. (1997). Relations of Five-
Factor Model antagonism facets with personality disorder symptomatology. Journal of
Batson, C. D. (1991). The altruism question: Toward a social-psychological answer.
Batson, C. D., Duncan, B. D., Ackerman, P., Buckley, T., & Birch, K. (1981). Is empathic
emotion a source of altruistic motivation? Journal of Personality and Social
Batson, C. D., O’Quin, K., Fultz, J., Vanderplas, M., & Isen, A. M. (1983). Influence of
self-reported distress and empathy on egoistic versus altruistic motivation to help.
Begue, L., Beauvois, J-L., Courbet, D., Oberle, D., Lapage, J., & Duke, A. A. (2015).
Personality predicts obedience in a Milgram paradigm. Journal of Personality, 83,
299–306.
Bellah, C. G., Bellah, L. D., & Johnson, J. L. (2003). A look at dispositional vengefulness
from the Three and Five-Factor Models of personality. Individual Differences
Research, 1, 6–16.
Bienvenu, O. J., Samuels, J. F., Costa, P. T., Reti, I. M., Eaton, W. W., & Nestadt, G.
(2004). Anxiety and depressive disorders and the Five-Factor Model of personality: A
higher- and lower-order personality trait investigation in a community sample.
Depression and Anxiety, 20, 92–97.
Biesanz, J. C., West, S. G., & Millevoi, A. (2007). What do we learn about someone over
time? The relationship between length of acquaintance, consensus, and self-other
agreement in judgments of personality. Journal of Personality and Social Psychology,
92, 119–135.
Bilalic, M., McLeod, P., & Gobet, F. (2007). Personality profiles of young chess players.
Book, A. S., Anthony A., Volk, A. S., & Hosker, A. (2012). Adolescent bullying and
personality: An adaptive approach. Personality and Individual Differences, 52, 218–
223.
Brown, S. L., & Brown, R. M. (2006). Selective investment theory: Recasting the
functional significance of close relationships. Psychological Inquiry, 17, 1–29.
Carlo, G., Okun, M. A., Knight, G. P., & de Guzman, M. R. T. (2005). The interplay of
traits and motives on volunteering: Agreeableness, extraversion and prosocial value
motivation. Personality and Individual Differences, 38, 1293–1305.
Caspi, A., Block, J., Block, J. H., Kloop, B., Lynam, D., Moffitt, T. E., & Stouthamer-
Loeber, M. (1992). A “common language” version of the California Child Q-set for
personality assessment. Psychological Assessment, 4, 512–523.
Cattell, R. B. (1957) Personality & motivation: Structure & measurement. Yonkers on
Hudson, NY: World.
Christopher, A. N., Zabel, K. L., & Miller, D. E. (2013). Personality, authoritarianism,
social dominance, and ambivalent sexism: A mediational model. Individual Differences
Research, 11, 70–80.
Cole, P. M. (1986). Children’s spontaneous control of facial expression. Child
Costa, P. T., Jr., Herbst, J. H., McCrae, R. R., Samuels, J., & Ozer, D. J. (2002). The
replicability and utility of three personality types. European Journal of Personality, 16,
73–87.
Costa, P. T., Jr., Herbst, J. H., McCrae, R. R., & Siegler, I. C. (2000). Personality at
midlife: Stability, intrinsic maturation and response to life events. Assessment, 7, 365–
378.
Costa, P. T., Jr., & McCrae, R. R. (1988). Personality in adulthood: A six-year
longitudinal study of self-reports and spouse ratings on the NEO Personality Inventory.
Costa, P. T., Jr., & McCrae, R. R. (1992). Revised NEO personality inventory (NEO-PI-
R) and NEO Five-Factor Inventory (NEO-FFI) professional manual. Odessa, FL:
Costa, P. T., Jr., & McCrae, R. R. (1995). Domains and facets: Hierarchical personality
assessment using the revised NEO Personality Inventory. Journal of Personality
Costa, P. T., Jr., McCrae, R. R., & Dye, D. A. (1991). Facet scales for agreeableness and
conscientiousness: A revision of the NEO personality inventory. Personality and
Cumberland-Li, A., Eisenberg, N., & Reiser, M. (2004). Relations of young children’s
agreeableness and resiliency to effortful control and impulsivity. Social Development,
13, 193–212.
Davis, M. H. (1996). Empathy: A social psychological approach. Boulder, CO: Westview
Press.
Davis, M. H. (2015). Empathy and prosocial behaviour. In D. A. Schroeder & W. G.
Graziano (Eds.), Oxford handbook of prosocial behavior (pp. 282–306). Oxford,
England: Oxford University Press.
DeCuyper, M., De Pauw, S., De Fruyt, F., De Bolle, M., & De Clercq, B. J. (2009). A
meta-analysis of psychopathy-, antisocial PD- and FFM associations. European
De Fruyt, F., Bartels, M., Van Leeuwen, K. G., De Clercq, B., Decuuper, M., &
De Raad, B., Hendriks, A. A. J., & Hofstee, W. K. B. (1994). The Big Five: A tip of the
iceberg of individual differences. In C. F. Halverson, Jr., G. A. Kohnstamm, & R. P.
Martin (Eds.), The developing structure of temperament and personality from infancy
to adulthood (pp. 91–109). Hillsdale, NJ: Erlbaum.
De Raad, B., & Hofstee, W. K. B. (1993). A circumplex approach to the Five Factor
Model: A facet structure of trait adjectives supplemented by trait verbs. Personality
DeYoung, C., Peterson, J. B., & Higgins, D. M. (2002). Higher-order factors of the Big
Five predict conformity: Are there neuroses of health? Personality and Individual
DeYoung, C., Quilty, L. C., & Peterson, J. B. (2007). Between facets and domains: 10
Digman, J. M., & Takemoto-Chock, N. K. (1981). Factors in the natural language of
personality: Re-analysis, comparison, and interpretation of six major studies.
Multivariate Behavioral Research, 16, 149–170.
Dijker, A. J. M., & Koomen, W. (2007). Stigmatization, tolerance, and repair: An
integrative psychological analysis of responses to deviance. New York, NY:
Cambridge University Press.
Dobzhansky, T. (1955). Evolution, genetics, and man. New York, NY: Wiley.
Dovidio, J. F., Piliavin, J. A., Schroeder, D. A., & Penner, L. A. (2006). The social
psychology of prosocial behavior. Mahwah, NJ: Erlbaum.
Eastwick, P. W. (2009). Beyond the Pleistocene: Using phylogeny and constraint to
inform the evolutionary psychology of human mating. Psychological Bulletin, 135,
794–821.
Eisenberg, N., & Spinrad, T. L. (2004). Emotion-related regulation: Sharpening the
definition. Child Development, 75, 334–339.
Eisenberger, N. I., Lieberman, M. D., & Satpute, A. B. (2005). Personality from a
controlled processing perspective: An fMRI study of neuroticism, extraversion, and
self-consciousness. Cognitive, Affective, and Behavioral Neuroscience, 5, 169–181.
Ekehammar, B., & Akrami, N. (2007). Personality and prejudice: From Big Five
personality factors to facets. Journal of Personality, 75, 899–926.
Erdheim, J., Wang, M., & Zichar, M. J. (2006). Linking the Big Five personality
constructs to organizational commitment. Personality and Individual Differences, 41,
959–970.
Evans, D. E., & Rothbart, M. K. (2007). Developing a model of adult temperament.
Feigl, H. (1970). The “Orthodox” view of theories: Remarks in defense as well as
critique. In M. Radner & S. Winokur (Eds.), Minnesota studies in the philosophy of
science: Analyses of theories & methods of physics & psychology (pp. 3–16).
Minneapolis, MN: University of Minnesota Press.
Finch, J. F., Panter, A. T., & Caskie, G. I. L. (1999). Two approaches to identifying
personality dimensions across method. Journal of Personality, 67(3), 407–438.
Fisher, D. M., Bell, S. T., Dierdorff, E. C., &. Belohlav, J. A. (2012). Facet personality
and surface-level diversity as team mental model antecedents: Implications for implicit
coordination. Journal of Applied Psychology, 97, 825–841.
Furnham, A., Moutafi, J., & Chamorro-Premuzic, T. (2005). Personality and intelligence:
Gender, the Big Five, self-estimated and psychometric intelligence. International
Journal of Selection and Assessment, 13, 11–24.
Gillespie, W., & Myors, B. (2000). Personality of rock musicians. Psychology of Music,
28, 154–165.
Glinski, K., & Page, A. C. (2010). Modifiability of neuroticism, extraversion, and
agreeableness by group cognitive behaviour therapy for social anxiety disorder.
Behaviour Change, 27, 42–52.
Goldberg, L. (1990). An “alternative description of personality”: The Big Five factor
Goldberg, L. R. (1992). The development of markers of the Big Five factor structure.
Psychological Assessment, 4, 26–42.
Goldberg, L. R. (1999). A broad-bandwidth, public-domain, personality inventory
measuring the lower-level facets of several Five-Factor Models. In I. Mervielde, I.
& Gough, H. C. (2006). The International Personality Item Pool and the future of
public-domain personality measures. Journal of Research in Personality, 40, 84–96.
Gough, H. (1987). California Psychological Inventory administrator’s guide. Palo Alto,
CA: Consulting Psychologists Press.
Gow, A. J., Whiteman, M. C., Pattie, A., & Deary, A. J. (2005). Goldberg’s “IPIP” Big-
Five factor markers: Internal consistency and concurrent validation in Scotland.
Graziano, W. G., & Bruce, J. W. (2008). Attraction and the initiation of relationships: A
review of the empirical literature. In S. Sprecher, A. Wenzel, & J. Harvey (Eds.),
Handbook of relationship initiation (pp. 269–295). New York, NY: Psychology Press.
Graziano, W. G., Bruce, J. W., Sheese, B. E., & Tobin, R. M. (2007). Attraction,
personality, and prejudice: Liking none of the people most of the time. Journal of
Graziano, W. G., & Habashi, M. M. (2010). Motivational processes underlying both
prejudice and helping. Personality and Social Psychology Review, 14, 313–331.
Graziano, W. G., & Habashi, M. M. (2015). The quest for the prosocial personality. In D.
A. Schroeder & W. G. Graziano (Eds.), Oxford handbook of prosocial behavior.
Oxford, England: Oxford University Press.
Graziano, W. G., Habashi, M. M., Sheese, B. E., & Tobin, R. M. (2007). Agreeableness,
empathy, and helping: A person x situation perspective. Journal of Personality and
Graziano, W. G., Hair, E. C., & Finch, J. F. (1997). Competitiveness mediates the link
between personality and group performance. Journal of Personality and Social
Psychology, 73, 1394–1408.
Graziano, W. G., Jensen-Campbell, L. A., & Hair, E. C. (1996). Perceiving interpersonal
conflict and reacting to it: The case for agreeableness. Journal of Personality and
Graziano, W. G., Jensen-Campbell, L. A., Shebilske, L. J., & Lundgren, S. R. (1993).
Social influence, sex differences, and judgments of beauty: Putting the interpersonal
back in interpersonal attraction. Journal of Personality and Social Psychology, 65,
522–531.
Graziano, W. G., & Tobin, R. M. (2002). Agreeableness: Dimension of personality or
social desirability artifact? Journal of Personality, 70, 695–727.
Graziano, W. G., & Tobin, R. M. (2009). Agreeableness. In M. R. Leary & R. H. Hoyle
(Eds.), Handbook of individual differences in social behavior (pp. 46–61). New York,
NY: Guilford.
Graziano, W. G., & Tobin, R. M. (2013). The cognitive and motivational foundations
underlying agreeableness. In M. D. Robinson, E. R. Watkins, & E. Harmon-Jones
(Eds.), Handbook of cognition and emotion. (pp. 347–364). New York, NY: Guilford.
Griffiths, P. E., & Tabery, J. (2007). Behavioral genetics and development: Historical and
conceptual causes of controversy. New Ideas in Psychology, 26, 332–352.
Grimm, K. J., & Widaman, K. F. (2012). Construct validity. In H. Cooper (Editor-in-
chief), APA handbook of research methods in psychology: Vol. 1: Planning, measures
& psychometrics (pp. 621–642). Washington, DC: American Psychological
Association.
Gunderson, J. G., Shea, M. T., Skodal, A. E., McGlashan, T. H., Morey, L. C., Stout, R.
L., … Keller, M. B. (2000). The collaborative longitudinal personality disorders study:
Development, aims, design, and sample characteristics. Journal of Personality
Disorders, 14, 300–315.
Haas, B. W., Omura, K., Constable, R. T., & Canli, T. (2007). Is automatic emotion
regulation associated with agreeableness? A perspective using a social neuroscience
approach. Psychological Science, 18, 130–132.
Habashi, M. M., & Wegener, D. (2008). Preliminary evidence that agreeableness is more
closely related to responsiveness than conformity. Unpublished manuscript, Purdue
University, West Lafayette, IN.
Hair, E. C., & Graziano, W. G. (2003). Self-esteem, personality, and achievement in high
school: A prospective longitudinal study in Texas. Journal of Personality, 71, 971–994.
Hales, A., Kassner, M. P., Williams, K. D., & Graziano, W. G. (under review).
Disagreeableness as a cause and consequence of ostracism.
Heaven, P. C. L. (1996). Personality and self-reported delinquency: Analysis of the “Big
Five” personality dimensions. Personality and Individual Differences, 20, 47–54.
Hilbig, B. E., Zettler, I., Leist, F., & Heydasch, T. (2013). It takes two: Honesty–Humility
and Agreeableness differentially predict active versus reactive cooperation. Personality
Hofstee, W. K. B. (1994). Who should own the definition of personality? European
Hofstee, W. K. B., Ostendorf, F., & Boomsma, A. (1998). Trait structure: Abridged-
circumplex vs. hierarchical-structural conceptions. Groningen, The Netherlands:
University of Groningen.
Hogan, R. (1982). A socioanalytic theory of personality. In M. Page (Ed.), Nebraska
symposium on motivation (Vol. 30, pp. 55–89). Lincoln, NE: University of Nebraska
Press.
Hogan, R. (1986). Hogan Personality Inventory manual. Minneapolis, MN: National
Computer Systems, Inc.
Hogan, R., Hogan, J., & Roberts, B. W. (1996). Personality measurement and
employment decisions: Questions and answers. American Psychologist, 51, 469–477.
Hopwood, C. J., Morey, L. C., Skodol, A. E., Stout, R. L., Yen, S., Ansell, E. B., … &
McGlashan, T. H. (2007). Five-Factor Model personality traits associated with alcohol-
related diagnoses in a clinical sample. Journal of Studies on Alcohol and Drugs, 68,
455–460.
Jakobwitz, S., & Egan, V. (2006). The dark triad and normal personality. Personality and
Jefferson, T., Jr., Herbst, J. H., & McCrae, R. R. (1998). Associations between birth order
and personality traits: Evidence from self-reports and observer ratings. Journal of
Jensen-Campbell, L. A., Gleason, K. A., Adams, R., & Malcolm, K. T. (2003).
Interpersonal conflict, agreeableness, and personality development. Journal of
Jensen-Campbell, L. A., & Graziano, W. G. (2001). Agreeableness as a moderator of
interpersonal conflict. Journal of Personality, 69, 323–361.
Jensen-Campbell, L. A., & Graziano, W. G. (2005). Two faces of temptation: Differing
motives for self-control. Merrill-Palmer Quarterly, 51, 287–324.
Jensen-Campbell, L. A., & Malcolm, K. T. (2007). The importance of conscientiousness
in adolescent interpersonal relationships. Personality and Social Psychology Bulletin,
33, 368–383.
Jensen-Campbell, L. A., Rosselli, M., Workman, K. A., Santisi, M., Rios, J. D., & Bojan,
D. (2002). Agreeableness, conscientiousness and effortful control processes. Journal of
John, O. P. (1984). How shall a trait be called? A feature analysis of altruism. In A.
Angleitner, A. Furnham, & G. Van Heck (Eds.), Personality psychology in Europe: 1st
European conference on personality: Selected papers. Vol.1: Theoretical and empirical
developments (pp. 117–140). Lisse, The Netherlands: Swets & Zeitlinger.
John, O. P., & Robins, R. W. (1993). Gordon Allport: Father and critic of the Five-Factor
Model. In K. H. Craik, R. Hogan, & R. N. Wolfe (Eds.), Fifty years of personality
psychology (pp. 215–236). New York, NY: Plenum.
John, O. P., & Srivastava, S. (1999). The Big Five trait taxonomy: History, measurement,
and theoretical perspectives. In L. A. Pervin & O. P. John (Eds.), Handbook of
personality: Theory and Research (2nd ed., pp. 102–138). New York, NY: Guilford.
Jonas, E., McGregor, I., Klackl, J., Agroskin, D., Fritsche, I., Holbrook, C., … Quirin, M.
(2014). Threat and defense: From anxiety to approach. Advances in Experimental
Jones, S. E., Miller, J. D., & Lynam, D. R. (2011). Personality, antisocial behavior, and
aggression: A meta-analytic review. Journal of Criminal Justice, 39, 329–337.
Kassner, M. (2011). Personality moderators of the door-in-the-face compliance technique.
Unpublished master’s thesis. Purdue University, West Lafayette, IN.
Kausel, E. E., & Slaughter, J. E. (2011). Narrow personality traits and organizational
attraction: Evidence for the complementary hypothesis. Organizational Behavior and
Human Decision Processes, 114, 3–14
Kelley, H. H., Holmes, J. G., Kerr, N. L., Reis, H. T., Rusbult, C. E., & Van Lange, P. A.
M. (2003). An atlas of interpersonal situations. New York, NY: Cambridge University
Press.
Kern, M. L., Duckworth, A. L., Urzua, S. S., Loeber, R., Stouthamer-Loeber, M., &
Lynam, D. R. (2013). Do as you’re told! Facets of agreeableness and early adult
outcomes for inner-city boys. Journal of Research in Personality, 47, 795–799.
Kochanska, G., & Knaack, A. (2003). Effortful control as a personality characteristic of
young children: Antecedents, correlates, and consequences. Journal of Personality, 71,
1087–1112.
Kochanska, G., Murray, K., & Coy, K. C. (1997). Inhibitory control as a contributor to
conscience in childhood: From toddler to early school age. Child Development, 68,
263–277.
Koelsch, S., Enge, J., & Jentschke, S. (2012). Cardia signatures of personality. PLoS One,
7(2), e31441.
Kohnstamm, G., Halverson, C. F., Mervielde, I., & Havill, V. (1998). Parental
description of child personality: Developmental antecedents of the Big Five? Mahwah,
NJ: Erlbaum.
Krueger, R. F., Hicks, B. N., & McGue, M. (2001). Altruism and antisocial behavior:
Independent tendencies, unique personality correlates, distinct etiologies.
Laursen, B., Pulkkinen, L., & Adams, R. (2002). The antecedents and correlates of
agreeableness in adulthood. Developmental Psychology, 38(4), 591–603.
Lee, K., & Ashton, M. C. (2006). Further assessment of the HEXACO Personality
Inventory: Two new facet scales and an observer report form. Psychological
Lockenhoff, C. E., Terracciano, A., Patriciu, N. S., Eaton, W. W., & Costa, P. T., Jr.
(2009). Self-reported extremely adverse life events and longitudinal changes in Five-
Factor Model personality traits in an urban sample. Journal of Traumatic Stress, 22,
53–59.
Loehlin, J. C. (1992). Genes and environment in personality development. Newbury Park,
CA: Sage.
Luminet, O., Bagby, R. M., Wagner, H., Taylor, G. J., & Parker, J. D. A. (1999). Relation
between alexithymia and the Five-Factor Model of personality: A facet-level analysis.
MacCorquodale, K., & Meehl, P. E. (1948). On a distinction between hypothetical
constructs and intervening variables. Psychological Review, 55, 95–107.
MacDonald, K. B. (2008). Effortful control, explicit processing, and the regulation of
human evolved predispositions. Psychological Review, 115, 1012–1031.
MacLaren, V. V., Best, L. A., Dixon, M. J., & Harrigan, K. A. (2011). Problem gambling
and the Five Factor Model in university students. Personality and Individual
MacLaren, V., Ellery, M., & Knoll, T. (2015). Personality, gambling motives and
cognitive distortions in electronic gambling machine players. Personality and
Madsen, L., Parsons, S., & Grubin, D. (2006). The relationship between the Five-Factor
Model and DSM personality disorder in a sample of child molesters. Personality and
Matthews, G., & Deary, I. J. (1998). Personality traits. Cambridge, England: Cambridge
University Press.
McCrae, R. R., & Costa, P. T., Jr. (2003). Personality in adulthood (2nd ed.). New York,
NY: Guilford.
McCrae, R. R., & John, O. P. (1992). An introduction to the Five-Factor Model and its
applications. Journal of Personality, 60, 175–215.
Michalski, R. L., & Shackelford, T. K. (2002). An attempted replication of the
relationships between birth order and personality. Journal of Research in Personality,
36, 182–188.
Miller, J. D., Lynam, D., Zimmerman, R. S., Logan, T. K., Leukefeld, C., & Clayton, R.
(2004). The utility of the Five Factor Model in understanding risky sexual behavior.
Ode, S., & Robinson, M. D. (2008). Can agreeableness turn gray skies blue? A role for
agreeableness in moderating neuroticism-linked depression. Journal of Social and
Ode, S., Robinson, M. D., & Wilkowski, B. M. (2008). Can one’s temper be cooled? A
role for agreeableness in moderating Neuroticism’s influence on anger and aggression.
Otero-Lopez, J. M., & Pol, E. V. (2013). Compulsive buying and the Five Factor Model
of personality: A facet analysis. Personality and Individual Differences, 55, 585–590.
Parks, C. D. (2011). Personality influences on group processes: The past, present, and
future. In M. Snyder & K. Deaux (Eds.), Handbook of social and personality
psychology (pp. 517–544). New York, NY: Oxford University Press.
Paulhus, D. L. (2015). Toward a taxonomy of dark personalities. Current Directions in
Paulhus, D. L., & John, O. P. (1998). Egoistic and moralistic biases in self-perception:
The interplay of self-deceptive styles with basic traits and motives. Journal of
Paulhus, D. L., &. Jones, D. N. (2015). Measures of dark personalities. In G. J. Boyle, D.
H. Saklofske, & G. Matthews (Eds.), Measures of personality and social psychological
constructs (pp. 562–594). San Diego, CA: Elsevier Academic Press.
Pearce, P. L., & Amato, P. R. (1980). A taxonomy of helping: A multidimensional scaling
analysis. Social Psychology Quarterly, 43, 363–371.
Penner, L. A., Fritzsche, B. A., Craiger, J. P., & Freifeld, T. S. (1995). Measuring the
prosocial personality. In J. N. Butcher& C. D. Spielberger (Eds.), Advances in
personality assessment (Vol. 10, pp. 147–163). Hillsdale, NJ: Erlbaum.
Petty, R. E., & Wegener, D. T. (1999). The elaboration likelihood model: Current status
and controversies. In S. Chaiken & Y. Trope (Eds.), Dual-process theories in social
psychology (pp. 41–72). New York, NY: Guilford.
Piedmont, R. L., & Weinstein, H. P. (1993). A psychometric evaluation of the new NEO-
PIR facet scales for agreeableness and conscientiousness. Journal of Personality
Piedmont, R. L., & Weinstein, H. P. (1994). Predicting supervisor ratings of job
performance using the NEO Personality Inventory. The Journal of Psychology:
Interdisciplinary and Applied, 128, 255–265.
Porges, S. W., & Carter, C. S. (2012). Mechanisms, mediators, and adaptive
consequences of caregiving. In S. L. Brown, R. M. Brown, & L. A. Penner (Eds.),
Moving beyond self-interest: Perspectives from evolutionary biology, neuroscience,
and social science (pp. 53–74). New York, NY: Oxford University Press.
Pursell, G. R., Laursen, B., Rubin, K. H., Booth-LaForce, C., & Rose-Krasnor, L. (2008).
Gender differences in patterns of association between prosocial behavior, personality,
and externalizing behavior. Journal of Research in Personality, 42, 472–481.
Ricord, E. (1840). Elements of the philosophy of mind applied to the development of
thoughts and feelings (pp. 348–368). New York, NY: John N. Bogert.
Riffin, C., Lockenhoff, C. E., Pillemer, K., Friedman, B., & Costa, P. T., Jr. (2013). Care
recipient agreeableness is associated with caregiver subjective physical health status.
The Journals of Gerontology: Series B: Psychological Sciences and Social Sciences,
68B, 927–930.
Rojas, S. L., & Widiger, T. A. (2014). Convergent and discriminant validity of the Five
Factor Form. Assessment, 21, 143–157.
Rothbart, M. K., & Bates, J. E. (2006). Temperament. In N. Eisenberg, W. Damon, & R.
M. Lerner (Eds.), Handbook of child psychology: Vol. 3, Social, emotional, and
personality development (6th ed., pp. 99–166). Hoboken, NJ: John Wiley & Sons Inc.
Rothbart, M. K., & Posner, M. I. (1985). Temperament and the development of self-
regulation. In L. C. Hartlage & C. F. Telzrow (Eds.), The neuropsychology of
individual differences: A developmental perspective (pp. 93–123). New York, NY:
Plenum.
Rueda, M. R., Posner, M. I., & Rothbart, M. K. (2005). The development of executive
attention: Contributions to the emergence of self-regulation. Developmental
Neuropsychology, 28, 573–594.
Saarni, C. (1984). An observational study of children’s attempt to monitor their
expressive behavior. Child Development, 55, 1504–1513.
Salekin, R. T., Debus, S. A., & Barker, E. D. (2010). Adolescent psychopathy and the
Five Factor Model: Domain and facet analysis. Journal of Psychopathology and
Behavioral Assessment, 32, 501–514.
Salmon, W. C. (1984). Scientific explanation and the causal structure of the world.
Princeton, NJ: Princeton University Press.
Samuel, D. B., Mullins-Sweatt, S. N., & Widiger, T. A. (2013). An investigation of the
factor structure and convergent validity of the Five-Factor Model rating form.
Samuels, J., Eaton W. W., Bienvenu, O. J., Brown, C. H., Costa P. T., Jr., & Nestadt G.
(2002). Prevalence and correlates of personality disorders in the community. British
Journal of Psychiatry, 180, 536–542.
Savani, K., Morris, M. W., Naidu, N. V. R., Kumar, S., & Berlia, N. V. (2011). Cultural
conditioning: Understanding interpersonal accommodation in India and the United
States in terms of modal characteristics of interpersonal influence situations. Journal of
Scandell, D. J., Wlazelek, B. G., & Scandell, R. S. (1997). Personality of the therapist and
theoretical orientation. The Irish Journal of Psychology, 18, 413–418.
Schroeder, D. A., & Graziano, W. G. (2015). The field of prosocial behavior: An
introduction and overview. In D. A. Schroeder & W. G. Graziano (Eds.), Oxford
handbook of prosocial behavior (pp. 3–34). Oxford, England: Oxford University Press.
Seemann, E. A., Buboltz, W. C., Thomas, A., Soper, B., & Wilkinson, L. (2005). Normal
personality variables and their relationship to psychological reactance. Individual
Differences Research, 3, 88–98.
Shadish, W., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-
experimental designs for generalized causal inference. Boston, MA: Houghton Mifflin.
Shiner, R. L., Masten, A. S., & Roberts, J. M. (2003). Childhood personality foreshadows
adult personality and life outcomes two decades later. Journal of Personality, 71,
1145–1170.
Smits, D. J. M., & Boeck, P. D. (2006). From BIS/BAS to the Big Five. European
Sober, E., & Wilson, D. S. (1998). Unto others: The evolution and psychology of
unselfish behavior. Cambridge, MA: Harvard University Press.
Solomon, R. (1980). The opponent-process theory of acquired motivation: The costs of
pleasure and the benefits of pain. American Psychologist, 35, 691–712.
Solomon, R. L., & Corbit, J. D. (1974). An opponent-process theory of motivation: I.
Temporal dynamics of affect. Psychological Review, 57, 119–145.
Soto, C. J., & John, O. P. (2012). Development of Big Five domains and facets in
adulthood: Mean-level age trends and broadly versus narrowly acting mechanisms.
Journal of Personality, 80, 881–904
Stead, R., & Fekken, G. C. (2014). Agreeableness at the core of the dark triad of
personality. Individual Differences Research, 12(4-A), 131–141.
Stürmer, S., Benbow, A. E. F., Siem, B., Barth, M., Bodansky, A. N., & Lotz-Schmitt, K.
(2013). Psychological foundations of xenophilia: The role of major personality traits in
predicting favorable attitudes toward cross-cultural contact and exploration. Journal of
Thomas, A., Mittelman, M., Chess, S., Korn, S. J., & Cohen, J. (1982). A temperament
questionnaire for early adult life. Educational and Psychological Measurement, 42(2),
593–600.
Thomason, S. J., Weeks, M., Bernardin, H. J., & Kane, J. (2011). The differential focus of
supervisors and peers in evaluations of managerial potential. International Journal of
Selection and Assessment, 19, 82–97.
Timmerman, T. A. (2004). Relationships between NEO PI-R personality measures and
job performance ratings of inbound call center employees. Applied H.R.M. Research, 9,
35–38.
Tobin, R. M., & Graziano, W. G. (2011). The disappointing gift: Dispositional and
situational moderators of emotional expressions. Journal of Experimental Child
Tobin, R. M., Graziano, W. G., Vanman, E. J., & Tassinary, L. G. (2000). Personality,
emotional experience, and efforts to control emotions. Journal of Personality and
Trapnell, P. D., & Wiggins, J. S. (1990). Extension of the Inter-personal Adjective Scales
to include the Big Five dimensions of personality. Journal of Personality and Social
Van Iddekinge, C. H., McFarland, L. A., & Raymark, P. H. (2007). Antecedents of
impression management use and effectiveness in a structured interview. Journal of
Management, 33, 752–773.
Voller, E. K., & Long, P. J. (2010). Sexual assault and rape perpetration by college men:
The role of the Big Five personality traits. Journal of Interpersonal Violence, 25, 457–
480.
White, A. E., Kenrick, D. T., Li, Y. J., Mortensen, C. R., Neuberg, S. L., & Cohen, A. B.
(2012). When nasty breeds nice: Threats of violence amplify agreeableness at national,
individual, and situational levels. Journal of Personality and Social Psychology, 103,
622–634.
Wiggins, J. S. (1991). Agency and communion as conceptual coordinates for the
understanding and measurement of interpersonal behavior. In L. W. M. Gove & D.
Cicchetti (Eds.), Thinking clearly about psychology: Volume 2: Personality and
psychopathology (pp. 89–113). Minneapolis, MN: University of Minnesota Press.
Williams, P. G., Suchy, Y., & Kraybill, M. L. (2010). Five-Factor Model personality
traits and executive functioning among older adults. Journal of Research in
Wonderlic, E. F. (1992). Manual of the Wonderlic Personnel Test & Scholastic Level
Exam II: Wonderlic Personnel Test, Inc.
Xia, J., & Habashi, M. M. (2015). Personality and attitude change: The effect of
agreeableness on information processing. Annual Meeting of the Society for
Personality & Social Psychology, February 28, 2015, Long Beach, CA.
Conscientiousness
Joshua J. Jackson and Brent W. Roberts
Abstract
Conscientiousness refers to a broad swath of constructs that reflects the
propensity to be self-controlled, responsible to others, hardworking, orderly, and
rule abiding. To understand why conscientiousness is one of the best
psychological predictors of important outcomes (e.g., longevity; divorce), this
chapter provides a broad overview of the trait. First, the Sociogenomic model of
personality traits is briefly described as a means to provide a common language to
discuss the status of conscientiousness. Next, the hierarchical structure of
conscientiousness is described, including a description of common measures used
to assess conscientiousness, as well as constructs related to conscientiousness.
The development of conscientiousness is then discussed, followed by a review of
the predictive ability of conscientiousness. The potential mechanisms driving the
development of conscientiousness and the pathways that relate conscientiousness
to important outcomes are also examined.
Key Words: Big Five, personality traits, conscientiousness, facets, personality
assessment, predictive validity, personality development, self-control, grit
The personality trait of conscientiousness reflects a family of constructs that

describes individual differences in the propensity to be self-controlled,
responsible to others, hardworking, orderly, and rule abiding (Roberts, Jackson,
Fayard, Edmonds, & Meints, 2009). Although typically thought of as part of the
Big Five, constructs resembling conscientiousness are included in almost all of
the classical personality taxonomies, such as those put forth by Cattell (1957),
Eysenck (1947), Gough, (1956), D. Jackson (1976), Block and Block (1980),
and Tellegen (1982). The importance of conscientiousness as a psychological
construct is indisputable when considering the predictive ability of life’s
important outcomes, such as health and longevity (Jackson, Connolly, Levine, &
Garrison, 2015; Moffitt et al., 2011), occupational success (Dudley, Orvis,
Lebiecki, & Cortina, 2006), marital stability (Solomon & Jackson, 2014a),
academic achievement (Noftle & Robins, 2007) and even wealth (Mroczek,
Turiano, Chan, Hill, & Roberts, 2015). As a result, conscientiousness is used in
diverse fields such as economics (e.g., Almlund, Duckworth, Heckman, &
Kautz, 2011), political science (Gerber et al., 2011), and public policy (Tough,
2013).
Despite the historical and practical significance of conscientiousness, much is
unknown about the trait. As seen in the definition, conscientiousness is not a
singular dimension but instead encompasses a broad range of lower-order
individual differences. The Big Five are indeed “Big,” and as a result, there
exists confusion over what falls under the broad umbrella of conscientiousness
and how to effectively measure it. Confusion over the status of
conscientiousness is problematic as quality measurement of the trait is necessary
to help us understand the relationship between conscientiousness and important
life outcomes. Similar difficulties arise when considering the genetic and
developmental etiology of conscientiousness where accurate assessment of
conscientiousness is necessary to describe how it develops across the lifespan
and to pinpoint its genetic and biological bases. This chapter covers the
conceptual standing of conscientiousness as a personality trait, reviewing the
measurement and the nomological network in which conscientiousness is
embedded, after which the development of conscientiousness and its predictive
utility are discussed.
Conscientiousness as a Personality Trait

Conscientiousness is most often thought of as a personality trait, which
reflects the relatively enduring, automatic patterns of thoughts, feelings, and
behaviors that differentiate people from one another, and are elicited in trait-
evoking situations (Roberts, 2009; Roberts & Jackson, 2008).
Conscientiousness, like all personality traits, can be conceptualized as existing
within a hierarchical structure, in which broad constructs are structured above
more specific lower-order facets, with behavioral, cognitive, and affective
manifestations of these traits falling at even lower ends of the hierarchy (Jackson
et al., 2010; Markon, Krueger, & Watson, 2005; Roberts, Bogg, Walton,
Chernyshenko, & Stark, 2004). Recently, the authors of this chapter put forth a
model that capitalizes on this hierarchical structure of personality in the hopes of
bringing together diverse ways of thinking about personality traits. Described
herein is the Sociogenomic model of personality traits (Roberts, 2009; Roberts &
Jackson, 2008), which will help in better understanding the measurement,
development, and predictive power of conscientiousness.
The Sociogenomic model describes two different levels of analyses for
personality. The first level is a broad personality trait, which reflects the
relatively enduring manifestations typical to a person that are embedded within
neurophysiological structures as a way to provide consistency across time
(Jackson, Hill, & Roberts, 2012). This is the level in which conscientiousness is
typically discussed and measured. The second level is the state-level, which
reflects the moment-to-moment fluctuations in the manifestations of personality
(Fleeson, 2001). These state-level manifestations are made up of
thoughts/cognitions, feelings/emotions, and behavior, which are the result of
both an environmental stimulus (e.g., one is more likely to talk when at a party)
and personality traits (e.g., extraverts are more likely to talk more than
introverts). It is at this level that a personality trait becomes “visible,” and exists
in a way others and oneself can perceive. Specifically, it is at the state-level at
which the assessment and inference of a personality trait occur. Importantly, the
manifestation of a personality trait includes thoughts/cognitions and
feelings/emotions in addition to behavior, two important states that sometimes
are left out of definitions of personality traits (Bandura, 2012; Jackson et al.,
2012). A corollary of this part of the model is that a thorough assessment of a
personality trait needs to include all three state-level components.
A common misperception of personality traits is that they denote the existence
of highly heritable, and thus unchangeable and decontextualized, constructs. The
Sociogenomic model easily addresses questions concerning how context
influences coherence across situations and time by integrating two popular views
of personality that take opposite positions concerning the mutability of
personality (Roberts, 2009). Traditional trait models, such as the Five Factor
Model (FFM; McCrae & Costa, 2008), emphasize stability across time and
context, whereas social cognitive models of personality, such as the Cognitive-
Affective Processing system (CAPs; Mischel & Shoda, 2008), emphasize change
across situations and time. According to the Sociogenomic model, these two
positions are not in conflict, but simply reflect a focus on the two different levels
of analysis common to personality models. The FFM focuses on a broad trait
level that exists across multiple contexts and time, whereas CAPs focuses on the
state-level manifestations and emphasize the differences between states across
time or situations. When a broad level of analysis such as the FFM is
emphasized there is a natural tendency to focus on the stability of personality
given that the traits are inferred from a multitude of states across different
contexts. In contrast, CAPs focuses on the state variations that by definition are
more attuned to environmental influences, and thus vary more across contexts
and time. Together both the FFM and CAPs describe the same system, they just
emphasize different components of the system, much like in the proverb of the
blind men (from Indostan) describing different features of the same elephant.
When describing the development of conscientiousness it is necessary to take
into account the existence of both a relatively stable trait, as well as
behaviors/thoughts/feelings that reflect the existence of a trait, but vary due to
contextual factors. To repeat a sentiment echoed for decades: behaving
unconscientiously does not invalidate the possibility that a person is highly
conscientious (Jackson, Hill, & Roberts, 2011). And, of course, as many parents
will note, one act of probity does not make for a conscientious child.
Part of the disagreement between various models and conceptualizations of
personality comes from the disagreements over the contextualized and/or
decontextualized nature of personality traits. Again, the different level of
analysis helps clarify this issue. An important contribution of the Sociogenomic
model is that the environment is unlikely to directly influence personality traits
(Roberts & Jackson, 2008). Rather, the environment can modify state-levels
directly (e.g., being at work is associated with more conscientious behaviors).
Personality traits are inferred from repeated state-level manifestations, each of
which necessarily has a context that it manifests within. Thus at the level of
broad traits, the trait-eliciting contexts are multifaceted because they necessarily
aggregate state-level experiences that range across the broader trait spectrum.
For example, self-control necessitates the presence of something tempting
(Tsukayama, Duckworth, & Kim, 2012); without the context it does not make
conceptual sense to describe (or infer) someone as being self-controlled. To infer
a trait level of self-control requires evidence about self-control states in a
number of contexts across time. In this sense, the broad trait is contextualized in
that each manifestation of the trait is necessarily embedded within a context. The
broad trait domain collapses across the many different contexts, resulting in what
can be thought of as a trait embedded within a generalized context, meaning that
the trait is not context-specific but extends to the many different domains
wherein it will most likely manifest. In the case of self-control, the dimension is
always contextualized as it cannot be inferred in the absence of a tempting
offering, such as food, spending, smoking, or sex. As a result, personality traits
can be thought of as being able to manifest in a number of different contexts,
though not so many as to suggest that context is separate from personality traits.
From the discussion of the Sociogenomic model it is possible to draw three
inferences that will guide the discussion of the structure, development, and
predictive power of conscientiousness. First, conscientiousness describes both a
higher-order, broad construct as well as more fine-grained, state-level thoughts,
feelings, and behaviors. As a result, a systematic description of
conscientiousness requires the successful integration of these two extremes.
Second, the trait of conscientiousness is relatively stable across time but the
manifestations may change across situations and time. Identifying the contexts
that can shift the state-level experiences may help us understand the mechanisms
responsible for the development of conscientiousness, given that the
environment cannot directly influence trait levels. Third, measures of
conscientiousness are necessarily contextualized, which means that contexts can
be used to help understand the mechanisms by which personality traits influence
important life outcomes.
The Hierarchical Structure of Conscientiousness

The two-level of division in the Sociogenomic model can be considered a
slight oversimplification. For example, there will be intermediary levels of
specificity, such as those found at the facet level of conscientiousness. As per the
definition, conscientiousness refers to many different and distinct components,
which can be ordered hierarchically from broad to narrow. The narrower traits
are referred to as facets, with each having a smaller set of state manifestations
and contexts. These facets can be broken down even further into the state-level
component of behaviors, thoughts, and feelings associated with
conscientiousness. Although a number of studies have begun work at identifying
the lower-order facet structure of conscientiousness, the best way to understand
the lower-order structure is not to identify a specific level of analyses as is
typically done, but rather to work down the level of the hierarchy from the broad
trait to state-level manifestations.
Moving from the broad trait of conscientiousness to a basic two-factor
structure, conscientiousness consists of a proactive and an inhibitory component
(Costa, McCrae, & Dye, 1991; Jackson et al., 2010; Roberts, Chernyshenko,
Stark, & Goldberg, 2005). The proactive component refers to being goal oriented
and striving to do well in your endeavors, while the inhibitory component refers
to being responsible, to delaying gratification, and to controlling impulses.
Together, these two overarching components provide a structure under which the
rest of the facets can be organized. It should be noted that an alternative
perspective finds that the two-facet structure of conscientiousness consists of
industriousness and orderliness (DeYoung, Quilty, & Peterson, 2007). Although
this two-factor solution may initially seem similar to the proactive and inhibitory
components of conscientiousness, empirical examinations of the lower-order
structure of conscientiousness shows that industriousness and orderliness fall in
the proactive component (Roberts et al., 2005).
As we move down the conscientiousness hierarchy, facets emerge that reflect
coherent factors found in many extant facet measures. Three factor solutions
typically consist of achievement, self-control, and responsibility (Jackson et al.,
2010; Roberts et al., 2005), though few studies have explicitly tested three-factor
solutions. Numerous studies find four-factor solutions consisting of orderliness,
industriousness, responsibility, and self-control (Jackson et al., 2010; Peabody &
De Raad, 2002; Perugini & Gallucci, 1997; Roberts, Lejuez, Krueger, Richards,
& Hill, 2014), indicating that a four-facet description is the minimum number of
facets needed to accurately describe the breadth of conscientiousness.
Two facets of the four-factor solution, orderliness and industriousness, fall
under the proactive domain of conscientiousness. Individuals high in orderliness
like to be neat and clean, and situate their belongings in an organized fashion
(Jackson et al., 2009). To a lesser extent, orderliness also includes a proclivity
toward making and adhering to plans, such as utilizing to-do lists and planners
that aide in a more organized and orderly day. The facet of industriousness
reflects how hard-working you are, as well as your ability to persevere in the
face of challenges and setbacks. Industrious individuals like to finish what they
start, going so far as to work long hours or even to take on extra work in order to
do a thorough job (Jackson et al., 2010). Individuals low in industriousness can
be considered more economical in the way they apply themselves, and may be
more likely to avoid work, procrastinate, and give up when facing a challenge.
The facet of responsibility has both inhibitory and proactive components and
describes a tendency to uphold obligations and follow rules. Those high in
responsibility are less likely to break promises, miss appointments, or oversleep
whereas those low in responsibility may be described as unreliable or flighty and
would be more likely to bend or break rules (Jackson et al., 2010).
Responsibility tends to be strongly associated with the trait of agreeableness,
likely due to the fact that many obligations and responsibilities manifest in
interpersonal contexts. In this respect, responsibility can be seen as an
“interstitial” trait that lies equidistant between conscientiousness and
agreeableness.
Self-control reflects the inhibitory side of conscientiousness and refers to the
ability of a person to inhibit impulses, to have forethought, and to avoid being
reckless or out of control. Individuals high in self-control are able to put off
immediate gratification in order to attain longer-term goals. For example, those
high in self-control are less likely to make impulsive purchases, to put off or
cancel plans at the last minute, and burst out in anger when feeling frustrated
(Jackson et al., 2010).
Additional facets of conscientiousness beyond these four also appear in the
literature, depending on the initial item pool and the number of factors extracted
(Roberts et al., 2014). Three additional facets of traditionality, formality, and
punctuality are of particular note as they replicate across two or more studies and
reflect domains mostly distinct from the four-factor solution (MacCann,
Duckworth, & Roberts, 2009; Perugini & Gallucci, 1997; Roberts et al., 2004,
2005). Traditionality is the inclination to support and follow the norms of society
in order to maintain good social environments. Individuals high in traditionality
are more likely to uphold family and cultural traditions, whereas individuals low
in traditionality can be described as nonconforming or unconventional (Roberts
et al., 2004). Traditionality also extends to having more conventional or
conservative beliefs, particularly within the political and spiritual spectrum, and
is closely related to low openness to experience (Johnson, 1994). Formality is
associated with the tendency to have a sense of propriety, including being polite
and proper, and keeping your physical appearance neat and clean. Those high in
formalness may call others by formal titles, say please and thank you, and are
less likely to swear (Jackson et al., 2010). Punctuality describes the practice of
doing things in a timely manner and could be considered the narrowest facet of
conscientiousness in terms of concrete behaviors. Interestingly, punctuality is
equally related to each of the other facets of conscientiousness, unlike the other
facets of conscientiousness, and thus serves as a useful marker for overall levels
of conscientiousness (Jackson et al., 2010). That is to say, being punctual
appears important when considering your ability to plan (orderliness), work hard
to get somewhere (industriousness), avoid temptations that might lead to being
late (self-control), and care enough to meet other people on time (responsibility).
Further down the hierarchy five, six, eight, and even nine components have
been proposed (Costa & McCrae, 1992; Jackson et al., 2009; MacCann et al.,
2009; Peabody & De Raad, 2002; Roberts et al., 2004, 2005; Saucier &
Ostendorf, 1999). Part of the difficulty in interpreting these factor structures is
that the names used for each facet differ across measures; one researcher’s
traditionality is another’s conventionality. Moreover, beyond four factors the
structure does not replicate well, most likely because the initial item pools vary
in breadth of content. Despite these differences there has been more work on the
facet structure of conscientiousness than on any other Big Five trait.
Moving to narrower levels of the hierarchy leads to the state-level experiences
associated with conscientiousness. Typical measures of conscientiousness
collapse items that assess the different state-level manifestations of thoughts,
feelings, and behaviors (Pytlik Zillig, Hemenover, & Dienstbier, 2002; Werner
& Pervin, 1986). It is possible that two people score the same on a broad
measure of conscientiousness but manifest it in different ways depending on the
relative contributions of the facets of conscientiousness and their respective
thoughts, feelings, and behaviors. Moreover, the pathways and associations
between conscientiousness and important life outcomes may vary by
manifestation (Roberts, Jackson, Burger, & Trautwein, 2009). As a result, it can
be informative to investigate the structure of conscientiousness states separately
from the other manifestations. Is there more breadth in the types of conscientious
behavior compared to conscientious thoughts? Or does the structure of
conscientious behaviors mirror the structure of conscientious questionnaires?
Thus far, only the hierarchical structure of conscientious behaviors has been
investigated, with the results largely mirroring the structure found in
questionnaires (Jackson et al., 2010). However, some important differences
emerge, such that laziness overtakes industriousness as a major factor. This is
partially due to industriousness being contextualized within work, whereas
behaviors associated with laziness may exist across multiple domains. It is
possible that existing conscientiousness measures do not adequately assess a
wide-enough set of contexts or that they oversample personality states within the
context of a particular job.
As seen with laziness, the contextual aspect of conscientiousness begins to
emerge more clearly at this level of analysis. For example, orderliness manifests
itself at home and at the workplace through cleanliness behaviors, whereas
responsible behaviors typically include other persons. Understanding the
contexts associated with conscientious behaviors is theoretically important
because it helps identify the experiences that lend themselves to the expression
of conscientiousness. Understanding the experiences that matter for
conscientiousness can help inform a number of important areas of inquiry:
person–environment transaction, if–then contingencies, processes of personality
development, and the causal factors between personality and important life
outcomes.
Measures of Conscientiousness
Many measures of conscientiousness exist, ranging from broad omnibus
inventories that include facets to brief assessments of a few items designed to
assess the broad trait (Mike, Harris, Roberts, & Jackson, 2015). Although the
choice of which measure depends upon the researcher’s needs and time
constraints, we should emphasize the gains in specificity and reliability when
using broad measures (see also the chapter by L. Simms, Williams, and E.
Simms). As seen below, some existing measures of conscientiousness do not
assess all facets of conscientiousness or weight them equally; the unfortunate
result is that measures of conscientiousness cannot be adequately compared with
one another. Adding to the confusion is that many omnibus measures that
include facet measures have names that are specific to the scale, making
comparisons across measures difficult.
Likely, the most common measure of conscientiousness is the 48-item NEO
Personality Inventory-Revised (NEO PI-R) and the shorter 12-item version
within the NEO-Five Factor Inventory (NEO-FFI; Costa & McCrae, 1992). The
NEO PI-R measures six facets: Competence (closely related to the facet of
industriousness), Order (orderliness), Dutifulness (responsibility), Achievement
Striving (industriousness), Self-Discipline (orderliness), and Deliberation (self-
control). The NEO-FFI was not designed to assess facets, but broadly captures
the facets of orderliness, industriousness, and responsibility (Chapman, 2007;
Saucier, 1998).
The facets of conscientiousness found within the NEO PI-R and NEO-FFI
tend to focus on industriousness and orderliness (Roberts et al., 2005), and thus
may not necessarily reflect the best assessment in terms of breadth. A number of
other measures describe various facet-level conceptualizations of
conscientiousness, each of which offers some advantages. The Abridged Big
Five Circumplex (AB5C; Hofstee, de Raad, & Goldberg, 1992) includes nine
facets of conscientiousness but, unfortunately, does not adequately cover all the
facets of conscientiousness discussed above (Roberts et al., 2005). Instead, the
facets of the AB5C mainly assess industriousness and orderliness with a single
facet measure designed to tap self-control. In contrast, the Chernyshenko
Conscientiousness Scales (CCS) consists of six facets with 10 items per facet
(Hill & Roberts, 2011). Five of the six facets assess previously described facets
(orderliness, self-control, responsibility, traditionalism, and industriousness),
making it one of the only measures that adequately taps the major facets of
conscientiousness. The sixth facet, virtue, assesses the tendency to be
responsible members of the community by being honorable and rule abiding. It
shares some overlap with the responsibility facet but is broader in scope and
concerns issues of right and wrong rather than the tendency to follow through
with personal obligations. Another quality facet measure is the 20-item measure
from the HEXACO-Personality Inventory (Lee & Ashton, 2008). In this scale,
conscientiousness is measured by assessing four facets: Organization
(orderliness), Diligence (self-control/industriousness), Perfectionism
(orderliness/industriousness), and Prudence (self-control). Finally, the
Conscientiousness Adjective Checklist (CAC; Jackson et al., 2009) measures
most of the identified facets of conscientiousness, including Orderliness,
Industriousness, Self-Control, Responsibility, Conventionality, and Decisiveness
(Roberts et al., 2004).
Many options also exist for medium to brief measures of conscientiousness,
where there is less need for a facet level of analysis. The most common is the
nine items that assess conscientiousness using the Big Five Inventory (BFI;
John, Donahue, & Kentle, 1991), which appears to mainly assess the self-control
and order components of conscientiousness (Soto & John, 2009). Other medium-
length options include 10- and 20-item International Personality Item Pool
conscientiousness measures (Goldberg et al., 2006), as well as the 8-item
minimarker scale (Saucier, 1994) and the 6-item Five Factor Model Rating Form
(Mullins-Sweatt, Jamerson, Samuel, Olson, & Widiger, 2006). In terms of short
measures, the most popular is the Ten-Item Personality Inventory (TIPI),
consisting of just two items (Gosling, Rentfrow, & Swann, 2003). Though
longer measures are generally preferable (Credé, Harms, Niehorster, & Gaye-
Valentine, 2012), short measures such as the TIPI do provide an adequate
assessment of conscientiousness if you are under extreme time constraints.
Conscientiousness by Another Name

Numerous constructs in related fields are associated with the broad trait and
facets of conscientiousness. Many of these constructs have rich literatures that
could inform the measurement, development, and predictive ability of
conscientiousness. However, as is typical with the narrow focus of academia,
these literatures mostly remain separate from one another with some researchers
suggesting that constructs related to conscientiousness are meaningfully different
in terms of malleability or causal status (e.g., Bogg & Roberts, 2013). Many of
these variables though should be considered related to or even synonymous with
conscientiousness.
Impulsivity (or impulse control) is strongly related to the facet of self-control,
and thus should be considered a facet of conscientiousness, though it is not
always viewed this way. One reason that impulsivity is not seen as a component
of conscientiousness is that many measures of conscientiousness neglect the self-
control facet and instead overly weight the orderliness and industriousness facets
(Roberts et al., 2005). Another reason for this confusion is the multidimensional
nature of impulsivity, such that impulsivity is often linked with the Big Five
traits of extraversion and neuroticism in addition to conscientiousness
(DeYoung, 2011; Whiteside & Lynam, 2001). The relationship with multiple
traits stems from the plurality of reasons we may behave impulsively. One
possibility is through sensation seeking, reflecting the approach component of
extraversion (Revelle, 1997; Zuckerman, Eysenck, & Eysenck, 1978). Impulsive
behavior may also manifest as neuroticism, in which impulsive actions are likely
to occur because of emotional instability (Costa & McCrae, 1992). However,
continuing evidence indicates that most measures of impulsivity are primarily
related to conscientiousness (Paunonen, 1998; Roberts et al., 2004). Going by
the four-factor structure of impulsivity—consisting of sensation seeking, felt-
urgency, premeditation, and perseverance—it appears that the latter two most
closely align with conscientiousness whereas the first two tap extraversion and
neuroticism, respectively (Whiteside & Lynam, 2001).
A closely related construct to impulse control is constraint, which is typically
measured by the Multidimensional Personality Questionnaire (Tellegen, 1982).
Temperament researchers have defined constraint as individual differences in the
tendency toward planfulness and a focus on the future rather than on risk taking
and recklessness (Clark & Watson, 2008). Constraint is associated with most
facets of conscientiousness (Gaughan, Miller, Pryor, & Lynam, 2009), indicating
that constraint can be viewed as a slightly broader version of conscientiousness
that may also include aspects of agreeableness (Markon et al., 2005).
The self-control facet of conscientiousness is often used interchangeably with
impulsivity and constraint, though it is necessary to be careful as a slightly
broader definition of self-control is used within the self and identity field. Here
self-control is defined as “the capacity for altering one’s own responses,
especially to bring them into line with standards such as ideals, values, morals,
and social expectations, and to support the pursuit of long-term goals”
(Baumeister, Vohs, & Tice, 2007, p. 351). In addition to items that tap
impulsivity, this measure of self-control (Tangney, Baumeister, & Boone, 2004)
also includes items that assess facets of industriousness, responsibility, and
punctuality. As a result, it can be conceptualized as a broad measure of
conscientiousness, though typical studies using this measure distance themselves
from the Big Five literature and perceived baggage that comes with a personality
trait model.
The more recently developed construct of grit has also gained attention due to
the ability to predict school achievement (Duckworth, Peterson, Matthews, &
Kelly, 2007; Tough, 2013). Grit is defined as the ability to sustain interest and
effort toward long-term goals and thus is conceptually similar to the facet of
perseverance, industriousness and self-control. Given that conscientious
individuals tend to strive toward goals and achievements (Roberts, Jackson,
Fayard, et al., 2009), it is not surprising that an extremely high correlation
between grit and conscientiousness exists, nearing maximum levels possible
based on reliability (Duckworth & Quinn, 2009).
Another developmental construct related to conscientiousness is effortful
control, which refers to the ability to inhibit a dominant response to preform a
subdominant response, to detect errors, and to engage in planning (Kochanska,
Murray, & Harlan, 2000; Rothbart & Rueda, 2005). Importantly, effortful
control is thought of as an intentional process aimed at regulating goal-directed
processes (Eisenberg, Smith, Sadovsky, & Spinrad, 2004). As a result, effortful
control should be related to the inhibitive components of the two-factor structure
of conscientiousness. Similarly, others have viewed effortful control in the
context of two-mode models, with effortful control indexing the reflective
control aspect of self-control, in contrast to the more reactive quick mode that
drives impulses (Carver, Johnson, & Joormann, 2008). In line with these ideas,
various measures of effortful control are associated with measures of self-control
and conscientiousness (Evans & Rothbart, 2007; MacDonald, 2008). Although
effortful control is often discussed alongside the construct of executive
functioning, it should be noted that most tasks that assess executive functioning
are not strongly associated with conscientiousness broadly construed
(Duckworth & Kern, 2011; Edmonds, Bogg, & Roberts, 2009).
Self-regulation, in contrast, is broader than effortful control and can be
thought of as including regulation of both emotion and behaviors. Similar to
effortful control, behavioral self-regulation is defined as being able to control
your thoughts, feelings, and behaviors in an effort to act in goal-directed ways
(e.g., Hoyle, 2010). In contrast to conscientiousness, most research on self-
regulation focuses on the state levels of the Sociogenomic model. Accordingly,
self-regulation relates closely to the state-level manifestations of the facets of
self-control and industriousness. One interesting aspect of the self-regulation
literature is the tendency to relate the behaviors to motivation, a label often used
to describe conscientiousness. For example, pursuing personal goals and giving
up on unattainable goals are associated with adaptive self-regulation (Wrosch,
Scheier, Miller, Schulz, & Carver, 2003).
Delay of gratification is another term that is definitionally similar to
components of conscientiousness, notably self-control (e.g., Mischel, Shoda, &
Rodriguez, 1992). A number of studies have demonstrated that children differ in
their ability to refrain from immediate gratification in order to receive a more
desirable outcome in the future (Mischel et al., 1992). A recent study finds that
the marshmallow task, the quintessential measure of delay of gratification,
correlates strongly with conscientiousness and influences life outcomes
primarily through indexing the facet of self-control (Duckworth, Tsukayama, &
Kirby, 2013).
Development of Conscientiousness over the Lifespan

Conscientiousness, like all personality traits, is a developmental construct that
is relatively consistent across the lifespan, but also undergoes change. Part of the
apparent contradiction between consistency and change stems from the multiple
ways to define change and stability (Caspi, Roberts, & Shiner, 2005). In terms of
rank order consistency, assessed by the correlation between two assessments, the
consistency of conscientiousness increases steadily with age until plateauing
between the ages of 50 and 70 years (Roberts & DelVecchio, 2000), and then
declining in old age (Lucas & Donnellan, 2011). This means that people who are
above average on conscientiousness as a child will likely be above average on
conscientiousness as an adult.
In contrast to the relative rank orderings between people, mean level, or
absolute changes may also occur. Across a number of cross-sectional and
longitudinal studies conscientiousness tends to increase from young adulthood
up to the age of 60 years (Allemand, Zimprich, & Hendriks, 2008; Roberts,
Jackson, Burger, et al., 2009; Soto & John, 2012; Terracciano, McCrae, Brant, &
Costa, 2005). This normative pattern of development replicates across a number
of different cultures and countries, including Germany, Italy, Portugal, Croatia,
and South Korea (McCrae et al., 1999), as well as Vietnam (Walton et al., 2013).
Out of all of the Big Five traits, conscientiousness is the trait that evidences the
largest absolute increase across the lifespan.
Despite the general pattern to increase in conscientiousness across the
lifespan, the rate and direction of change differ from decade to decade. In late
childhood and adolescence, conscientiousness declines during puberty and the
teenage years before beginning to increase in young adulthood (Soto, John,
Gosling, & Potter, 2011). Most of the increases in conscientiousness occur
during young adulthood with the increases subsequently tapering off during
middle age (Roberts, Walton, & Viechtbauer, 2006). The pattern of change shifts
during older adulthood, to declines in conscientiousness, perhaps due to the
onset of retirement or a byproduct of decreased health (Lucas & Donnellan,
2011; Wortman, Lucas, & Donnellan, 2012).
Viewing conscientiousness at a facet level reveals a more nuanced
developmental story. Self-control and responsibility increase in young adulthood
and continue to increase into old age, mirroring the trends found at the broad
trait level (Jackson et al., 2009; Soto et al., 2011). Industriousness tends to
change the most during young adulthood, with some evidence of increases
continuing to occur throughout the lifespan (Jackson et al., 2009; Soto & John,
2012; Terracciano et al., 2005). Two facets prove unique. Conventionality does
not change from young adulthood to middle adulthood, but appears to change
during older adulthood (Jackson et al., 2009). Orderliness, on the other hand,
undergoes almost no changes across the lifespan compared to the other facets
(Jackson et al., 2009; Soto et al., 2011; Soto & John, 2012). Overall, increases in
conscientiousness earlier in life can be thought of as mainly resulting from
increases in impulse control, responsibility, and industriousness, whereas
increases in later adulthood are driven by changes in impulse control, reliability,
and conventionality.
Despite these mean level patterns, a significant amount of people do not
change at all on conscientiousness, or perhaps even decrease (Lüdtke, Roberts,
Trautwein, & Nagy, 2011). These individual differences in change indicate that
the environment plays a role in personality development (Roberts, Wood, &
Caspi, 2008). Again, the Sociogenomic model can help understand how the
environment can “get under the skin” and influence personality trait
development. Environments typically influence the state-level manifestations of
personality—the behaviors, thoughts, and feelings of personality—rather than
directly influence trait levels. However, if the environment presses for long-term
shifts in states related to personality, it may lead to changes to personality traits
in a bottom-up fashion (Roberts, 2009; Roberts & Jackson, 2008). For example,
being invested in your performance at school likely results in expectations to
behave and potential punishments if you do not achieve at high levels. These
expectations lead to experiences that shape and change your daily state-level
manifestations. Over time, success at school—due to following through with
expectations such as going to class or doing homework—would lead to shifts in
how you view yourself. These views may be internalized and translate into other
domains, not just the school context. Furthermore, the skills accrued within one
domain would likely translate to other domains (e.g., the ability to manage
responsibilities from multiple classes is a skillset applicable to work and
personal life). Researchers are just beginning to document the experiences
responsible for the development of conscientiousness. Progress is slow because
of the difficulties in assessing experiences and teasing apart the influence of
selection bias (Jackson & Allemand, 2014). However, a number of studies find
theoretically meaningful life experiences associated with changes in
conscientiousness.
Some of the most influential experiences relate to achievement, particularly
within academic and occupational settings. For example, students tend to
increase in conscientiousness during the final year of high school, likely in
preparation for the new responsibilities they anticipate taking on in college or in
the workforce (Bleidorn, 2012). However, individual experiences are important
as not everyone follows this pattern. Those who are not invested in school tend
to change less compared to those who are invested in school (Bleidorn, 2012;
Jackson, Lüdtke, & Trautwien, 2015). Leaving school and obtaining your first
job is also associated with increases in conscientiousness (Specht, Egloff, &
Schmukle, 2011), presumably due to new responsibilities associated with your
first job. In line with this idea, individuals who invest in and are satisfied with
their occupation evidence increases in conscientiousness (Roberts, Caspi, &
Moffitt, 2003), whereas people who deinvest in work, such as committing
counterproductive work behaviors, demonstrate declines (Hudson, Roberts, &
Lodi-Smith, 2012).
Change in conscientiousness also relates to health and health care behaviors.
For example, not surprisingly, enacting negative health behaviors is detrimental
to your health. Substance abuse is associated with nonnormative decreases in
conscientiousness (Roberts & Bogg, 2004). In contrast, those who demonstrate
the normative increases in conscientiousness will have better self-reported health
and fewer sick days at work (Magee, Heaven, & Miller, 2013; Takahashi,
Edmonds, Jackson, & Roberts, 2013; Turiano et al., 2012). The increases in
conscientiousness likely reflect increases in the ability and tendency to take care
of oneself. Consistent with this notion, increases in conscientiousness are
associated with increases in preventive health behaviors, such as healthy eating,
exercising, and safe driving (Takahashi et al., 2013). Similarly, decreases in
alcohol usage are related to increases in conscientiousness (Littlefield, Sher, &
Wood, 2009, 2010). Changes in health behaviors also are related to changes in
health status, suggesting that changes in conscientiousness may be able to lead to
changes in physical health (Takahashi et al., 2013).
Romantic relationships also appear important for the development of
conscientiousness. Becoming involved in your first romantic relationship is
associated with increases in conscientiousness (Neyer & Asendorpf, 2001; Neyer
& Lehnart, 2007). The length of a relationship is also associated with increases
in conscientiousness, such that women who were married longer showed greater
increases in conscientiousness (Roberts & Bogg, 2004). Increases in
conscientiousness likely reflect the maintenance of daily behaviors that keep a
relationship healthy, given that conscientiousness influences the quality of
romantic relationships (Solomon & Jackson, 2014a). Ending a long-term
relationship may also shape conscientiousness, though evidence suggests the
effects depend on gender. One study found that conscientiousness increases after
a divorce (Specht et al., 2011), whereas another study consisting solely of
women found that women decreased in conscientiousness after a divorce
(Roberts & Bogg, 2004). Though further research is needed to determine the
nature of these findings, one possible explanation is that divorced women may
have fewer responsibilities because they no longer have to take care of their
husbands, resulting in lower levels of conscientious, whereas men’s
responsibilities may increase after a divorce. Extending beyond romantic
relationships, there also appears to be an association between social relationships
and the development of conscientiousness (Hill, Payne, et al., 2014), but further
work is needed to tease apart the mechanisms responsible.
The Predictive Utility of Conscientiousness
Much of the interest in conscientiousness stems from its ability to predict a
wide swath of important life outcomes, sometimes decades in advance. Even
more astonishing is that the strength of the association is better than or equal to
other potential predictors, such as intelligence and socioeconomic status, which
are considered the best psychological predictors (Roberts, Kuncel, Shiner, Caspi,
& Goldberg, 2007). The pathways linking conscientiousness to life outcomes are
typically thought to result from behaviors associated with conscientiousness
(e.g., Bogg & Roberts, 2004), though many other pathways may exist (e.g.,
Shanahan, Hill, Roberts, Eccles, & Friedman, 2014). In this section, many of the
outcomes predicted by conscientiousness are presented and some possible
mechanisms are discussed.
Although not without some negative ramifications (Samuel & Widiger, 2011),
conscientiousness has a positive impact across a wide array of important life
outcomes. For instance, conscientiousness is one of the best predictors of
academic success (Bratko, Chamorro-Premuzic, & Saks, 2006; Noftle & Robins,
2007). Conscientious students tend to have better grades (Poropat, 2009;
Trautwein, Lüdtke, Roberts, Schnyder, & Niggli, 2009), especially those high on
the facet of self-control (Noftle & Robins, 2007; Paunonen & Ashton, 2001;
Trautwein et al., 2009). In general, the association with academic success is due
to conscientious students devoting more time to studying, completing projects,
and turning assignments in on time (Credé & Kuncel, 2008; Duckworth &
Carlson, 2013). The behaviors likely result from high levels of motivation to
succeed and the belief that you will be successful within school settings
(Caprara, Vecchione, Alessandri, Gerbino, & Barbaranelli, 2011). The repertoire
of successful school behaviors and the identity of being a successful student aid
conscientious students in persevering when facing difficult or stressful courses
(Corker, Oswald, & Donnellan, 2012).
Conscientiousness also relates to success in the transition from school to the
work force (see also the chapter by Sutin). Highly conscientious people earn
more money (Roberts, Jackson, Duckworth, & Von Culin, 2011), are more
satisfied with their jobs (Ng, Eby, Sorensen, & Feldman, 2005), and are more
likely to receive promotions (Ng et al., 2005; Solomon & Jackson, 2014b).
These benefits occur because effective school behaviors translate to effective
workplace behaviors that result in greater occupational success. For example,
conscientious individuals do not demonstrate many counterproductive workplace
behaviors such as stealing, drinking on the job, being tardy, or missing work
(Mount, Ilies, & Johnson, 2006; Roberts, Harms, Caspi, & Moffitt, 2007). The
effective behavior of highly conscientious individuals likely results from having
greater career achievement goals and better attitudes toward their workplace
environment (Ng et al., 2005). In general, highly conscientious individuals are
more committed and invested in their jobs (Sutin & Costa, 2010). Finally, there
is evidence that the relationship between conscientiousness and achievement
domains like school and work extends into older adulthood. Conscientious
individuals are more likely to volunteer when they are retired compared to when
they are working (Mike, Jackson, & Oltmanns, 2014). These findings suggest
that volunteering changes meaning for conscientious individuals: when
conscientious individuals retire they lose an important niche and hence are more
likely to fill their time with a meaningful activity, such as volunteering.
Conscientiousness is one of the strongest predictors of physical health
(Hampson, 2012). Conscientiousness predicts fewer self-reported health
problems (Lodi-Smith et al., 2010; Turiano et al., 2012) as well as more
objective markers, such as physician-rated health status (Chapman, Lyness, &
Duberstein, 2007) and various biomarkers associated with health status
(Hampson, Edmonds, Goldberg, Dubanoski, & Hillier, 2013; Israel et al., 2014;
Moffitt et al., 2011), a decrease in onset of major diseases such as a stroke or a
heart attack (Sutin, Ferrucci, Zonderman, & Terracciano, 2011; Weston, Hill, &
Jackson, 2015), Alzheimer’s disease (Terracciano et al., 2013; Wilson,
Schneider, Arnold, Bienias, & Bennett, 2007), and even increased longevity
(Hill & Roberts, 2011; Jackson, Connolly, et al., 2015; Jokela et al., 2013; Kern
& Friedman, 2008; Roberts, Harms, et al., 2007). The relationship between
conscientiousness and health exists across the entire lifespan, as childhood and
young adult-rated conscientiousness predicts health and mortality decades later
(Freidman et al., 1993; Hampson et al., 2013; Jackson, Connolly, et al., 2015).
The primary mechanism is an increase in health-related behaviors, though
many other pathways exist (Shanahan et al., 2014). Conscientious individuals
engage in more health prevention behaviors, such as exercising and eating
healthy (Lodi-Smith et al., 2010), and are also less likely to smoke, use drugs,
abuse alcohol, or engage in risky driving or sexual behaviors (Bogg & Roberts,
2004; O’Connor, Conner, Jones, McMillan, & Ferguson, 2009; Turiano et al.,
2012; Weston & Jackson, 2015). If health issues do arise, conscientious
individuals are more likely to adhere to their medical regimens (Hill & Roberts,
2011). By following their doctors’ advice, and taking prescribed medications as
needed, they are better able to overcome health ailments and prevent future poor
health. Furthermore, conscientiousness helps people effectively respond to
health issues, such as the onset of a major disease, by modifying their previous
health behaviors (Weston & Jackson, 2015).
High levels of conscientious can even seep over and influence the health of
your family. For example, diabetic children whose mothers are high in
conscientiousness tend to have better blood sugar control (Vollrath, Landolt,
Gnehm, Laimbacher, & Sennhauser, 2007). Marrying a partner high in
conscientiousness also leads to better health, presumably through the ability of
your spouse to shape health-related behaviors such as improved eating and
exercise habits (Roberts, Smith, Jackson, & Edmonds, 2009).
Conscientiousness also plays a role in the success and maintenance of
romantic relationships. Conscientious people are less likely to get divorced
(Roberts, Harms, et al., 2007). This is likely a result of the fact that people high
in conscientiousness have higher levels of relationship satisfaction and
commitment levels (Dyrenforth, Kashy, Donnellan, & Lucas, 2010; Solomon &
Jackson, 2014a). High levels of satisfaction and commitment reflect an
accumulation of day-to-day experiences that strengthen the relationship along
with fewer experiences that jeopardize the relationship (Hill, Nickel, & Roberts,
2014). In terms of strengthening experiences, conscientiousness is associated
with relationship maintenance behaviors (Baker & McNulty, 2011), such as
engaging in constructive problem solving. Conscientious individuals are also
more forgiving, which would help resolve long-term grudges among couples
(Hill & Allemand, 2012). In terms of negative behaviors, conscientiousness is
associated with lower levels of cheating (Buss & Shackelford, 1997) and
physical and mental abuse (Buss, 1991; Hines & Saudino, 2008).
The influence of conscientiousness stretches beyond the broad domains of
work, health, and relationships and into less desirable outcomes. Low levels of
conscientiousness predict greater engagement in criminal behaviors, greater
arrest rates, and longer sentences (Clower & Bothwell, 2001; Jones, Miller, &
Lynam, 2011; Wiebe, 2004). Levels of conscientiousness also have financial
repercussions. Those low in conscientiousness have difficulties saving money,
and as a result have worse credit scores (Bernerth, Taylor, Walker, & Whitman,
2012; Webley & Nyhus, 2006) and lower net worth (Mroczek et al., 2015).
People low in conscientiousness are also more prone to gamble, particularly if
they are highly impulsive (Bagby et al., 2007).
Summary
Conscientiousness is a personality trait that taps into the regulatory system in
humans and appears, therefore, to be a critical focal point for optimal
functioning. Best thought of as a domain rather than a unitary entity,
conscientiousness subsumes multiple components or facets, such as self-control,
orderliness, industriousness, and responsibility. The faceted structure alludes to
the fact that conscientiousness is a hierarchical domain that can be examined and
manifested from broad to narrow levels and encompasses state as well as trait
levels of analysis. The multifaceted structure must be taken into consideration
when choosing assessment methods, as different measures emphasize some
facets over others. Consistent with the emerging consensus on personality traits,
conscientiousness is a developmental construct and therefore shows both
continuity and change throughout the life course. Finally, conscientiousness
appears to play a positive role in the major life domains of work, love, and
health. Although not without some negative ramifications, conscientiousness is
clearly a personality trait that deserves greater attention given the pervasive
positive impact it has on people’s lives.
References
Allemand, M., Zimprich, D., & Hendriks, A. A. J. (2008). Age differences in five
personality domains across the lifespan. Developmental Psychology, 44, 758–770.
Almlund, M., Duckworth, A. L., Heckman, J. J., & Kautz, T. D. (2011). Personality
psychology and economics (No. w16822). National Bureau of Economic Research.
Bagby, R., Vachon, D. D., Bulmash, E. L., Toneatto, T., Quilty, L. C., & Costa, P. T.
(2007). Pathological gambling and the five-factor model of personality. Personality
and Individual Differences, 43(4), 873–880.
Baker, L. R., & McNulty, J. K. (2011). Self-compassion and relationship maintenance:
The moderating roles of conscientiousness and gender. Journal of Personality and
Bandura, A. (2012). On the functional properties of perceived self-efficacy revisited.
Journal of Management, 38, 9–44.
Baumeister, R. F., Vohs, K. D., & Tice, D. M. (2007). The strength model of self-control.
Current Directions in Psychological Science, 16, 351–355.
Bernerth, J. B., Taylor, S. G., Walker, H., & Whitman, D. S. (2012). An empirical
investigation of dispositional antecedents and performance-related outcomes of credit
scores. Journal of Applied Psychology, 97(2), 469–478.
Bleidorn, W. (2012). Hitting the road to adulthood—short-term personality development
during a major life transition. Personality and Social Psychology Bulletin, 38, 1594–
1608.
Block, J. H., & Block, J. (1980). The role of ego-control and ego-resiliency in the
organization of behavior. In Development of cognition, affect, and social relations: The
Minnesota symposia on child psychology (Vol. 13, pp. 39–101). Hillsdale, NJ:
Erlbaum.
Bogg, T., & Roberts, B. W. (2004). Conscientiousness and health related behaviors: A
meta-analysis of the leading behavioral contributors to mortality. Psychological
Bulletin, 130, 887–919.
Bogg, T., & Roberts, B. W. (2013). Duel or diversion? Conscientiousness and executive
function in the prediction of health and longevity. Annals of Behavioral Medicine,
45(3), 400–401
Buss, D. M. (1991). Conflict in married couples: Personality predictors of anger and
upset. Journal of Personality, 59(4), 663–688.
Buss, D. M., & Shackelford, T. K. (1997). Susceptibility to infidelity in the first year of
marriage. Journal of Research in Personality, 31(2), 193–221.
Bratko, D., Chamorro-Premuzic, T., & Saks, Z. (2006). Personality and school
performance: Incremental validity of self-and peer-ratings over intelligence.
Personality and Individual Differences, 41(1), 131-142.
Caprara, G., Vecchione, M., Alessandri, G., Gerbino, M., & Barbaranelli, C. (2011). The
contribution of personality traits and self-efficacy beliefs to academic achievement: A
longitudinal study. British Journal of Educational Psychology, 81(1), 78–96.
Carver, C. S., Johnson, S. L., & Joormann, J. (2008). Serotonergic function, two-mode
models of self-regulation, and vulnerability to depression: What depression has in
common with impulsive aggression. Psychological Bulletin, 134(6), 912.
Cattell, R. B. (1957). Personality and motivation structure and measurement. New York,
NY: World Book.
Chapman, B. P. (2007). Bandwidth and fidelity on the NEO-Five Factor Inventory:
Replicability and reliability of Saucier’s (1998) item cluster subcomponents. Journal of
Personality Assessment, 88(2), 220–234.
Chapman, B. P., Lyness, J. M., & Duberstein, P. R. (2007). Personality and medical
illness burden among older adults in primary care. Psychosomatic Medicine, 69, 277–
282.
Clark, L. A., & Watson, D. (2008). Temperament: An organizing paradigm for trait
psychology. In R. W. Roberts, O. P. John, & L. A. Pervin (Eds.), Handbook of
personality psychology: Theory and research (3rd ed., pp. 265–286). New York, NY:
Guilford Press.
Clower, C. E., & Bothwell, R. K. (2001). An explanatory study of the relationship
between the Big Five and inmate recidivism. Journal of Research in Personality, 35,
231–237.
Corker, K. S., Oswald, F. L., & Donnellan, M. B. (2012). Conscientiousness in the
classroom: A process explanation. Journal of Personality, 80(4), 995–1028.
Costa, P. T., Jr., & McCrae, R. R. (1992). NEO PI-R Professional Manual. Odessa, FL:
Costa, P. T., McCrae, R., & Dye, D. A. (1991). Facet scales for agreeableness and
conscientiousness: A revision of the NEO personality inventory. Personality and
Credé, M., Harms, P., Niehorster, S., & Gaye-Valentine, A. (2012). An evaluation of the
consequences of using short measures of the Big Five personality traits. Journal of
Credé, M., & Kuncel, N. R. (2008). Study habits, skills, and attitudes: The third pillar
supporting collegiate academic performance. Perspectives on Psychological Science,
3(6), 425–453.
DeYoung, C. G. (2011). Impulsivity as a personality trait. In K. D. Vohs & R. F.
Baumeister (Eds.), Handbook of self-regulation: Research, theory, and applications
(2nd ed., pp. 485–502). New York, NY: Guilford Press.
Duckworth, A. L., & Carlson, S. M. (2013). Self-regulation and school success. In B. W.
Sokol, F. M. E. Grouzet, & U. Müller (Eds.), Self-regulation and autonomy: Social and
developmental dimensions of human conduct. New York, NY: Cambridge University
Press
Duckworth, A. L., & Kern, M. L. (2011). A meta-analysis of the convergent validity of
self-control measures. Journal of Research in Personality, 45, 259–268.
Duckworth, A. L., Peterson, C., Matthews, M. D., & Kelly, D. R. (2007). Grit:
Perseverance and passion for long-term goals. Journal of Personality and Social
Psychology, 92, 1087–1101.
Duckworth, A. L., & Quinn, P. D. (2009). Development and validation of the Short Grit
Scale (Grit-S). Journal of Personality Assessment, 91, 166–174.
Duckworth, A. L., Tsukayama, E., & Kirby, T. (2013). Is it really self-control?
Examining the predictive power of the delay of gratification test. Personality and
Dudley, N. M., Orvis, K. A., Lebiecki, J. E., & Cortina, J. M. (2006). A meta-analytic
investigation of conscientiousness in the prediction of job performance: Examining the
intercorrelations and the incremental validity of narrow traits. Journal of Applied
Dyrenforth, P. S., Kashy, D. A., Donnellan, M. B., & Lucas, R. E. (2010). Predicting
relationship and life satisfaction from personality in nationally representative samples
from three countries: The relative importance of actor, partner, and similarity effects.
Edmonds, G. E., Bogg, T., & Roberts, B. W. (2009). Are personality and behavioral
measures of impulse control convergent or distinct predictors of health behaviors?
Eisenberg, N., Smith, C. L., Sadovsky, A., & Spinrad, T. L. (2004). Effortful control. In
R. F. Baumeister & K. D. Vohs (Eds.), Handbook of self-regulation: Research, theory,
and applications (pp. 259–282). New York, NY: Guilford Press.
Evans, D. E., & Rothbart, M. K. (2007). Developing a model for adult temperament.
Eysenck, H. J. (1947). Dimensions of personality. Oxford, England: Kegan Paul.
Fleeson, W. (2001). Toward a structure- and process-integrated view of personality:
Traits as density distributions of states. Journal of Personality and Social Psychology,
80, 1011–1027.
Freidman, H. S., Tucker, J. S., Tomlinson–Keasey, C., Schwartz, J. E., Wingard, D. L., &
Criqui, M. H. (1993). Does childhood personality predict longevity? Journal of
Gaughan, E. T., Miller, J. D., Pryor, L. R., & Lynam, D. R. (2009). Comparing two
alternative measures of general personality in the assessment of psychopathy: A test of
the NEO PI-R and the MPQ. Journal of Personality, 77, 965–996.
Gerber, A. S., Huber, G. A., Doherty, D., Dowling, C. M., Raso, C., & Ha, S. E. (2011).
Personality traits and participation in political processes. The Journal of Politics, 73(3),
692–706
public-domain personality measures. Journal of Research in Personality, 40(1), 84–96
Gosling, S. D., Rentfrow, P. J., & Swann, W. B. (2003). A very brief measure of the Big-
Five personality domains. Journal of Research in Personality, 37(6), 504–528.
Gough, H. G. (1956). California Psychological Inventory. Palo Alto, CA: Consulting
Psychologists Press.
Hampson, S. E. (2012). Personality processes: Mechanisms by which personality traits
“get outside the skin.” Annual Review of Psychology, 63, 315.
Hampson, S. E., Edmonds, G. W., Goldberg, L. R., Dubanoski, J. P., & Hillier, T. A.
(2013). Childhood conscientiousness relates to objectively measured adult physical
health four decades later. Health Psychology, 32, 925–928.
Hill, P. L., & Allemand, M. (2012). Explaining the link between conscientiousness and
forgiveness. Journal of Research in Personality, 46, 497–503.
Hill, P. L., Nickel, L. B., & Roberts, B. W. (2014). Are you in a healthy relationship?:
Linking conscientiousness to health via implementing and immunizing behaviours.
Hill, P. L., Payne, B. R., Jackson, J. J., Stine-Morrow, E. A., & Roberts, B. W. (2014).
Perceived social support predicts increased conscientiousness during older adulthood.
The Journals of Gerontology Series B: Psychological Sciences and Social Sciences,
69(4), 543-547.
Hill, P. L., & Roberts, B. W. (2011). The role of adherence in the relationship between
conscientiousness and perceived health. Health Psychology, 30(6), 797–804.
Hines, D. A., & Saudino, K. J. (2008). Personality and intimate partner aggression in
dating relationships: The role of the “Big Five”. Aggressive Behavior, 34(6), 593-604.
Hofstee, W. K. B., de Raad, B., & Goldberg, L. R. (1992). Integration of the Big Five and
63, 146–163.
Hoyle, R. H. (2010). Personality and self-regulation. In R. H. Hoyle (Ed.), Handbook of
personality and self-regulation (pp. 1–18). Malden, MA: Wiley-Blackwell.
Hudson, N. W., Roberts, B. W., & Lodi-Smith, J. (2012). Personality trait development
and social investment in work. Journal of Research in Personality, 46(3), 334–344.
Israel, S., Moffitt, T. E., Belsky, D. W., Hancox, R. J., Poulton, R., Roberts, B., … Caspi,
A. (2014). Translating personality psychology to help personalize preventive medicine
for young adult patients. Journal of Personality and Social Psychology, 106(3), 484.
Jackson, D. N. (1976). Basic personality inventory. Port Huron, MI: Sigma Assessment
Systems.
Jackson, J. J., & Allemand, M. (2014). Moving personality development research
forward: Utilizing structural equation models. European Journal of Personality, 28,
300–310.
Jackson, J. J., Bogg, T., Walton, K. E., Wood, D., Harms, P. D., Lodi-Smith, J., …
Roberts, B. W. (2009) Not all conscientiousness scales change alike: A multimethod,
multisample study of age differences in the facets of conscientiousness. Journal of
Jackson, J. J., Connolly, J. J., Levine, M., & Garrison, M. (2015). Your friends know how
long you will live: A 75 year study of peer-rated personality traits. Psychological
Science, 26, 335–340.
Jackson, J. J., Hill, P. L., & Roberts, B. W. (2011). Sociogenomic theory as an answer to
the heritability problem. European Journal of Personality, 25, 274–276.
Jackson, J. J., Hill, P. L., & Roberts, B. W. (2012). Misconceptions of traits continue to
persist: A response to Bandura. Journal of Management, 38, 745–752.
Jackson, J. J., Lüdtke, O., & Trautwien, U. (2015). What do you gain from school besides
smarts? Educational experiences and personality trait development. Unpublished
manuscript.
What do conscientious people do? Development and validation of the behavioral
indicators of conscientiousness (BIC). Journal of Research in Personality, 44(4), 501–
511.
John, O. P., Donahue, E. M., & Kentle, R. L. (1991). The Big Five Inventory—Versions
4a and 54. Berkeley, CA: University of California, Berkeley, Institute of Personality
and Social Research.
Johnson, J. A. (1994). Clarification of factor five with the help of the AB5C model.
Jokela, M., Batty, G. D., Nyberg, S. T., Virtanen, M., Nabi, H., Singh-Manoux, A., &
Kivimäki, M. (2013). Personality and all-cause mortality: Individual-participant meta-
analysis of 3,947 deaths in 76,150 adults. American Journal of Epidemiology, 178(5),
667–675.
Jones, S. E., Miller, J. D., & Lynam, D. R. (2011). Personality, antisocial behavior, and
aggression: A meta-analytic review. Journal of Criminal Justice, 39(4), 329–337.
Kern, M. L., & Friedman, H. S. (2008). Do conscientious individuals live longer? A
quantitative review. Health Psychology, 27(5), 505.
Kochanska, G., Murray, K. T., & Harlan, E. T. (2000). Effortful control in early
childhood: Continuity and change, antecedents, and implications for social
development. Developmental Psychology, 36, 220–232.
Lee, K., & Ashton, M. C. (2008). The HEXACO personality factors in the indigenous
personality lexicons of English and 11 other languages. Journal of Personality, 76,
1001–1053.
Littlefield, A. K., Sher, K. J., & Wood, P. K. (2009). Is “maturing out” of problematic
alcohol involvement related to personality change? Journal of Abnormal Psychology,
118(2), 360–374.
Littlefield, A. K., Sher, K. J., & Wood, P. K. (2010). A personality-based description of
maturing out of alcohol problems: Extension with a Five-Factor model and robustness
to modeling challenges. Addictive Behaviors, 35, 948–954.
Lodi-Smith, J., Jackson, J., Bogg, T., Walton, K., Wood, D., Harms, P., & Roberts, B. W.
(2010). Mechanisms of health: Educations and health-related behaviors partially
mediate the relationship between conscientiousness and self-reported physical health.
Psychology and Health, 25, 305–319.
to university life. Journal of Personality and Social Psychology, 101(3), 620–637.
MacCann, C., Duckworth, A. L., & Roberts, R. D. (2009). Empirical identification of the
major facets of conscientiousness. Learning and Individual Differences, 19(4), 451–
458.
MacDonald, K. B. (2008). Effortful control, explicit processing, and the regulation of
human evolved predispositions. Psychological Review, 115, 1012–1031.
Magee, C. A., Heaven, P. C., & Miller, L. M. (2013). Personality change predicts self-
reported mental and physical health. Journal of Personality, 81(3), 324–334.
McCrae, R. R., & Costa, P. T., Jr. (2008). The five-factor theory of personality. In O. P.
John, R. W. Robins, & L. A. Pervin (Eds.), Handbook of personality: Theory and
research (pp. 159–181). New York, NY: Guilford Press.
McCrae, R. R., Costa, P. T., Pedroso de Lima, M., Simões, A, Ostendorf, F., Angleitner,
A, … Piedmont, R. L. (1999). Age differences in personality across the adult life span:
Mike, A., Harris, K., Roberts, B.W., & Jackson, J. J. (2015). Conscientiousness: Lower-
order structure, life-course consequences and development. In J. Wright (ed.),
International encyclopedia of social and behavioral sciences (2nd ed., pp. 658-665).
Mike, A., Jackson, J. J & Oltmanns, T. F., (2014). The Conscientious Retiree: The
relationship between conscientiousness, retirement, and volunteering. Journal of
Research in Personality, 52, 68–77
Mischel, W., & Shoda, Y. (2008). Toward a unified theory of personality: Integrating
dispositions and processing dynamics within the cognitive-affective processing system.
In O. P. John, R. W. Robins, & L. A. Pervin (Eds.), Handbook of personality
psychology: Theory and research (3rd ed., pp. 208–241). New York, NY: Guilford
Press.
Mischel, W., Shoda, Y., & Rodriguez, M. L. (1992). Delay of gratification in children. In
G. Loewenstein & J. Elster (Eds.), Choice over time (pp. 147–164). New York, NY:
Russell Sage Foundation.
Moffitt, T. E., Arseneault, L., Belsky, D., Dickson, N., Hancox, R. J., Harrington, H. L.,
… Caspi, A. (2011). A gradient of childhood self-control predicts health, wealth, and
public safety. Proceedings of the National Academy of Sciences, 108, 2693–2698.
Mount, M., Ilies, R., & Johnson, E. (2006). Relationship of personality traits and
counterproductive work behaviors: The mediating effects of job satisfaction. Personnel
Mroczek, D., Turiano, N. A., Chan, W., Hill, P. L., & Roberts, B. W. (2015). Personality
traits predict net worth. Manuscript under review.
(2006). Psychometric properties of an abbreviated instrument of the Five-Factor model.
Assessment, 13(2), 119–137.
Neyer, F. J., & Asendorpf, J. B. (2001). Personality-relationship transaction in young
adulthood. Journal of Personality and Social Psychology, 81(6), 1190–1204.
Neyer, F. J., & Lehnart, J. (2007). Relationships matter in personality development.
Evidence from an 8-year longitudinal study across young adulthood. Journal of
Ng, T. H., Eby, L. T., Sorensen, K. L., & Feldman, D. C. (2005). Predictors of objective
and subjective career success. A meta-analysis. Personnel Psychology, 58(2), 367–408.
five correlates of GPA and SAT scores. Journal of Personality and Social Psychology,
93(1), 116–130.
O’Connor, D. B., Conner, M., Jones, F., McMillan, B., & Ferguson, E. (2009). Exploring
the benefits of conscientiousness: An investigation of the role of daily stressors and
health behaviors. Annals of Behavioral Medicine, 37(2), 184–196.
Paunonen, S. V. (1998). Hierarchical organization of personality and prediction of
Paunonen, S. V., & Ashton, M. C. (2001). Big Five predictors of academic achievement.
Peabody, D., & De Raad, B. (2002). The substantive nature of psycholexical personality
factors: A comparison across languages. Journal of Personality and Social Psychology,
83(4), 983–997.
Perugini, M., & Gallucci, M. (1997). A hierarchical faceted model of the Big Five.
Poropat, A. E. (2009). A meta-analysis of the five-factor model of personality and
academic performance. Psychological Bulletin, 135(2), 322.
when we assess a Big 5 trait? A content analysis of the affective, behavioral and
cognitive processes represented in the Big 5 personality inventories. Personality &
Social Psychology Bulletin, 28, 847–858.
Revelle, W. (1997). Extraversion and impulsivity: The lost dimension? In H. Nyborg
(Ed.), The scientific study of human nature: Tribute to Hans J. Eysenck at eighty (pp.
189–212). Amsterdam, the Netherlands: Pergamon/Elsevier Science.
Roberts, B. W. (2009). Back to the future: Personality and assessment and personality
development. Journal of Research in Personality, 43, 137–145.
Roberts, B. W., & Bogg, T. (2004). A longitudinal study of the relationships between
conscientiousness and the social-environmental factors and substance-use behaviors
that influence health. Journal of Personality, 72(2), 325–354.
Roberts, B. W., Bogg, T., Walton, K. E., Chernyshenko, O. S., & Stark, S. E. (2004). A
lexical investigation of the lower-order structure of conscientiousness. Journal of
Roberts, B. W., Caspi, A., & Moffitt, T. E. (2003). Work experiences and personality
development in young adulthood. Journal of Personality and Social Psychology, 84(3),
582–593.
Roberts, B. W., Chernyshenko, O. S., Stark, S., & Goldberg, L. R. (2005). The structure
of conscientiousness: An empirical investigation based on seven major personality
questionnaires. Personnel Psychology, 58(1), 103–139.
Roberts, B. W., Harms, P. D., Caspi, A., & Moffitt, T. E. (2007). Predicting the
counterproductive employee in a child-to-adult prospective study. The Journal of
Applied Psychology, 92(5), 1427–1436.
Roberts, B. W., & Jackson, J. J. (2008). Sociogenomic personality psychology. Journal of
Personality, 76(6), 1523–1544.
Roberts, B. W., Jackson, J. J., Burger, J., & Trautwein, U. (2009). Conscientiousness and
externalizing psychopathology: Overlap, developmental patterns, and etiology of two
related constructs. Development and Psychopathology, 21(3), 871–888.
Roberts, B., Jackson, J. J., Duckworth, A. L., & Von Culin, K. (2011). Personality
measurement and assessment in large panel surveys. Forum for Health Economics &
Policy, 14(3), 1–32.
Roberts, B. W., Jackson, J. J., Fayard, J. V., Edmonds, G., & Meints, J. (2009).
Conscientiousness. In M. R. Leary & R. H. Hoyle (Eds.), Handbook of individual
differences in social behavior (pp. 369–381). New York, NY: Guilford Press.
Roberts, B. W., Lejuez, C., Krueger, R. F., Richards, J. M., & Hill, P. L. (2014). What is
conscientiousness and how can it be assessed? Developmental Psychology, 50, 1315–
1330.
Roberts, B. W., Smith, J., Jackson, J. J., & Edmonds, G. (2009). Compensatory
conscientiousness and health in older couples. Psychological Science, 20, 6–10.
Roberts, B. W., Walton, K., & Viechtbauer, W. (2006). Patterns of mean-level change in
personality traits across the life course: A meta-analysis of longitudinal studies.
Roberts, B. W., Wood, D., & Caspi, A. (2008). The development of personality traits in
adulthood. In O. P. John, R. W. Robins, & L. A. Pervin (Eds.), Handbook of
personality: Theory and research (3rd ed., pp. 375–398). New York, NY: Guilford
Press.
Rothbart, M. K., & Rueda, M. R. (2005). The development of effortful control. In U.
Mayr, E. Awh, & S. Keele (Eds.), Developing individuality in the human brain: A
tribute to Michael I. Posner (pp. 167–188). New York, NY: Guilford Press.
Samuel, D. B., & Widiger, T. A. (2011). Conscientiousness and obsessive-compulsive
personality disorder. Personality Disorders: Theory, Research, and Treatment, 2(3),
161.
Saucier, G. (1994). Mini-markers: A brief version of Goldberg’s unipolar Big-Five
markers. Journal of Personality Assessment, 63(3), 506–516
Saucier, G. (1998). Replicable item-cluster subcomponents in the NEO Five-Factor
Inventory. Journal of Personality Assessment, 70(2), 263–276.
Saucier, G., & Ostendorf, F. (1999). Hierarchical subcomponents of the Big Five
personality factors: A cross-language replication. Journal of Personality and Social
Shanahan, M. J., Hill, P. L., Roberts, B. W., Eccles, J., & Friedman, H. S. (2014).
Conscientiousness, health, and aging: The life course of personality model.
Solomon, B., & Jackson, J. J. (2014a). Why does personality predict divorce? Multiple
pathways through satisfaction. Journal of Personality and Social Psychology. 106, 978-
996.
Solomon, B. C., & Jackson, J. J. (2014b). The long reach of one’s spouse spouses’
personality influences occupational success. Psychological Science, 25, 2189-2198.
Soto, C. J., & John, O. P. (2009). Ten facet scales for the Big Five Inventory:
Convergence with NEO PI-R facets, self-peer agreement, and discriminant validity.
Soto, C. J., & John, O. P. (2012). Development of Big Five domains and facets in
adulthood: Mean-level age trends and broadly versus narrowly acting mechanisms.
Specht, J., Egloff, B., & Schmukle, S. C. (2011). Stability and change of personality
across the life course: The impact of age and major life events on mean-level and rank-
order stability of the Big Five. Journal of Personality and Social Psychology, 101(4),
862–882.
Sutin, A. R., & Costa, P. R. (2010). Reciprocal influences of personality and job
characteristics across middle adulthood. Journal of Personality, 78(1), 257–288.
Sutin, A. R., Ferrucci, L., Zonderman, A. B., & Terracciano, A. (2011). Personality and
obesity across the adult life span. Journal of Personality and Social Psychology, 101,
579–592.
Takahashi, Y., Edmonds, G. W., Jackson, J. J., & Roberts, B. W. (2013). Longitudinal
correlated changes in conscientiousness, preventative health-related behaviors, and
self-perceived physical health. Journal of Personality, 81, 417–427.
Tangney, J. P., Baumeister, R. F., & Boone, A. L. (2004). High self-control predicts good
adjustment, less pathology, better grades, and interpersonal success. Journal of
Tellegen, A. (1982). Brief manual for the multidimensional personality questionnaire.
Terracciano, A., Iancono, D., O’Brien, R., Troncoso, J. C., An, Y., Sutin, A. R., …
Resnick, S. M. (2013). Personality and resilience to Alzheimer’s disease
neuropathology: A prospective autopsy study. Neurobiology of Aging, 34, 1045–1050.
Terracciano, A., McCrae, R. R., Brant, L. J., & Costa, P. T. (2005). Hierarchical linear
Aging. Psychology and Aging, 20(3), 493–506.
Tough, P. (2013). How children succeed. Chicago, IL: Random House.
Trautwein, U., Lüdtke, O., Roberts, B. W., Schnyder, I., & Niggli, A. (2009). Different
forces, same consequence: Conscientiousness and competence beliefs are independent
predictors of academic effort and achievement. Journal of Personality and Social
Psychology, 97(6), 1115–1128.
Tsukayama, E., Duckworth, A. L., & Kim, B. (2012). Resisting everything except
temptation: Evidence and an explanation for domain-specific impulsivity. European
Turiano, N. A., Pitzer, L., Armour, C., Karlamangla, A., Ryff, C. D., & Mroczek, D. K.
(2012). Personality trait level and change as predictors of health outcomes: Findings
from a national study of Americans (MIDUS). The Journals of Gerontology Series B:
Psychological Sciences and Social Sciences, 67, 4–12.
Vollrath, M. E., Landolt, M. A., Gnehm, H. E., Laimbacher, J. J., & Sennhauser, F. H.
(2007). Child and parental personality are associated with glycaemic control in Type 1
diabetes. Diabetic Medicine, 24(9), 1028–1033.
Walton, K. E., Huyen, B. T. T., Thorpe, K., Doherty, E. R., Juarez, B., D’Accordo, C., &
Reina, M. T. (2013). Cross-sectional personality differences from age 16–90 in a
Vietnamese sample. Journal of Research in Personality, 47(1), 36–40.
Webley, P., & Nyhus, E. K. (2006). Parents’ influence on children’s future orientation
and saving. Journal of Economic Psychology, 27, 140–164.
Werner, P. D., & Pervin, L. A. (1986). The content of personality inventory items.
Journal of Personality & Social Psychology, 51, 622–628.
Weston, S., Hill, P. L., & Jackson, J. J. (2015). Personality traits predict the onset of
major disease. Social Psychological and Personality Science, 6, 309–317.
Weston, S., & Jackson, J. J. (2015). Identification of the healthy neurotic: Personality
traits predict smoking after disease onset. Journal of Research in Personality, 54, 61–
69.
Wiebe, R. P. (2004). Delinquent behavior and the Five-Factor Model: Hiding in the
adaptive landscape? Individual Differences Research, 2, 38–62.
Wilson, R. S., Schneider, J. A., Arnold, S. E., Bienias, J. L., & Bennett, D. A. (2007).
Conscientiousness and the incidence of Alzheimer disease and mild cognitive
impairment. Archives of General Psychiatry, 64, 1204–1212.
Wortman, J., Lucas, R. E., & Donnellan, M. B. (2012). Stability and change in the Big
Five personality domains: Evidence from a longitudinal study of Australians.
Wrosch, C., Scheier, M. F., Miller, G. E., Schulz, R., & Carver, C. S. (2003). Adaptive
self-regulation of unattainable goals: Goal disengagement, goal reengagement, and
subjective well-being. Personality and Social Psychology Bulletin, 29, 1494–1508.
Zuckerman, M., Eysenck, S. B. G., & Eysenck, H. J. (1978). Sensation seeking in
England and America: Cross-cultural, age, and sex comparisons. Journal of Consulting
and Clinical Psychology, 46, 139–149.
SECTION 2
Construct Validity
Robustness
Brian P. O’Connor
Abstract
This article examines the effectiveness of the Five Factor Model (FFM) in
capturing or duplicating the scales and primary dimensions found in other
personality inventories. It considers the robustness—or “comprehensiveness”—of
the FFM at both the scale and dimensional structure levels, as well as the nature
and extent of the evidence for the FFM as an integrative, organizational
framework for other personality tests. “Robustness” here refers to the tendency
for the FFM dimensions to keep showing up in a wide range of old and new
measures that were designed to assess supposedly unique and important other
constructs. This article begins with a review of the primary findings that were
reported by O’Connor (2002) before discussing the nature of dimensions in
personality psychology data. It then evaluates the robustness of the FFM at the
dimensional structure level and suggests directions for further research at the
scale and dimensional structure levels.
Key Words: Five Factor Model, personality, scales, dimensions, personality
inventories, robustness, personality tests, personality psychology,
comprehensiveness
This chapter focuses on the degree to which the Five Factor Model (FFM)
captures the scales and primary dimensions that exist in other personality
inventories. Many hundreds of personality measures have been developed in the
history of our discipline. There must surely be a map or a family tree that
organizes and simplifies this complex world of tests. Can the FFM fulfill such a
role? McCrae and Costa (1986) claimed that FFM “offers a universal and
comprehensive framework for the description of individual differences in
personality” (p. 1001), and there have been many empirical reports on the
overlap between the FFM and various other measures of personality. Many
reviewers have also used the FFM to organize, summarize, and integrate
personality research literatures (e.g., Feingold, 1994; Funder, 2001; Ozer &
Benet-Martinez, 2006; Ozer & Reise, 1994; Roberts & Del Vecchio, 2000;
Segerstrom, 2000; Shiner & Caspi, 2003). This chapter provides an overview of
the nature and extent of the evidence for the FFM as an integrative,
organizational framework for other personality tests.
The term “robustness” is here used to refer to the tendency of the FFM
dimensions to keep showing up in a wide range of old and new measures of
personality that were designed to assess supposedly unique and important other
constructs. “Robustness” is clearly related to “comprehensiveness.” These three
terms will sometimes be used interchangeably with the understanding that the
sole focus is on the FFM in relation to other measures. The question of whether
the FFM is also comprehensive with regard to lexical personality data is beyond
the scope of the present review (see Almagor, Tellegen, & Waller, 1995;
Goldberg, 1993; John, Naumann, & Soto, 2008; McCrae & Costa, 1995; see also
the chapter by de Raad and Mlačić).
It is also important to distinguish between robustness at the scale level and
robustness at the dimensional structure level. This distinction has rarely been
made in previous work and it is important both empirically and theoretically.
The previous neglect of this distinction has apparently led to misunderstandings
and ongoing controversies.
This chapter begins with a reprise of the primary findings that were reported
in a previous article, “A quantitative review of the comprehensiveness of the
FFM in relation to popular personality inventories” (O’Connor, 2002). This is
followed by suggestions for further research at both the scale and dimensional
structure levels.
One possible interpretation of the claim that the FFM “offers a universal and
comprehensive framework for the description of individual differences in
personality” (McCrae & Costa, 1986, p. 1001) is that the FFM should capture
most of the variance in other personality scales. This was my own initial
understanding as a student and there seemed to be a close analogy with the
primary colors in perception. Just as the many different colors of light can be
produced from different blends of the three primary colors (red, green-yellow,
and blue), different personality traits should perhaps be reducible to different
blends of a small number of primary traits, such as the FFM dimensions. If the
FFM is universal and comprehensive, then it should be able to recreate the
scores on other measures. In other words, it should capture most of the reliable
variance in other measures. In perception, the three primary colors are 100%
effective in reproducing all the other colors from the spectrum. How effective is
the FFM in capturing or duplicating other personality constructs? This question
is important for illustrative purposes, but it raises expectations that are not
realistic or fair. First to be reviewed will be the relevant evidence.
Data Sets and Analytic Methods

Much of the work on the appearance of the FFM in other measures has been
conducted by McCrae and Costa. In a series of investigations, they have reported
associations between their own carefully developed measures of the five factors
and the scales of popular personality inventories. To estimate the percentages of
variance in other scales that are captured by the FFM, all previous reports of
correlations or factor loadings from joint analyses of FFM variables and other
personality inventory scales were gathered. When only factor loadings were
available from previous reports, the interbattery correlations were obtained by
multiplying a loading matrix by its transpose (Bernstein, 1988; Gorsuch, 1983).
The reports were peer-reviewed journal articles that were identified from
extensive PSYCHLIT searches (using the terms “five” “factor” “model” and
“personality” “test,” “measure,” or “inventory”). The inventories and data
sources are listed in Table 8.1. For some inventories, more than one report of
associations between the inventory scales and FFM variables was available.
Every published data set known to the author was included in the analyses.
The percentage of variance in each personality inventory scale accounted for
by the FFM was computed using matrix algebra formulas for R-squared that can
be found in most multivariate statistics texts (see Bernstein, 1988, p. 104). The
computations required both the interbattery correlations and the FFM
intercorrelations. In cases in which the FFM intercorrelations were not reported
by investigators, the FFM intercorrelations from the relevant manual for their
FFM measure [typically a version of the NEO Personality Inventory-Revised
(NEO PI-R), Costa & McCrae, 1989, 1992b] were used instead. This introduces
a degree of imprecision, but the FFM intercorrelations in the manuals should
nevertheless be similar to the unreported correlations in the data from the
original articles (any imprecision will serve to attenuate the statistics in the
present study). In some cases, investigators reported correlations between
inventory scales and orthogonal factor scores for the FFM. A correlation matrix
of ones and zeros was therefore used to represent the FFM intercorrelations
instead of intercorrelations from manuals. After the R-squared values for the
individual scales were computed, the mean R-squared values for the scales in
each inventory were computed.
Variance in Other Measures Accounted for by the FFM

The mean percentages of variance accounted for by the FFM variables for the
scales in each inventory are provided in the “Mean Rsq.” column in Table 8.1.
These values have typically not been reported in previous work and they are
noteworthy because most are in the 30% to 50% range. It should be noted that
the mean values hide the fact that the R-squared values for some scales in an
inventory were small and others were large.
It should also be noted that 100% is not the proper target of the reference
point. The column of e2 values in Table 8.1 (which will be discussed in greater
detail below) indicates that most of the percentages of error variance in other
scales ranged between 15% and 25%. The FFM cannot be expected to account
for the error variances in other tests, which means that the percentages of
variance accounted for by the FFM should be revised upward. So, for example,
if the FFM accounts for 50% of the total variation in another measure that has a
15% error variance, then the more accurate percentage of variance is 50/(100 –
15) = 59%. When the FFM accounts for 50% of the total variation and the error
variance is 25%, the more accurate value is 67%. When the FFM accounts for
30% of the total variation and the error variance is 15%, the more accurate value
is 35%. And when the FFM accounts for 30% of the total variation and the error
variance is 25%, the more accurate value is 40%.
Table 8.1. Statistics for Data in Personality Inventory Manuals and for Associations
with the FFM
Personality Inventory Scale Intercorrelations Associations with the FFM
Inventory Data N Number h2 s2 Data N Mean Rv Rt
e2
Source of Source Rsq.
Factors
ACL Gough & 591 5 73 03 24 Craig et al. 147 39 86 91
Heilbrun (1998)
(1983)
Piedmont et 414 62 95 97
al. (1991)
BPI Maraun & 404 2 43 28 29 Costa & 109 23 86 86
Chrisjohn McCrae
(1995) (1992a)
Levin & 457 25 94 94
Montag
(1991)
Montag & 527 38 84 87
Levin
(1994)
CPI Gough 1000 3 65 02 33 McCrae et 348 33 74 92
(1987) al. (1993)
DOTS Angleitner 323 2 25 52 23 Angleitner 323 23 99 99
& &
Ostendorf Ostendorf
(1994) (1994)
EASI Angleitner 323 3 50 08 41 Angleitner 323 43 99 99
& &
Ostendorf Ostendorf
(1994) (1994)
EPPS Edwards 1509 5 52 24 24 Piedmont et 164 21 60 78
(1959) al. (1992)
EPPS-N Sherman 315 NA NA NA NA Piedmont et 166 32 77 81
& Poe al. (1992)
(1970)
GZTS Guilford 2465 3 49 32 19 McCrae 180 47 96 96
et al. (1989)
(1976)
IAS-R Wiggins 132 2 66 20 13 McCrae & 315 49 83a 99
et al. Costa
(1983) (1989a)
ISI Lorr & 303 4 47 34 19 Lorr et al. 236 43 50a 57a
DeJong (1992)
(1986)
McCrae & 115 39 89 91
Costa
(1994)
JPI Jackson 115 4 49 29 22 Paunonen & 86 35 76 81
(1976) Jackson
(1996)
MBTI Myers & 55,971 1 24 57 20 Furnham 160 44 74 76
McCauley (1996)
(1985)
MacDonald 161 43 89 89
et al. (1994)
MacDonald 48 62 66 66
et al. (1994)
McCrae & 201 50 85 85
Costa
(1989b)
McCrae & 267 52 88 88
Costa
(1989b)
MCMI-I Millon 978 3 79 06 15 Costa & 207 39 89 91
(1983) McCrae
(1990)
Lehne 99 45 93 93
(1994)
MCMI-II Millon 769 3 70 20 10 Costa & 62 31 67 71b
(1987) McCrae
(1990)
Hyer et al. 80 14 66 69b
(1994)
MCMI-III Millon 398 2 65 12 23 O’Connor 614 47 97 97
(1994) & Dyce
(1998)
MMPI- Dahlstrom 340 3 60 14 26 McCrae 274 29 87 88
Basic et al. (1991)
(1975)
MMPI-J Edwards 842 5 NA NA NA McCrae 141 24 60 83
& (1991)
Edwards
(1991)
MMPI-C Costa et 1576 3 43 42 15 Costa et al. 141 41 69 93
al. (1985) (1986)
MMPI- Morey et 475 2 62 11 27 Costa & 274 41 96 97
PD al. (1985) McCrae
(1990)
Trull (1992) 54 46 96 97
MMPI Harkness 2567 2 45 36 20 Trull et al. 57 29 91b 91b
PSY-5 et al. (1995)
(1995)
MPQ Church & 300 5 28 57 14 Church 575 34 86 94
Burke (1994)
(1994)
PAI Morey 1000 3 62 19 19 Costa & 114 29 91 91
(1991) McCrae
(1992a)
Montag & 583 42 77 90
Levin
(1994)
Montag & 286 46 92 95
Levin
(1994)
Morey 95 42 89 92
(1991)
PRF Jackson 1862 5 47 42 11 Costa & 296 37 91 94
(1984) McCrae
(1988)
SCL-90-R Levenson 1324 1 61 23 16 Smith & 109 19 99 99
et al. Snell (1996)
(1988)
SDS De Fruyt 934 3 44 45 12 De Fruyt & 934 21 85b 87b
(Dutch) & Mervielde
Mervielde (1997)
(1997)
SDS Holland 470 3 54 34 12 Schinka et 645 20 75 88
(1985): al. (1997)
females
Holland 297 2 51 37 12 Schinka et 389 16 63ns 75ns
(1985): al. (1997)
males
SNAP Clark 222 3 42 38 19 Clark 194 33 93 93
(1996) (1996)
TCI Cloninger 300 1 25 53 22 Cloninger 136 45 95b 95b
et al. & Svrakic
(1993) (1994)
16PF Caprara et 614 58 Barbaranelli 608 24 96 97
(Italian) al. (1993) & Caprara
(1996)
Note. Decimals are omitted; h2 = mean proportion of common variance; s2 = mean proportion of specific
variance; e2 = mean proportion of error variance; NA = data not available; Rv = congruence for varimax
rotations; Rt = congruence for targeted rotations; all congruences are significant at the .0001 level, except
“a” = p > .01; “b” = p > .001; “ns” = not significant; the full inventory names appear below, at the end of
this table.
Mean Rsq. = mean R squared; ACL = Adjective Checklist (Gough & Heilbrun, 1983); BPI = Basic
Personality Inventory (Jackson, 1989); CPI = California Psychological Inventory (Gough, 1987); DOTS-R
= Dimensions of Temperament Survey (Windle & Lerner, 1986); EASI = EASI Temperament Inventory
(Buss & Plomin, 1975); EPPS = Edwards Personal Preference Schedule (“N” = normative version;
Edwards, 1959); GZTS = Guilford–Zimmerman Temperament Survey (Guilford, Zimmerman, & Guilford,
1976); IAS-R = Revised Interpersonal Adjective Scales (Wiggins, Trapnell, & Phillips, 1988); ISI =
Interpersonal Style Inventory (Lorr, 1986); JPI = Jackson Personality Inventory (Jackson, 1976); MCMI =
Millon Clinical Multiaxial Inventory (Millon); MMPI = Minnesota Multiphasic Personality Inventory (“J”
= Johnson et al., 1984, scales; “C” = Costa et al., 1985, scales); MMPI-PD = MMPI Personality Disorder
Scales (Morey, Waugh, & Blashfield, 1985); MMPI-PSY-5 = MMPI Psychopathology Five (Harkness,
McNulty, & Ben-Porath, 1995); MPQ = Multidimensional Personality Questionnaire (Tellegen, 1985);
MBTI = Myers-Briggs Type Indicator (Myers & McCauley, 1985); PAI = Personality Assessment
Inventory (Morey, 1991); PRF = Personality Research Form (Jackson, 1984); SCL-90-R = Symptom
Checklist (Derogatis, 1983); SDS = Self-Directed Search (Holland, 1985); SNAP = Schedule for
Nonadaptive and Adaptive Personality Clark (1996); TCI = Temperament and Character Inventory
(Cloninger et al., 1994).
Although effect sizes in the 30% to 50% range, or, more accurately, in the
35% to 67% range, would be considered high and rare in most psychological
investigations, the values do not seem particularly impressive when compared to
100%. Although any particular color from the color spectrum can be perfectly
replicated by combining the three primary colors, scores on a personality
inventory subscale cannot be so precisely replicated by the FFM. This apparently
justifies doubts about the comprehensiveness of the FFM (e.g., Butcher &
Rouse, 1996, p. 100; Mershon & Gorsuch, 1988). Proponents of the FFM can
point to the considerable associations (percentages of variance accounted for,
correlations, or factor loadings) between personality inventory scales and
measures of the five factors, whereas critics can point to percentages of variance
accounted for that are noticeably lower than 100%.
The FFM is undoubtedly robust in the sense of accounting for noteworthy
portions of the variance in most other popular measures. But nonnegligible
portions of variance remain unaccounted for. Something is missing. Are there
major, hidden, non-FFM branches of the family tree that, if measured, would
raise the percentages of variance accounted for to the 100% range? The answer
to this question requires an understanding of the nature of dimensions (big
branches) in personality psychology data, and it requires a review of the
evidence for the robustness of the FFM at the level of dimensional structures.
The Nature of Dimensions in Personality Psychology Data

Is the unaccounted for variance in other measures due to systematic trait
dimensions that are not tapped by the FFM? The five dimensions resulted from
attempts to simplify the complex world of personality descriptors. Attempts to
extract a smaller number of dimensions from a large set of variables necessarily
involve a loss of information. Critics are therefore justified in believing that five
dimensions are not sufficient to capture the richness and subtleties of specific
personality variables or inventory scales. However, “completely accounting for”
and “precisely differentiating between” specific personality variables are not
goals that can be attained by data reduction models such as the FFM (Costa &
McCrae, 1995a, p. 218). Consider the following analogy. When the correlations
between personality inventory scales [such as the Minnesota Multiphasic
Personality Inventory (MMPI) basic scales] are factor analyzed and no other
correlations are included in the analyses, three or four factors typically emerge.
These factors describe or summarize the dimensions that underlie the scales, but
there will nevertheless be a loss of information and detail about specific scales in
this exercise. However, it would obviously be wrong to dismiss the obtained
scale factors as inadequate (as critics have dismissed the FFM) because of this
loss of information. The FFM cannot be expected to provide representations of
personality constructs that are substantially superior to those provided by the
data reduction (e.g., factor analytic) solutions for those constructs. Perfect
representations could be obtained only by extracting as many dimensions as
there are scales (e.g., in principal components analyses), but this would defeat
the purpose of data reduction models. The fact that the FFM contains one or two
dimensions more than the number that typically exists in personality inventories
suggests the potential for enhanced representations. However, the extra
dimensions have clearly not resulted in a complete accounting for every bit of
personality variance in every scale. The FFM will probably never achieve this
goal, and should not be expected to achieve this goal (Costa & McCrae, 1995a,
p. 218).
Concepts from the factor analysis literature can be used to make these points
more clear. Distinctions are often made between common, specific, and error
variance (Bernstein, 1988, p. 187; Gorsuch, 1983, p. 109). Common variance is
the portion of variance in a variable that is shared with other variables. Specific
variance is the portion of variance in a variable that is reliably measured but that
is unique to the variable. Error variance is also unique to a variable, but is not
reliably measured. The FFM as a comprehensive data reduction model should
capture the common variance in personality scales that accumulates to form
dimensions. However, the FFM will not also capture all the remaining specific
and error variances in personality scales and these portions of variance may be
substantial. Unfortunately, the total variation in personality inventories has not
been broken down and reported in these ways in previous work. Readers of the
literature must therefore not only discern, on their own, what portions of
variance are accounted for by the FFM, they must also somehow guess what
portions of common variance exist in personality inventories, given that they are
aware of the distinctions between common, specific, and error variance in the
first place.
Costa and McCrae’s position on this issue was made clear:
The claim that the FFM is comprehensive does not mean that it exhaustively
measures individual differences in personality, any more than a comprehensive
examination asks every single question a student should be able to answer on a topic.
What the model hypothesizes is that almost every personality trait is substantially
related to one or more of the five factors, and that any remaining traits … form a
miscellaneous category rather than covarying to define a sixth or subsequent factor.
(Costa & McCrae, 1995a, p. 218)
Trait variance that is not associated with the FFM is clearly believed to take the
form of specific variance. However, this clarification appeared in only a footnote
in one of McCrae and Costa’s papers. Although they have occasionally, briefly
added the “common variance” qualification (e.g., Costa & McCrae, 1990, p.
363; McCrae & Costa, 1995, p. 62, 1997, p. 509), many of their descriptions of
the FFM as “universal” and “truly comprehensive” have not been accompanied
by this qualification (e.g., Costa & McCrae, 1992a, p. 5; McCrae, 1989, p. 243;
1991, p. 399; McCrae & Costa, 1986, p. 1001, 1990, p. vi, 1991, p. 367). Critics
have therefore apparently assumed that “comprehensive” means that their
favorite personality constructs should be completely captured by the FFM, and
the debate has been ongoing. However, progress in the debate over the number
and nature of personality dimensions is not likely to be made when there is a
preoccupation with capturing specific scales. Progress is more likely to be made
by seeking answers to a more basic question that addresses the data reduction
nature of the FFM. Does the FFM capture the dimensions that exist in the scales
of other personality inventories, or do personality dimensions other than those
tapped by the FFM (“a different universe of psychological constructs,” Butcher
& Rouse, 1996, p. 100) exist in those inventories? Surprisingly, there have been
no quantitative evaluations of this issue in previous work.
At this point, it is useful to recognize what is required for a “dimension” to
exist and emerge in data analyses. Dimensions emerge when they summarize
substantial portions of variance in the data. Most personality inventories have
between 10 and 20 scales and the variation associated with any single scale is
not likely to be substantial in this context, unless it is shared with other scales.
Bits of scale-specific variance may help precisely describe individual scales, but
they will remain crumbs and will not be identified as dimensions unless the same
variation also appears on other scales and accumulates into a dimension (i.e.,
surpasses cut-off criteria that are used to identify dimensions). What seems to be
required to make progress in the current debate are ways of determining whether
the variation in personality inventories that is not accounted for by the FFM
forms “dimensions” of this kind.
The problem has been that identifying dimensions that are not associated with
the FFM is not a simple or obvious exercise. When the correlations between
personality inventory scales are factor analyzed, the solutions that emerge are
often ambiguous and are open to differing interpretations. This is because the
various instruments tap some of the five factors more than others. The solutions
are blended or aligned differently than FFM solutions and there is too much
subjectivity involved in determining whether factors are from the FFM or not.
Alternatively, proponents of the FFM have reported correlations between the
five factors and personality inventory scales, or they have reported results from
joint factor analyses. Proponents are impressed by the magnitudes of the
associations, and by how scales are spread across the five factor space. They also
claim that no other factors seem to exist in the data. However, critics may remain
unimpressed by the same numbers. It thus seems that reports of associations
between the five factors and other personality tests, for all to see and judge with
their own eyes, have not been as convincing or conclusive as Costa, McCrae,
and others probably expected. Quantitative indices of the degree to which the
FFM captures the dimensions in popular inventories are required, yet none has
so far been reported in the literature.
Robustness at the Dimensional Structure Level

O’Connor (2002) reported findings from relatively stringent quantitative tests
that were designed to help take a small step forward in this debate. The factor
structures that exist in previously published associations between the FFM and
other inventories were identified and then statistically compared to the factor
structures that exist in the inventory scale intercorrelations reported by the test
developers (or other independent sources). The tests were thus simple, familiar
comparisons of two sets of factor loadings. However, the tests were unique and
perfectly suited for the robustness (comprehensiveness) debate because one set
of the factor loadings was based solely on the associations between the FFM and
other inventory scales. Specifically, interbattery factor analyses (Tucker, 1958)
were conducted on previously published data to determine whether the factor
structures in popular personality inventories could be replicated using only the
known associations between the inventory scales and FFM variables. The
statistical method (see Finch, Panter, & Caskie, 1999, for an overview) is easily
understood by reference to an example. Figure 8.1 portrays a 10-by-10 matrix of
correlations between FFM variables and the 5 MMPI PSY-5 scales.
Figure 8.1. Partitioned supermatrix of correlations between FFM scales and MMPI PSY-
5 scales.
There are four obvious sections of the “supermatrix” in Figure 8.1: the 5 × 5
submatrix of FFM intercorrelations, the 5 × 5 submatrix of MMPI scale
intercorrelations, the 5 × 5 submatrix of correlations between the FFM variables
and the MMPI variables, also known as the “interbattery correlations,” and the 5
× 5 transpose of the interbattery correlations.
Factor analytic techniques are typically applied to complete correlation
matrices, that is, to square matrices that contain all of the correlations between
variables. For example, regular factor analyses could be conducted on the above
supermatrix (which would be a joint factor analysis of the FFM and MMPI
variables), or on the 5 × 5 matrix of FFM intercorrelations, or on the 5 × 5
matrix of MMPI scale intercorrelations. In contrast, interbattery factor analysis
can be conducted on less than complete correlation matrices, such as the above 5
× 5 matrix of interbattery correlations. The focus on such partial or
“submatrices” also involves an adjustment in the interpretations of the results.
Regular factor analytic procedures reveal the dimensions that underlie the
variables in complete correlation matrices, whereas interbattery factor analyses
reveal only the dimensions that underlie the associations between the two sets of
variables. The variance that is shared or is common to two sets of variables (in
this case, the variation that is contained in interbattery correlations) may or may
not provide an adequate representation of the variation in the original two sets of
variables. The factor structure that exists in the interbattery correlations will
represent the structure that exists in one set of the original variables (e.g., the
inventory scale intercorrelations) only if the variables in the other set (the FFM
intercorrelations) are sufficiently comprehensive to capture its structure. If the
variables in one set are not comprehensive, then the structure in the interbattery
correlations will not greatly resemble the structure in other set of variables.
The exclusive focus in interbattery factor analysis on the variation that is
shared among two sets of variables thus provides a uniquely useful tool for
assessing the comprehensiveness of the FFM, as it appears in other inventories.
The factor structures that are identified in interbattery correlations are direct,
unencumbered, and pure representations of the ability of the FFM to capture the
dimensions in other inventories. Once identified, these structures can be
validated by comparisons with independently obtained factor structures for the
inventory scales.
The usefulness of the technique is perhaps best illustrated through contrasts
with the procedures that have typically been used in the literature. Researchers
have most commonly reported only the interbattery correlations. Some of the
coefficients in these matrices may seem large, but the overall picture is
ambiguous with regard to the comprehensiveness debate. This is because readers
are unable to determine whether the reported interbattery correlations capture the
dimensions that exist in the other personality inventories. Furthermore, the
coefficients for some inventory scales in interbattery correlation matrices are
often modest, which prompts readers to suspect that other dimensions exist in
the data. Interbattery factor analysis can provide direct, unambiguous
information on whether the data in interbattery correlations are comprehensive
with regard to the factor structures that exist in other inventories.
Researchers have also sometimes reported the results of joint factor analyses
of FFM and other inventory scales. This analytic technique has been much less
common than the reporting of interbattery correlations, but it is also problematic.
In joint factor analysis, information about battery membership is disregarded and
factors that are specific to each battery are confounded with between-battery
factors (Cudeck, 1982, p. 48; see also Panter, Tanaka, & Hoyle, 1994, p. 119).
“The interbattery model, by virtue of its ability to separate battery-specific
(method) factors from interbattery (trait) factors, frees the researcher from
making interpretations based on what is essentially method variability” (Finch et
al., 1999, p. 435). In summary, joint factor analyses are slightly more
informative than simple interbattery correlations, but they leave readers
wondering about method variance contamination. The reported loading matrix
results are also ambiguous because descriptive information about associations
between two sets of scales is provided without precise statistical indices of the
degree to which the variables in one set capture the structure in the other set.
In the O’Connor (2002) review, interbattery factor analyses were conducted
on previously published interbattery correlations involving FFM variables and
the scales of other inventories. The results were then compared to the results of
factor analyses of inventory scale intercorrelations obtained from independent
sources (typically from the inventory manuals published by the test developers).
The tests were therefore stringent attempts to provide answers to a simple
question: To what extent do factor structures based solely on associations with
FFM variables resemble the factor structures in the original personality
inventory scale intercorrelations? High degrees of resemblance will suggest that
the FFM “captures” the structural dimensions in other inventories, whereas low
degrees of resemblance would indicate that the FFM does not sufficiently
capture the dimensions in other inventories. In other words, if the structures
based solely on the variance that is shared with the FFM are highly similar to the
factor structures that exist in the real inventory scale intercorrelations, then
untapped dimensions probably do not exist in the data and the FFM can be said
to capture the dimensions in other scales. These predictions were tested using
data from the published literature for 28 personality inventories.
In summary, controversy has been generated by strong, often unqualified
statements regarding the comprehensiveness of the FFM, which have led critics
to focus on the degree to which the FFM completely captures specific
personality scales. Greater attention must be given to the distinctions among
common, specific, and error variances, and data on these important sources of
variation in personality inventories have not been reported or discussed in FFM
debates. Readers have also been forced to conduct eyeball assessments of the
variation in personality scales that is accounted for by the FFM, because the
actual numbers have typically not been reported in previous work. These
percentages of variance are clearly less than 100%, yet the seriousness of the
apparent deficit cannot be gauged without knowledge of the (unreported)
portions of common variance that exist in personality inventories. There have
also been no quantitative, statistical tests of the comprehensiveness of the FFM.
Reanalyses of published data were therefore conducted to provide (1) the
proportions of common, specific, and error variance that exist in the inventory
scales; (2) the percentages of variance in inventory scales that is accounted for
by the FFM; and (3) quantitative assessments of the degree to which the FFM
captures the dimensions that exist in other inventories.
Data Sets and Analytic Methods

The proportions of common, specific, and error variance in personality
inventory scale intercorrelations were estimated from the matrices reported by
the original test developers. These matrices were typically reported in the test
manuals, but sometimes they appeared in other sources (see Table 8.1 for
references). Estimations of the specific and error variance components required
reliability estimates for the scales, and the internal consistency values reported
by the test developers were used for these purposes. In some cases, data were
available for both males and females, or for clinical and nonclinical samples.
The analyses were conducted on whatever data set was larger and
demographically most similar to the samples from which the interbattery
correlations were obtained. The interbattery correlations analyzed in this review
were from previous reports of correlations or factor loadings from joint analyses
of FFM variables and other personality inventory scales, as described above (see
Table 8.1).
Number of factors. Parallel analyses (Longman, Cota, Holden, & Fekken,
1989) were conducted to determine the number of factors in the personality
inventory scale intercorrelations published by the tests developers (typically in
the test manuals). A total of 1,000 random data analyses were conducted for
each correlation matrix, and the 95th percentile eigenvalues were used to
determine the number of factors.
Common, specific, and error variance. The proportions of common
variance in the personality inventory scale intercorrelations published by the test
developers were derived from common factor analyses (CFAs). Principal axis
factor analyses were conducted on the correlation matrices using the squared
multiple correlations as the initial communality estimates on the main diagonal
(Bernstein, 1988, p. 187; Gorsuch, 1983, p. 107). The means of the final
communalities were then computed, thus providing estimates of the average
proportions of common variance in the inventory scales. The mean proportions
of error variance were computed by subtracting the scale reliabilities from one,
and finding the average of the remainders. The mean proportions of specific
variance in the scales were computed by subtracting the communality estimates
from the reliabilities and finding the average of the remainders (Gorsuch, 1983,
p. 109).
Interbattery factor analyses and varimax rotation. The interbattery factor
analysis technique described by Tucker (1958) was used in the present analyses.
The computations for Tucker’s method are performed on just the interbattery
correlations. Tucker’s procedure yields a matrix of loadings for each set of
variables in the analyses, and the interbattery loading matrices for the personality
inventory scales were orthogonally rotated to a simple structure.
Principal axis factor analyses and varimax rotation. Principal axis factor
analyses (PAFs) with varimax rotation were then conducted on the correlation
matrices reported by the test developers. Principal components analyses were
used in the O’Connor (2002) analyses. PAF was used for the present analyses in
order to confirm that the findings are essentially the same and do not vary across
these two analytic methods.
Congruences between factor solutions. The Tucker–Burt–Wrigley–Neuhaus
congruence coefficient was computed to index the degree of similarity between
the varimax-rotated loading matrices (see Guadagnoli & Velicer, 1991, and ten
Berge, 1986, for reviews). To save space, the Tucker et al. congruence
coefficient was computed for all corresponding elements in the matrix
comparisons (this procedure was recommended and used by McCrae,
Zonderman, Costa, Bond, & Paunonen, 1996). The result is an overall fit
coefficient, based on all factor loadings, instead of multiple congruence
coefficients each based on the loadings for single factors. The overall
coefficients are roughly equivalent to the mean of the factor congruence
coefficients and are efficient summaries of the total degree of fit between two
loading matrices (the individual factor congruences provide little additional
useful information for the present purposes).
Targeted rotations. Targeted rotations of the PAF loading matrices were also
conducted because the structure in a correlation matrix can be viewed from
different vantage points and the views provided by popular rotational methods
(e.g., varimax) are not necessarily the only or the best vantage points (Watson,
Clark, & Harkness, 1994, p. 20). Procrustes targeted rotations are considered
most appropriate for testing complex models and for testing models that are not
consistent with a simple structure (McCrae & John, 1992, p. 189). Targeted
rotations were used to answer the following question: When the interbattery
factor structures are examined from the vantage point that maximizes
congruence with the structures in the inventory scale intercorrelations, do the
two structures resemble one another?
Tests of statistical significance. For every comparison of an interbattery
factor structure with a personality inventory factor structure, 10,000 random data
analyses were performed to assess the probability of the obtained congruences
emerging solely on the basis of chance. This procedure was described by
Paunonen (1997) and is a random data permutation test of statistical significance
(Edgington, 1995). This Monte Carlo procedure was also used to assess the
statistical significance of the congruence coefficients for the varimax-rotated
(non-Procrustes) factor solution comparisons.
Overview of the methods and illustrative data. The primary steps in the
analyses are easily summarized. For each inventory:
1. Compute the mean proportions of common, specific, and error variance in the scale
intercorrelations reported by the test developers. Parallel analysis was used to determine
the number of factors.
2. Conduct an interbattery factor analysis on the interbattery correlations.
3. Perform a varimax rotation on the interbattery factor loadings, and compute the
overall congruence between the rotated loadings and the varimax-rotated loadings derived
from a PAF of the scale intercorrelations reported by the test developers.
4. Rotate the interbattery factor loadings to maximum congruence with the varimax-
rotated loadings derived from a PAF of the scale intercorrelations reported by the test
developers.
Illustrative data for the Guilford–Zimmerman Temperament Survey (GZTS;

Guilford, Zimmerman, & Guilford, 1976) are provided in Table 8.2. Parallel
analysis revealed that there were three factors in the GZTS scale
intercorrelations published by Guilford et al. (1976). The varimax-rotated
loadings are displayed in the left-most panel of Table 8.2. The Pearson
correlations between the GZTS scales and the FFM variables, derived from data
published by McCrae (1989), are presented in the second panel. Interbattery
factor analysis was performed on these interbattery correlations, and the
varimax-rotated loadings are presented in the third panel of Table 8.2. The
overall congruence between the rotated interbattery factor loadings (Panel 3) and
the rotated loadings from the data of Guilford et al. (Panel 1) was very high (Rv
= .978). The interbattery factor loadings were then rotated to maximum
congruence with the loadings from the data of Guilford et al. (Panel 1), and these
targeted rotation loadings are displayed in Panel 4. The overall congruence
between these loadings and those in Panel 1 was once again very high (Rt =
.981).
Results
Number of factors. The results of the parallel analyses for determining the
number of factors in the personality inventory correlation matrices are provided
in Table 8.1. There were five or fewer factors in every case. The parallel
analysis-based numbers of factors for the inventories were used in all subsequent
analyses in the present study.
Variance components. The estimated proportions of common, specific, and
error variance in the personality inventories, based on data published by the test
developers, are also reported in Table 8.1. The proportions of common variance
were generally in the .35 to .75 range. The variation in portions of specific
variance across inventories is more striking, ranging from close to zero for some
inventories and as high as .57 for others. The mean percentages of variance
accounted for by the FFM variables (mean R-squared values) for the scales in
each inventory (see Table 8.1) are in the 30% to 50% range. The values do not
seem particularly impressive when compared to 100%, but they seem much
stronger when they are more properly compared to the proportions of common
variance in the scales. The mean proportion of common variance was 50.1%,
whereas the mean proportion of variance accounted for by the FFM was 38.8%.
Congruences between factor solutions. To examine whether the FFM
captures the dimensions that exist in other personality inventories, the
interbattery factor loading matrices were compared to the loading matrices for
the original inventory scale intercorrelations. The congruence coefficients for the
varimax and targeted rotations are listed in Table 8.1, and they were generally
very high (over .90) and statistically significant.
The weakest congruences to emerge were for the Edwards Personal
Preference Schedule (EPPS; Edwards, 1959) and the Jackson Personality
Inventory (JPI; Jackson, 1976). However, in all other cases, whenever a
relatively low overall congruence coefficient emerged for an inventory based on
one set of data, high congruence coefficients were obtained for either other
versions of the same inventory or for the same version of the inventory based on
analyses of other data (the reasons for the problematic evidence for the EPPS
and JPI were described by O’Connor, 2002). For example, poor replication
results emerged for the Millon Clinical Multiaxial Inventory (MCMI-II; Millon,
1987), perhaps because the interbattery correlations for this inventory were
based on small samples (N = 62 and N = 80). Yet the congruence coefficients for
the MCMI-I (Millon, 1983) and MCMI-III (Millon, 1994) were quite high. The
factor structure in the Johnson, Null, Butcher, and Johnson (1984) alternative
MMPI scales was not well replicated, yet the factor structures in the MMPI basic
scales and in the Costa, Zonderman, McCrae, and Williams (1985) alternative
MMPI scales were well replicated. For the Self-Directed Search (SDS; Holland,
1985), the factor structure for the male data was not well replicated, but the
congruences for the female data were reasonably strong.
Table 8.2. Principal Axis Factor Loadings and Interbattery Correlations from the
Analyses for the Guilford Zimmerman Temperament Survey
Panel 1 Panel 2 Panel 3 Panel 4
Varimax-Rotated Interbattery Correlations Varimax- Procrustes-
Factor Loadings Derived From McCrae Rotated Rotated
From (1989) Interbattery Interbattery
Intercorrelations Factor Factor
Published By Loadings Loadings
Guilford et al.
(1976)
1 2 3 N E O A C 1 2 3 1 2 3
GZTS Scales:
General −09 −63 −06 −06 49 07 −47 29 02 −44 −07 −15 −61 −07
Activity
Restraint 17 15 −65 −13 −35 12 18 61 19 07 −85 19 14 −64
Ascendance 14 −65 −05 − 28 62 17 −21 16 11 −93 −05 07 −66 −06
Sociability 26 −49 24 −29 69 −09 06 −06 20 −66 08 21 −52 24
Emotional 50 −16 01 −59 15 −09 05 14 67 −27 04 49 −21 01
Stability
Objectivity 60 −12 −00 −67 14 −08 15 13 85 −23 08 58 −18 −00
Friendliness 51 31 −19 −44 −24 −04 49 13 63 19 −18 54 25 −19
Thoughtfulness −13 −15 −64 13 05 58 24 29 −15 −11 −48 −14 −13 −64
Personal 53 −00 −04 −52 08 −14 29 15 61 −08 −04 53 −06 −03
Relations
Masculinity 20 −02 04 −39 −14 12 −22 −08 37 −06 09 −15 −61 −07
Overall Congruences Between Panel 1: The Varimax-Rotated Factor Loadings from Guilford et al. (1976)
and Panel 3: The Varimax-Rotated Interbattery Factor Loadings: Rv = 96 and Panel 4: The Procrustes-
Rotated Interbattery Factor Loadings: Rt = 96
Discussion
In sum, the factor structures that exist in the scales of many popular
personality inventories can be closely replicated using data derived solely from
scale associations with the FFM. The basic dimensions that exist in other
personality inventories can thus be considered “well captured” by the FFM. If
we are going to be concerned with percentages of variance, then distinctions
should be made between the common, specific, and error variance components
in personality scales. The percentages of variance accounted for by the FFM
should be evaluated in relation to the proportions of common variance and not in
relation to the total variance in personality scales. These distinctions have not
been made in previous work, and relevant data have not been reported. The data
from the present review indicate that the portions of scale variance accounted for
by the FFM (the mean across inventories was 38.8%) are substantial when
evaluated in relation to the portions of common variance that exist in most
personality inventories (mean = 50.1%), although there is still room for
improvement (see Table 8.1).
More importantly, the preoccupation with percentages of variance and
individual scales seems unwarranted when it is recognized that the FFM is a data
reduction model designed to simplify the complex world of personality
descriptors. The primary question is not whether the FFM completely accounts
for, or differentiates between, specific personality constructs, but whether the
FFM captures the dimensions that exist in other personality constructs. This
question has not been quantitatively evaluated in previous work. The present
findings are much less ambiguous than those of past studies, and indicate that the
FFM generally does capture the dimensions in other inventories. Block (1995)
identified biases and methodological problems with previous research on the
FFM, but the present findings indicate that the consequences of these biases
could not have been severe. The FFM may not always be fully recovered in
other personality inventories, but it can apparently recover the dimensions in
most other inventories.
Given that factor solutions usually involve a loss of information, how can
differentiation and variance accounted for be increased? The most obvious
solution would be to lower the stringency of dimensional cut-off criteria to
increase the number of dimensions permitted in the analyses (a practice that is
clearly not recommended). The addition of minifactors would raise an
interesting question for further research: Do the minidimensions that exist in
personality inventories resemble the (excluded) minidimensions that exist in the
lexical and factor analytic data that led to the FFM? In summary, the FFM
cannot be expected to provide representations of personality constructs that are
noticeably superior to the representations provided by the data reduction
solutions for those constructs. Facet-level improvements to FFM measures may
help reach these ceilings, but they will not help surpass them. Further
improvements require consideration of the “crumbs” from data reduction
models, and these crumbs in the factor solutions for personality inventory scales
may or may not be similar to the crumbs that have been discarded by FFM
researchers.
We can readily see how a neglect of these considerations, combined with
some strong statements by McCrae and Costa, have fueled the ongoing debate.
Statements such as “the five factor model offers a universal and comprehensive
framework for the description of individual differences in personality” (McCrae
& Costa, 1986, p. 1001), and “by assessing traits from each of the five factors,
the clinician can obtain a comprehensive portrait of the client’s personality”
(Costa & McCrae, 1992a, p. 5), should be qualified. The FFM captures the
primary dimensions, but not all the variation, in other personality constructs. The
absence of this qualification, combined with the absence of quantitative tests of
whether dimensions have been captured, has unfortunately led some readers to
focus on specifics, such as the inability of the FFM to capture particular scales
completely. For example, Butcher and Rouse (1996) and Ben-Porath and Waller
(1992) dismissed the FFM because it is too broad and does not provide enough
specific information for the needs of clinical applications (i.e., relative to clinical
personality inventories such as the MMPI). However, data reduction models will
always be broad and cannot be expected to provide such specific information.
The factors structures that exist in clinical personality measures do not provide
such specific information either, yet they are well captured by the FFM. It would
thus seem that progress in the FFM debate could be made if proponents toned
down and qualified some of their strong statements (as they sometimes have,
e.g., Costa & McCrae, 1990, p. 363, 1995a, p. 218; McCrae & Costa, 1995, p.
62, 1997, p. 509; McCrae & John, 1992, p. 177), and if critics realized that the
FFM is about dimensions and not about scores on specific measures. Progress
could also be made if more recognition and attention were given to the non-
FFM, “nondimensional,” scale-specific variance in personality scales that is of
interest to clinicians and FFM critics.
The scales within many personality inventories vary in breadth of
measurement. Some scales tap broad dimensions of individual differences
(neuroticism, extraversion), whereas other scales tap narrow, specialized
constructs (e.g., self-harm, intellectual efficiency). These levels of measurement
differences were ignored in the present mathematical focus on replicating the
factor structures that exist in other measures. However, it is possible for a scale
in a personality inventory to assess a broad dimension of individual differences
and, at the same time, not show up in the factor structure for the inventory. This
could occur if no other scales in the inventory tapped into the same dimension,
causing the variance in the broad scale to appear only in the form of scale-
specific variance. The broad dimension, perhaps unrelated to the FFM, would
thus have been missed in the present analyses. Understanding the naturally
occurring factor structures in personality inventories has nevertheless been a
long-standing concern to researchers. The present study focused on these
structures, and neglected the possibility of broad dimensions simultaneously
existing and not showing up in factor analyses.
Directions for Further Research at the Scale Level

Facets of the FFM
If measures of the FFM were further improved, what likely findings would
emerge for the comprehensiveness debate? Most discussions of this issue focus
on facets of the five dimensions (Costa & McCrae, 1995b; Widiger & Trull,
1992). Additions and refinements at the facet level of measurement would likely
further increase the degree to which the FFM captures the dimensions in other
inventories, and increase the degree of differentiation between personality
constructs. However, there might be only modest increases in variance
accounted for in individual scales. The facet-level analyses that have been
conducted have helped differentiate between inventory scales, but they have
apparently not dramatically increased the variance accounted for (e.g., Church,
1994; Costa & McCrae, 1988; Dyce & O’Connor, 1998; Hahn & Comrey, 1994;
McCrae et al., 1993; Montag & Levin, 1994; Trull et al., 1995). The FFM is a
data reduction model and, for most inventories, the ceiling that can be reached
by the FFM is roughly the amount of variance accounted for in the factor
solutions for the original scale intercorrelations.
Validity of the Non-FFM Portions of Variance in Other Measures

As described above, the FFM typically accounts for between 30% and 50% of
the total variance, or between 35% and 67% of the reliable, nonerror variance, in
other measures. Ignoring their sizes, do the portions of variance that are not
accounted for by the FFM matter? This is a question about the validity of the
unaccounted for variance and the question remains (generally) unanswered and
open for investigation. The research and analytic methods required to answer the
question are simple. Measures of the FFM could be included in studies of the
concurrent or predictive validity of non-FFM measures. The question then
becomes how much variance in the relevant outcome variables is explained by
the non-FFM measures after the variation that is accounted for by the FFM is
removed. It would be difficult to argue that a non-FFM measure is unique or
important if the FFM dimensions account for most of the validity evidence for
the measure, a result that is empirically possible. If the non-FFM measure
accounts for a nonnegligible portion of the variance in relevant outcome
variables, then its specific variance clearly matters and requires more careful
attention. A broad, quantitative review of the validity of the non-FFM portions
of variance in other measures is required, but I suspect that sufficient relevant
data are not available from previously published reports.
Locations in Five-Dimensional Space

The inherent potential of the FFM as an integrative, organizational family tree
for other personality tests has not been fully realized. The many reports of
associations between individual scales from other measures and the FFM remain
scattered and unintegrated. One consequence is that the field remains unaware of
which sections of the family tree are full and which sections are unpopulated.
For simplicity, assume that the five dimensions are orthogonal. This means that
the space in which people and other personality tests exist is five dimensional.
What needs to be known are the relative population densities of the many
regions of this space.
The point is perhaps more clearly made via comparisons with a two-
dimensional model, such as the interpersonal circumplex. The two orthogonal
dimensions of dominance–submissiveness and friendliness–hostility together
form a two-axis plot on which all interpersonal traits and behaviors can
presumably be located. Other interpersonal traits are just blends of the two
interpersonal dimensions. This simply means that other traits are vectors that run
through the origin of the two-dimensional space, with each vector being defined
by its weight or angular location with reference to each axis. The same principles
apply just as well to five-dimensional spaces. The interpersonal circumplex can
be described as having four broad regions (quadrants) stemming from the origin
based on the two axes. When the five-dimensional space of personality is
similarly broken up into high–low chunks based on its axes, there are 32 regions
(e.g., + N + E + O + A + C, + N + E + O + A–C, +N + E + O–A + C, + N + E +
O–A–C). These broad regions exist even though we are not able to visualize
them (we are unable to visualize our own family tree!). Non-FFM personality
measures can be thought of (if the FFM is indeed robust and comprehensive) as
vectors that run through the origin of the five-dimensional space. The vectors are
like FFM regression equations. A more detailed presentation of important
geometric properties of the FFM was provided by O’Connor and Dyce (2001) in
a report on personality disorders.
What is needed is a review that identifies the regions of the FFM space that
have many or few vectors. The universe of 32 regions is perhaps too crude, but it
would be a good start (also, given that the vectors run through the origin and are
bipolar, the focus could be on 16 regions instead of 32). Regions that do not
have vectors based on non-FFM measures are nevertheless still captured by the
FFM measures. Scores on such vectors could easily be created. Regions left
blank by other measures would raise questions about the reasons for the neglect,
and about the validity or usefulness of the respective regions. Having a full index
of the FFM vector locations would also be useful in evaluating new personality
measures. Does a new measure merely give us another vector through the same
region that has been assessed by previous measures? In what way is a measure
new? It is difficult to answer such questions without the full index. Fortunately,
the data for creating such an index already exist in the coefficients provided in
previously published studies.
Corresponding important questions can be asked about the locations of
people, rather than scales, in the five-dimensional personality space. The five
dimensions are presumably orthogonal and have normal distributions, but we do
not know if people are evenly spread out or dispersed in this space. Some
regions may be more populated with people than others. Do homogeneous
groupings of people exist? If so, how many groupings are there and where are
they located? Are there regions with people but with no vectors (personality
tests)? Are there personality tests (vectors) through regions that have few
people? Are there groupings of persons with distinct forms of psychopathology
for which there are no personality tests (vectors)? Large, representative data sets
would be required to answer such questions, along with creative statistical
methods for revealing the population densities of different regions of the family
tree.
The Five Dimensions as Moderator Variables

There is yet another way in which the FFM is robust. It is very common for
researchers who are primarily concerned with nonpersonality questions to
wonder whether the effects they are investigating depend on personality or
individual difference variables in some way. Does the effect of an experimental
manipulation or a particular kind of psychotherapy vary depending on the
participants’ personalities? This kind of question commonly arises in research
proposals by students, by colleagues, and in grant applications. What personality
measure should be used? A common strategy is to be safe, that is, be
comprehensive and use a measure of the FFM dimensions. But although the use
of measures of the FFM as potential personality moderator variables is an
undoubtedly robust phenomenon, the analyses that are typically conducted once
the data are collected are crude. This is another case of the potential of the FFM
as an integrative model not being fully exploited.
Specifically, the problem is that the analyses of the FFM personality variables
as moderators are typically conducted on just one dimension at a time.
Researchers will test for treatment by personality variable interactions in the
prediction of outcomes, and they will do so using each FFM dimension in turn,
by itself. The full FFM space has five dimensions, 32 broad regions, and many
hundreds of possible vectors running through it. Every vector is a possible
moderator. When researchers assess the moderating role of just one dimension at
a time, they are testing just one of the many possible vectors; that is, they are
testing only the vector that corresponds to the FFM axis in question. When
researchers run the search for interactions five times, that is, for each dimension
in turn, they are examining just five of the many hundreds of possibilities. In
many data sets, it seems likely that the largest personality moderator effect is for
a FFM vector that is a particular blend of the FFM dimensions. In other words,
researchers measure the FFM dimensions in their work in order to be careful and
comprehensive, but they then use analytic methods that do not permit the
strongest moderator effects to be found in their data.
Fortunately, there is a solution to this problem. A simple but crude and time-
consuming approach would involve compiling a long list of possible vectors and
their FFM coordinates, computing participants’ scores on each vector, and then
sequentially testing each vector for an interaction. A more precise and elegant
approach is to use an optimization routine, such as the FMINCON procedure in
MATLAB or corresponding routines in R or SAS. There are three simple steps:
(1) convert the scores on the FFM dimensions to orthogonal factor scores; (2)
specify the function to be optimized; and (3) specify the constraints.
The function to be optimized is the f-squared value for the interaction term in
a moderated regression. F-squared is the effect size for an interaction over and
above the main effects. It is the proportion of systematic variance accounted for
by the interaction relative to the unexplained variance in the outcome variable
(Aiken & West, 1991, p. 157). The optimization routine scans through the FFM
space looking for the strongest interaction, represented by the f-squared effect
size, while respecting the constraints placed on the search.
The constraints that are placed on the optimization routine pertain to the
coordinates of possible FFM vectors. The optimization routine must be provided
with geometrically meaningful bounds for the coordinates of the FFM vectors.
The boundary values that can be used in computations are the cosines of the
angles, which are 1.0 for zero degrees and –1.0 for 180 degrees. One further
constraint must be placed on the model: The sum of the squares of the cosines
for any vector must equal one (Green & Carroll, 1976, p. 87). This constraint is a
geometric necessity. A vector cannot have just any combination of cosines
between 1.0 and –1.0 on the various dimensions.
This scanning-optimization procedure can be conducted for searches through
the entire FFM space and through specified regions of the FFM space. The
strongest pattern that exists for the whole space does not preclude the existence
of different, weaker patterns in other regions. The FFM space can be divided
into broad regions based on either high (positive) or low (negative) values for
each dimension. Five dimensions, each with two regions (high and low), result
in 32 possible combinations or regions, as described above. The searches for
interactions within each of these FFM regions involve simple adjustments to the
upper and lower boundary constraints on the FFM weights. Searches within the
high or positive side of a dimension involve constraining the weights for the
dimension to fall between zero and 1.0, and searches within the low or negative
side of a dimension involve constraining the weights to fall between zero and –
1.0.
Only minimal time and effort are required to set up these analyses and the
potential payoff is great. The procedure will immediately reveal the FFM vector
that has the strongest moderating effect in a data set. Researchers no longer need
to be puzzled by having used personality moderator variables in their data
collections and then finding nothing or wondering if the data have been
thoroughly analyzed. I have used the optimization method in two previous
studies. One study focused on the FFM as the universe of personality moderators
of the validity of measures of social desirability responding (O’Connor, 2006).
The other study focused on the FFM as the universe of personality moderators of
the effects of parenting styles on adolescent psychopathology (O’Connor &
Dvorak, 2001). Very many moderated relationships were found in both
investigations, and almost none of the effects was for an FFM axis by itself.
Most of the interactions in the data sets would have been missed by searches for
interactions that proceeded one FFM dimension at a time.
It is also possible, and easier, to use the FFM in tests of theoretically or
empirically based a priori predictions of moderated effects. For example, a
researcher may suspect that a particular non-FFM measure could moderate the
effect of a treatment or contextual variable on an outcome, but the researcher
may have used measures of the FFM and not a direct measure of the (FFM)
vector in question for the data collection. Scores on this vector could
nevertheless be created from the FFM scores on the basis of theory or previous
research.
Two examples were provided in the study by O’Connor and Dvorak (2001),
who had scores on the FFM, parenting styles, and problem behaviors for a
sample of 402 adolescents. The correlations between callous–unemotional traits
and the FFM dimensions that were reported in a separate investigation by
Harpur, Hart, and Hare (1994) were used to produce regression equation weights
in the creation of proxy scores for callous–unemotional tendencies (an aspect of
psychopathy). In the O’Connor and Dvorak (2001) data set, males who scored
higher on the FFM callous–unemotional vector displayed relatively high levels
of delinquency regardless of the degree of maternal control they received. In
contrast, males who were low on the callous–unemotional FFM vector were
delinquent only in the context of excessive maternal control.
Similarly, proxy scores for personality “resiliency” were produced using the
correlations between ego resiliency and FFM scores that were reported by Huey
and Weisz (1997). Numerous interactions between the FFM proxy resiliency
scores and parenting styles in the prediction of adolescent problems emerged in
the O’Connor and Dvorak (2001) data. For example, males with high proxy
resiliency scores were low in aggression regardless of the degree of maternal
harshness they received. In contrast, males with low proxy resiliency scores
were increasingly aggressive as maternal harshness increased. The obtained
interactions were theoretically meaningful and consistent with previous findings
that used specific, non-FFM measures rather than FFM proxy scores. Such
findings suggest that the FFM may indeed capture the high validity portions of
variance in non-FFM measures.
Directions for Further Research at the Dimensional Structure

Level
An Unused Statistical Method for Clarifying Dimensional Structures
An unused statistical method can provide enhanced depictions of the
associations between FFM and non-FFM measures. As described above, when
researchers report Pearson correlations between FFM and non-FFM measures,
readers are often left wondering if something is missing. The values of the
correlations are rarely very high, and it is impossible to determine whether
additional dimensions would help capture more variance. Alternatively, when
the loadings from joint factor analyses of the FFM and other measures are
reported, the factors may not directly align with the FFM dimensions, rendering
the results difficult to interpret.
ten Berge, Kiers, and Commandeur (1993) described a factor analysis
procedure that can assist in such situations. The correlations between the FFM
and non-FFM measures can be factor analyzed in relation to a researcher-
specified target matrix of fixed and free-to-estimate loadings. For example,
assume a researcher had data for the five FFM variables and for 10 non-FFM
scales. To more clearly determine where the 10 non-FFM variables are located
in the FFM space, the loadings in a five factor target matrix could be arranged so
as to force the five factors to align with the FFM dimensions. This would be
accomplished by using 1s and 0s for the FFM loadings on the five factors. The
loadings for the 10 non-FFM variables in the analysis would be left free and
would be estimated by the factor analysis procedure. This analysis could then be
followed by a second run that involved a sixth factor on which the FFM
variables had 0s for loadings and with the loadings for the non-FFM variables on
all six factors left free to be estimated by the procedure. It would then be clearer
as to whether the non-FFM variables fall within the space that is defined by the
FFM dimensions.
The Search for Missed Additional Dimensions

Do one or more additional dimensions of personality exist that have somehow
been missed in the many years of lexical and personality inventory research?
The consistent power and effectiveness of the FFM in capturing the dimensions
that exist in other measures, as reviewed above, make it difficult to imagine that
additional dimensions have been missed all of this time. Yet claims of additional
dimensions are sometimes made. For example, Ashton and Lee (2007) claim that
honesty–humility is a major, previously neglected, sixth dimension of
personality. Establishing the existence of additional dimensions requires at least
two categories of empirical evidence. First, there must be validity evidence
(concurrent or predictive), above and beyond the FFM dimensions, for any
additional dimensions. And second, there must be evidence that the additional
dimensions are broad enough to summarize substantial portions of common
variance in personality data. Specific variances cannot be promoted to the status
of dimensions, as described above.
Evidence for additional dimensions can be arranged. A researcher simply
needs to identify a construct that is not strongly captured by the FFM and then
build perhaps five or six new measures of different, related facets of this
construct. When the correlations for the additional measures are included in joint
factor analyses with FFM measures, there will be enough common variance
among the new scales for the additional dimension to surpass the cutoff criteria
for the existence of a sixth dimension. Build it, and you and others will find it.
Additional, non-FFM dimensions did not naturally emerge in the history of
research of personality inventories. If additional dimensions are now going to be
deliberately built into new measures, then the evidence for their existence
deserves careful scrutiny because it can be arranged.
Anchoring the Dimensions to Facilitate Research Integration

A variety of different measures of the FFM are now available and in use (see
the chapter by Simms). The measures vary greatly in length, depth, and breadth.
Some of the measures were designed to assess conceptualizations of the FFM
dimensions that differ from other conceptualizations and measurements. But
even scores on measures that were based on the same conceptualizations are not
necessarily equivalent and may result in differing patterns of associations with
non-FFM measures and with nonpersonality outcome variables. It is as if FFM
researchers are all dedicated to roughly the same five-dimensional model but
disagree on the preferred rotation or location of the five axes. Variability in the
measurement of the FFM dimensions has surely resulted in apparently
inconsistent or conflicting findings. The research literature would be more
consistent and much easier to integrate if the field could agree on the nature,
location, or rotation of the five axes.
Cartographers struggled with this same problem hundreds of years ago (this
analogy was first provided by Gifford & O’Connor, 1987, and Bob Gifford
deserves all the credit). The concepts of latitude and longitude were proposed in
the second century A.D. Latitude was easily measured, but not longitude.
Navigation at sea in particular was a real problem until the chronometer, for
longitude, was perfected in the 1750s. But then there were big squabbles, for
another hundred years, over where the prime meridian should be located. Each
nation wanted it drawn through one of its own cities. Map making was a mess.
In 1884, an international conference placed it in Greenwich, England. Maps then
became more coherent and sea travel much safer.
Our own FFM literature could become more coherent if we were similarly
able to agree on the locations of our prime meridians. In cartography, the
location of the prime meridian was arbitrary, just as the many possible factor
rotations of a dimensional structure in personality are somewhat arbitrary and
equally valid. Or maybe not. Perhaps the FFM dimensions should be anchored in
particular patterns of brain activity. In any case, it seems clear that the field now
has a powerful, robust dimensional model, that the disagreements are over
specifics, and that even more progress could be made if the field agreed on the
measurement anchoring points for the dimensions. It would then become even
more clear where other measures of personality are located within our five-
dimensional map.
References
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting
interactions. Newbury Park, CA: Sage.
Almagor, M., Tellegen, A., & Waller, N. G. (1995). The big seven model: A cross-
cultural replication and further exploration of the basic dimensions of natural language
trait descriptors. Journal of Personality and Social Psychology, 69, 300–307.
Angleitner, A., & Ostendorf, F. (1994). Temperament and the big five factors of
69–90). Hillsdale, NJ: Erlbaum.
11(2), 150–166.
Barbaranelli, C., & Caprara, G. V. (1996). How many dimensions to describe
personality? A comparison of Cattell, Comrey, and the big five taxonomies of
personality traits. European Review of Applied Psychology, 46, 15–24.
Ben-Porath, Y. S., & Waller, N. G. (1992). “Normal” personality inventories in clinical
assessment: General requirements and the potential for using the NEO personality
inventory. Psychological Assessment, 4, 14–19.
Bernstein, I. H. (1988). Applied multivariate analysis. New York, NY: Springer-Verlag.
Block, J. (1995). A contrarian view of the five-factor approach to personality description.
Buss, A. H., & Plomin, R. (1975). A temperament theory of personality development.
New York, NY: Wiley Interscience.
Butcher, J. N., & Rouse, S. V. (1996). Personality: Individual differences and clinical
assessment. Annual Review of Psychology, 47, 87–111.
Caprara, G. V., Barbaranelli, C., Borgogni, L., & Perugini, M. (1993). The “big five
questionnaire”: A new questionnaire to assess the five-factor model. Personality and
Church, A. T. (1994). Relating the Tellegen and five-factor models of personality
Church, A. T., & Burke, P. J. (1994). Exploratory and confirmatory tests of the big five
and Tellegen’s three- and four-dimensional models. Journal of Personality and Social
Clark, L. A. (1996). Schedule for maladaptive and adaptive personality. Minneapolis,
MN: University of Minneapolis Press.
Cloninger, C. R., Przybeck, T. R., Svrakic, D. M., & Wetzel, R. D. (1994). The
temperament and character inventory (TCI): A guide to its development and use. St.
Louis, MO: Washington University School of Medicine.
Cloninger, C. R., & Svrakic, D. M. (1994). Differentiating normal and deviant personality
by the seven-factor personality model. In S. Strack & M. Lorr (Eds.), Differentiating
normal and abnormal personality (pp. 40–64). New York, NY: Springer.
Cloninger, C. R., Svrakic, D. M., & Przybeck, T. R. (1993). A psychobiological model of
temperament and character. Archives of General Psychiatry, 50, 975–990.
Costa, P. T., & McCrae, R. R. (1988). From catalog to classification: Murray’s needs and
the five-factor model. Journal of Personality and Social Psychology, 55, 258–265.
Costa, P. T., & McCrae, R. R. (1989). NEO-PI/FFI manual supplement. Odessa, FL:
Costa, P. T., & McCrae, R. R. (1990). Personality disorders and the five-factor model of
personality. Journal of Personality Disorders, 4, 362–371.
Costa, P. T., & McCrae, R. R. (1992a). Normal personality assessment in clinical
practice: The NEO personality inventory. Psychological Assessment, 4, 5–13.
Costa, P. T., & McCrae, R. R. (1992b). Revised NEO personality inventory (NEO-PIR)
and NEO five-factor inventory (NEO-FFI): Professional manual. Odessa, FL:
Costa, P. T., & McCrae, R. R. (1995a). Solid grounds in the wetlands of personality: A
reply to Block. Psychological Bulletin, 117, 216–220.
Costa, P. T., & McCrae, R. R. (1995b). Domains and facets: Hierarchical personality
assessment using the revised NEO personality inventory. Journal of Personality
Costa, P. T., Busch, C. M., Zonderman, A. B., & McCrae, R. R. (1986). Correlations of
MMPI factor scales with measures of the five factor model of personality. Journal of
Costa, P. T., Zonderman, A. B., McCrae, R. R., & Williams, R. B. (1985). Content and
comprehensiveness in the MMPI: An item factor analysis in a normal adult sample.
Craig, R. J., Loheidi, R. A., Rudolph, B., Leifer, M., & Rubin, N. (1998). Relationship
between psychological needs and the five-factor model of personality classification.
Cudeck, R. (1982). Methods for estimating between-battery factors. Multivariate
Dahlstrom, W. G., Welsh, G. S., & Dahlstrom, L. E. (1975). An MMPI handbook (vol.
II): Research applications. Minneapolis, MN: University of Minnesota Press.
De Fruyt, F., & Mervielde, I. (1997). The five-factor model of personality and Holland’s
RIASEC interest types. Personality and Individual Differences, 23, 87–103.
Derogatis, L. R. (1983). SCL-90-R revised manual. Baltimore, MD: John Hopkins
University School of Medicine.
Dyce, J. A., & O’Connor, B. P. (1998). Personality disorders and the five-factor model: A
test of facet-level predictions Journal of Personality Disorders, 12, 31–45.
Edgington, E. S. (1995). Randomization tests. New York, NY: Marcel Dekker.
Edwards, A. L. (1959). Edwards Personal Preference Schedule manual. New York, NY:
Psychological Corporation.
Edwards, L. K., & Edwards, A. L. (1991). A principal-components analysis of the
Minnesota Multiphasic Personality Inventory factor scales. Journal of Personality and
Bulletin, 116, 429–456.
Finch, J. F., Panter, A. T., & Caskie, G. I. L. (1999). Two approaches for identifying
shared personality dimensions across methods. Journal of Personality, 67, 407–438.
Funder, D. C. (2001). Personality. Annual Review of Psychology, 52, 197–221.
Furnham, A. (1996). The big five versus the big four: The relationship between the
Myers-Briggs Type Indicator (MBTI) and NEO-PI five factor model of personality.
Gifford, R., & O’Connor, B. (1987). The interpersonal circumplex as a behavior map.
Gorsuch, R. L. (1983). Factor analysis. Hillsdale, NJ: Erlbaum.
Gough, H. G. (1987). California Psychological Inventory administrator’s guide. Palo
Alto, CA: Consulting Psychologists Press.
Gough, H. G., & Heilbrun, A. B. (1983). The Adjective Check List manual (1983 ed.).
Palo Alto, CA: Consulting Psychologists Press.
Green, P. E., & Carroll, J. D. (1976). Mathematical tools for applied multivariate
analysis. New York, NY: Academic Press.
Guadagnoli, E., & Velicer, W. (1991). A comparison of pattern matching indices.
Multivariate Behavior Research, 26, 323–343.
Guilford, J. S., Zimmerman, W. S., & Guilford, J. P. (1976). The Guilford-Zimmerman
Temperament Survey handbook: Twenty-five years of research and applications. San
Diego, CA: Edits.
Hahn, R., & Comrey, A. L. (1994). Factor analysis of the NEO-PI and the Comrey
Personality Scales. Psychological Reports, 75, 355–365.
Harkness, A. R., McNulty, J. L., & Ben-Porath, Y. S. (1995). The Personality
Psychopathology Five (PSY-5): Constructs and MMPI-2 scales. Psychological
Harpur, T. J., Hart, S. D., & Hare, R. D. (1994). Personality of the psychopath. In P. T.
Costa & T. A. Widiger (Eds.), Personality disorders and the five-factor model of
personality (pp. 149–173). Washington, DC: American Psychological Association.
Holland, J. L. (1985). Self-directed search—1985 Edition. Odessa, FL: Psychological
Assessment Resources.
Huey, S. J., & Weisz, J. R. (1997). Ego control, ego resiliency, and the five-factor model
as predictors of behavioral and emotional problems in clinic-referred children and
adolescents. Journal of Abnormal Psychology, 106, 404–415.
Hyer, L., Brawell, L., Albrecht, B., Boyd, S., Boudewyns, P., & Talbert, S. (1994).
Relationship of NEO-PI to personality styles and severity of trauma in chronic PTSD
victims. Journal of Clinical Psychology, 50, 699–707.
Jackson, D. N. (1976). Jackson Personality Inventory manual. Port Huron, MI: Research
Jackson, D. N. (1989). Basic Personality Inventory manual. Port Huron, MI: Sigma
Assessment Systems.
Johnson, J. H., Null, C., Butcher, J. N., & Johnson, K. N. (1984). Replicated item factor
analysis of the full MMPI. Journal of Personality and Social Psychology, 47, 105–114.
Lehne, G. K. (1994). The NEO-PI and MCMI in the forensic evaluation of sex offenders.
In P. T. Costa & T. A. Widiger (Eds.), Personality disorders and the five-factor model
of personality (pp. 175–188). Washington, DC: American Psychological Association.
Levenson, M. R., Aldwin, C. M., Bosse, R., & Spiro, A. (1988). Emotionality and mental
health: Longitudinal findings from the normative aging study. Journal of Abnormal
Levin, J., & Montag, I. (1991). Relationship between the basic personality inventory and
the NEO-personality inventory in a nonpatient sample. Psychological Reports, 69,
1176–1178.
Longman, R. S., Cota, A. A., Holden, R. R., & Fekken, G. C. (1989). PAM: A double-
precision FORTRAN routine for the parallel analysis method in principal components
analysis. Behavior Research Methods, Instruments, and Computers, 21, 477–480.
Lorr, M. (1986). Interpersonal Style Inventory (ISI) manual. Los Angeles, CA: Western
Psychological Services.
Lorr, M., & DeJong, J. (1986). A short form of the interpersonal style inventory (ISI).
Journal of Clinical Psychology, 42, 466–468.
Lorr, M., Youniss, R. P., & Kluth, C. (1992). The interpersonal style inventory and the
five-factor model. Journal of Clinical Psychology, 48, 202–206.
MacDonald, D. A., Anderson, P. E., Tsagarakis, C. I., & Holland, C. J. (1994).
Examination of the relationship between the Myers-Briggs Type Indicator and the NEO
personality inventory. Psychological Reports, 74, 339–344.
Maraun, M. D., & Chrisjohn, R. D. (1995). Radex structure of Jackson’s Basic
Personality Inventory. Personality and Individual Differences, 18, 881–892.
McCrae, R. R. (1989). Why I advocate the Five-Factor Model: Joint analyses of the NEO-
PI and other instruments. In D. M. Buss & N. Cantor (Eds.), Personality psychology:
Recent trends and emerging directions (pp. 237–245). New York: Springer-Verlag.
McCrae, R. R. (1991). The five factor model and its assessment in clinical settings.
McCrae, R. R., & Costa, P. T. (1986). Clinical assessment can benefit from recent
advances in personality psychology. American Psychologist, 41, 1001–1003.
McCrae, R. R., & Costa, P. T. (1989a). The structure of interpersonal traits: Wiggins’
circumplex and the five-factor model. Journal of Personality and Social Psychology,
56, 586–595.
McCrae, R. R. & Costa, P. T. (1989b). Reinterpreting the Myers-Briggs Type Indicator
from the perspective of the five-factor model of personality. Journal of Personality, 57,
17–40.
McCrae, R. R., & Costa, P. T. (1990). Personality in adulthood. New York, NY: Guilford
Press.
McCrae, R. R., & Costa, P. T. (1991). The NEO personality inventory: Using the five-
factor model in counseling. Journal of Counseling and Development, 69, 367–376.
McCrae, R. R., & Costa, P. T. (1994). Does Lorr’s Interpersonal Style Inventory measure
the five-factor model? Personality and Individual Differences, 16, 195–197.
McCrae, R. R., & Costa, P. T. (1995). Positive and negative valence within the five-factor
model. Journal of Research in Personality, 29, 443–460.
McCrae, R. R., & Costa, P. T. (1997). Personality structure as a human universal.
McCrae, R. R., Costa, P. T., & Piedmont, R. L. (1993). Folk concepts, natural language,
and psychological constructs: The California Psychological Inventory and the five-
factor model. Journal of Personality, 61, 1–26.
McCrae, R. R., & John, O. P. (1992). An introduction to the five-factor model and its
McCrae, R. R., Zonderman, A. B., Costa, P. T., Bond, M. H., & Paunonen, S. V. (1996).
Evaluating replicability of factors in the revised NEO personality inventory:
Mershon, B., & Gorsuch, R. L. (1988). Number of factors in the personality sphere: Does
increase in factors increase predictability of real-life criteria? Journal of Personality
Millon, T. (1983). Millon Clinical Multiaxial Inventory manual (3rd ed.). Minneapolis,
MN: National Computer Systems.
Millon, T. (1987). Manual for the MCMI-II. Minneapolis, MN: National Computer
Systems.
Millon, T. (1994). Manual for the MCMI-III. Minneapolis, MN: National Computer
Systems.
Montag, I., & Levin, J. (1994). The five factor model and psychopathology in nonclinical
samples. Personality and Individual Differences, 17, 1–6.
Morey, L. C. (1991). Personality Assessment Inventory: Professional manual. Odessa,
Morey, L. C., Waugh, M. H., & Blashfield, R. K. (1985). MMPI scales for DSM-III
personality disorders: Their derivation and correlates. Journal of Personality
Myers, I. B., & McCauley, M. H. (1985). A guide to the development of the Myers-Briggs
Type Indicator. Palo Alto, CA: Consulting Psychologists Press.
O’Connor, B. P. (2002). A quantitative review of the comprehensiveness of the five-
O’Connor, B. P. (2006). Social desirability measures and the validity of self-reports: A
comprehensive search for moderated relationships in five-factor model space. In S. P.
Shohov (Ed.), Advances in psychology research (Vol. 40, pp. 39–73). Hauppauge, NY:
Nova Science Publishers.
O’Connor, B. P., & Dvorak, T. (2001). Conditional associations between parental
behavior and adolescent problems: A search for personality-environment interactions.
O’Connor, B. P., & Dyce, J. A. (1998). A test of models of personality disorder
configuration. Journal of Abnormal Psychology, 107, 3–16.
O’Connor, B. P., & Dyce, J. A. (2001). Rigid and extreme: A geometric representation of
personality disorders in five-factor model space. Journal of Personality and Social
Psychology, 81, 1119–1130.
Ozer, D. J., & Reise, S. P. (1994). Personality assessment. Annual Review of Psychology,
45, 357–388.
Panter, A. T., Tanaka, J. S., & Hoyle, R. H. (1994). Structural models for multimode
designs in personality and temperament research. In C. F. Halverson, G. A.
Kohnstamm, & R. P. Martin (Eds.), The developing structure of temperament and
personality from infancy to adulthood (pp. 111–137). Hillsdale, NJ: Erlbaum.
Paunonen, S. V. (1997). On chance and factor congruence following orthogonal
Procrustes rotation. Educational and Psychological Measurement, 57, 33–59.
Paunonen, S. V., & Jackson, D. N. (1996). The Jackson Personality Inventory and the
five-factor model of personality. Journal of Research in Personality, 30, 42–59.
Piedmont, R. L., McCrae, R. R., & Costa, P. T. (1991). Adjective Check List scales and
the five-factor model. Journal of Personality and Social Psychology, 60, 630–637.
Piedmont, R. L., McCrae, R. R., & Costa, P. T. (1992). An assessment of the Edwards
Personal Preference Schedule from the perspective of the five-factor model. Journal of
Schinka, J. A., Dye, D. A., & Curtiss, G. (1997). Correspondence between five-factor and
RIASEC models of personality. Journal of Personality Assessment, 68, 355–368.
Segerstrom, S. C. (2000). Personality and the immune system: Models, methods, and
mechanisms. Annals of Behavioral Medicine, 22, 180–190.
Sherman, R. C., & Poe, C. A. (1970). Factor-analytic scales of a normative version of the
EPPS. Measurement and Evaluation in Guidance, 2, 243–248.
Smith, D. R., & Snell, W. E. (1996). Goldberg’s bipolar measure of the big-five
personality dimensions: Reliability and validity. European Journal of Personality, 10,
283–299.
Tellegen, A. (1985). Structures of mood and personality and their relevance to assessing
anxiety, with an emphasis on self-report. In A. H. Tuma & J. Maser (Eds.), Anxiety and
the anxiety disorders (pp. 681–706). Hillsdale NJ: Erlbaum.
ten Berge, J. M. F. (1986). Some relationships between descriptive comparisons of
components from different studies. Multivariate Behavioral Research, 21, 29–40.
ten Berge, J. M. F., Kiers, H. A. L., & Commandeur, J. J. F. (1993). Orthogonal
Procrustes rotation for matrices with missing values. British Journal of Mathematical
and Statistical Psychology, 46, 119–134.
Trull, T. J. (1992). DSM-III-R personality disorders and the five-factor model of
personality: An empirical comparison. Journal of Abnormal Psychology, 101, 553–560.
Tucker, L. R. (1958). An inter-battery method of factor analysis. Psychometrika, 23, 111–
136.
Watson, D., Clark, L. A., & Harkness, A. R. (1994). Structures of personality and their
relevance to psychopathology. Journal of Abnormal Psychology, 103, 18–31.
Widiger, T. A., & Trull, T. J. (1992). Personality and psychopathology: An application of
the five-factor model. Journal of Personality, 60, 363–393.
Wiggins, J. S., Steiger, J. H., & Gaelick, L. (1983). Evaluating circumplexity in
personality data. Multivariate Behavioral Research, 16, 263–289.
Wiggins, J. S., Trapnell, P., & Phillips, N. (1988). Psychometric and geometric
characteristics of the Revised Interpersonal Adjective Scales (IAS-R). Multivariate
Windle, M., & Lerner, R. M. (1986). Reassessing the dimensions of temperament
individuality across the life span: The Revised Dimensions of Temperament Survey
(DOTS-R). Journal of Adolescent Research, 1, 213–230.
Universal and Specific in the Five Factor Model of
Personality
Jüri Allik and Anu Realo
Abstract
Personality psychologists—perhaps even more than in some other disciplines—
are deeply interested in what is common to personality descriptions in all cultures
and societies. The purpose of this chapter is to discuss the potential universality of
the Five Factor Model (FFM) of general personality structure. The chapter begins
with a discussion of what is meant, or should be meant, by a universal. Discussed
then is the empirical support, as well as the conceptual and empirical difficulty, in
establishing universality in personality structure, for the FFM as well as other
dimensional models. The chapter then considers different levels of analysis
(including cultural and intraindividual analyses), higher-order invariants
(including sex differences, age differences, and differences in perspective), and
whether mean levels are universal. The chapter concludes with a discussion of the
basis for personality universals, as well as addressing the common challenges to
universality.
Key Words: Five Factor Model, universal, culture, personality structure,
differences in perspective
There are many endemic diseases such as monkey fever or Russian

encephalitis. One of them, Kyasanur forest disease, is specific only to some
forested parts of India and the tick-borne encephalitis is widely distributed
across Eurasia. In spite of their endemic character they both have an independent
category in the International Classification of Diseases (ICD-10). Although these
two diseases are encountered in only very specific places they are still universal
in the sense that every human who contracts these infections develops the same
Kyasanur forest disease or tick-borne encephalitis, each with their typical
symptoms. Their symptoms look very similar because different human
organisms react to these viruses in a highly similar, if not to say in a universal,
manner.
Although sometimes challenged (Patel & Winston, 1994), mental disorders
are comparable to physical disorders (as their inclusion in the ICD-10 implies)
because their neurological and psychological substrates are likewise universal.
Personality traits provide the substrate of personality disorders (PDs), so the
question of their pan-cultural invariance is crucial. This chapter continues to
elaborate our previous arguments that in many respects personality traits are
indeed universal, clearly dominating over specific aspects (Allik, Realo, &
McCrae, 2013).
Personality Universals
When people describe their own personality or that of someone they know
well, many of the descriptors typically go hand-in-hand. For example,
individuals who are described by themselves or by their close acquaintances as
talkative are also believed to experience positive emotions frequently, and those
who are reported to be modest often describe themselves as willing to assist
others in need of help. These covariations tend to group around the same five
basic themes—Neuroticism, Extraversion, Openness to Experience,
Agreeableness, and Conscientiousness—(Goldberg, 1993; McCrae & Costa,
1997), which seem to transcend languages and cultures, giving a good reason to
suggest that this structure of covariation may be a human universal (McCrae &
Costa, 1997).
Strictly speaking, universal means that something is characteristic to all
members of a certain class, without any limits or exceptions. Very little in nature
and even less in culture meet this absolute criterion, but many characteristics
appear to be relatively invariant. Psychologists are interested in universals at the
level of both the individuals (e.g., “all people are mortal”) and the group (e.g.,
“women are more agreeable”), but these levels must be distinguished. The claim
that gender differences in personality are universal does not mean that every
female on earth scores higher (or lower) than every male on a certain trait; rather
it means that in all groups of people, the same degree or direction of gender-
related trait differences is found on average. Because only very few things exist
without exceptions, the observed regularities are expected to hold not in all
groups of people but in most of them. In this case it is more appropriate to talk
about near-universals. Most of the properties of personality traits discussed in
this chapter are (potentially) universal at the group level and consequently are
near-universals. Thus, we are adopting a theoretical position according to which
the concept of psychological universals is not a dichotomous distinction between
universals and nonuniversals. Between these two polarities there is a gradation
of near-universals with various degrees of generality (Norenzayan & Heine,
2005).
The metric system, with its base of 10 units, was devised mainly because
humans started to use their fingers for counting. If pigs could have developed
their own counting system, they would probably have preferred an octal system
because they have only four digits per hoof (Leroi, 2005). However, readers may
be very surprised to learn that five digits per limb is an anatomical near-
universal, not something that is characteristic of all humans. Surprisingly many
people are born with extra digits. About 1 in 3,000 Europeans and about 1 in 300
Africans are born with extra fingers or toes (or both) (Leroi, 2005). Like many
human traits, polydactyly represents a substantial genetic heterogeneity that
varies across different populations and ethnic groups (Malik, 2014). So, if even
fundamental anatomical features such as the number of fingers—which has
influenced the whole history of civilization—are only near-universals, then it is
more than expected that the majority of psychological universals simply cannot
belong to the absolute category of universals.
Linguists were probably among the first who faced the problem of
universality. Currently, the list of the world’s languages, called Ethnologue,
contains 7,106 living languages (Lewis, Simons, & Fennig, 2014). Most of these
languages are distinctive so that without proper learning they are
incomprehensible to speakers of other languages, sometimes living only a few
miles away. In spite of this enormous variety, languages have features that occur
systematically across all of them. For example, if a language is spoken, it has
consonants and vowels. Even further, all known languages seem to have
minimally three vowels including/i/,/a/, and/u/and most languages, not all,
contain nasals (Burquest & Payne, 1993). Thus, in addition to absolute
universals—true for all languages, living or extinct—there are general
tendencies that hold for most languages and that can pretend to have the status of
near-universals in most. Noam Chomsky (1965) has famously argued that there
is a universal grammar—general generative and combinatorial mechanisms that
have an innate biological basis—used by all spoken languages. But even this one
of the strongest universalist claims may has been violated since at least one of
the world’s living languages, Pirahã (spoken by about 300 Pirahãs living in
Brazil’s Amazonas state), arguably lacks some of the features of universal
grammar (Everett, 2005).
In response to a dominant paradigm of the social sciences—cultural relativism
—Donald Brown, a professor of anthropology at the University of California,
started to compose a list of human universals (Brown, 1991). As it turned out,
ethnographers have noticed many features common to all known human
societies studied so far. For example, everywhere people live they have baby
talk, jokes, magic, and a preference for sweet tastes, to say nothing, of course,
about language. In the long list of universals or near-universals, many are
associated with what we can call personality dispositions. For example,
according to this list, in all human societies people have childhood fears,
classification of behavioral propensities and inner states, and facial expression of
anger; and in all known human populations males are more aggressive than
females. Steven Pinker (2002) extended this list by adding, among other
features, fear of death, tickling, and a desire to have a positive self-image.
The status of many of these and other putative human universals is still
uncertain. For example, anger—mentioned in the above list of universals—is
one of the fundamental human emotions that has emerged consistently across
time and culture (Chon, 2002), and there are equivalents for the word anger in
all major languages of the world (Mesquita, Frijda, & Scherer, 1997). Yet, there
is one society—the Utkuhikhalingmiut (“Utku”) Eskimos—that does not have a
special word for anger. Over 40 years ago, anthropologist Jean Briggs described
the Utku society (which at the time of her research consisted of 35 individuals—
the only inhabitants in an arctic area of more than 35,000 square miles) in her
book expressively titled Never in Anger (Briggs, 1970). If these ethnographic
observations are valid, the experience and expression of anger may be
disqualified from the rank of the absolute universals and degraded to the rank of
the near-universals. Similar fates, however, could happen to even more
fundamental constituents of the human society: contrary to what Claude Lévi-
Struss (1969/1949) has claimed, there is at least one society in which there is no
concept of fathers or husbands (Hua, 2001).
A Universal Structure of Personality

When it comes to personality dispositions it is not a trivial task to determine
the rank of their universality. Even linguists acknowledge that there are still a
large number of unidentified languages, and among the above-mentioned 7,106
living languages only a minority is thoroughly described. Psychological research
rooted in Western culture—sometimes called WASP (Western Academic
Scientific Psychology)—is believed by some scholars to be of little relevance to
the majority of the world (Berry, Poortinga, Segall, & Dasen, 2002), reflecting
only a small minority of WEIRD (Western, Educated, Industrialized, Rich, and
Democratic) people (Henrich, Heine, & Norenzayan, 2010a, 2010b; Jones,
2010). A recent survey of the top psychological journals found that 96% of all
research participants were from Western industrialized countries (Henrich et al.,
2010b). However, unlike many fields in psychology, personality research has
been a truly international enterprise for a number of years (Allik, 2012). Even if
questionnaires were mainly devised by WASP or WEIRD researchers, they were
soon translated into many different languages. For example, Indian researchers
translated the Revised NEO Personality Inventory (NEO PI-R; McCrae & Costa,
2010), developed in Baltimore, the United States, into Telugu and Marathi
(Lodhi, Deo, & Belhekar, 2002; McCrae, 2002). This is a major advance,
because there are 74 and 90 million Telugu and Marathi speakers, respectively,
occupying the sixteenth and fifteenth positions in the list of the most spoken
languages (Lewis et al., 2014).
Based on what was said above, it is unrealistic to expect that any aspect of
personality dispositions would be absolutely universal or even very strongly
near-universal. When McCrae and Costa (1997) proposed the bold hypothesis
that the pattern of covariation among personality traits is a human universal they
were able to rely on only six translations of the NEO PI-R, into the German,
Portuguese, Hebrew, Chinese, Korean, and Japanese languages. Nevertheless,
data from these diverse cultures with languages from five distinct language
families were persuasive enough to suggest that the observed regularity in the
pattern of covariation among personality traits will be not violated when more
and more new cultures and languages are subjected to a critical examination
(McCrae & Allik, 2002).
In addition to occasional failures to reproduce the basic five factor structure in
certain cultures (Cheung, van de Vijver, & Leong, 2011; Panayiotou, Kokkinos,
& Spanoudis, 2004; Zecca et al., 2013), there was a highly acclaimed study of
Tsimane forager-horticulturalist men and women of Bolivia (Gurven, von
Rueden, Massenkoff, Kaplan, & Lero Vie, 2013). The authors of this paper
suggested that the Five Factor Model (FFM) may not be a human universal. At
variance from a typical five factor structure, Tsimane personality variation—
both for the self-ratings and observer-ratings—displays only two principal
factors that may reflect socioecological characteristics common to small-scale
societies (Gurven et al., 2013). No doubt, this is an admirable study of a large (N
= 632) illiterate, indigenous population and there must be a good reason why a
habitual five factors did not emerge either in self-ratings or in informant-ratings.
The same 44-item Big Five Inventory (BFI; Benet-Martinez & John, 1998) was
successfully used for the study of 56 nations or territories and in each case the
same relatively invariant factor structure was identified (Schmitt et al., 2007).
Against this massive evidence of universality it is obviously too early to declare
the FFM dead. Before the final verdict is pronounced it is necessary to eliminate
all simpler or even “technical” explanations. For example, perhaps the five
factors failed to emerge due to translation problems. Similarly, the BFI is not
balanced in terms of positively and negatively worded items and it is
characteristic of less educated samples to respond more to wording than to the
content of items (Schmitt et al., 2007). Thus, before declaring that the structure
of personality is not invariant across human societies it is more productive to use
informants who are not subjected to the same literacy and educational
constraints (Allik & McCrae, 2004a). It is not excluded that Tsimanes also have
a rather conventional personality structure but they simply do not have enough
skill to describe and report details of this structure.
However, there is no need to visit exotic cultures to find exceptions from
universality. For example, Toomela (2003) proposed that those individuals who
primarily use everyday concepts in thinking do not reveal a coherent Big Five
personality structure. In this sense the FFM is a product of cultural development
that can be achieved only when thinking in scientific concepts is accomplished.
Participants who predominantly think in everyday concepts have a tendency to
produce a pattern of covariation in which, similar to Tsimanes, not all five
factors have been equally represented. However, a reanalysis of Toomela’s data
demonstrated that a targeted rotation may be a remedy for most structural
problems (Allik & McCrae, 2004a). Although people usually manage to make
personality judgments that are accurate enough for navigation of the complex
social world, the accuracy is achieved when relevant behavioral information is
available to and detected by a judge who then utilizes that information correctly
(Funder, 2012). As we suggested above, the problem may be in the available
personality information and how this information is used.
At least formally, universality of the FFM seems to signify that the same
pattern of covariation between personality traits is observed in all age groups
without exceptions. Even if personality appears to be preserved more or less
intact through very old age (Martin, Long, & Poon, 2002; Mõttus, Johnson, &
Deary, 2012) many researchers are sceptical about personality in very young
age. Analogous to cognitive abilities, it is tempting to believe that personality
traits emerge somehow in the process of development and it may take a
considerable time to mature before they finally become identical to the adult
personality structure. However, when well-informed adults were asked to rate
kindergarten children (aged 4–6 years) four of the five personality factors were
easily recovered from these ratings (Mervielde, Buyst, & De Fruyt, 1995). When
children approach school age they became able to describe their own personality
and their ratings in many countries group around the typical five themes
(Bleidorn & Ostendorf, 2009; De Fruyt, Mervielde, & Van Leeuwen, 2002;
Rossier, Quartier, Enescu, & Iselin, 2007). Twelve-years-old children are almost
ready to answer adult personality questionnaires and the common five factor
structure of personality is clearly recognizable in their self-ratings (Allik, Laidra,
Realo, & Pullmann, 2004; McCrae et al., 2002). Nevertheless, the personality
structure of 12-year-old children demonstrates only an approximate congruence
with the adult structure, suggesting that not all children of that age have
developed the abilities required for observing their own personality dispositions
and for giving reliable self-reports on the basis of these observations (see also
the chapter by De Pauw). The self-reported personality trait structure matures
and becomes sufficiently differentiated around age 14–15 years and grows to be
practically indistinguishable from adult personality by the age of 16 years (Allik
et al., 2004). Thus, on the basis of self-reports alone it is not possible to maintain
that younger children have a personality structure that is dramatically different
from that of adults.
Large-Scale Cross-Cultural Studies

Although it is not entirely clear how to establish universality, it is inevitable
that large-scale cross-cultural studies are necessary to provide evidence that
something is indeed characteristic of all or nearly all human beings. On the basis
of data collected in three or four cultures it is impossible even to guess what is
universal and what is specific in personality. However, the collection of
personality data from many cultures is even more expensive than gathering data
about all spoken languages. There are only a few ways to collect personality data
from a sufficient number of countries. The first is to develop a popular inventory
that will be translated into a large number of languages by enthusiastic
colleagues. The Eysenck Personality Questionnaire (EPQ; Eysenck & Eysenck,
1975) and the NEO PI-R (McCrae & Costa, 2010) are good examples of this
relatively slow and complicated method of collecting data (Lynn & Martin,
1995; McCrae, 2002; van Hemert, van de Vijver, Poortinga, & Georgas, 2002).
Another way is to form an international research syndicate that is held together
by the promise that the first two or three papers are co-authored by all those who
participate in collecting data. For instance, David Schmitt, one of the most
successful elaborators of this research mechanism, was able to obtain personality
data from 56 countries or territories using the above mentioned BFI (Schmitt et
al., 2007). In this study the BFI was translated from English into 28 languages
and administered to 17,837 individuals from 56 nations (Schmitt et al., 2007).
Although the differences in personality mean values and structure were small,
their geographic pattern was replicable to what was previously established by
more sophisticated measurement instruments (cf. Allik & McCrae, 2004b).
Exploiting the same principle, with the help of a large number of collaborators
McCrae and Terracciano gathered observer-reported personality data and
national character ratings from 50 cultures (McCrae, Terracciano, & 78
Members of the Personality Profiles of Cultures Project, 2005; McCrae,
Terracciano, & 79 Members of the Personality Profiles of Cultures Project,
2005).
Another important development in research technology is, of course, the
widespread use of the Internet, which allows the collection of huge samples
during a relatively short period of time. Perhaps one of the best examples is the
BBC Internet study, which examined sex differences on three personality traits
—Extraversion, Agreeableness, and Neuroticism—for over 200,000 participants
from 53 nations (Lippa, 2010). In another impressive study data were collected
from more than a half million participants using the BFI (Rentfrow, 2010;
Rentfrow et al., 2013; Rentfrow, Gosling, & Potter, 2008). Participants in this
study were distributed very closely to the percentage of the total U.S. population
for each state. Not only do worldwide distributions of personality traits
demonstrate a meaningful pattern (Allik & McCrae, 2004b) but statewide
personality differences across the United States are also linked to a variety of
important social indicators. Although it has been argued that Internet findings
are consistent with results from traditional methods (Gosling, Vazire, Srivastava,
& John, 2004), there is indisputable evidence that self-recruited Internet data
may sometimes be biased compared to random sampling (Pullmann, Allik, &
Realo, 2009), which may constrain their potential value.
Perhaps the most important lesson from all these large-scale comparative
studies is the ease with which personality instruments can transcend language
and culture barriers. The same basic pattern of covariations—the FFM—has
been replicated, more or less accurately, in almost every language and culture
studied so far (Kallasmaa, Allik, Realo, & McCrae, 2000; Rolland, 2002;
Schmitt et al., 2007). Essentially the same factor structure was recovered from
self-ratings (McCrae, 2002) and from observer ratings (McCrae, Terracciano, &
78 Members of the Personality Profiles of Cultures Project, 2005; McCrae,
2005). Thus, the FFM appears to be reasonably invariant across methods of
measurement.
There are only a few studies of personality in which geographic
representativeness inside one country was achieved (Rentfrow, 2010; Rentfrow
et al., 2013; Rentfrow, Mellander, & Florida, 2009). Perhaps it is not so urgent to
achieve a sufficient geographic coverage for a small country such as Estonia
(Allik et al., 2004), but it is desirable, if not imperative, for large countries such
as the United States, China, or Russia. Recently, 7,065 participants from 39
samples in 33 administrative areas of the Russian Federation identified an
ethnically Russian adult or college-aged man or woman whom they knew well
and rated the target using the Russian observer rating paper-and-pencil version
of the NEO PI-R (Allik, Realo, et al., 2009). The expected FFM structure was
clearly replicated in the full sample, with factor congruence coefficients of
.95–.96 for all five factors. When these analyses were repeated within the 39
samples, all showed reasonable to good replications of the FFM, with average
factor congruence coefficients ranging from .90 to .98. In the first few studies
published two other geographic giants—the United States and the People’s
Republic of China—are treated as consisting of different subregions (Rentfrow,
2010; Rentfrow et al., 2009, 2013; Van de Vliert, Yang, Wang, & Ren, 2013).
Although the five factor structure is clearly recognizable in nearly every
language into which the NEO PI-R or BFI has been translated (for an exception
see Gurven et al., 2013), in less developed countries the data seem to fit the FFM
less perfectly than in industrialized, less agrarian countries (Piedmont, Bain,
McCrae, & Costa, 2002; Zecca et al., 2013). Research suggests that the degree of
fit to the FFM depends primarily on the quality of the data (McCrae,
2005). For example, one indicator of data quality is negative item bias (Schmitt
& Allik, 2005). In less developed countries in which people live in economic
need and access to education is limited, respondents are inclined to answer
negatively worded items slightly differently than they answer directly
formulated statements. The fit of the FFM may also depend on the cultural
relevance of specific items. In several cultures the Openness factor has proved to
be the weakest, especially in African countries (Piedmont et al., 2002). It seems
that NEO PI-R items such as “Poetry has little or no effect on me” or “I often try
new and foreign foods” may represent the Openness concept less clearly in
Burkina Faso and Zimbabwe than they do in Western countries (Rossier,
Dahourou, & McCrae, 2005; Zecca et al., 2013). This implies that a more
appropriate selection of items is needed to optimize the translation of the FFM
into more exotic languages and cultures.
Alternatives to the FFM

However, the universality of the FFM does not rule out the possibility that
some other covariance patterns, with smaller or larger numbers of factors, may
also be replicable across many languages and cultures. For example, Eysenck’s
three-factor (van Hemert et al., 2002) and psycholexical six-factor (Ashton, Lee,
& de Vries, 2014; Lee & Ashton, 2008) structures have also been replicated in
many cultures. The compatibility of structures with different numbers of factors
becomes understandable within a structural framework based on a hierarchy of
personality traits (Markon, Krueger, & Watson, 2005). The hierarchy of traits
can be cut on different levels of generalization and as a result has three, five, or
six relatively stable factors.
It has been argued that imposing a factor structure derived from Western
samples on non-Western cultures may leave unnoticed unique personality factors
specific to these cultures alone, called emic dimensions of personality (Cheung
et al., 2011). For example, it was noted that the FFM ignores an Interpersonal
Relatedness factor that is unique to the Chinese (or more generally Asian)
personality (Cheung, Cheung, Wada, & Zhang, 2003; Cheung et al., 2001).
Because ren qing—friendly person—is believed to be deeply rooted in the
Chinese mentality, European concepts are not sufficient to describe this specific
aspect of Chinese culture. However, it was soon discovered that this supposedly
specific Chinese or Asian personality trait could be reproduced fairly well in a
European-American sample, indicating that the Interpersonal Relatedness factor
in not unique to Asian populations (Lin & Church, 2004).
Travelers were probably the first to notice cross-cultural differences. Those
who were raised with the Protestant working spirit may feel confused in their
first encounter with the mañana cultures (House, Hanges, Javidan, Dorfman, &
Gupta, 2004). Friendliness, politeness, and sympathy are highly valued in the
Spanish culture in which there is even a special word for it—simpatico
(Ramirez-Esparza, Gosling, & Pennebaker, 2008; Triandis, Lisansky, Marin, &
Betancourt, 1984). It is also believed that Portuguese saudade—a deep
emotional state of nostalgic or profound melancholic longing for an absent
something or someone that is loved—has no appropriate translation into English
or any other language (Bułat Silva, 2012). In spite of a premature optimism that
amae—the pattern of attachment and dependence between mother and child—is
unique to Japanese culture alone (Behrens, 2004), it is more realistic to assume
universal mechanisms behind this concept (Cheung et al., 2011). Although it
may not be easy to translate German Schadenfreude into English, nobody really
doubts that feeling gloating pleasure, Dutch comfort, or mischief-joy is totally
alien to Anglo-American culture (Leach, Spears, Branscombe, & Doosje, 2003;
Smith et al., 1996).
In a similar vein, many domestic and foreign observers have claimed that
Russians have a unique constellation of personality traits that mirrors their
distinctive historical and cultural experience. In contrasting themselves with the
industrialized and materialistic cultures of the West, Russians in the nineteenth
century began to define themselves in terms of their spiritual qualities, their
distinctive “Russian soul” or dusha (Allik et al., 2011). To capture distinctive,
emic aspects of Russian personality beyond the familiar Big Five dimensions, a
set of emic personality items was developed. For instance, the widely perceived
inclination of Russians toward fatalism was measured by items such as
“Believes that he/she cannot escape his/her fate” and “Believes that he/she is an
architect of one’s own fortunes” (reversed). As it turned out, most of the
variance in the emic items could be explained by the known Big Five factors
(Allik et al., 2011). There is no question that cultures may have their unique
traits such as ren qing or simpatico. The problem is always about the exact
proportions between specific and universal. If personality traits are understood
as enduring tendencies to feel, think, and act in a characteristic way then
universal aspects are clearly dominant over specific aspects in their worldwide
distribution. It was observed, for example, that the mean differences between
cultures or geographic regions are typically 10 times smaller than an average
interindividual variance within those cultures or regions (Allik, 2005). Individual
differences we encounter in our everyday experience are by an order of
magnitude larger than typical differences between the mean values on
personality traits of a group of people occupying a large geographic territory.
The FFM Structure at Different Levels of Analysis
The pattern of covariation in the FFM was established by means of factor
analyses based on interindividual differences. From that level it is possible to
move either higher to the level of cultures or lower to the level of single
individuals.
Culture-Level Analysis
Collecting NEO PI-R self-report data from 36 cultures or territories was
difficult (McCrae, 2002). However, it was still too small a dataset to subject the
mean, culture-level, values on 30 facet scales to a factor analysis. A solution was
to split each culture into four subgroups according to sex and age (females
versus males; college age people versus adults), increasing subsamples to 114.
With minor variations, the culture-level analysis of the means of these 114
samples replicated the five factor individual level factor structure (see McCrae,
2002, Table 2). These findings were subsequently replicated in culture-level
analyses of observer ratings of college-age and adult targets from 51 cultures
(McCrae, Terracciano, & 78 Members of the Personality Profiles of Cultures
Project, 2005), and of adolescents from 24 cultures (McCrae et al., 2010).
Initially this replication of the FFM structure was interpreted as an empirical
finding about the structure of personality on the culture level, but it soon became
clear that is a statistical necessity. Assigning individuals randomly to an arbitrary
114 subsamples would have resulted in an even better replication of the
individual level factor structure (McCrae, Terracciano, & 78 Members of the
Personality Profiles of Cultures Project, 2005). Thus, only deviations, not
resemblance, between individual and group-level factor structures are indicative
of the possible involvement of culture. When culture-level factor structure
replicates the individual-level structure it means that the influence of culture on
personality traits (and their assessment) is negligible. Existing data show that
culture contributes consistently but rather modestly to the pattern of covariances
(McCrae & Terracciano, 2008).
Intraindividual Level of Analysis

Another direction of generalizability is to move from the level of the group to
that of the individual. According to some researchers, personality trait
covariation models such as the FFM provide information that holds true at the
level of groups or populations but may not apply to the level of the individual
(Borsboom, 2005). For example, it was demonstrated that if a latent factor model
fits a given population, this does not guarantee the fit of the same model for each
or even any individual subjects from that group, assuming that intraindividual
variation is measured by repeated administration of the same instrument
(Borsboom, Mellenbergh, & Van Heerden, 2003; Molenaar & Campbell, 2009):
The structure of traits across individuals is not necessarily the same as the
structure of states within the individual. Borsboom and colleagues concluded
from this that models derived from between-subject variation cannot provide
causal explanations for the behavior of the individual—a conclusion that has
been challenged by others (McCrae & Costa, 2008). However, this work raises
the question of whether, and in what sense, the FFM can be said to characterize
individuals.
Allik et al. (2012) argued that the FFM characterizes an individual if scores on
each of the indicators of a factor (e.g., the six facets that define each factor in the
NEO PI-R) are at similar levels (especially in contrast to variations in levels of
facets across different factors). A person who is high on Anxiety and Angry
Hostility and Depression and Self-Consciousness and Impulsiveness and
Vulnerability can meaningfully be said to be characterized by the Neuroticism
factor, whereas an individual who is high on the first three facets and low on the
last three does not show a coherent Neuroticism factor. Allik and colleagues
(2012) operationalized this concept using the intraclass correlation and
concluded that most individuals are reasonably well characterized by the FFM
structure.
Although the common pattern of covariations was dominant (e.g., individuals
who are described as talkative are also believed to experience a need for
excitement, and those who are reported to be modest are often described as
amenable), this does not exclude the possibility that in infrequent cases some of
the normally incompatible personality traits can coexist. It is possible, for
instance, that individuals who are often described by themselves or by their close
acquaintances as angry and hostile are very seldom believed to experience
shame or embarrassment, which are other indicators of Neuroticism (Allik et al.,
2012, see Fig. 4C). Similarly, it is possible to find a person who is highly dutiful
and carefully deliberates his or her actions but has a relative low need for
achievement (Allik et al., 2012, see Fig. 4B). In most cases dutifulness and
achievement striving go hand in hand, but infrequently these two traits are
disassociated or are even opposite. Despite being infrequent, these cases suggest
that some atypical combinations of personality traits exist. Unfortunately, we
know very little about these atypical combinations and even less about
circumstances that make their occurrence possible.
Higher-Order Invariants
Debates concerning FFM focused mainly on the proper number of factors.
Perhaps one of the most famous statements about the reality of five factors was
formulated as follows: “We believe it is simply an empirical fact, like the fact
that there are seven continents on earth or eight American presidents from
Virginia. Biologists recognize eight classes of vertebrates (mammals, birds,
reptiles, amphibians, and four classes of fishes, one extinct), and the theory of
evolution helps to explain the development of these classes. It does not,
however, explain why eight classes evolved, rather than four or eleven, and no
one considers this a defect in the theory” (McCrae & John, 1992, p. 194).
Indeed, one of the most vocal criticisms was that there is no good explanation as
to why there are five and not, for example, seven factors (Block, 1995, 2010).
Although a hierarchical approach to the number of factors (Markon et al., 2005)
considerably relieved the tension, the five factors is still perceived to be the core
of the FFM. However, the most significant progress has been achieved
establishing universals or near-universals that are independent of the number of
factors. We can call them higher-order invariants because their existence does
not depend critically on how many personality factors are there.
Sex Differences
Lynn and Martin (1995) were among the first who reported a systematic
pattern of gender differences: Women obtained higher mean scores than men on
Neuroticism scales in all 37 nations in which the results of the EPQ were
available; men scored higher than women on Extraversion in 30 countries and on
Psychoticism in 34 countries. Subsequent studies using measures of the FFM
show that women in most countries are higher in several traits related to
Neuroticism, Agreeableness, warmth, and openness to feelings, whereas men
score higher on scales measuring assertiveness and openness to ideas (Costa,
Terracciano, & McCrae, 2001; Lippa, 2010; Schmitt, Realo, Voracek, & Allik,
2008). These differences are consistent with universal gender stereotypes
(Löckenhoff et al., 2014), but the measured differences are generally rather
small.
Although the direction of gender differences is near-universal, the magnitude
shows systematic variation: These differences systematically increase with the
level of development—including a long and healthy life, equal access to
knowledge and education, and economic wealth (Costa et al., 2001; Lippa, 2010;
Schmitt et al., 2008). This finding was counterintuitive, because most people
assumed that gender equality would lead to diminished sex differences in
personality. Several explanations have been offered to explain this puzzling
finding. One explanation is that gender differences are illusory as a by-product
of self-stereotyping, which occurs when between-gender social comparisons are
made. Because these social comparisons are more likely to exert a greater impact
in Western nations it is expected that the disparity between men and women
appears to increase with the level of development (Guimond et al., 2007). In a
similar direction, Costa and colleagues (2001) speculated that it reflected
different processes of attribution in traditional and modern cultures. Schmitt and
colleagues (2008) proposed that heightened levels of sexual dimorphism result
from personality traits of men and women being less constrained and more able
to naturally diverge in developed nations. In less fortunate social and economic
conditions, innate personality differences between men and women may be
attenuated.
Do men vary more than women in personality? Data collected from 51
cultures or territories suggested that in most cultures, male targets varied more
than female targets, and ratings by female informants varied more than ratings
by male informants, which may explain why higher variances for men are not
found in self-reports (Borkenau, McCrae, & Terracciano, 2013). Variances were
higher in more developed societies and effects of target sex were stronger in
more individualistic societies. It seems that individualistic cultures enable a less
restricted expression of personality, resulting in larger variances, particularly
among men (Borkenau et al., 2013).
Age Differences
Although the same Five Factor structure appears to persist through the major
part of the human lifespan the mean traits change slowly but relentlessly. For
example, it seems to be a universal rule that younger people are considerably
more extraverted and open than older people, whereas older people are perceived
to be more agreeable and conscientious than younger people (Allik, Realo, et al.,
2009; McCrae, Terracciano, & 78 Members of the Personality Profiles of
Cultures Project, 2005). Existing data seem to favor an explanation according to
which personality development through the lifespan follows a universal pattern
that is largely independent of the economic and political environment (Costa,
McCrae, et al., 2000). Consider a cross-sectional study of the observer-rated
personality traits of 7,065 Russians. Most adult targets were born before the first
satellite Sputnik was launched in 1957, and approximately 10% were born
before Stalin’s purges in 1937; there were even five targets born before the
Bolshevik revolution led by Vladimir Lenin in 1917 (Allik, Realo, et al., 2009).
In contrast, the college-age targets had lived most of their lives in the post-Soviet
era. These major historical events, very different from those experienced
elsewhere in the world, might have left their imprints on the personality of
targets, and uniquely Russian cohort effects might have created a distinctive
pattern of Russian age differences. Instead, age differences in general showed
the same pattern seen elsewhere: The difference profile between younger and
older Russians across the 30 NEO PI-R facet scales has almost exactly the same
shape as it has in the United States, Portugal, or Korea. These findings seem to
support the hypothesis that intrinsic maturational changes in the mean level of
personality traits are most likely genetically determined and relatively immune
to social and historical influences (Allik, Realo, et al., 2009).
According to the social investment theory (Helson, Kwan, John, & Jones,
2002; Roberts, Wood, & Smith, 2005), the normative personality trait
development in adulthood can be explained not by biological processes but by
societal demands and universal social roles faced in young adulthood. It is true,
as many researchers have noted (Donnellan, Conger, & Burzette, 2007; Soto,
John, Gosling, & Potter, 2011), that most personality changes are in the direction
of increased maturity. As a result, personality pathology tends to decline with
age, notably in the case of borderline PD. Although plausible, the social
investment theory is unable to explain some of the most basic facts about
personality development. An almost complete lack of susceptibility to social and
historic changes, which could radically modify social demands and roles, makes
explanations based on intrinsic maturation more plausible.
Differences in Perspective
Social psychologists have conducted countless numbers of laboratory
experiments to explore the idea that there is a fundamental disparity between the
way people perceive themselves and the way they are perceived by others (Jones
& Nisbett, 1971; Nisbett, Caputo, Legant, & Marecek, 1973; Watson, 1982).
This disparity is believed to originate from an inevitable asymmetry between
internal and external viewpoints: People are immersed in their own sensations,
emotions, and cognitions at the same time that their experience of others is
dominated by what can be observed externally (Pronin, 2008).
However, all these arguments about systematic differences between how
people see the personality traits of others and their own personality traits are
suspicious in view of the fact that normative self-rated personality mean scores
converge almost perfectly with normative observer-rated mean scores on
personality questionnaires (Allik, Realo, et al., 2010). There are truly miniature
differences between self-rated and observer-rated mean raw scores, amounting to
less than one-quarter standard deviation for most traits.
Nevertheless, the small differences that are seen demonstrate a cross-
culturally replicable pattern of difference between internal and external
perspectives for Big Five personality traits. People everywhere see themselves
(on average) as more neurotic and open to experience than they are seen by other
people. External observers generally hold a higher opinion of an individual’s
Conscientiousness than he or she does about himself or herself. As a rule, people
think that they have more positive emotions and excitement seeking but less
assertiveness than it seems from the vantage point of an external observer. This
cross-culturally replicable disparity between internal and external perspectives is
not consistent with predictions based on the actor-observer hypothesis, because
the size of the disparity was unrelated to the visibility of personality traits.
Surprisingly, a relatively strong negative correlation (r = –.53) between the
average self-minus-observer profile and social desirability ratings suggests that
people in most of the cultures studied view themselves less favorably than they
are perceived by others (Allik, Realo, et al., 2010). It is clear that our current
theories cannot explain the direction of the small differences in personality trait
perception, but once again personality processes have shown themselves to be
universal.
Are Mean Levels Universal?

There are several lines of evidence that suggest that the observed differences
in trait levels across cultures are real. As noted above, the FFM can be replicated
in culture-level analyses, because cultures that score higher on some definers of
a factor usually score higher on others: Variations in trait levels are thus not due
merely to random fluctuations introduced by translation. Culture-level means
show construct validity in a number of ways, including correlations of means
based on self-reports versus observer ratings (McCrae, Terracciano, & 78
Members of the Personality Profiles of Cultures Project, 2005; McCrae et al.,
2010) and correlations with other culture-level variables, such as individualism-
collectivism (Hofstede & McCrae, 2004).
Perhaps one of the most intriguing observations is that the geographic
distribution of mean scores of personality traits has a systematic pattern for both
self and observer ratings (Allik & McCrae, 2004b; McCrae, Terracciano, & 79
Members of the Personality Profiles of Cultures Project, 2005; Rentfrow et al.,
2008; Schmitt et al., 2007) across the world and with countries (Allik, Realo, et
al., 2009; Rentfrow, 2010). For example, it seems to be a general rule that
extraverted and open-minded people live predominantly in economically
prosperous, democratic, and individualistic countries (Allik & McCrae, 2004b).
However, the ranking of cultures on some personality traits may seem
counterintuitive. Indeed, it is not highly expected that the most conscientious—
determined, strong-willed, organized, dutiful, and deliberate—people live in
Burkina Faso and Congo whereas the least conscientious, according to the self-
reports at least, live in Japan and Korea (Mõttus, Allik, & Realo, 2010). These
informal observations can be supported by more rigorous analyses. For instance,
Heine and colleagues (2008) reanalyzed published data and showed that
aggregate national scores of self-reported Conscientiousness were, contrary to
the authors’ expectations, negatively correlated with various country-level
behavioral and demographic indicators of Conscientiousness, such as postal
workers’ speed, accuracy of clocks in public banks, accumulated economic
wealth, and life expectancy at birth. Oishi and Roth (2009) extended the list of
paradoxical findings by showing that nations with high self-reported
Conscientiousness were not less but more corrupt.
What should be done if we encounter a paradoxic relationship between
aggregate national scores of personality and some independent societal
measures? Before we can come to a final verdict on faults of culture-level
personality scores, we need to follow a simple list of prescriptions on how to
react in situations in which trait measures are not related to external criteria in an
expected manner (Mõttus et al., 2010). First, it is possible that the personality
traits used in predictive validity studies are sometimes too broad and only some
of their aspects are related to the expected criterion variable. Second, going for a
refined description of personality should be coupled with a rigorous and
comprehensive choice of external validity criteria. Ideally, the choice of criteria
should be based on a clear, theoretically sound account of the causal chain of
events that connects the ways of responding on personality scales to variations in
the expected external criterion variable. For instance, it has been tempting to
hypothesize that high culture-level means of Conscientiousness should yield
high accuracy of bank clocks (Heine et al., 2008). In fact, we need a causal
explanation as to how a greater proportion of conscientious people in a given
population help to get bank clocks, monitored by very small and probably very
unrepresentative fractions of populations, more accurate.
The third prescription is to consider alternative ways of conceptualizing the
puzzling relationships. The assumed links between external criteria and
personality mean that scores are often based on broad theoretical generalizations.
For instance, it is appealing to believe that the maintenance of democracy
presupposes not only efficient regulation and a transparent legal system but also
competent and responsible people. Therefore, it could be expected that in more
democratic countries citizens are more responsible and disciplined, resulting in a
positive correlation between the level of democracy and the mean national
scores of Conscientiousness. However, the relationship may almost equally well
go the other way around—it is a dictatorship that better enforces hard work,
discipline, and order in society. In reality, an effective democracy is much more
likely to be found in cultures with a strong emphasis on self-expression values,
whereas dutifulness, order, and hard work are the correlates of survival, the
opposite of self-expression (Inglehart & Welzel, 2005). Thus, it can be argued
that in countries with higher scores of Conscientiousness—that is, where people
are rule abiding, inhibited by social constraints, and keen on keeping order—
people are not able to realize their potential for freedom and autonomy, which,
in turn, are the cornerstones of democracy (Mõttus et al., 2010).
Scepticism concerning the validity of the country mean scores was also
stimulated by the publication of surprising findings that indicated that national
character ratings did not converge with assessed personality traits (Hřebíčková &
Graf, 2013; McCrae et al., 2013; McCrae, Terracciano, Realo, & Allik, 2007;
Realo et al., 2009; Terracciano et al., 2005). The lack of correspondence
between national stereotypes and assessed personality traits elicited a vigorous
debate (Ashton, 2007; McGrath & Goldberg, 2006; Perugini & Richetin, 2007;
Robins, 2005). Among critical comments, Leon Festinger’s social comparison
processes—the idea that people estimate their attitudes or dispositions relative to
social standards (Festinger, 1954)—was repeatedly mentioned. These frame-of-
reference explanations are very seductive in their simplicity, but they are
certainly much easier to propose than prove. For example, Heine and colleagues
(2008) proposed that people likely bring to mind a standard that lies outside their
own culture, for example, a perceived international norm (Heine et al., 2008), in
making their own ratings. Yet how could laypersons have such an extraordinary
ability to obtain accurate information about the mean levels of personality traits
across many countries when psychologists find it so difficult?
Other fields, also facing the reference-level problem, have learned to cope
with it. Health studies, for example, are familiar with the paradox that in those
countries and regions in which people complain more about serious health
problems, people are in fact healthier and live longer (Sen, 2002). Another
example are Americans who are decidedly less satisfied with their income than
the Dutch even if their real incomes are more or less comparable (Kapteyn,
Smith, & Van Soest, 2013). A Harvard political scientist, Gary King, proposed
the use of anchoring vignettes—brief descriptions of hypothetical persons—
along with self-reports (Hopkins & King, 2010; King, Murray, Salomon, &
Tandon, 2004; King & Wand, 2007). Provided that the anchoring vignettes
display various levels of the same characteristic that is being measured by self-
reports, it is possible to determine the relative position of self-ratings among the
hypothetical persons depicted in the vignettes. In a recent study, anchoring
vignettes were used to test whether people from 21 countries (Australia, Benin,
Burkina Faso, People’s Republic of China, Estonia, Germany, Hong-Kong,
Japan, Lithuania, Malaysia, Mali, Mauritius, Philippines, Poland, Russia,
Senegal, South Africa, South Korea, Sweden, Switzerland, and the United
States) have different standards for Conscientiousness (Mõttus, Allik, Realo,
Pullmann, et al., 2012). All participants rated their own Conscientiousness and
that of the 30 hypothetical persons portrayed in the short vignettes, with the
latter ratings expected to reveal individual differences in standards of
Conscientiousness. Contrary to expectations of the reference-level theorists, the
vignettes were rated relatively similarly in all cultures, suggesting no substantial
culture-related differences in standards for Conscientiousness. Controlling for
the small differences in these standards did not substantially change the rankings
of countries on mean self-ratings and the predictive validities of these rankings
for objective criteria. These findings lend little support to the hypothesis that
mean self-rated Conscientiousness scores are influenced by culture-specific
standards, considerably restricting the range of potential explanations for the
puzzling country rankings in Conscientiousness (Mõttus, Allik, Realo, Pullmann,
et al., 2012). Although it is premature to draw any firm conclusions, it may be
that personality traits are estimated in absolute rather than in relative terms. It is
possible, for example, that people have developed a more robust and
unconditional way of judging their basic tendencies to feel, think, and behave
than for judging the level of political freedom in their society or their work
satisfaction (Kapteyn et al., 2013; King, Murray, Salomon, & Tandon, 2003;
Kristensen & Johansson, 2008).
What Is Behind Personality Universals?

Searching for personality universals cannot be a goal in itself; they need to be
explained, because they may reveal the most fundamental properties of human
personality. The observation of universal properties has already stimulated some
theoretical explanations. Five Factor Theory (FFT) emerged as a response to
challenges posed by these recently discovered universalities (McCrae & Costa,
1996, 1999, 2003). In contrast to the FFM, which is an empirical summary about
the covariation of personality traits, FFT is an attempt to explain universal or
near-universal properties of human personality. How can we understand the
extraordinary stability of personality traits across the human lifespan (McCrae &
Costa, 2003)? Why does the same pattern of covariation among personality traits
emerge in countries with completely different economic prosperity, historical
experience, and cultural traditions (McCrae, Terracciano, & 78 Members of the
Personality Profiles of Cultures Project, 2005; Schmitt et al., 2007)? Why are the
effects of heritability overwhelming compared to the vanishingly small effects of
the shared environment (McCrae, Jang, Livesley, Riemann, & Angleitner, 2001;
Riemann, Angleitner, & Strelau, 1997; Yamagata et al., 2006)? Or why do life
events show very little influence on the levels of personality traits (Costa,
Herbst, McCrae, & Siegler, 2000)? Or why a utopia to create a new human
being, homo sovieticus, was a total failure (Angleitner & Ostendorf, 2000)? The
FFT was a response to these and other challenges and provides an explanation
for most of these personality universals or near-universals established during the
past few decades. According to the FFT, personality traits are basic tendencies,
deeply rooted in the organism, that are relatively immune to influences from the
environment (McCrae & Costa, 1996, 1999, 2003).
The main purpose of that central “dogma”—in the sense that Francis Crick
(1990) used that word—of FFT was to provide a clear basis for formulating a
testable hypothesis (Allik & McCrae, 2002). Postulating a general heuristic
principle that there is no “transfer” from culture and life experience to basic
personality traits obviously stimulates the search for conditions—certainly not
very frequent ones—in which this general postulate is violated (Allik & McCrae,
2002). However, finding these violations may be more difficult than is claimed
by the critics of FFT. Typically those who oppose the central “dogma”
concentrate on one particular detail, forgetting about the whole picture. Another
source of opposition is a conflicting “dogma” according to which personality,
like the human mind in general, is almost entirely shaped by culture (Barkow,
Cosmides, & Tooby, 1992). Many critics of FFT seem to suffer from a false
impression that it is an easy task to find evidence demonstrating how culture
determines personality.
Challenges to Universality
More than 10 years ago Allik and McCrae (2002) reviewed evidence that
could challenge the central “dogma” of the FFT in the context of cross-cultural
research. There are four groups of findings that appear to be inconsistent with
the basic hypothesis about the immunity of personality traits to cultural
influences (Allik & Realo, 2013).
Unique traits. The first is related to an aspiration to find unique personality
traits or exceptional combinations of common traits that are specific to one
culture alone (Cheung et al., 2011). As we have argued above, all these attempts
to establish unique or indigenous traits were mainly futile because what was
believed to be a unique trait was found to be applicable to other cultures as well,
although perhaps not exactly in the same degree of salience.
Acculturation. The second involves personality differences associated with
acculturation. McCrae and colleagues examined Chinese undergraduates living
in Hong Kong and Vancouver (McCrae, Yik, Trapnell, Bond, & Paulhus, 1998).
Canadian-born Chinese shared many features of the Hong Kong personality
profile, but they also demonstrated significant acculturation effects: Their
profiles were more similar to European Canadians than the personality profiles
of recent emigrants from Hong Kong. It is quite appalling that it took so long to
use the same study design to examine how acculturation shapes personality traits
(Güngör et al., 2013; Söldner, 2013; Zimmermann & Neyer, 2013). This new
wave of studies provided support for the view that personality can be subjected
to cultural influence. For example, Japanese Americans became more
“American” and less “Japanese” in their personality as they reported higher
participation in the U.S. culture (Güngör et al., 2013). However, these studies
also showed that a more sophisticated methodology is needed to study
acculturation. To make firm conclusions about acculturation, it is necessary to
determine personality traits before sojourning in a new cultural environment. It
was shown that many acculturation effects can be explained by self-selection:
Open and extraverted people are more likely to contact new cultures in which
they become even more open and experience more positive emotions (Söldner,
2013; Zimmermann & Neyer, 2013). Thus, acculturation is a specific case in
which the central “dogma” could be violated. It seems, however, that the size of
this violation is smaller than initially expected.
Language. Evidence from the third group attempts to demonstrate that
language influences the personality of the speaker. When a speaker switches
from one language to another, his or her gestures may change. Is switching
languages accompanied by a change of personality? This is, of course, a part of a
more general question: Does the language you speak affect the way you think
(Deutscher, 2010)? Some researchers are optimistic in answering this question;
they believe that bilinguals may have two personalities that they can switch with
the change of language (Ramirez-Esparza et al., 2006). For example, when
German and Spanish versions of the NEO Five Factor Inventory were
administered to two groups of bilinguals of these two languages both groups,
regardless of the individual’s first language, scored higher on Extraversion and
Neuroticism when Spanish was the test language (Veltkamp, Recio, Jacobs, &
Conrad, 2013). This is a sufficient reason to conclude that language indeed
modulates the way in which respondents answer personality items. It is less
certain that these two shifts in scores for Extraversion and Neuroticism indicate
the individual’s access to multiple cultural meaning systems and the ability of
the individual to switch between different culturally appropriate behaviors. For
example, it is possible that these two versions of the questionnaire were not fully
identical. To achieve the same level of expression on some trait it is necessary to
use more extreme answers in one language than another. Nobody can exclude
the possibility that switching languages results in small changes in the speaker’s
personality. To reach this conclusion, however, it is necessary to eliminate a
number of more trivial explanations.
Unfortunately, the number of studies in which two versions of the same
personality questionnaire are administered to bilinguals is regrettably small. In
one of these relatively rare studies, Konstabel (1999) asked bilinguals to answer
the NEO PI-R questionnaire both in Estonian and Russian. Except for some
trivial differences these two versions were practically identical. Although
personality stereotypes of these two ethnic groups are dramatically contrasting
(Realo et al., 2009), the language in which personality items were formulated
has only a negligible effect on answers (Konstabel, 1999). Likewise, the
occurrence of cultural frame shifting was mostly negligible when Swedish-Finns
switched their answers from one language to another (Lönnqvist, Konstabel,
Lönnqvist, & Verkasalo, 2014).
Cohort effects. When the same questionnaire is used without change over a
considerable period of time it is possible to determine whether men and women
of the same age but born in different years have identical personality scores. Age
differences in the mean level of adult personality traits are rather small and
longitudinal changes of individual scores are even smaller (McCrae & Costa,
2003). Nevertheless, Jean Twenge has reported several meta-analyses showing
dramatic cohort effects (Twenge, 2000, 2001; Twenge & Campbell, 2002, 2008;
Twenge & Im, 2007). For example, American college students have increased in
both Neuroticism and Extraversion by nearly a full standard deviation over the
past half century (Twenge, 2000, 2001).
Surveys and polls show that people’s values, habits, and social practices
change all the time (Inglehart, Basanez, Diez-Medrano, Halman, & Luijkx,
2004; Putnam, 2000). It is tempting to believe that these changes also affect
personality traits. Do dramatic cohort effects demonstrate the way in which
culture and society modify personality? Not necessarily. Personality scores are
always a mixture of various components. For example, if respondents were
asked to describe an ideal person their ratings were highly correlated with their
self-description. These descriptions were, in turn, correlated with personality
descriptions attributed to the typical representative of their own nation (Allik,
Mõttus, & Realo, 2010; Allik, Mõttus, et al., 2009). Thus, along with the
distinctive personality traits ratings there are also components representing
information about an average person and social desirability. As already noted
(Allik & McCrae, 2002), scales analyzed by Twenge are mainly keyed in a
positive direction. It is well documented that cultures vary considerably based on
extreme and neutral responding (Mõttus, Allik, Realo, Rossier, et al., 2012).
Together with extreme and neutral responding it is possible that the
acquiescence bias—a tendency to agree with all the questions (Smith, 2004)—
also decreases with time. Because artifactual explanations of the cohort effects
cannot be automatically discarded, it is possible that the “true” personality
scores, which remain after all potential biases are partialled out, have not
changed much over the past years.
It is also important to remember that the interpretation of cross-sectional data
is tricky. Secular trends reported by Twenge were difficult to replicate in more
representative samples and the inspection of effect sizes provided little evidence
for strong or widespread cohort-linked changes (Terracciano, 2010;
Trzesniewski & Donnellan, 2010). Moreover, separation of cohort effects from
the age of participants and study period effects is not a trivial exercise (e.g.,
Realo & Dobewall, 2011). Usually there is simply not enough data to separate
the cohort effect from the effect of age and study moment of participants. In
these singular cases and acknowledging these distinctions it is indeed possible to
demonstrate that people who were born in a particular year are on average less
satisfied with their lives than those who were born only a decade earlier or later
(Realo & Dobewall, 2011). Again, that the cohort effects on personality exist is
not a real problem. The problem is their size, which might be too small to have
any theoretical or practical consequences.
Conclusions
It is occasionally argued that personality psychologists are more interested in
individual differences than in searching for universals as is done in physics or
chemistry. Nobody seems to question, publicly at least, that it is an essential
function of science to seek universals and if personality psychologists are
determined to follow a scientific pursuit then they also need to look for invariant
properties in their data. When Galileo dropped objects of different materials and
weight from the Leaning Tower of Pisa he looked for a property that is common
to all matter (Allik et al., 2013). In the same way, personality psychologists—
perhaps even more than in some other disciplines—are deeply interested in what
is common to personality descriptions in all cultures and societies (McCrae,
2009). Only these universal features could provide a satisfying answer to the
question of why some people are happy and others feel miserable, why some are
laborious and others are lackadaisical, and why some are inquisitive while others
are satisfied with what they already have.
References
Allik, J. (2005). Personality dimensions across cultures. Journal of Personality Disorders,
19(3), 212–232.
Allik, J. (2012). Personality psychology in the first decade of the new millennium: A
bibliometric portrait. European Journal of Personality, 27, 5–14.
Allik, J., Laidra, K., Realo, A., & Pullmann, H. (2004). Personality development from 12
to 18 years of age: Changes in mean levels and structure of traits. European Journal of
Personality, 18(6), 445–462.
Allik, J., & McCrae, R. R. (2002). A Five-Factor Theory perspective. In R. R. McCrae &
J. Allik (Eds.), The Five Factor Model of personality across cultures (pp. 303–322).
Allik, J., & McCrae, R. R. (2004a). Escapable conclusions: Toomela (2003) and the
universality of trait structure. Journal of Personality and Social Psychology, 87, 261–
265.
Allik, J., & McCrae, R. R. (2004b). Toward a geography of personality traits: Patterns of
profiles across 36 cultures. Journal of Cross-Cultural Psychology, 35(1), 13–28.
Allik, J., Mõttus, R., & Realo, A. (2010). Does national character reflect mean personality
traits when both are measured by the same instrument? Journal of Research in
Allik, J., Mõttus, R., Realo, A., Pullmann, H., Trifonova, A., McCrae, R. R., … 55
Members of the Russian Character and Personality Survey. (2009). How national
character is constructed: Personality traits attributed to the typical Russian.
Psychological Journal of International University of Nature, Society and Human
“Dubna,” 2, 1–23.
Allik, J., & Realo, A. (2013). Kultuur ja isiksus/Culture and personality/. Tuna:
Ajalookultuuri Ajakiri, 16(4), 2–25.
Allik, J., Realo, A., & McCrae, R. R. (2013). Universality of the Five-Factor Model of
personality. In P. T. Costa & T. Widiger (Eds.), Personality disorders and the Five
Factor Model of personality (pp. 61–74). Washington, DC: American Psychological
Association.
How people see others is different from how people see themselves: A replicable
pattern across cultures. Journal of Personality and Social Psychology, 94, 168–182.
Person-fit to the Five Factor Model of personality. Swiss Journal of Psychology, 71(1),
35–45.
Allik, J., Realo, A., Mõttus, R., Pullmann, H., Trifonova, A., McCrae, R. R., …
Korneeva, E. E. (2011). Personality profiles and the “Russian Soul”: Literary and
scholarly views evaluated. Journal of Cross-Cultural Psychology, 42(3), 372–389.
Allik, J., Realo, A., Mõttus, R., Pullmann, H., Trifonova, A., McCrae, R. R., & 56
Members of the Russian Character and Personality Survey. (2009). Personality traits of
Russians from the observer’s perspective. European Journal of Personality, 23, 567–
588.
Angleitner, A., & Ostendorf, F. (2000). The FFM: A comparison of German speaking
countries (Austria, Former East and West Germany, and Switzerland). Paper presented
at the XXVIIth International Congress of Psychology, Stockholm, Sweden.
Ashton, M. C. (2007). Self-reports and stereotypes: A comment on McCrae et al.
Ashton, M. C., Lee, K., & de Vries, R. E. (2014). The HEXACO Honesty-Humility,
Agreeableness, and Emotionality Factors: A review of research and theory. Personality
and Social Psychology Review, 18(2), 139–152.
Barkow, J. H., Cosmides, L., & Tooby, J. (1992). The Adapted mind: Evolutionary
psychology and the generation of culture. New York, NY: Oxford University Press.
Behrens, K. Y. (2004). A multifaceted view of the concept of amae: Reconsidering the
indigenous Japanese concept of relatedness. Human Development, 47(1), 1–27.
Benet-Martinez, V., & John, O. P. (1998). Los Cinco Grandes across cultures and ethnic
groups: Multitrait multimethod analyses of the Big Five in Spanish and English.
Berry, J. B., Poortinga, Y. H., Segall, M. H., & Dasen, P. R. (2002). Cross-cultural
psychology: Research and applications (2nd ed.). Cambridge, England: Cambridge
University Press.
Bleidorn, W., & Ostendorf, F. (2009). A Big Five inventory for children and adolescents:
The German version of the Hierarchical Personality Inventory for Children (HiPIC).
Diagnostica, 55(3), 160–173.
Block, J. (1995). A contrarian view of the 5-Factor approach to personality description.
Block, J. (2010). The Five-Factor framing of personality and beyond: Some ruminations.
Psychological Inquiry, 21(1), 2–25.
Borkenau, P., McCrae, R. R., & Terracciano, A. (2013). Do men vary more than women
in personality? A study in 51 cultures. Journal of Research in Personality, 47(2), 135–
144.
Borsboom, D. (2005). Measuring the mind: Conceptual issues in contemporary
psychometrics. New York, NY: Cambridge University Press.
Borsboom, D., Mellenbergh, G. J., & Van Heerden, J. (2003). The theoretical status of
latent variables. Psychological Review, 110(2), 203–219.
Briggs, J. L. (1970). Never in anger: Portrait of an Eskimo family. Cambridge, MA:
Harvard University Press.
Brown, D. (1991). Human universals. New York, NY: McGraw-Hill.
Bułat Silva, Z. (2012). Saudade: A key Portuguese emotion. Emotion Review, 4(2), 203–
211.
Burquest, D. A., & Payne, D. (1993). Phonological analysis: A functional approach.
Dallas, TX: Summer Institute of Linguistics.
Cheung, F. M., Cheung, S. F., Wada, S., & Zhang, J. X. (2003). Indigenous measures of
personality assessment in Asian countries: A review. Psychological Assessment, 15(3),
280–289.
Cheung, F. M., Leung, K., Zhang, J. X., Sun, H. F., Gan, Y. Q., Song, W. Z., & Xie, D.
(2001). Indigenous Chinese personality constructs—Is the Five-Factor Model
complete? Journal of Cross-Cultural Psychology, 32, 407–433.
Cheung, F. M., van de Vijver, F. J. R., & Leong, F. T. L. (2011). Toward a new approach
to the study of personality in culture. American Psychologist, 66(7), 593–603.
Chomsky, N. (1965). Aspects of a theory of syntax. Cambridge, MA: MIT Press.
Chon, K. K. (2002). Cultural aspects of anger. In C. von Hofsten & L. Bäckman (Eds.),
Psychology at the turn of the millennium: Social, developmental, and clinical
perspectives (Vol. 2, pp. 323–346). East Sussex, England: Psychology Press.
Costa, P. T., Herbst, J. H., McCrae, R. R., & Siegler, I. C. (2000). Personality at midlife:
Stability, intrinsic maturation, and response to life events. Assessment, 7(4), 365–378.
Costa, P. T., McCrae, R. R., Martin, T. A., Oryol, V. E., Senin, I. G., Rukavishnikov, A.
A., … Molfese, D. L. (2000). Personality development from adolescence through
adulthood: Further cross-cultural comparisons of age differences. In V. J. Molfese & D.
L. Molfese (Eds.), Temperament and personality development across life span (pp.
235–252). Mahwah, NJ: Lawrence Erlbaum Associates.
Costa, P. T., Terracciano, A., & McCrae, R. R. (2001). Gender differences in personality
traits across cultures: Robust and surprising findings. Journal of Personality and Social
Crick, F. (1990). What mad pursuit. A personal view of scientific discovery. London,
England: Penguin.
De Fruyt, F., Mervielde, I., & Van Leeuwen, K. (2002). The consistency of personality
type classification across samples and Five-Factor measures. European Journal of
Personality, 16, S57–S72.
Deutscher, G. (2010). Through the language glass: Why the world looks different in other
languages. London, England: Arrow Books.
Donnellan, M. B., Conger, R. D., & Burzette, R. G. (2007). Personality development
from late adolescence to young adulthood: Differential stability, normative maturity,
and evidence for the maturity-stability hypothesis. Journal of Personality, 75(2), 237–
263.
Everett, D. L. (2005). Cultural constraints on grammar and cognition in Piraha—Another
look at the design features of human language. Current Anthropology, 46(4), 621–646.
Questionnaire. London, England: Hodder and Stoughton.
Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7, 117–
140.
Funder, D. C. (2012). Accurate personality judgment. Current Directions in
Gosling, S. D., Vazire, S., Srivastava, S., & John, O. P. (2004). Should we trust web-
based studies? A comparative analysis of six preconceptions about Internet
questionnaires. American Psychologist, 59(2), 93–104.
Guimond, S., Branscombe, N. R., Brunot, S., Buunk, A. P., Chatard, A., Desert, M., …
Yzerbyt, V. (2007). Culture, gender, and the self: Variations and impact of social
comparison processes. Journal of Personality and Social Psychology, 92(6), 1118–
1134.
Güngör, D., Bornstein, M. H., De Leersnyder, J., Cote, L., Ceulemans, E., & Mesquita, B.
(2013). Acculturation of personality: A three-culture study of Japanese, Japanese
Americans, and European Americans. Journal of Cross-Cultural Psychology, 44(5),
701–718.
Gurven, M., von Rueden, C., Massenkoff, M., Kaplan, H., & Lero Vie, M. (2013). How
universal is the Big Five? Testing the Five-Factor Model of personality variation
among forager-farmers in the Bolivian Amazon. Journal of Personality and Social
Psychology, 104(2), 354–370.
Heine, S. J., Buchtel, E. E., & Norenzayan, A. (2008). What do cross-national
comparisons of personality traits tell us? The case of conscientiousness. Psychological
Science, 19(4), 309–313.
Helson, R., Kwan, V. S. Y., John, O. P., & Jones, C. (2002). The growing evidence for
personality change in adulthood: Findings from research with personality inventories.
Henrich, J., Heine, S. J., & Norenzayan, A. (2010a). Most people are not WEIRD.
Nature, 466(7302), 29.
Henrich, J., Heine, S. J., & Norenzayan, A. (2010b). The weirdest people in the world?
Behavioral and Brain Sciences, 33(2–3), 1–75.
Hofstede, G., & McCrae, R. R. (2004). Personality and culture revisited: Linking traits
and dimensions of culture. Cross-Cultural Research, 38, 52–88.
Hopkins, D. J., & King, G. (2010). Improving anchoring vignettes designing surveys to
correct interpersonal incomparability. Public Opinion Quarterly, 74(2), 201–222.
House, R. J., Hanges, P. J., Javidan, M., Dorfman, P. W., & Gupta, V. (2004). Culture,
leadership, and organizations. The GLOBE study of 62 societies. Thousand Oaks, CA:
Sage.
Hřebíčková, M., & Graf, S. (2013). Accuracy of national stereotypes in central Europe:
Outgroups are not better than ingroups in considering personality traits of real people.
Hua, C. (2001). A society without farther or husband: The Na of China. New York, NY:
Zone Book.
Inglehart, R., Basanez, M., Diez-Medrano, J., Halman, L., & Luijkx, R. (2004). Human
beliefs and values: A cross-cultural sourcebook based on the 1999–2002 values
surveys. Mexico: Siglo XXI Editores.
Inglehart, R., & Welzel, C. (2005). Modernization, cultural change, and democracy: The
human development sequence. New York, NY: Cambridge University Press.
Jones, D. (2010). A WEIRD view of human nature skews psychologists’ studies. Science,
328(5986), 1627.
Jones, E. E., & Nisbett, R. E. (1971). The actor and the observer: Divergent perceptions
of the causes of behavior. Morristown, NJ: General Learning Press.
Kallasmaa, T., Allik, J., Realo, A., & McCrae, R. R. (2000). The Estonian version of the
NEO-PI-R: An examination of universal and culture-specific aspects of the Five-Factor
Model. European Journal of Personality, 14, 265–278.
Kapteyn, A., Smith, J. P., & Van Soest, A. (2013). Are Americans really less happy with
their incomes? Review of Income and Wealth, 59(1), 44–65.
King, G., Murray, C. J. L., Salomon, J. A., & Tandon, A. (2003). Enhancing the validity
and cross-cultural comparability of measurement in survey research. American
Political Science Review, 97(4), 567–583.
King, G., Murray, C. J. L., Salomon, J. A., & Tandon, A. (2004). Enhancing the validity
and cross-cultural comparability of measurement in survey research (Vol 97, Pg 567,
2003). American Political Science Review, 98(1), 191–207.
King, G., & Wand, J. (2007). Comparing incomparable survey responses: Evaluating and
selecting anchoring vignettes. Political Analysis, 15(1), 46–66.
Konstabel, K. (1999). A bilingual retest study of the Revised NEO Personality Inventory:
A comparison of Estonian and Russian versions. Unpublished Master’s Thesis:
University of Tartu, Tartu, Estonia.
Kristensen, N., & Johansson, E. (2008). New evidence on cross-country differences in job
satisfaction using anchoring vignettes. Labour Economics, 15(1), 96–117.
Leach, C. W., Spears, R., Branscombe, N. R., & Doosje, B. (2003). Malicious pleasure:
Schadenfreude at the suffering of another group. Journal of Personality and Social
Psychology, 84(5), 932–943.
personality lexicons of English and 11 other languages. Journal of Personality, 76(5),
1001–1053.
Leroi, A. M. (2005). Mutants: On the form, varieties and errors of the human body.
London, England: Harper Perennial.
Lévi-Strauss, C. (1969/1949). The elementary structures of kinship. Boston, MA: Beacon
Press.
Lewis, P. M., Simons, G. F., & Fennig, C. D. (2014). Ethnologue: Languages of the
world (17th ed.). Dallas, TX: SIL International.
Lin, E. J. L., & Church, A. T. (2004). Are indigenous Chinese personality dimensions
culture-specific? An investigation of the Chinese Personality Assessment Inventory in
Chinese American and European American samples. Journal of Cross-Cultural
Lippa, R. A. (2010). Sex differences in personality traits and gender-related occupational
preferences across 53 nations: Testing evolutionary and social-environmental theories.
Archives of Sexual Behavior, 39(3), 619–636.
Löckenhoff, Corinna E., Chan, Wayne, McCrae, Robert R., De Fruyt, Filip, Jussim, Lee,
De Bolle, Marleen, … Terracciano, Antonio. (2014). Gender stereotypes of personality
universal and accurate? Journal of Cross-Cultural Psychology, 45(5), 675–694.
Lodhi, P. H., Deo, S., & Belhekar, V. M. (2002). The Five-Factor Model of personality:
Measurement and correlates in the Indian context. In R. R. McCrae & J. Allik (Eds.),
The Five Factor Model of personality across cultures (pp. 227–248). New York, NY:
Kluwer Academic/Plenum Publishers.
Lönnqvist, J. E., Konstabel, K., Lönnqvist, N., & Verkasalo, M. (2014). Accuracy,
consensus, in-group bias, and cultural frame shifting in the context of national
character stereotypes. Journal of Social Psychology, 154(1), 40–58.
Lynn, R., & Martin, T. (1995). National differences for thirty-seven nations in
extraversion, neuroticism, psychoticism and economic, demographic and other
correlates. Personality and Individual Differences, 19, 403–406.
Malik, S. (2014). Polydactyly: Phenotypes, genetics and classification. Clinical Genetics,
85(3), 203–212.
Martin, P., Long, M. V., & Poon, L. W. (2002). Age changes and differences in
personality traits and states of the old and very old. Journals of Gerontology Series B-
Psychological Sciences and Social Sciences, 57(2), P144–P152.
McCrae, R. R. (2002). NEO-PI-R data from 36 cultures: Further intercultural
comparisons. In R. R. McCrae & J. Allik (Eds.), The Five-Factor Model of personality
McCrae, R. R. (2009). The physics and chemistry of personality. Theory & Psychology,
19(5), 670–687.
McCrae, R. R., & Allik, J. (2002). The Five-Factor Model of personality across cultures.
McCrae, R. R., Chan, W., Jussim, L., De Fruyt, F., Löckenhoff, C. E., De Bolle, M., …
Terracciano, A. (2013). The inaccuracy of national character stereotypes. Journal of
McCrae, R. R., & Costa, P. T. (1996). Towards a new generation of personality theories:
Theoretical context for the Five-Factor Theory. In J. S. Wiggins (Ed.), The Five-Factor
Model of personality: Theoretical perspectives (pp. 51–87). New York, NY: Guilford
Press.
McCrae, R. R., & Costa, P. T. (1997). Personality trait structure as a human universal.
McCrae, R. R., & Costa, P. T. (1999). A Five-Factor Theory of personality. In L. A.
Pervin & O. P. John (Eds.), Handbook of personality: Theory and research (Vol. 2, pp.
McCrae, R. R., & Costa, P. T. (2003). Personality in adulthood: A Five-Factor Theory
perspective. New York, NY: Guilford Press.
McCrae, R. R., & Costa, P. T. (2008). Empirical and theoretical status of the Five-Factor
Model of personality traits In G. Boyle, G. Matthews, & D. Saklofske (Eds.), Sage
handbook of personality theory and assessment (Vol. 1, pp. 273–294). Los Angeles,
CA: Sage.
McCrae, R. R., & Costa, P. T., Jr. (2010). NEO Inventories professional manual. Odessa,
McCrae, R. R., Costa, P. T., Terracciano, A., Parker, W. D., Mills, C. J., De Fruyt, F., &
Mervielde, I. (2002). Personality trait development from age 12 to age 18:
covariation of personality traits. Journal of Personality, 69(4), 511–535.
McCrae, R. R., & John, O. P. (1992). An introduction to the Five-Factor Model and its
McCrae, R. R., & Terracciano, A. (2008). The Five-Factor Model and its correlates in
individuals and cultures. In F. van de Vijver, D. A. van Hemert, & Y. H. Poortinga
(Eds.), Multilevel analysis of individuals and cultures (pp. 249–283). New York, NY:
Lawrence Erlbaum.
Project. (2005). Personality profiles of cultures: Aggregate personality traits. Journal of
McCrae, R. R., Terracciano, A., De Fruyt, F., De Bolle, M., Gelfand, M. J., Costa, P. T.,
& Project, 42 Collaborators of the Adolescent Personality Profiles of Cultures. (2010).
The validity and structure of culture-level personality scores: Data from ratings of
young adolescents. Journal of Personality, 78(3), 815–838.
McCrae, R. R., Terracciano, A., Realo, A., & Allik, J. (2007). On the validity of culture-
level personality and stereotype scores. European Journal of Personality, 21(8), 987–
991.
McCrae, R. R., Yik, M. S. M., Trapnell, P. D., Bond, M. H., & Paulhus, D. L. (1998).
Interpreting personality profiles across cultures: Bilingual, acculturation, and peer
rating studies of Chinese undergraduates. Journal of Personality and Social
Psychology, 74(4), 1041–1055.
McGrath, R. E., & Goldberg, L. R. (2006). How to measure national stereotypes?
Science, 311(5762), 776–777.
Mervielde, I., Buyst, V., & De Fruyt, F. (1995). The validity of the Big Five as a model
for teachers ratings of individual differences among children aged 4-12 years.
Mesquita, B., Frijda, N. H, & Scherer, K. R. (1997). Basic processes and human
development. In J. Berry, P. R. Dasen, & T. S. Saraswathi (Eds.), Handbook of cross-
cultural psychology (Vol. 2, pp. 255–297). Boston, MA: Allyn & Bacon.
Molenaar, P. C. M., & Campbell, C. G. (2009). The new person-specific paradigm in
psychology. Current Directions in Psychological Science, 18(2), 112–117.
Mõttus, R., Allik, J., & Realo, A. (2010). An attempt to validate national mean scores of
conscientiousness: No necessarily paradoxical findings. Journal of Research in
Mõttus, R., Allik, J., Realo, A., Pullmann, H., Rossier, J., Zecca, G., … Tseung, C. N.
(2012). Comparability of self-reported conscientiousness across 21 countries.
Mõttus, R., Allik, J., Realo, A., Rossier, J., Zecca, G., Ah-Kion, J., … Johnson, W.
(2012). The effect of response style on self-reported conscientiousness across 20
countries. Personality and Social Psychology Bulletin, 38(11), 1423–1436.
Mõttus, R., Johnson, W., & Deary, I. J. (2012). Personality traits in old age: Measurement
and rank-order stability and some mean-level change. Psychology and Aging, 27(1),
243–249.
Nisbett, R. E., Caputo, C., Legant, P., & Marecek, J. (1973). Behavior as seen by the
actor and as seen by the observer. Journal of Personality and Social Psychology, 27,
154–164.
Norenzayan, A., & Heine, S. J. (2005). Psychological universals: What are they and how
can we know? Psychological Bulletin, 131(5), 763–784.
Oishi, S., & Roth, D. P. (2009). The role of self-reports in culture and personality
research: It is too early to give up on self-reports. Journal of Research in Personality,
43(1), 107–109.
Panayiotou, G., Kokkinos, C. M., & Spanoudis, G. (2004). Searching for the “Big Five”
in a Greek context: The NEO-FFI under the microscope. Personality and Individual
Differences, 36(8), 1841–1854.
Patel, V., & Winston, M. (1994). Universality of mental-illness revisited: Assumptions,
artifacts and new directions. British Journal of Psychiatry, 165, 437–440.
Perugini, M., & Richetin, J. (2007). In the land of the blind, the one-eyed man is king.
Piedmont, R. L., Bain, E., McCrae, R. R., & Costa, P. T. (2002). The applicability of the
Five-Factor Model in sub-Saharan culture. In R. R. McCrae & J. Allik (Eds.), The
Five-Factor Model of personality across cultures (pp. 155–173). New York, NY:
Kluwer Academic/Plenum Publishers.
Pinker, S. (2002). The blank slate: The modern denial of human nature. London,
England: Penguin Books.
Pronin, E. (2008). How we see ourselves and how we see others. Science, 320(5880),
1177–1180.
Pullmann, H., Allik, J., & Realo, A. (2009). Global self-esteem across the life span: A
cross-sectional comparison between representative and self-selected Internet samples.
Experimental Aging Research 35, 20–44.
Putnam, R. D. (2000). Bowling alone: The collapse and revival of American community.
New York, NY: Simon & Schuster.
Ramirez-Esparza, N., Gosling, S. D., Benet-Martinez, V., Potter, J. P., & Pennebaker, J.
W. (2006). Do bilinguals have two personalities? A special case of cultural frame
switching. Journal of Research in Personality, 40(2), 99–120.
Ramirez-Esparza, N., Gosling, S. D., & Pennebaker, J. W. (2008). Paradox lost:
Unraveling the puzzle of simpatia. Journal of Cross-Cultural Psychology, 39(6), 703–
715.
Realo, A., Allik, J., Verkasalo, M., Lönnqvist, J. E., Kwiatkowska, A., Kööts, L., …
Renge, V. (2009). Mechanisms of the national character stereotype: How people in six
neighboring countries of Russia describe themselves and the typical Russian. European
Realo, A., & Dobewall, H. (2011). Does life satisfaction change with age? A comparison
of Estonia, Finland, Latvia, and Sweden. Journal of Research in Personality, 45(3),
297–308.
Rentfrow, P. J. (2010). Statewide differences in personality toward a psychological
geography of the United States. American Psychologist, 65(6), 548–558.
Psychology, 105(6), 996–1012.
Rentfrow, P. J., Gosling, S. D., & Potter, J. (2008). A theory of the emergence,
persistence, and expression of geographic variation in psychological characteristics.
Perspectives on Psychological Science, 3(5), 339–369.
Rentfrow, P. J., Mellander, C., & Florida, R. (2009). Happy states of America: A state-
level analysis of psychological, economic, and social well-being. Journal of Research
in Personality, 43(6), 1073–1082.
FFI scales. Journal of Personality, 65(3), 449–475.
Roberts, B. W., Wood, D., & Smith, J. L. (2005). Evaluating Five Factor Theory and
social investment perspectives on personality trait development. Journal of Research in
Personality, 39(1), 166–184.
Robins, R. W. (2005). The nature of personality: Genes, culture, and national character.
Science, 310, 62–63.
Rolland, J. P. (2002). The cross-cultural generalizability of the Five-Factor Model of
personality. In R. R. McCrae & J. Allik (Eds.), The Five-Factor Model of personality
Rossier, J., Dahourou, D., & McCrae, R. R. (2005). Structural and mean-level analyses of
the Five-Factor Model and locus of control—Further evidence from Africa. Journal of
Cross-Cultural Psychology, 36(2), 227–246.
Rossier, J., Quartier, V., Enescu, R., & Iselin, A. (2007). Validation of the French version
of the hierarchical personality inventory for children (HiPIC)—Influence of gender and
age on personality traits in 8- to 12-year-olds. European Journal of Psychological
Assessment, 23(2), 125–132.
Schmitt, D. P., & Allik, J. (2005). Simultaneous administration of the Rosenberg Self-
Esteem Scale in 53 nations: Exploring the universal and culture-specific features of
global self-esteem. Journal of Personality and Social Psychology, 89, 623–642.
Schmitt, D. P., Allik, J., McCrae, R. R., Benet-Martinez, V., Alcalay, L., Ault, L., …
Zupancic, A. (2007). The geographic distribution of Big Five personality traits:
Patterns and profiles of human self-description across 56 nations. Journal of Cross-
Cultural Psychology, 38(2), 173–212.
Schmitt, D. P., Realo, A., Voracek, M., & Allik, J. (2008). Why can’t a man be more like
a woman? Sex differences in Big Five personality traits across 55 cultures. Journal of
Sen, A. (2002). Health: Perception versus observation—Self reported morbidity has
severe limitations and can be extremely misleading. British Medical Journal,
324(7342), 860–861.
Smith, P. B. (2004). Acquiescent response bias as an aspect of cultural communication
style. Journal of Cross-Cultural Psychology, 35, 50–61.
Smith, R. H., Turner, T. J., Garonzik, R., Leach, C. W., UrchDruskat, V., & Weston, C.
M. (1996). Envy and schadenfreude. Personality and Social Psychology Bulletin,
22(2), 158–168.
Söldner, T. M. L. (2013). Personality, values, and cultural perception in the sojourner
context: A new perspective on culturation in Germany, Japan, and the US. (Doctor
rerum naturalium), Humboldt-Universität zu Berlin, Berlin.
Terracciano, A. (2010). Secular trends and personality: Perspectives from longitudinal
and cross-cultural studies-commentary on Trzesniewski & Donnellan (2010).
Perspectives on Psychological Science, 5(1), 93–96.
Terracciano, A., Abdel-Khalek, A. M., Adam, N., Adamovova, L., Ahn, C., Ahn, H. N.,
… McCrae, R. R. (2005). National character does not reflect mean personality trait
levels in 49 cultures. Science, 310(5745), 96–100.
Toomela, A. (2003). Relationships between personality structure, structure of word
meaning, and cognitive ability: A study of cultural mechanisms of personality. Journal
Triandis, H. C., Lisansky, J., Marin, G., & Betancourt, H. (1984). Simpatia as a cultural
script of Hispanics. Journal of Personality and Social Psychology, 47(6), 1363–1375.
Trzesniewski, K. H., & Donnellan, M. B. (2010). Rethinking “Generation Me”: A study
of cohort effects from 1976–2006. Perspectives on Psychological Science, 5(1), 58–75.
Twenge, J. M. (2000). The age of anxiety? Birth cohort change in anxiety and
neuroticism, 1952–1993. Journal of Personality and Social Psychology, 79(6), 1007–
1021.
Twenge, J. M. (2001). Birth cohort changes in extraversion: A cross-temporal meta-
analysis, 1966–1993. Personality and Individual Differences, 30, 735–748.
Twenge, J. M., & Campbell, W. K. (2002). Self-esteem and socioeconomic status: A
meta-analytic review. Personality and Social Psychology Review, 6(1), 59–71.
Twenge, J. M., & Campbell, W. K. (2008). Increases in positive self-views among high
school students: Birth-cohort changes in anticipated performance, self-satisfaction,
self-liking, and self-competence. Psychological Science, 19(11), 1082–1086.
Twenge, J. M., & Im, C. (2007). Changes in the need for social approval, 1958–2001.
Van de Vliert, E., Yang, H., Wang, Y., & Ren, X.-P. (2013). Climato-economic imprints
on Chinese collectivism. Journal of Cross-Cultural Psychology, 44, 589–605.
van Hemert, D. A., van de Vijver, F. J. R., Poortinga, Y. H., & Georgas, J. (2002).
Structural and functional equivalence of the Eysenck Personality Questionnaire within
and between countries. Personality and Individual Differences, 33, 1229–1249.
Veltkamp, G. M., Recio, G., Jacobs, A. M., & Conrad, M. (2013). Is personality
modulated by language? International Journal of Bilingualism, 17(4), 496–504.
Watson, D. (1982). The actor and the observer: How are their perceptions of causality
divergent. Psychological Bulletin, 92(3), 682–700.
Psychology, 90(6), 987–998.
Zecca, G., Verardi, S., Antonietti, J.-P., Dahourou, D., Adjahouisso, M., Ah-Kion, J., …
Rossier, J. (2013). African cultures and the Five-Factor Model of Personality: Evidence
for a specific Pan-African structure and profile? Journal of Cross-Cultural Psychology,
44(5), 684–700.
Zimmermann, J., & Neyer, F. J. (2013). Do we become a different person when hitting
the road? Personality development of sojourners. Journal of Personality and Social
Psychology, 105(3), 515–530.
The Lexical Foundation of the Big Five Factor Model
Boele de Raad and Boris Mlačić
Abstract
A dictionary is the tangible repository of the common stock of words, although
dictionaries comprise at best 10% of the full lexicon. Part of the lexicon is made
up of the words used to describe what people do and what people are like. The
psycholexical approach to personality focuses on this subset of words and on its
exploitation, or what can be said to be the glossary of personality. This chapter is
concerned with the history of the psycholexical approach to personality
description, from ancient history to the more recent efforts, albeit focusing in
particular on its modern history. Psycholexical taxonomies from around the world
will be considered, as well as taxonomies based on nouns, verbs, adverbs, and
their combinations. Ongoing controversies, difficulties, and disputes regarding
alternative psycholexical personality structures will be considered, as well as
recommendations for future research.
Key Words: Big Five, Five Factor Model, lexicon, psycholexical, personality
trait structure, taxonomy, etic, emic
In the history of personality psychology, the two related versions of the five-
dimensional model of personality traits, the Big Five Model and the Five Factor
Model (FFM), are probably most prominent because they are both built on the
understanding of having virtually exhausted the full domain of traits. The broad
acceptance of the model is certainly a popularity index, but more important, it is
an index of its authoritative nature. The origin of the model is characterized by
the divide between the psycholexically based Big Five approach and the
questionnaire-based approach of the FFM.
The five-dimensional model of personality traits is the trait model constituted
by the five factors of Extraversion, Agreeableness, Conscientiousness,
Emotional Stability, and Intellect or Openness to Experience. The two different
wordings for the fifth factor are symbolic of one of the differences between the
psycholexical approach and the questionnaire approach, respectively.
In this chapter, the history of the model is covered, going back to Cattell’s
pioneering work to which the two related model versions are linked, and before,
but also briefly reviewing some competing models, most notably that of
Eysenck, because they add to the discussion of the best way to structure the vast
domain of traits. Moreover, the ideas from the history of (personality)
psychology that gave floor, flesh, and flavor to the concept of having a
comprising system of traits are also considered. An effort is also made to sketch
the context in which ideas about persons are formed and find their way into the
lexicon in a summarized format, most notably through trait-descriptive adjectival
expressions. The term “lexical” in the title of this chapter implies that the
building blocks of the model are to be found in the catalogue of a language’s
words, the collection of all words of a language community. For practical
purposes, we might refer to the dictionary as the tangible repository of that
common stock of words, although dictionaries comprise at best 10% of the full
lexicon. Part of the lexicon is made up of the words used to describe what people
do and what people are like. The psycholexical approach to personality focuses
on this subset of words and on its exploitation; we might call it the glossary of
personality.
Introduction
Exploiting the Lexicon for Scientific Use
It took a long time for the scientific study of personality to recognize the value
of natural language for advancing personality psychology. With that growing
insight, the notion increased that everyday language had shortcomings. Taking
the shortcomings into account, it would, however, be foolish “to ignore such a
storehouse of accumulated wisdom as a natural starting-point for the study of
behavioral attributes” (Wiggins, 1973, p. 329).
There are different possible ways to exploit everyday language. The
systematic study of person-talk (De Raad, 1985) is one option. This could
involve recording actual conversations, and studying those conversations for
personality-relevant utterances. Another option is tracing personages and their
trait-attributions in literary works (cf. Bromley, 1977; McAdams, 1994). Other
possibilities may be found in letters and films. The psycholexical approach
usually involves the use of a dictionary as the tangible representation of the
lexicon. The advantage of this latter method is the systematic representation of
the more useful part of a lexicon by generations of lexicographers. Yet, choosing
the dictionary over other lexical resources may have consequences in terms of
collecting lexical items with a specific momentum of representation and of
function.
The Lexical Hypothesis

Before the two versions of the five factorial trait model started playing a
significant role in the psychological literature, there is a rather clear timeline
from Galton (1884) to Cattell (1943a, 1943b) in which ideas for a glossary of
personality were given form and in which the large list of personality trait terms
was summarized in an orderly system. During that timeline, the main principles
of the psycholexical approach were formulated, starting with Galton and
elaborated especially by Allport and Odbert (1936), resulting, among other
things, in a rationale that had, according to Allport and Odbert, a “portion of
plausibility”:
Linguistic symbols have demonstrated utility; they have been tested throughout the
ages for their power of representing stable facts of experience… . If traits exist at all
it is natural and proper to name them… . Naturally the more often a disposition … is
encountered in the population the more chance it has of being christened.
(Allport & Odbert, 1936, p. 10)
The purport of this rationale was described as the so-called lexical hypothesis
by Goldberg (1981), who stated that traits or individual differences found to be
important by people are represented or will be represented in language. The
significance of this hypothesis was independently phrased by the philosopher of
language Austin (1970), the writer and semantic poet Themerson (1974), and the
psycholinguist Miller (1991).
The Context of Everyday Differentiation

People form an important topic in daily conversations. What people do or not
do, their achievements and failures, what happened to them, their emotional
reactions in regard to situational demands, etc. are talked over, evaluated, and
explained.
In a cafe, possibly two-thirds of the conversation is person-talk, about who
does what with whom, and whether that is good or bad; who is popular and who
is not, and why; how we deal with a difficult child, a difficult partner, or
colleague. The most frequently sold books are about how the main characters
cope with their intimate experiences and how they react to changes of life.
Everyday conversations at work or at home relate for a large part to behavior and
traits of self or other. Ideas about people are put into words, as much as possible
in a language that fits the idea. This is not only a matter of content—is the
observation represented well? It is also a matter of use—how do people react to
it? When there is a lack of words or when words fall short, new words or more
prosaic expressions are invented or metaphors are used—and sometimes these
new expressions are adopted by others. The practical usage provides feedback on
the effectiveness of the expression: are feelings relieved, did we strike the right
note, did we pluck someone’s heartstrings? Words are continuously tested for
their usefulness.
This everyday person-talk, and also written text, does form a kind of
continuous “survival test.” Lingual expressions (e.g., egoistic, aggressive, or
shy) that have proven their use in often longtime testing for their capacity to
represent facts of experience, and to communicate about those experiences, have
a good opportunity to create a firm position in the lexical storehouse. The more
often a trait or disposition in the behavior of people is observed, the greater the
chance that the trait is labeled and that it becomes a communicative commodity.
By focusing on what the lexical storehouse has in reserve for personality,
expression is given to the appreciation of the descriptive and detailing potential
of the lexicon of personality. That archive of everyday language with that
enormous differentiating potential forms the rough material for the construction
of a scientifically acceptable medium for the description of personality traits.
A problem of course is that everyday language is not a neutral language. The
lingual elements that stood the test of survival are loaded in all kinds of respects.
“Everyday interactions can thus be considered as the stage where conceptions
about people, particularly concerning these peoples’ personality or character,
come into being, are maintained, or are changed” (De Raad, 1985, p. 1).
Sedimentation of Person-Talk
Interestingly, spontaneous everyday conversation of persons is not typically
characterized by the types of words that are used in personality trait research. In
addition to the fact that everyday person-talk is often incomplete, without
explicit intention, and serves transient goals, most utterances are in behavioral
wordings. Perhaps in 10% of the cases trait-descriptive adjectives are used (De
Raad, 1985). Of immediate interest for the psycholexical procedure is not so
much the single utterances that people may use to make sense of behaviors, but
rather how they summarize and communicate those meanings. Hampson (1982,
1984) referred to this as a constructivist process: when it is observed that a
person gives money to charity, remits someone a debt, and manages matters for
someone else, for example, this may lead to the impression that the person is
generous. We focus on the sedimentation (cf. Berger & Luckmann, 1966) of
such impressions from person-talk into the lexicon of a language.
The Study of Abstract Trait Words

When the abstract trait-descriptive adjective forms the unit in studying
personality, we may ask indeed whether we are studying words and their
semantics or the behavior referred to by those words. A pressing issue in the
psycholexical research, almost from the very beginning, has been the question:
“What are we really studying, human behavior or relations between words on the
basis of semantic similarity?” Both methods have explicitly been used, the one
involving ratings of people and their behaviors on sets of trait-descriptive
adjectives, and the other involving judgments of similarity between trait
descriptive words. Wiggins (1973) suggested that the first method produced so-
called external structures, and the second produced internal structures.
D’Andrade (1965), a fervent critic of personality, equated the two methods,
claiming that inferences on human behavior are derived from the semantic
similarity judgments. D’Andrade (1965) was drawing on his study using
Norman’s 20 Big Five scales. He found the same factors in studies of judgments
about people and in judgments of similarity of meaning. He repeated his
conclusion in a subsequent publication (Shweder & D’Andrade, 1980). Peabody
(1987) also studied internal structure using a much larger set of personality
descriptors with 57 scales and 114 adjectives, totaling 6,498 judgments. He
found six factors, much like the Big Five; the factors were, however, uneven in
size, with three bigger factors (Agreeableness, Conscientiousness, and Surgency)
and three smaller ones (Intellect, Emotional Stability, and Values). Opposed to
D’Andrade, Peabody (1987) argued that the internal structure is simpler and that
it is derived from external judgments. In a follow-up study, Peabody and
Goldberg (1989) compared the results based on five external data sets and two
internal data sets and found variants of the Big Five in all seven data sets.
However, in every instance there were three larger factors, Surgency,
Agreeableness, and Conscientiousness, thus giving birth to the Big Three model.
Peabody and Goldberg (1989) also corroborated the finding that conceptual
judgments or semantic similarity ratings were simpler than judgments of people.
Church and Katigbak (1989) studied the internal and external structure in the
Filipino language, found factors similar to the Big Five, and supported the cross-
language generalizability of the Peabody and Goldberg (1989) study, and also
further refuted D’Andrade’s claims that judgments of people simply reflect
semantic similarity or implicit personality notions.
The Questionable Nature of Everyday Language

Trait words are often contaminated by their context of development and
application. Traits are named in terms of standards and interests of a certain
space of time. For some words the meaning is particularly articulated under the
influence of Christian values, such as “devout,” “compassion,” and “patience.”
The term “jovial” comes from astrology (under the influence of Jupiter),
“radical” from politics, and “neurotic” from psychology. Not only the time
frame, but also the spirit of the age is of influence. The extravert was once
described as spiritually poor and superficial and the introvert as high and
inwardly rich. Now the introvert is rather the eccentric and individualistic and
the extrovert is energetic and sociable. Some words are used only in specific
contexts (they are region specific), belong to a certain jargon, or serve a
fashionable purpose. Words can reveal secrets, frustrate, or hurt. Trait words can
be used as weapons. As invectives they can become the verbal vitriol with which
reputations are sometimes ruined and people are brought to stand apart.
John, Angleitner, and Ostendorf (1988) summarized the main criticisms
regarding the use of ordinary trait language for scientific purposes, including the
laypeople origin (cf. Block, 1995), the contextual loading, and the additional
complication of communicability across languages. Saucier and Goldberg (1996)
agreed that the natural language has shortcomings, but they argued that it
represents the best approximation of the total population of personality variables.
In addition, the lexical approach minimizes a prestructuring of variables
according to theorists’ preconceptions of personality (De Raad, 2000; Saucier &
Goldberg, 1996). Hofstee (2003) argued that we can dispute that ordinary
language is not subtle enough for scientific purposes, yet, in studies with
questionnaires built on expert language (e.g., Digman & Inouye,1986), the Big
Five were also recovered.
History
Ancient History
Trait archives from the past. Plato (427–349 bce) seems to be responsible for
the earliest known register of traits of the human personality, a list of no more
than four cardinal traits: courage, justice, temperance, and prudence. These traits
may be seen as the result of a virtual dispute over the most important traits: basic
traits needed for the foundation of an ideal society. People who were endowed
with prudence expressed in traits of wisdom and intellect would be well-
qualified for leading positions; those who were endowed with traits of courage,
bravery, and duty would be fit for the protective and executive tasks in a society;
and the masses of workers would ideally be equipped with temperance expressed
through moderation and self-control. Finally, the basic trait, justice, was
considered as a general trait important for the different layers of a society.
Plato’s student Aristotle (384–322 bce; Ross, trans., 1988) amplified this trait
register to a system with character traits, and provided each with its extremes in
which its deficiency or excess was expressed, thus accounting for 39 distinct
character traits. In turn, a student of Aristotle, Tyrtamus (called “Theophrastus”
by Aristotle because of his divine writing style, 371–287 bce), undertook the task
of describing 30 characters, seemingly to give a fuller portrait of moral
character, each provided with examples of strengths and weaknesses in a
contextualized sketch. These ancient character descriptions all implied
suggestions of what kind of persons we ought to be. The type of character
writing introduced by Theophrastus became very popular since the renaissance.
Aldington (1925) has brought together some 500 short character studies,
including that of Theophrastus, thus providing a rich resource of moral character
traits.
De Raad and Ceulemans (2001) studied the semantics of the 30 characters of
Theophrastus, by identifying 345 typical actions in the character sketches and
classifying them using the Abridged Big Five Circumplex (AB5C) model (De
Raad, Hendriks, & Hofstee, 1992) as the accommodative system. Both the 345
typical actions and the 30 characters were best described by combinations of the
negative poles of the factors, Agreeableness, Conscientiousness, and Intellectual
Autonomy, thus supporting the often heard conjecture that Theophrastus’
characters conveyed a moral message.
Recently, Dahlsgaard, Peterson, and Seligman (2005) have taken up the study
of moral character traits again by examining philosophical and religious
traditions around the world. They constructed a long catalogue of positive traits
(character strengths, virtues) and classified them into six core virtues. De Raad
and Van Oudenhoven (2011; see also Cawley, Martin, & Johnson, 2000) studied
the Dutch lexicon of virtues from a psycholexical perspective, and concluded
that six factors of virtues covered a lot of ground of the Big Five (Emotional
Stability excluded). The lexically based virtue factors showed only partial
overlap with the core virtues described by Dahlsgaard et al. (2005).
Characters in text in ancient times. Another interesting resource for trait
descriptors, and the use of those descriptors in characterizing people, is evident
in the literary text, as in a novel. In some novels certain psychological qualities
are explicitly staged through the novel’s protagonist, often with the function of
furthering the plot of the story. Examples are greed, a main theme in The Great
Gatsby by F. Scott Fitzgerald, and apathy and drowsiness, forming the theme in
Ivan Goncharov’s Oblomov. Going far back in history, an intriguing case can be
found in Homer’s Iliad, because of the hundreds of personages playing a role in
the story, many of whom are provided with character descriptions. Passakos and
De Raad (2009) identified 1,713 so-called epithets in the Iliad, which are
adjectival phrases in which an adjective is combined with the name of a
personage. A total of 1,057 of these epithets could be identified as trait-
descriptive epithets. Those epithets were classified in the AB5C segments of the
Big Five model. Whereas the Iliad is generally understood to form a display of
the heroic character, the Big Five viewpoint specifies more than before the
various sides of heroism as recorded in the lexicon of the time. Facets from all
Big Five factors were used to capture the meaning of heroism, as in
venturesome, good, powerful, bellicose, resourceful, and fearless. For treatises
on how personality is conceived of in ancient Greek times, see, for example, Gill
(1996) and Adkins (1970).
Modern History
The German inception of an alphabetical psychology: from Galton to
Baumgarten. While Germany became the main region for the development of
psychology at the end of the nineteenth century and different psychological
subdisciplines emerged, the lexical approach commenced in this fertile
environment as well. The approach was later called “alphabetical psychology”
by Kouwer (1963). The first who pointed to a dictionary as a valid resource was
possibly Galton (1884), who examined Roget’s Thesaurus and estimated that it
contained “fully one thousand words expressive of character” (p. 181), with
“character” referring to moral qualities. Some years after Galton, Rümelin
(1890) pointed again at the availability of hundreds of character traits in
language waiting to be exploited for scientific use. It took a few decades before
Klages (1926) again took up the rationale of the lexical approach for the study of
personality, and he roughly estimated that around 4,000 words in the German
language could be useful for that purpose. Just a few years later Baumgarten
(1933) made an effort to empirically test Klages’ hypothesis by examining
various German dictionaries and other publications, using frequency criteria for
the occurrence of trait descriptive words, and she came up with a list of 941
adjectives and 688 nouns useful for describing personality.
Fruits from Webster’s Unabridged Dictionary: from Allport and Odbert to
Cattell. In the years during which the center of the psychological evolution
shifted from Germany to the United States, the lexical approach moved along.
Allport and Odbert (1936) undertook the task of scrutinizing Webster’s
unabridged dictionary and they were able to select almost 18,000 words that
might be useful for the description of personality. Those words were classified
into four categories, differing in level of descriptiveness for everyday
conversation and differentiation, thus taking into account the contextual issues
mentioned earlier. Those categories were “Neutral terms designating possible
personality traits,” “Temporary moods or activity,” “Evaluations,” and a
“Miscellaneous” category. The first category contained “most clearly ‘real’ traits
of personality” (Allport & Odbert, 1936, pp. 24–38), with 4,504 lexical items.
Interestingly, lexical items symbolizing abilities (e.g., able, competent, gifted,
talented) ended up in the third or fourth category.
Starting with this first category of “real” personality traits, Cattell (1943a,
1943b) stretched the criterion for this category by adding a few hundred terms
mainly from the second category of “temporary state” terms, following by
removing rare and archaic terms and doubles that differed only through prefixes,
thus yielding a list of between 2,100 and 2,200 terms, which were classified into
160 clusters on the basis of synonymity. Cattell next searched the psychological
literature for personality-relevant terms that psychologists had used during a
century or so, which led to an amplification with 11 new categories, not well
represented in the dictionary, mainly in the areas of abilities and interests. The
final set of 171 categories or trait variables was supposed to comprise the
complete personality sphere. On these 171 trait variables peer ratings from 100
participants were collected. Based on empirical correlations among these ratings,
Cattell dropped quite a few variables and clustered the remaining ones into a
final set of 35 traits.
Cattell’s trait sphere summarized: 35 trait variables in 12 or 5 primary
factors. The set of 35 trait variables or somewhat reduced subsets has been used
by different investigators who arrived at different sets of factors. Cattell (1945,
1947) conducted factor analyses using ratings on the 35 personality variables,
and identified 11 to 12 factors, similar across samples. Cattell added four
questionnaire-specific scales to complete a system with 16 primary personality
factors (16PF). Fiske (1949), Tupes and Christal (1958, 1961), and Norman
(1963) each collected their own ratings on the 35 Cattell variables and none of
them was successful in replicating that many factors but rather arrived at a much
simpler structure with five independent dimensions. Since Norman’s (1963)
study, those five factors had been referred to as the “Norman five” and were
labeled Extraversion, Agreeableness, Conscientiousness, Emotional Stability,
and Culture. The factors were dubbed the “Big Five” by Goldberg (1981; cf.
Goldberg, 1990) in whose work the fifth factor shifted labels from Culture to
Intellect (Goldberg, 1993a).
Norman’s 2,800 stable traits. The drastic reduction of the personality lexicon
to a set of only 35 trait variables was “a matter of unhappy necessity” to Cattell
(1945, p. 70), due to the technical limits of the time. Tupes and Christal (1961)
thought that it was possible to find fundamental concepts other than those
described by the five factors based on the set of 35 trait variables. Norman
(1963) suggested going back to the total pool of trait names in the natural
language. Norman (1967) actually started building a new taxonomy that would
be “sufficiently exhaustive, precise and well structured to be useful for purposes
of scientific communication and assessment” (p. 2). He used a new edition of
Webster’s International Dictionary, searching for trait terms not yet included in
the Allport-Odbert list. To his surprise, he found only 171 new terms, yielding a
complete set of 18,125 terms. In several rounds of reduction, Norman (1967)
excluded about 11,000 terms according to criteria adopted by most lexical
researchers to date, and sorted the remaining 7,000 terms into the three main
rubrics of stable “biophysical” traits, temporary states and activities, and social
roles, relationships, and effects, more or less similar to the classification made
by Allport and Odbert (1936). Next, Norman collected a massive set of data for
the 2,800 terms, or more precisely 2,797, from the category of biophysical traits,
including ratings of their functionality, familiarity ratings, self and peer ratings,
and social desirability ratings. Those terms were first clustered into categories
defined by the Big Five poles, and subsequently into more narrow synonym
clusters, thus forming 75 clusters of trait terms.
Goldberg’s taxonomy of personality descriptive terms. About a decade after
Norman instigated his ambitious project, Goldberg embarked on bringing the
project “to its logical fruition” (Goldberg, 1975). Goldberg (1981) laid the
foundation for the contemporary methodology of the lexical approach through a
series of careful analyses and by giving some first answers to important
considerations: descriptions along a continual dimension are to be preferred over
a typological approach; unipolarity and bipolarity of dimensions should both be
given a role; orthogonality of dimensions has certain advantages but oblique
dimensions should be considered because they seem to be more realistic; and the
level of description should be neither too abstract nor too specific.
Goldberg (1982, 1990) shortened Norman’s 2,797 list to 1,710 trait-
descriptive adjectives using criteria of familiarity, removing dialect terms, and
adding 40 adjectives, mostly from Gough’s Adjective Check List. Goldberg
(1990) aggregated the self ratings collected by Norman on the 1,710 adjectives
into the 75 Norman clusters, thus obtaining cluster-scale scores, and applied
different methods of factor extraction and rotation on those cluster scores, and
repeatedly obtained five factors. Goldberg (1982, 1990) further reduced the
1,710 set, using various exclusion criteria, to the 479 most commonly used trait
adjectives. A clustering of these 479 terms into 133 synonym clusters took place,
and factor analyses on these 133 cluster variables in four samples of participants,
including two self-rating samples and two peer-rating samples, again yielded
clearly five factors. When the analysis of a further reduced set with 100
synonym clusters based on 339 adjectives produced again the Big Five,
Goldberg (1990, p. 1223) was enthusiastic of Big Five breadth: “Consequently,
it now seems reasonable to conclude that analyses of any reasonable large
sample of English trait adjectives in either self- or peer-descriptions will elicit a
variant of the Big Five factor structure and therefore that virtually all such terms
can be represented within this model.” Goldberg (1992) eventually developed a
set of 100 unipolar and 50 bipolar scales, calling them “standard markers of the
Big Five factor structure.” Goldberg’s taxonomic endeavor continued with the
development of an evaluation-explicit taxonomy, and three preliminary
taxonomies, namely a taxonomy of nouns, one of temporary moods, states, and
activities, and one of social roles and relationships or effects (Goldberg, 1982).
Cattell revisited: three secondary factors. The early studies of Cattell (1943a,
1947) were also a valuable source of information for Costa and McCrae (1985)
who were interested in using a measure of personality in the context of aging
studies. A first cluster analysis of the Cattell’s 16PF (Costa & McCrae, 1976)
scales revealed two recurring clusters, i.e., Extraversion and Neuroticism, but
also a hint of their future important dimension of Openness to Experience. The
origin of Costa and McCrae’s (1985) addition of Openness to Experience was
Coan’s (1972, 1974) Experience Inventory developed to measure components of
Openness to Experience. The instrument formed part of Coan’s strategy to study
the humanistic-oriented concept of an optimal personality, which was felt to
have openness to experience as a core quality. Coan (1974) agreed that Cattell
had done a remarkable job in describing the common personality trait
vocabulary. That vocabulary made it possible to differentiate people by their
observable behavior and by their ways of expressing emotions. According to
Coan, however, our language would be deficient in describing unexpressed
experiences and thoughts (Coan, 1974, p. 21). The Experience Inventory was
developed to fill that gap, and contained seven scales to measure the various
facets of experiences. For the original NEO Personality Inventory (Costa &
McCrae, 1985) six facets were selected to measure openness to experience, and
subsequently also six facets for both the Neuroticism and Extraversion scales.
The correlations of the three NEO scales with Goldberg’s Agreeableness and
Conscientiousness scales turned out to be essentially zero (McCrae & Costa,
1985a, 1987).
From the NEO-plus-two to the Five Factor theory. Costa and McCrae
(1992b) later added Agreeableness and Conscientiousness to their model, which
then constituted the FFM, so crucial for the popularization of the model and its
expansion to the domain of questionnaires. In subsequent years Costa and
McCrae conducted numerous studies (e.g., Costa, Busch, Zonderman, &
McCrae, 1986; McCrae & Costa, 1985b, 1989a, 1989b) relating the five factors
to other prominent personality instruments and models such as MMPI, MBTI,
Eysenck’s model, and Wiggins’ circumplex. Those studies found substantial and
meaningful relations between all the investigated personality instruments and the
FFM dimensions and convinced Costa and McCrae (1992a) that five factors are
basic dimensions of personality. This was corroborated by Goldberg and
Rosolack (1994) who viewed the Big Five as an integrative framework for
personality research. In recent years Costa and McCrae (e.g., McCrae, 2010;
McCrae & Costa, 2008) built a Five Factor Theory (FFT), going a step further
from the usual Big Five interpretations, and claiming that five factors have not
only descriptive but causal status, and that they are universal with strong genetic
and biological bases.
Trait Taxonomies Around the World
A Diversion of Lexical Methods
Some 40 years after departing from Germany to the United States, the lexical
approach to personality found its way back to Europe, mainly in the Netherlands
and again in Germany. The first two studies followed different procedures. The
first study was conducted in Dutch by Brokken (1978) who followed a
straightforward procedure of selecting terms from the lexicon that had trait
descriptive meaning. Those terms were identified using a practical definition of
traits in the form of identification sentences. This was referred to later as the
Dutch method. Angleitner, Ostendorf, and John (1990) performed the first
German psycholexical study for which they developed an explicit and detailed
schedule involving a variety of characterizational terms, including, for example,
traits, roles, and reputations. The distinctive significance of the German project
is in the detailed explanation on which terms are trait relevant, the minute
development of the trait classification system, and the elaboration of the
importance of the word class of nouns for personality description.
Central issues in these selection procedures involve (1) the definition of the
lexical documentation used for finding trait terms; (2) the definition of the
domain of characteristics that defines personality; (3) the definition of a trait to
be selected from those documents; and (4) defining the lingual categories of
description. With respect to the first, typically dictionaries are used as the type of
lexical documentation that is probably most comprehensive. However, it is also
possible to use novels, or other types of documentation in which personality
characterizations are provided. The use of dictionaries has been such that each
and every page was scanned for personality descriptive terms. Tellegen and
Waller (1987), however, sampled pages from a dictionary to arrive at a full and
unrestricted set of traits. Regarding the second, the German lexical team, for
example, defined personality in a very broad sense, including not only traits but
also roles, attitudes, reputations, etc., because all these concepts may convey
information on personality. With respect to the third, a general theoretical
definition that is mostly used, explicitly or implicitly, is “disposition” or another
term with the same intent (e.g., inclination). In the Dutch lexical program,
practical definitions were used in the form of sentences in which a trait term
should fit. Different practical definitions apparently produce different sets of
traits (Brokken, 1978). Related to both the second and third issue, Tellegen and
Waller (1987) proposed broadening the definitions of traits so as to also include
evaluative descriptors.
Regarding the fourth, several studies have been performed with word
categories different than adjectives, with De Raad and Barelds (2008) providing
the ultimate study with word sorts from all relevant categories in one single
study. The main idea was to capture all personality descriptive content in a
lexicon, not only the semantics conveyed through adjectives.
The Dutch Taxonomic Project

Brokken (1978), in a major phase of reducing the starting set of 8,690
adjectives, devised two criteria for deciding what personality descriptors are the
so called “Nature criterion” and the “Person criterion.” According to the former,
the Nature criterion, an adjective is useful for personality description if it could
be meaningfully inserted in the following sentence: “He (she) is … by nature.”
In accordance with the Person criterion, an adjective serves the purpose of
personality description if it answers the question: “Mr/Mr X., what kind of
person is he/she?” (De Raad, 2000). For the development of the taxonomy of
personality verbs, a similar identification sentence was used: “If someone
[verb]s more often than others, then that behavior shows his/her personality” (De
Raad, Mulder, Kloosterman, & Hofstee, 1988). For the taxonomy of personality
descriptive nouns for each noun judges were asked its usefulness in “describing,
typifying, characterizing, etc. a person” (De Raad & Hoskens, 1990). In a study
comparing the replicability of the Big Five using different trait descriptive word
classes, De Raad (1992) concluded that the five factors are most clearly found in
the structure of adjectives, but the structure of nouns and verbs can be also
interpreted in terms of some of the Big Five factors or their blends. The Dutch
method was also proven useful in lexical studies in other languages, such as in
Italian by Caprara and Perugini (1994). Aspects of both the Dutch and the
German methods were used in, for example, the Hungarian taxonomy (De Raad
& Szirmák, 1994; Szirmák & De Raad, 1994).
The German Taxonomic Project

Angleitner et al. (1990) elaborated on Allport and Odbert’s (1936) definition
of personality relevance by dividing it into three main steps. The first step
related to their expectation of six categories of person description to appear in
the dictionary, such as (1) stable traits, (2) states and moods, (3) activities, (4)
social aspects of personality, (5) abilities and talents, and (6) appearance. The
second step related to exclusion criteria indicating that terms are not personality
relevant if they could apply to all individuals or if they could be related to
geographical origin, to occupational identity, only to a part of the person, or be
doubtful and metaphorical. In the final and third step the Dutch method was
drawn on and terms from the lexicon were considered personality relevant if
they could be meaningfully inserted in any of two identification sentences.
The German team also included other word classes beyond adjectives, and
distinguished between attribute nouns and type nouns. The former are abstract
words and relate indirectly to people, describing their traits, behavior, and
experience, such as friendliness (Angleitner et al., 1990, p. 93). The latter refer
directly to people, characterizing their personality types, such as nerd.
The German method is reflected in the majority of the personality taxonomies
developed later, including another Italian Roman taxonomy (Di Blas & Forzi,
1998), Czech (Hřebíčková, 2007), Polish (Szarota, 1996), Filipino (Church,
Katigbak, & Reyes, 1996), Croatian (Mlačić & Ostendorf, 2005), Slovak
(Ruisel, 1997), and Spanish (Quevedo-Aguado, Iraegui, Anivarro, & Ross,
1996).
The Replicability of the Big Five in Germanic and Slavic Languages

In the late 1980s and 1990s, the lexically based taxonomy approach was
followed in many languages, mainly in Europe, with the replication of the Big
Five being most successful in Germanic and Slavic languages (Saucier, 2009).
The Dutch and German findings corroborated the Big Five structure. The
Dutch studies (De Raad, 1992; De Raad & Hoskens, 1990; De Raad et al., 1988)
found the Big Five in the structure of adjectives, but also to a greater or lesser
extent in the structure of nouns and verbs. The noun and verb structures also
suggest new meaningful dimensions or facets not covered by the adjective
structure. Whereas the German structure gave a clear confirmation of the
American-English-based Big Five structure, including an articulate intellect
factor, in the Dutch taxonomy the fifth adjective-based factor not only covered
more typical intellect facets but also those of rebelliousness and progressiveness,
as opposed to conventionality, thus emphasizing intellectual independence (cf.
De Raad, 1994).
The first study using the detailed German method clearly replicated the Big
Five in the German language (Ostendorf, 1990), using self and peer ratings and
corroborating Norman’s (1967) finding that Agreeableness is the largest factor in
the natural language, and Emotional Stability belongs to the smallest. Studies in
the Slavic languages that followed the German method were also supportive of
the Big Five. Szarota (1996) replicated the Big Five in Polish. The five factor
structure found in Czech (Hřebíčková, 2007) showed four factors similar to the
first four of the Big Five, but the fifth factor was peculiar in that it referred to
both abilities and dexterity. When exploring the trait structure in Czech with
personality-descriptive verbs, Hřebíčková, Osecká, and Čermák (1999) found
Big Five elements, but Intellect was absent. The Croatian personality taxonomy
(Mlačić & Ostendorf, 2005) replicated the Big Five using self and peer ratings
based on adjectives with three larger factors (Agreeableness, Extraversion, and
Conscientiousness) and two smaller ones (Intellect and Emotional Stability),
thus supporting the Big Five as well as the Big Three model. The Croatian study
also yielded a clear relationship between Croatian emic dimensions and the
respective imported American Big Five factors (Mlačić & Ostendorf, 2005).
Even the study in Russian (Shmelyov & Pokhil’ko, 1993) that investigated the
internal structure of traits found substantial congruencies between the Russian
factors and the Big Five, but it should be noted that those five Russian factors
were identified in a Six-Factor structure.
The Replicability of the Big Five in Other Indo-European Languages

Unlike Germanic and Slavic languages, the replication of the Big Five in other
Indo-European languages was far from perfect. The first study that appeared in
Romance languages was done in Italian by Caprara and Perugini (1994), and
they settled on the five factor solution. The factors, interpreted as
Extraversion/Energy, Quietness versus Irritability, Conscientiousness,
Selfishness versus Altruism, and Conventionality, did not align clearly with the
Anglo-Germano-Slavic Big Five. The fifth factor in Caprara and Perugini
(1994), Conventionality, was much more like the fifth factor in the Dutch
structure, emphasizing the rebelliousness and progressiveness at the
nonconventionality pole. In addition to this there was a clear shift in content
between the Italian factors Quietness and Altruism and the classic Big Five
Agreeableness and Emotional Stability. The other study in Italian language, done
by Di Blas and Forzi (1998), also partially replicated the Big Five in two sets of
data in a structure with a clear set of three factors (Conscientiousness,
Extraversion, and Agreeableness), and the Emotional Stability dissipated into
two factors, while a five factor solution did not yield an Intellect factor. In a
study by De Raad, Di Blas, and Perugini (1988) combining the data sets of the
two Italian taxonomies, and comparing the two structures, it turned out that the
two taxonomies had produced essentially the same Five Factor structure, with
four typical Big Five factors and the fifth factor emphasizing Integrity versus
Nurturance.
The study in Spanish (Benet-Martinez & Waller, 1997) was one of the first to
demonstrate the so-called Big Seven model with factors interpreted as Positive
Valence, Negative Valence, Pleasantness, Engagement, Temperamence,
Agreeableness, and Openness. Because Pleasantness was based on sociability,
joy, energy, and self-assurance, Saucier and Goldberg (2001) saw a clear
replication of the Big Three in the Spanish data, in addition to the Positive and
Negative Valence factors. It should be noted that Benet-Martinez and Waller
(1997) followed a specific methodology in which every fourth page of the
Spanish dictionary was inspected for trait descriptors. That kind of sampling
may have systematically influenced the exclusion of terms beginning with, for
example, “un” or “in” that denote the relative lack of a certain quality; this
could, for example, have influenced the size of the negative poles of resulting
factors.
Another taxonomy in the Romance language, French (Boies, Ashton, Pascal,
& Nicol, 2001), in Canada was one of the first Indo-European studies distinctive
in two aspects, namely the extraction of six factors, rather than five, and the
interpretation of one of the factors as Honesty. The six factors in that study
reported by their relative size were Agreeableness, Emotional Stability,
Extraversion, Conscientiousness, Honesty, and Imagination (with the notable
absence of Intellect-related terms). However, it must be noted that the factors of
Agreeableness and Honesty shared some of the aspects regarding their cross-
correlations with the adjective markers of noncorresponding factors, that is, the
factor of Agreeableness had the highest correlation with the adjective markers of
Agreeableness, but also a substantial secondary correlation with markers of
Honesty and vice versa.
A study in the modern Greek language (Saucier, Georgiades, Tsaousis, &
Goldberg, 2005) that used a modified German method examined both ipsatized
and raw-data factors and it was concluded that only one- and two-factor
solutions were invariant. The one factor was labeled as Evaluation and the two
factors as Morality/Social Propriety and Dynamism (Saucier et al., 2005).
Beyond that, the authors considered that the five- and Six-Factor solutions were
also stable, but the factors did not resemble the classical Big Five interpretation.
For example, the five factors in the ipsatized data set were labeled as Negative
Valence, Morality/Agreeableness, Conscientiousness, Prowess/Heroism, and
Positive Affect/Sociability whereas in the Six-Factor solution the
Morality/Agreeableness factor broke into Honesty and Even Temper.
A recent taxonomy in Hindi (Singh, Misra, & De Raad, 2013) confirmed the
stability of three factors in self ratings and peer ratings, but not the so-called Big
Three. The three Hindi factors reflected the ancient Hindi cultural concept of
Triguna, namely rajasic, tamasic, and sattvic. According to Singh et al. (2013)
the rajasic factor represents ambition, friendliness, and humility, tamasic
represents egoism, mean mindedness, and concealment, and sattvic represents
competence, organization, and soberness. Singh et al. (2013) also concluded that
the triguna factors covered much of the Big Five content and variance, which
prevented the Big Five from appearing beyond triguna.
In conclusion, the studies in Indo-European languages other than Germanic
and Slavic gave varying structures, with three, five, and up to seven factors, and
the replication of the Anglo-Germano-Slavic Big Five was sporadic, usually
with the lack of the Intellect factor and a dispersion of the Emotional Stability
factor. Those studies also offered first versions of models that later became the
competitors of the Big Five, such as the Big Six (Ashton & Lee, 2001) and the
Multi-Language 7 (ML7; Saucier, 2003a) models. It seems, however, that the
kernel structure that could be drawn from most of these studies, if not all, is the
Big Three model with often broad versions of Agreeableness, Extraversion, and
Conscientiousness.
The Big Five in Non-Indo-European Languages

Studies in Hungarian, a Finno-Ugric language (De Raad & Szirmák, 1994;
Szirmák & De Raad, 1994), were not directly supportive of the Big Five. A Five
Factor solution did not contain Intellect, but a factor called Integrity (De Raad &
Szirmák, 1994); only a Six-Factor solution additionally gave the Intellect factor.
In a Five Factor solution Agreeableness was well covered but had split into two
factors, respectively called Agreeableness and Integrity. The Six-Factor solution
formed the first appearance of the structure later confirmed in work of Ashton as
the Big Six.
Studies in Turkish, a member of the Altaic language family, were more
supportive of the Big Five (Goldberg & Somer, 2000; Somer & Goldberg,
1999). Somer and Goldberg (1999) reported a clear Five Factor structure in both
self and peer ratings with the first four factors being very similar to the
corresponding factors of the Big Five, but the fifth factor being closer to an
Openness reading rather than to an Intellect reading. In a study with a restricted
item pool, Goldberg and Somer (2000) replicated these findings, however, with
the fifth factor combining Intellect content with Unconventionality.
A study in Hebrew (Almagor, Tellegen, & Waller, 1995) led to the conclusion
of seven factors, labeled Agreeability, Dependability, Negative Valence, Positive
Valence, Positive Emotionality/Agentic, Negative Emotionality, and Positive
Emotionality/Communal. Although this finding seems to corroborate the Big
Seven model found in Spanish with the dissipation of Emotional Stability, it
must be noted that, unlike the Spanish study, Extraversion divided into two
factors of Positive Emotionality; the Intellect terms were attached to the Positive
Valence factor and Negative Valence was somewhat bipolar (Almagor et al.,
1995). The study in Hebrew also suffered from the same limitation as the
Spanish study; that is, it inspected only every fourth page of the dictionary,
which could have systematically affected the outcomes.
Studies in Filipino, or Tagalog (Church, Katigbak, & Reyes, 1998; Church,
Reyes, Katigbak, & Grimm, 1997) also settled for a Seven-Factor structure but,
unlike the above Big Seven, the factors were interpreted as Concern for Others,
Conscientiousness, Gregariousness, Intellect, Self-Assurance,
Temperamentalness, and Negative Valence. The authors concluded that the first
five factors were similar to the Big Five (with Agreeableness,
Conscientiousness, Extraversion, Intellect, and Neuroticism, respectively),
whereas the factor Temperamentalness covered aspects of three Big Five factors:
Agreeableness, Conscientiousness, and Neuroticism (Church et al., 1997). A
new study (Church et al., 1998) replicated this Seven-Factor structure, albeit
with the positive evaluation or positive valence terms blending with the Intellect
factor. However, the seven factors were found only in an Eight-Factor solution
with an additional small factor labeled as Uninhibited or Flamboyant (Church et
al., 1998). Church et al. (1998) concluded that the cross-cultural generalizability
of the Spanish Big Seven model was not complete; however, it must be noted
that the fusion of Positive Valence and Intellect also occurred in Hebrew.
Saucier (2003a) compared the Hebrew and Filipino structures. He emphasized
the convergence between the studies, which he expressed in the so-called Multi-
Language Seven (ML7), which he subsequently found to be recoverable in
English and in Italian.
A study in Korean (Hahn, Lee, & Ashton, 1999) could be seen as supportive
of the Big Three, like many studies discussed so far, but also of the model with
six factors. Hahn et al. (1999) interpreted Three-Factor to Six-Factor solutions
where the Three-Factor solution closely aligned to the classical Big Three; the
Four-Factor solution correlated strongly with the first four Big Five factors and
the five factors represented the Big Five with some rotation of
Conscientiousness and Intellect. The Six-Factor solution unveiled the additional
factor of Truthfulness, similar to the Hungarian Integrity factor (Hahn et al.,
1999), a factor that later become known as Honesty (Ashton & Lee, 2001).
The first study in Chinese (Yu et al., 2009) selected a set of 6,000 personality
descriptive adjectives, reduced them to 650, and interpreted five factors in self-
ratings of 610 students: Intelligent, Emotional, Conscientious, Unsocial, and
Agreeable. Although the labels of the Chinese factors seem supportive of the Big
Five, it must be noted that there was a lack of descriptors describing the positive
pole of Extraversion at the factor Unsociable, while the same factor gathered
some adjectives describing the negative pole of Intellect. Nevertheless, based on
the results of that study, Yu et al. (2009) developed a set of 100 marker
adjectives and interpreted the same five factors in self-ratings of 720 students.
A second study in Chinese (Zhou, Saucier, Gao, & Liu, 2009) was less
supportive of the Big Five. Zhou et al. (2009) started with 3,150 personality-
descriptive adjectives, reduced them to 413, and applied them to obtain self
ratings from 451 students as well as 500 peer ratings. Zhou et al. (2009)
interpreted a hierarchical pattern of structures from one to seven factors and
concluded that the One-Factor (Evaluation/Social Propriety) and Two-Factor
(Social Propriety and Dynamism) structures agreed with those from previous
studies. Moreover, Zhou et al. (2009) concluded that the Seven-Factor structure,
consisting of the factors Extraversion, Conscientiousness/Diligence,
Unselfishness, Negative Valence, Emotional Volatility, Positive
Valence/Intellect, and Dependence/Fragility, was most stable across self ratings
and peer ratings as well as across ipsatized and original data.
Judging by the lexical studies in non-Indo-European languages, it must be
concluded that the results were mixed in terms of the support for the Big Five
model. Although in some languages structures close to the Big Five model were
found, in other languages only traces of the Big Five were identified, and some
studies settled on interpreting various structures, from two up to seven factors.
Some additional support was found for the Big Six-Factor and the Seven-Factor
structures (e.g., ML7), but reservations remain about the methodology of some
studies. Interestingly, just as was the case for the Indo-European languages, the
Big Three factors seem to be identifiable in almost all non-Indo-European
languages.
Exploiting the Lingual Means to Characterize Personality
Grammatical Categories and Semantic Coverage
From a linguistic viewpoint it seems logical to use the category of adjectives
for the description of personality traits, because adjectives describe qualities of
objects and persons, as in “an honest person.” The psycholexical approach has
indeed been largely dominated by the use of trait-descriptive adjectives.
However, communication concerning persons allows for a great variety of
linguistic forms for person characterization, including single nouns (he/she is a
comedian), single adverbs (he/she behaves aggressively), single verbs (he/she
influences people), but also more natural or even poetic expressions (he/she is a
person who looks at things from different angles). Ryle (1949) argued that the
personality vocabulary does not consist only of dispositional words: “The judge,
the teacher, the novelist, the psychologist, and the man in the street are bound
also to employ a large battery of episodic words when talking about how people
do, or should, act and react” (p. 113). Episodic acts (is smoking a cigarette) may
be used to derive tendency statements (is a cigarette smoker) (see also Osgood,
1970). Psychologically, all such characterizing sentences or derivative sentences
are useful as long as they fulfill the “adjectival function”; that is, they
communicate on dispositional qualities of persons. Allport and Odbert (1936)
did not exclude other grammatical categories for personality description, and
actually included adjectives, adverbs, nouns, and participial terms. To avoid
duplication of semantics, nouns and other word classes appeared in their list only
if no corresponding adjective existed. Allport and Odbert’s (1936) first column,
with terms that “seem to symbolize most clearly ‘real’ traits of personality,” was
suggested not to be the final list (p. 26). Compounds and idioms might well be
added as “useful and apt phrases” (p. 30) for personality characterization.
In sum, there is more to language than just adjectives for personality
descriptive purposes, and the lexical approach would do a “suboptimal” job (De
Raad, 2000) if it would exclude classes of potentially relevant terms describing
personality. The most obvious candidates in this respect would be adverbs, many
of which appear in both adjectival and in adverbial format (e.g., aggressive and
aggressively), verbs, expressions, and nouns. For the latter category, Angleitner
et al. (1990) distinguished between type nouns (conservative) and attribute
nouns (conservatism).
Nouns
The first to explore a different domain of lingual descriptors was probably
Goldberg (1982) who studied the use of nouns especially within the framework
of developing an evaluation-explicit taxonomy of trait terms. Goldberg (1982)
catalogued 1,947 nouns from different sources, of which 1,342 were commonly
used terms. Goldberg (1982) concluded that, compared to adjectives, nouns
carry more negative implications and they are more colloquial and slangy.
Although this observation generally holds, De Raad and Hoskens (1990)
reasoned that it is probably a matter of degree and that there may be nouns for
which the evaluative component does not dominate the potential descriptive
component.
De Raad and Hoskens (1990) selected 8,450 possible personality-descriptive
nouns from a database comprising two comprehensive Dutch dictionaries,
reduced them to 3,200 nouns, using criteria of familiarity and of usefulness in
personality description, and finally reduced them to a list of 755 nouns, on which
self and peer ratings were collected from Dutch and Belgian participants. De
Raad and Hoskens (1990) concluded that in the different data sets the Big Five
factors can be easily identified with the noun-factor Anxiety representing Big
Five Emotional Stability, the noun-factor Perseverance representing Big Five
Conscientiousness, Antagonism representing Big Five Agreeableness, Culture
representing Big Five Intellect, and same-named Extraversion. In addition, they
found a factor they labeled Malignity, the largest in terms of accounted variance,
a finding that corroborates Goldberg’s (1982) observation that nouns carry more
negative implications.
Henss (1995, 1998) investigated type nouns in the German language, starting
with the list of around 5,500 type nouns, reduced it to 192, and used them in
ratings of prominent persons as stimuli. Henss (1998) concluded that his noun
factors are related to the Big Five factors, with an additional Physical
Attractiveness factor, which may make sense considering the stimulus persons
that were rated (cf. De Raad & Ostendorf, 1996).
Saucier (2003b) investigated the structure of English personality type nouns,
starting with Goldberg’s (1982) list of 1,947 type nouns. He made a reduction to
the 557 nouns with the highest frequency of use and a further reduction to 372
by excluding role nouns and profane terms. Analyses of self ratings and peer
ratings of those 372 type nouns yielded one- and two-factor structures robust
across samples, while within-gender analyses yielded a structure with eight
orthogonal factors. The first of these was labeled Social Unacceptability, a factor
closely related to the Malignity factor in Dutch. The second factor, labeled
Autonomous Intellect, was similar to the Dutch Culture factor, and the third
factor, Egocentrism, was related to the Dutch Antagonism. The seventh factor,
called Liveliness, was related to the Dutch noun factor Extraversion. The sixth
factor, Attractiveness, was related to the German factor with the same name. The
remaining factors, Masculinity, Delinquency, and Disorientation, seemed to be
more specific to the English study.
Di Blas (2005) investigated attribute nouns in the Italian language, starting
with a list of 3,200 potentially personality-relevant attribute nouns, reduced to a
final list with 447 nouns, using ratings of familiarity, frequency of use, and
personality relevance. Analyses of self-rating data revealed a stable Three-Factor
solution, corroborating the Big Three model, while the search for a stable
solution beyond the three factors was unsuccessful (Di Blas, 2005). An unstable
Six-Factor solution represented deviations of the Big Five with factors of
Conscientiousness, Self-Assurance, Sociability, Placidity, Honesty/Humility,
and Cleverness/Sophistication, which made Di Blas (2005) conclude that in the
Italian context personality language reflects the Big Three and the Big Six, but
not the Big Five.
Verbs
Apart from adjectives and nouns, the third word class that was sometimes
investigated by the psycholexical researchers is the class of verbs. De Raad
(2000) suggested that verbs somehow dropped from sight because they refer to
phenomena of a transitory nature and most of the lexical researchers were
interested in phenomena of a more permanent nature. However, De Raad et al.
(1988) considered that verbs are important for personality description since they
could be seen to be relevant not only for specific behavior but also for general
behavior if coupled with an adverb “often.” De Raad et al. (1988) started with a
list of 1,557 potentially personality-descriptive verbs and reduced the list to 543,
making use of ratings of usefulness for personality descriptions by indicating
whether a verb would fit in the following sentence: “If someone [verb]s more
often than others, then that behavior shows his/her personality” (p. 85). De Raad
et al. (1988) interpreted 10 factors in both self ratings on verbs and in partner
ratings, with six factors recurring in both self and other ratings, namely factors
called Malignity, Support, Antagonism, Verbal Aggression, Perseverance, and
Suppression. The largest factor in terms of accounted variance was Malignity,
just as in the case of nouns (De Raad & Hoskens, 1990). De Raad (1992)
compared the adjective, noun, and verb structure, using only self ratings and
comparable criteria of factor extraction. Moreover, the ratings were all ipsatized.
The direct visible effect was the removal of the Malignity factors from both the
noun structure and the verb structure. The ratings on adjectives provided the Big
Five structure with the possibility of an Agreeableness split-off in the form of
sixth factor Boastfulness versus Sincerity. The noun structure consisted of the
Big Five minus Emotional Stability. The verb structure was most optimal in the
form of a Two-Factor structure with an Agreeableness factor and a factor then
called Emotional Stability. This second factor may, however, be better
understood in terms of being active and decisive, emphasizing striving features.
Hřebíčková et al. (1999) investigated the structure of Czech personality-
descriptive verbs, starting with a list of 2,374 personality-relevant verbs and
reduced them to 1,530 by eliminating archaic, dialectal, and rarely used verbs.
Subsequently, judges rated clarity of meaning and personality relevance of those
verbs, after which they were classified into a category system that was
developed by Semin and Fiedler (1988). That system distinguished descriptive
action verbs (e.g., to call, to kiss, and to talk) from interpretative action verbs
(e.g., to help, to cheat, and to patronize) and state verbs (e.g., to like, to hate, and
to trust). These latter steps yielded a final list of 289 verbs from the latter two
categories. Those 289 verbs were used to obtain self ratings from 473
participants, and factoring after ipsatization provided solutions with two to six
factors, without preference for any of the structures. A Two-Factor solution
turned out to be much like the corresponding solution in Dutch. Hřebíčková et
al. (1999) concluded that the structure of Czech personality descriptive verbs
resembled the structure produced by adjectives, with the exception of content
referring to Intellect or Openness to Experience. Such an absence was also the
case in the Dutch study.
Adjectives, Nouns, Verbs, Adverbs, and Other Expressions Combined

Using this so-called nonrestrictive approach in selecting personality
descriptors, allowing also the inclusion of evaluative terms and state terms,
Almagor et al. (1995) succeeded in producing the Big Seven model, which
included versions of the Big Five plus two additional factors called Negative
Valence and Positive Valence (cf. Benet & Waller, 1995; Tellegen & Waller,
1987). Regarding the use of word categories, a truly unrestricted approach would
be one that makes use of all lingual forms in a single study. The adjectives may
form the most typical carriers of trait meaning, but De Raad and Barelds (2008)
argued that the exclusion of other word categories does not agree with the
psycholexical aim to arrive at full coverage of personality semantics in ordinary
language and it may also produce a biased and incomplete personality structure.
For this reason De Raad and Barelds (2008) once more undertook the task of
going through the Dutch lexicon to register all words and (standard) expressions
that could be used to characterize persons. All those units of description were put
in a brief sentence format before they were administered to participants to obtain
ratings. A total of 2,365 such items were finally administered to 1,466
participants for self ratings and for other ratings. Examples of items were “a
friendly person,” “someone one can trust,” “someone who helps people,” “a
balanced person,” “someone who rouses public sentiment,” “a horny person,” “a
fortune hunter,” and “someone who has wanderlust.” Eight factors were
ultimately extracted, were stable across subsamples, and were well interpretable.
Those eight factors included four factors that related well to four of the Big Five,
namely, Extraversion, Agreeableness, Conscientiousness, and Emotional
Stability; a broad factor called Competence and a smaller factor called
Conventionality, both reflecting aspects of Intellect; a factor called Virtue
relating to both Big Five Agreeableness and HEXACO Honesty-Humilitty; and a
last factor Hedonism capturing aspects of Sensation Seeking.
Other Categories of Words and Other Aspects of Personality

There is more to personality than dispositional traits. In the “constructive”
process leading to conceptions of peoples’ personalities (cf. Hampson, 1982), a
variety of features may contribute to how persons think that people are and how
they differ. This may include the physical appearance, the attitudes people have,
their roles, their relationships to others, and so forth. The German taxonomic
project has been most explicit in selecting these terms from the lexicon, covering
most or all classes of person characteristics (see Angleitner et al., 1990).
Physical appearance. Ostendorf and Angleitner (1994) investigated the
structure of adjectives describing overt characteristics and appearance (one of
the subordinate categories in the German classification system). Analyses of self
ratings and peer ratings yielded six factors: Fatness, Height, Sturdiness, Stylish
Appearance, Untidiness versus Well-Groomed Appearance, and Body
Deformations versus Harmonious Body. Relations between these factors and the
NEO-FFI scales showed that Conscientiousness was positively related with
Well-Groomed Appearance, and to some extent with Stylish Appearance and
Sturdiness; Neuroticism was negatively related to Sturdiness and Well-Groomed
Appearance; Agreeableness was positively related to Well-Groomed
Appearance.
Saucier (1997) investigated a range of terms, including references to
dispositions, but also to state terms, social evaluation terms, and physical
appearance terms. The terms were selected from pools of terms constructed by
Norman (1967) and by Goldberg (1982). Ratings on the finally chosen set of 525
terms were factored and seven factors were extracted: the Big Five factors plus
two additional factors, one describing Attractiveness and the other resembling
Negative Valence.
Imperio, Church, Katigbak, and Reyes (2008) also conducted a study
involving, among other things, 268 terms representing social and physical
attributes, and markers of the Big Five, HEXACO, and the ML7. Considerable
redundancy was found between the social and physical attribute dimensions and
trait and evaluative dimensions. The conclusion was that social and physical
attributes communicate information on personality, and vice versa.
Attitudes. Ostendorf (1996) investigated the structure of the subcategory of
Attitudes and Worldviews from the German classification system in relation to
measures of the Big Five, with both lexically based dispositional measures and
the NEO PI-R scales. An interesting question put by Ostendorf was whether
attributes such as conservative, liberal, and traditional, co-defining the Openness
to Experience scale, should be regarded as traits or as attitudes. Both self and
peer ratings of 104 prototypical attitude terms yielded two dominant factors
labeled as (1) Religiousness and (2) Conservatism versus Radicalism. A joint
factor analysis with both dispositional terms and attitude terms showed that the
two attitude factors were independent of the Big Five in a Seven-Factor solution.
Moreover, correlations between these seven factors and the NEO PI-R scales
revealed that the Conservatism versus Radicalism factor was substantially
negatively related to both the Openness to Experience domain scale and to all its
facet scales, thus giving flesh to Ostendorf’s proposal.
Saucier (2000) studied the structure of social attitudes by selecting terms from
the dictionary that represented attitudes and beliefs, almost all attribute nouns,
and almost all ending with “ism” (e.g., liberalism, utilitarianism, Marxism),
henceforth called “Isms.” An initial list contained 721 terms, with often more
than one definition, totaling 1,208 definitions. These definitions were rated on
relevance to the domain, and the relevance ratings were used to reduce the list to
374 definitions. After an adaptation of this latter list to a questionnaire format,
ultimately 335 definitions remained as distinct ism-variables. Ratings of
agreement–disagreement with the constructs were factor-analyzed. Three broad
factors were obtained, of which the first two were suggested to have some cross-
cultural generality. The first related to traditional religiosity, the second was
defined by items containing egoistic and materialistic attitudes, and the third
emphasized liberal, humanitarian, and enlightenment values (see also Saucier,
2013).
Roles and effects. Saucier (2010) investigated social effects, a subcategory in
the German classification system, interpreted as “an individual’s footprint on the
social world” (p. 224). Saucier (2010) started with an initial pool of 326 terms
for social roles and effects. This list went through different rounds with
familiarity ratings, and frequency of use ratings, ultimately ending in a
classification with categories that described not only social effects but also social
roles and relationships, social evaluations, and attitudes and worldviews. The
result was a list of 27 prototypical terms describing social effects. Factoring of
peer and self ratings of those terms resulted in a Two-Factor structure with one
factor representing whether a person is a source of pleasure to others and the
other factor representing whether a person is a source of pain to others.
Mlačić (2016) studied the terms belonging to a German superordinate
category of social and reputational aspects of personality that includes roles and
relationships, social effects, evaluations, and attitudes and worldviews. An
analysis of self-ratings using 532 terms belonging to that larger more inclusive
category of social and reputational aspects resulted in a Four-Factor structure
with factors labeled as Religiosity and Patriotism, Nurturance, General Social
Effects, and Modernism (Mlačić, 2016). An investigation of relations of those
factors with measures of personality (IPIP-NEO, Mlačić & Goldberg, 2007) and
social attitudes (as measured by ISMS, Saucier, 2000, and by SAS_G, a social
attitude scale, Milas, Mlačić, & Mikloušić, 2013) led to the conclusion that the
lexical factors describing social and reputational aspects of personality are partly
rooted in social attitudes and partly to dispositions (“personality proper;” Mlačić,
2016). For example, the factor General Social Effects was strongly related to Big
Five Extraversion and the factor Nurturance was substantially related to Big Five
Agreeableness. The factor Religiosity and Patriotism was strongly related to
traditional religiosity factors from both social attitude measures whereas the
factor. Modernism was related to the Modern Technology attitudes factor and
the Big-Five Intellect.
Intellect or Openness to Experience

John (1990) observed that there seemed to be no single Big Five. Referring to
a list of studies all reporting on Big Five dimensions, the variation in factor
naming was striking, yet understandable considering the broadness and
inclusiveness of the Big Five factors. Differences were clearest for factor five,
referred to as Culture, Intellect, and Openness to Experience. Of special interest
here are the two labels Intellect and Openness to Experience, of which the first,
proceeding from lexical studies, is said to emphasize “intellectual” traits such as
intelligent and insightful, and the second, boosted by Costa and McCrae (1985)
using the questionnaire approach, emphasizes “open” traits such as imaginative
and artistic. The question about the precise meaning of the fifth factor had led to
a special issue of the European Journal of Personality (De Raad & Van Heck,
1994).
The underlying issue here is the understanding of the lexical hypothesis,
which was introduced to encode distinctions made in daily transactions. McCrae
(1990) argued that everyday language “does not register all the significant ways
in which individuals differ in regard to Openness” (p. 123), and the concept
Openness “appears to knot together a wide variety of traits and topics of interest
to personality psychologists” (p. 123). This may well be true. De Raad et al.
(2010), for example, found that Emotional Stability (or its opposite Neuroticism)
appeared relatively weak in cross-cultural lexical studies, which contrasts with
its historical prominence. Emotional Stability may have less representation in
most natural languages, in comparison to clinical language, in which its
appreciation has led to many nuances of emotional experiences. Questionnaire
developers may sample widely from the rich variety of specific variables from
those clinical contexts.
McCrae’s (1990) argument that Openness is inadequately represented in
natural language trait terms was tested by Mulder (2006), who demonstrated
that, in Dutch, there is no problem finding a sufficient number of descriptors in
the natural language to reliably define both the domain of the Openness to
Experience and the corresponding six facets. Making use of De Raad and
Barelds’ (2008) unrestricted lexically derived list of 2,365 descriptors, 127 items
could be identified to represent Openness to Experience and its different facets.
Factoring the ratings on those 127 items produced a lexically based Openness to
Experience structure that matched the NEO PI-R Openness structure, with the
exception of the Esthetics facet.
The Dutch lexical fifth factor has been described as capturing traits such as
original, philosophical, broadminded, rebellious, constructive, nonconformist,
critical, poetic, and creative (De Raad et al., 1992; cf. De Raad & Doddema-
Winsemius, 2006). It is a factor capturing a rich variety of facets, of which some,
for example the rebellious connotation, are not represented in the NEO PI-R
Openness factor. Taking Openness instead of this Intellect factor would
impoverish the complexity of everyday language, just as taking Openness
instead of the lexical Intellect might impoverish psychological language. Hofstee
(2003) argued that it is possible to dispute at length that ordinary language is not
subtle enough for scientific purposes. Yet, in studies with questionnaires built on
expert language (e.g., Digman & Inouye, 1986), the Big Five were also
recovered.
The difference of opinion here lies in what should be considered as important
or relevant differentiating traits. Costa and McCrae (1985) adopted Openness to
Experience from Coan (1972), which they felt would capture the contents of a
cluster of traits in their own work that they had referred to as an Experiential
Style dimension. Coan (1972) had developed the Openness to Experience scale
on the basis of a study using a battery of existing instruments (Coan, 1974).
Coan (1972, 1974) distinguished various facets for his Openness to Experience
scale for the purpose of expanding the humanistic psychology concept of the
optimal personality, and Costa and McCrae (1985) made extensive use of those
facets for their own purpose. That questionnaire origin in Coan’s work may at
least in part be held responsible for the expansion of Openness to Experience as
the fifth factor, in comparison to what the lexical approach had been said to have
generated (see also McCrae, 1994).
Block (1995) also criticized the ordinary language origin of the concepts
proceeding from the lexical approach. Costa and McCrae (1992a) and Eysenck
(1992) disputed the criteria to arrive at basic dimensions of personality. Much of
these disputes relate to personality dimensions found elsewhere in the
personality literature, which were hardly or not at all visible in the Big Five
framework. Some of those dimensions covered specific areas of interest, such as
locus of control (Rotter, 1954) and ego development (Loevinger, 1976); other
dimensions were part of multidimensional systems, thus offering alternatives to
the Big Five.
Competition and Validation

Competing Systems
Assuming the relevance of both the comprehensiveness and the trait semantic
coverage of the Big Five approach, an important question for competing systems
is whether they account for more or for less of the domain of traits (see also
Chapter 19 by O’Connor). Are certain traits possibly theoretically inevitable?
Just as in the case of Neuroticism and Openness to Experience, certain other
concepts may be found to be underrepresented or completely missing in the
presentation of a trait model. Zuckerman (1984), for example, made a strong
point in showing that sensation seeking is a basic trait, particularly because of its
temperamental meaning. Others may find that, for example, moral characteristics
are underrepresented. Some other such “devoted” constructs are achievement
motivation, field dependence, and locus of control. Zuckerman (2002) suggested
an alternative five-factorial model, starting with the assumption that important
traits are assumed to have a biological-evolutionary basis. Eysenck (1967) and
Strelau (1983) have also emphasized the temperamental connotation of
personality.
It is important to realize that the Big Five system does not reveal the whole
story of personality. It is a comprehensive and differential descriptive system
that is understood to form a rich starting point for further specification. Issues,
such as change, unconscious processes, and heredity of traits, are not contained
in the lexical trait vocabularies. Yet trait taxonomy helps to semantically
understand the different features of those issues; moreover, empirical research
around those issues often makes use of instruments based on the Big Five.
Psycholexically Based Departing Models

In recent years there has been an increase in propositions concerning the
proper structure of lexically derived traits. Some focused on replicability in
almost all languages or cultures around the world, and others focused on
maximal semantic representation for which replicability might be found in as
many languages or cultures as possible. Propositions that focus on cross-cultural
validity contain structures that tend to have fewer factors than the Big Five, most
often two or three, and propositions that focus on optimal semantic coverage
contain structures that tend to have more factors than the Big Five (see also
Chapter 20 by Wright).
Big Two. Digman (1997) was possibly the first to suggest a recurrent Two-
Factor solution. Digman made use of 14 studies of which the correlation
matrices for Big Five scales were available, nine based on ratings from adult
participants and five using children and adolescent participants. Digman (1997)
found that his first factor, which he named α, was related to the Big Five factors
Agreeableness, Conscientiousness, and Emotional Stability, whereas the second
factor, labeled β, was related to the Big Five factors Extraversion and Intellect.
Digman (1997) interpreted the α factor in terms of a socialization process and
the β factor as personal growth. Those higher-order factors were related to the
metaconcepts of Communion and Agency, respectively, understood to underlie
the domain of interpersonal behavior (Bakan, 1966; Wiggins, 1991). Digman’s
factors found support in DeYoung (2006), who used the term Stability for the α
factor and the term Plasticity for the β factor. Stability was described as “the
need to maintain a stable organization of psychosocial function,” and Plasticity
as “the need to explore and incorporate novel information into that organization”
(DeYoung, 2006, p. 1149). DeYoung also offered a neurobiological
interpretation of those two factors, linking Stability to the functioning of the
serotonergic system and Plasticity to the functioning of the dopaminergic
system. A Two-Factor structure was also found in modern Greek (Saucier et al.,
2005), with the factors being interpreted as Morality/Social Propriety and
Dynamism, and in Chinese (Zhou et al., 2009) with factors named Social
Propriety and Dynamism.
Big Three. The Big Three model, with broad versions of the first three Big
Five factors, Extraversion, Agreeableness, and Conscientiousness, which
originated from the work of Peabody and Goldberg (1989), has proven itself to
be a more serious competitor to the Big Five in terms of cross-cultural
replicability. Throughout the text above, the Big Three, or its kernel aspects,
were found in studies of adjectives in Italian (Di Blas & Forzi, 1998), Spanish
(Benet-Martinez & Waller, 1997), Croatian (Mlačić & Ostendorf, 2005), Korean
(Hahn et al., 1999), and Hungarian (De Raad & Szirmák, 1994; Szirmák & De
Raad, 1994), in a study of personality-descriptive nouns (Di Blas, 2005), and in
studies with internal data (Peabody, 1987). A study in Hindi (Singh et al., 2013)
also referred to three factors, but those three reflected the Indian Triguna instead
of the Big Three.
Big Six. Judging by the number of publications in the past 10 years and by the
apparent interest of the scientific community in finding out what is beyond the
Big Five, another serious competitor for the Big Five is a model with six factors,
in two versions: one is Ashton and Lee’s (2007) HEXACO model and the other
is Saucier’s (2009) Big Six.
A first report of a lexically based Six-Factor structure with an additional factor
beyond the Big Five was made for Hungarian by De Raad and Szirmák (1994).
That additional factor, called Integrity (veracious, just, trustworthy), was
considered as an incidental finding, tentatively explained in terms of the
pertaining sociopolitical system. A similar factor (called Trustworthiness) was
later found in Italian by Di Blas and Forzi (1999), and also in Korean (Hahn et
al., 1999) and in French (Boies et al., 2001). In a discussion of various lexical
studies, Ashton and Lee (2001) were led to the conclusion “that a six-
dimensional taxonomy of personality variation provides the most parsimonious
and comprehensive framework for understanding the structure of personality
characteristics” (p. 350), and this HEXACO consists essentially of the Big Five
factors (with rotated versions of Agreeableness and Emotional Stability) plus an
additional factor describing Honesty-Humility. The Honesty-Humility factor,
defined by traits of sincerity, unassumingness, and fairness versus slyness,
pretentiousness, and greed, appeared to be a split off from Big Five
Agreeableness. This split had the effect of a reduced HEXACO-Agreeableness,
defined by traits of patience, gentleness, and flexibility versus ill-temper,
quarrelsomeness, and stubbornness. Another effect was that HEXACO-
Emotional Stability lost the ill-temper content, which had moved to HEXACO-
A. The later Openness naming of the HEXACO-O is peculiar, since the main
defining traits in Ashton et al. (2004) stem from lexical studies with
Intellect/Imagination factors. This change is based on their belief that intellectual
ability lies outside personality proper (Lee & Ashton, 2004), and it implies that
the HEXACO model is not directly rooted in the psycholexical approach to
personality.
The additional factor in Saucier (2009) is not Honesty-Humility but a
Negative Valence factor. According to Saucier (2009), Ashton and Lee’s Six-
Factor model should be seen as a “narrowband” model, because it was based on
lexical studies that primarily excluded highly evaluative and emotional state
terms. Consistent with the suggestions made by Almagor et al. (1995), Saucier
followed a “wideband” approach, and analyzed studies from seven languages,
namely Chinese, English, Filipino, Greek, Hebrew, Spanish, and Turkish, that all
followed an inclusive variable selection approach, meaning that terms included
in those studies did not reflect just dispositions, but also state terms, evaluative
terms, and physical appearance. Saucier (2009) concluded with a “wideband
cross-language six” structure on the basis of those studies with the factors
Conscientiousness (Consistency/Organization), Negative Valence (versus Non-
Violativeness), Agreeableness, Resiliency versus Internalizing Negative
Emotionality, Gregariousness/Cheerfulness, and Originality/Talent. According
to Saucier, those factors corresponded well with the disposition terms of the
“narrowband,” cross-language six of Ashton et al. (2004). Specifically,
“wideband” Negative Valence and “narrowband” Honesty corresponded in their
emphasis on amoral/moral traits. This correspondence agrees with the finding in
De Raad and Barelds (2008), in which a Virtue factor (sincerity, honesty,
friendly versus unfair, indecent, dishonest), correlated substantially with both
Honesty-Humility and Negative Valence (negatively).
Big Seven. The inclusion of evaluative terms and mood state terms in the
Hebrew taxonomy (Almagor et al., 1995) led to the proposal of the Big Seven
model. A similar study was done in Spanish (Benet-Martinez & Waller, 1997).
The Big Seven forms the final Big Five competitor in the realm of the
psycholexical approach, culled Multi-Language seven or ML7 by Saucier
(2003a). ML7, with the factors Gregariousness, Self-Assurance, Even Temper
versus Temperamentalness, Concern for Others, Conscientiousness,
Originality/Virtuosity, and Negative Valence or Social Unacceptability, is based
on the study in Hebrew and one in Filipino (Church et al., 1997) that also
comprised seven factors. Saucier (2003a, p. 186) translated markers of the terms
from the Hebrew and Filipino studies, applied them to the American sample, and
concluded with a synthesis in the form of the ML7. None of the Seven-Factor
studies was clearly fully replicated in any of the other languages; only some four
or five of the Big Five factors recurrently appeared.
Cross-Cultural Findings
Cross-cultural psychologists have often endorsed the universality of
psychological characteristics, as can be seen in cross-cultural studies on
dimensions such as achievement motivation, anxiety, and authoritarianism
(Church, 2000). Mayer, Lin, and Korogodsky (2011) studied the possibility of
universality of personality conceptions in cultural traditions as different as
Confucianism, Buddhism, and Judaism. They concluded that judging personality
was an important aspect of all those different traditions.
With respect to the psycholexical approach, systematic cross-cultural studies
using a good variety of languages or cultures have been performed with a focus
on two factors (Saucier et al., 2014), on three factors (De Raad & Peabody,
2005; De Raad et al., 2010, 2014; Peabody & De Raad, 2002), on five factors
(De Raad, Perugini, Hrebícková, & Szarota, 1998; De Raad, Perugini, &
Szirmák, 1997; Hofstee, Kiers, De Raad, Goldberg, & Ostendorf, 1997), and on
six factors (Ashton et al., 2004; De Raad et al., 2010). The emphasis here will be
on the replicability of five factors (the Big Five). Other studies are referred to
when they contribute to information on the cross-cultural validity of the Big
Five.
The Big Five model has found cross-cultural support generally in two ways;
these two are often linked to the distinction between emic and etic (Berry, 1969;
see also Chapter 23 by Allik and Realo). The etic approach is typically followed
by constructing a trait system (a questionnaire) in one language, and then
translating and applying it in another language or culture. Studies with the Five
Factor Personality Inventory (FFPI; Hendriks, Hofstee, & De Raad, 1999) in 13
languages provided evidence that the FFPI was a reliable and valid measure in a
large variety of countries (Hendriks et al., 2003). Similarly, studies with the
NEO PI-R (Costa & McCrae, 1992b) showed replicability of the five factors in
most cultures (McCrae et al., 2005). A lesson drawn by Allik, Realo, and
McCrae (2013) from studies such as these was that it is apparently easy to
transcend language barriers using such personality instruments. Notwithstanding
such excellent cross-cultural findings, criticism has been expressed repeatedly,
especially from the side of the cross-cultural methodologists (e.g., Berry,
Poortinga, Segall, & Dasen, 2002). An important issue is that translated
instruments tend to be relatively insensitive in detecting individual differences of
interest in the target language. Ashton and Lee (2001), for example, found that
that certain FFM-Openness to Experience facets were not very applicable in
many Asian samples.
The emic approach in this case aims at finding a trait structure that best
summarizes the trait domain of a particular language or culture. Linked to this
first approach has been the repeated finding that independent psycholexical
studies in various Western languages led to the Big Five structure. Studies
comparing lexical Big Five structures from different languages, and leading to
the conclusion that the Big Five is replicable across those languages, are often
done through the analysis of the contents of the structures. Notwithstanding the
recurrence of the Big Five, the structure is reproduced better in some languages
than in others (Saucier & Goldberg, 2001).
Cheung, Van de Vijver, and Leong (2011) proposed an approach that would
integrate etic and emic. Such a combined approach can be found in some studies
that investigated the replicability of the Big Five across languages (De Raad et
al., 1997, 1998; Hofstee et al., 1997), and in other studies that were not
particularly restricted to only five factors (De Raad et al., 2010; Peabody & De
Raad, 2002). De Raad et al. (2010) summarized those earlier studies focusing on
the Big Five and concluded that the Big Five were not all cross-culturally
replicable, a finding that was corroborated in Peabody and De Raad (2002), who
found instead more support for a recurrent three-factorial structure (cf. also De
Raad & Peabody, 2005).
De Raad et al. (2010) compared 14 independently developed trait structures.
For each structure, the starting point was taken in the lexicon of the pertaining
language. The trait terms of that language were used to obtain ratings, and those
were factored to arrive at a trait structure meaningful to that language. The 14
taxonomies were pairwise compared, after finding a common part of the factor
structures on the basis of acceptable translations of items into the languages of a
pair. On average, the results indicated that not five, but rather three factors were
well replicable across the languages under study. Beyond the first three factors
(with traits that are typical of Extraversion, Agreeableness, and
Conscientiousness, respectively), the equivalence of factors across languages
tends to diverge.
When focusing on replicability of psycholexically based factors across most
languages or cultures around the world, the Big Five tends to lose in competition
with structures that have just two or three factors. Recently, Saucier et al. (2014)
investigated whether a hypothesized Two-Factor structure (Saucier & Goldberg,
2001), with one factor describing Dynamism and the other describing Social
Propriety or Social Self-Regulation, could be detected in nine diverse languages.
These included seven data sets that were previously published taxonomies,
namely Chinese, Korean, Filipino, Turkish, Greek, Polish, and Hungarian, and
two new data sets, namely for Maasai and Senoufo. Those nine languages also
represented, in addition to Indo-European languages, seven other language
families: Sino-Tibetan, Korean, Austronesian, Altaic, Finno-Ugric, Nilotic, and
Niger-Congo. Saucier et al. (2014) extracted two factors for each data set that
could be interpreted as Dynamism and Social Self-Regulation in each language
involved. They subsequently selected 50 markers for each of the two factors per
language, totaling 900 markers. Of the translatable terms (into English),
ultimately 10 turned out to mark Social Self-Regulation in most of the
languages, and seven were found to mark Dynamism. These two factors are
easily identified as being similar to Digman’s (1997) socialization and personal
growth and to DeYoung, Peterson, and Higgins’ (2002) Stability and Plasticity
factors, respectively. Saucier et al. (2014) noted that those two factors
represented a “common denominator necessary-but-not-sufficient model” (p.
12), meaning there was substantial within-language personality variation that
could not be covered by those Big Two factors.
A strong case for a cross-culturally tenable Three-Factor model was made by
De Raad et al. (2014). In a previous study, De Raad et al. (2010) concluded that
only three factors are replicable across languages. In a subsequent study De
Raad et al. (2014) looked for the joint structure of personality descriptors from
11 psycholexical studies by merging all the data in a “super matrix” with 1,993
trait terms and 7,104 participants. The results of a simultaneous component
analysis (Kiers & ten Berge, 1994) again yielded the Big Three, interpreted as
Dynamism, Affiliation, and Order, with traits from Extraversion, Agreeableness,
and Conscientiousness, respectively, as kernel traits. Drawing on the results of
those two studies, De Raad et al. (2014) referred to those factors as the “Pan-
cultural trait structure.”
Structuring the Big Five Trait Domain

The majority of factor structures of personality offered in the literature are
presenting independent factors, whether they are three, five, eight, or another
number. Such a structure is imposed upon the data through the application of the
psychometric technique, most typically Principal Components Analysis followed
by Varimax rotation (see also Chapter 20 by Wright). The related dominant
viewpoint in factor analysis has been the “vertical viewpoint” (Goldberg, 1993b;
McCormick & Goldberg, 1997). Under this viewpoint each factor is interpreted
in terms of the variables that have high loadings on it. The underlying ideal is the
simple structure, in which each trait variable loads substantially on only one
factor and the loadings on the other factors are close to zero. The simple
structure form is also hierarchical, since trait variables are specifications of
factors. Whereas those factors are superordinate to the more specific traits, the
factors are rather narrow in meaning, with high internal consistency and a rather
fixed angular position in the trait space (Hofstee, 2003). A simple conception of
the Big Five trait factors is thus the vertical view including a representation with
five independent dimensions. Although there are trait variables that load on one
factor only, the majority of trait variables tends to have substantial loadings on
two factors (Hofstee, De Raad, & Goldberg, 1992); variables that load on more
than two factors tend to be vague in meaning.
Given the observation that clear and meaningful traits load on no more than
one or two factors, traits can be represented in two-dimensional arrangements,
by using the pairs of loadings for all trait variables as coordinate values. The
arrangement of traits that is thus formed in the two-dimensional space is called a
circumplex (Guttman, 1954). In contrast to a simple structural understanding of
traits, with narrow coverage of meaning, a circular arrangement of traits
demonstrates the breadth in semantic coverage through the dispersion of the trait
items along the circle (cf. Gurtman, 1997). Hofstee et al. (1992) integrated the
Big Five model and the circumplex understanding of traits into a comprehensive
representation. The interesting thing about the circumplex representation is that
the positions of the trait variables relative to each other become clear, to
variables with similar meaning, to variables belonging to adjacent clusters of
meaning, and to variables with opposite meaning.
The importance of hierarchy has been discussed repeatedly by researchers in
the field of personality (e.g., Cattell, 1947; Eysenck, 1970), with relations among
traits running from more abstract to more specific (cf. John, Hampson, &
Goldberg, 1991). The behavior “laughing at jokes” is more specific than “being
lively,” which in turn is more specific than “extraversion.” This example from
Eysenck (1970) represents what might be called a strict hierarchy (cf. De Raad,
2009). Most typically, hierarchies are studied in two ways, a bottom-up approach
and a top-down approach (see Goldberg, 2006). The bottom-up hierarchies start
with individual items, which are then clustered into semantic groups, which in
turn are clustered into larger groups of traits, until the highest level is reached.
The more recently developed top-down approach has been applied in Principal
Components analyses of traits, in which structures are considered with one factor
or component, two factors, up to a structure with the maximum number of
factors considered relevant. The factors from adjacent levels of factor extraction
are then correlated yielding the hierarchical configuration (cf. De Raad &
Szirmák, 1994; Zuckerman, Kuhlman, & Camac, 1988). If all levels of
extraction from one to the maximum number of interest are included in the
hierarchy, the result may be conceived of as a continuum of abstraction
(Goldberg, 2006). Since, theoretically or empirically, the interest has been in
specific sets of recurrent factors, two, three, five, or six, we may well present
only those levels of interest. An example can be found in De Raad et al. (2014),
in which the relations between three and six factors are given.
Final Comments
The FFM and the Big Five model do not tell the full story of personality, but
they do tell an important story. The potential of the psycholexical approach to
taxonomize the trait domain has not been exploited to its fullest. Yet, the
approach has made good progress. In this chapter the thoughts, procedures, and
findings, particularly in reference to its main finding, the Big Five, have been
reviewed and put in a context of criticism.
The psycholexical approach has been seriously questioned from the very
beginning, for several reasons. Criticisms have been recurrent regarding the use
of ordinary language for scientific purposes, and the fact that the most exploited
tangible form of a lexicon, the dictionary, contains single words, and not
sentences in terms of which people spontaneously convey their opinions of
persons. A convincing argument to exploit the ordinary language documentation
of trait terms is its overwhelming richness. Generations of lexicographers have
scrutinized and documented on a regular basis the many, many ways people
have communicated during generations, and still tend to communicate on things
and people, and this includes the vast domain of person-talk. It is almost
impossible to generate, for example in one or more scientific sessions, a
vocabulary of traits that is as complete and as rich as what has already been
sedimented and documented in a lexicon. Yet, as has been observed for
Neuroticism and for Openness to Experience, professionals may build up their
own vocabulary for certain trait domains that agrees with their research or with
their theory, and the specifics of such a vocabulary may or may not correspond
to what can be found in ordinary language. In each case, it may be important to
check such specific vocabularies for its ordinary language equivalence.
Ultimately, psychological constructs with specific technical meaning may have
to be translated into everyday language, if only for the purpose of arriving at
intelligible items in a questionnaire that should cover the meaning of such
constructs. As a more general note, when professional psychologists aim at
generating a specific trait vocabulary, they are actually drawing upon their own
individual (and restricted, and possibly even biased) lexicon, and are therefore
operating along the lines of psycholexical procedures.
A dictionary as the tangible representation of a lexicon consists of single
words and their explanations. The single words of interest, namely trait-
descriptive words, are taken to contain abstracted information based on
observations of actions and events in which people participated. Single trait-
descriptive adjectives, for example, have been used extensively in research and
they usually function well in communicating on traits. However, improvement is
very possible, by turning each single trait-adjective into one or more behavioral
sentences that cover the meaning of the trait well (cf. Goldberg, 2014; Hendriks,
Hofstee, & De Raad, 1998).
The psycholexical approach has too extensively been restricted to trait-
adjectives. Other word categories with adjectival potential should be exploited
more systematically; in particular, unrestricted procedures should be followed,
possibly in the format reported by De Raad and Barelds (2008). That study
included the various word categories, used brief sentences, and was less
restrictive in the coverage of trait phenomena than most other lexical studies.
An issue that has not been dealt with directly in this chapter is the use of
facets. Yet, implicitly, it has been discussed in relation to the full exploitation of
ordinary language and the many distinctions that are found in the lexicon for the
description of certain traits. Hierarchies imply specificity, and circumplexes
explicitly deal with trait nuances. A question is how many facets may be
necessary or desired to communicate efficiently and economically on the many
trait nuances. Professional psychologists may tend to exaggerate distinctions as
compared to what laypeople do. A large part of the problem may be solved
empirically, by factoring within specific trait domains; this may lead to different
numbers of facets per trait domain. The other possibility is the use of
circumplexes. The Abridged Big Five Circumplex (Hofstee et al., 1992), for
example, defines a fixed set of facets related to combinations of pairs of factors.
Some of those facets may turn out to be practically or nearly empty.
With respect to coverage of the domain of traits, particularly in relation to
evaluative terms, there has been interest in subareas of traits that could serve the
description of devoted research questions. Some studies have, for example,
recently been performed to provide a full portrait of virtues, a specific subarea of
traits that is relevant to discussions about morality. Such specific vocabularies
indeed tend to provide “decomposed” representations in greater detail than the
full trait taxonomy gives as the ultimate result. Examples include De Raad and
Van Oudenhoven (2011) and Morales-Vives, De Raad, and Vigil-Colet (2014).
The 30 or so trait taxonomies that have been performed thus far do cover a
rather restricted number of languages, with many belonging to the Indo-
European language family (see De Raad et al., 2014). Many more lexical studies
are needed, particularly in non-Indo-European languages, and in Asia and
Africa, to arrive at a stage at which we might start to talk of global research
findings, and draw tentative conclusions on a proper cross-culturally valid trait
structure. More indigenous studies in Africa, Asia, and also South America may
also provide more insight on more culturally typical trait characteristics.
There is as yet little reason to expect some canonical solution for a trait
structure; it seems to make more sense to focus on a consensually acceptable
model that does justice to central trait concerns in most languages, and that may
play a role in the development of instruments that are useful in integrated emic–
etic research and practice (cf. Cheung et al., 2011). For such an approach it is
crucial that large pools of trait descriptors are collected from the languages of
the world and put in a joint catalogue after being translated into a common
language, expectedly English. Such a pool of globally relevant trait words could
then be used in a process of reduction taking the various cultural trait interests
into account as much as possible.
References
Adkins, A.W. H. (1970). From the many to the one: A study of personality and views of
human nature in the context of Ancient Greek Society, values and beliefs. London,
England: Constable.
Aldington, R. (1925). A book of “characters.” London, England: Routledge.
Allik, J., Realo, A., & McCrae, R. R. (2013). Universality of the Five-Factor model of
personality. In T. A. Widiger & P. T. Costa, Jr. (Eds.), Personality disorders and the
five-factor model of personality (pp. 61–74). Washington, DC: American Psychological
Association.
Allport, G. W., & Odbert, H. S. (1936). Trait-names. A psycho-lexical study.
Psychological Monographs, 47 (Whole No. 211).
Almagor, M., Tellegen, A., & Waller, N. (1995). The Big Seven model: A cross-cultural
replication and further exploration of the basic dimensions of natural language of trait
descriptions. Journal of Personality and Social Psychology, 69, 300–307.
Angleitner, A., Ostendorf, F., & John, O. P. (1990). Towards a taxonomy of personality
descriptors in German. A psycho-lexical study. European Journal of Personality, 4,
89–118.
Aristotle (D. Ross, Trans.). (1988). The nicomachean ethics. Oxford, England: Oxford
University Press.
personality. European Journal of Personality, 15, 327–353.
11, 150–166.
Ashton, M. C., Lee, K., Perugini, M., Szarota, P., De Vries, R. E., Di Blas, L., … De
Raad, B. (2004). A six-factor structure of personality-descriptive adjectives: Solutions
Austin, J. L. (1970). Philosophical papers. London, England: Oxford University Press.
Bakan, D. (1966). The duality of human existence: Isolation and communion in Western
man. Boston, MA: Beacon.
Baumgarten, F. (1933). Die Charactereigenschaften. [The character traits]. In Beiträge zur
Charakter-und Persönlichkeitsforschung (Heft 1). Bern, Switzerland: Verlag A
Francke A.-G.
Benet, V., & Waller, N. G. (1995). The Big Seven factor model of personality
description: Evidence for its cross-cultural generality in a Spanish sample. Journal of
Benet-Martinez, V., & Waller, N. G. (1997). Further evidence for the cross-cultural
generality of the Big Seven factor model: Indigenous and imported Spanish personality
constructs. Journal of Personality, 65, 567–598.
Berger, P. L., & Luckmann, T. (1966). The social construction of reality. New York, NY:
Doubleday.
Berry, J. W. (1969). On cross-cultural comparability. International Journal of
Berry, J. W., Poortinga, Y. H., Segall, M. H., & Dasen, P. R. (2002). Cross-cultural
psychology: Research and applications. New York, NY: Cambridge University Press.
Block, J. (1995). A contrarian view of the Five-Factor approach to personality
description. Psychological Bulletin, 117, 187–215.
Boies, K., Lee, K., Ashton, M. C., Pascal, S., & Nicol, A. A. M. (2001). The structure of
the French personality lexicon. European Journal of Personality, 15, 277–295.
Brokken, F. B. (1978). The language of personality. Unpublished doctoral dissertation.
Meppel, The Netherlands: Kripps.
Bromley, D. B. (1977). Personality description in ordinary language. London, England:
John Wiley.
Caprara, G. V., & Perugini, M. (1994). Personality described by adjectives: The
generalizability of the Big Five to the Italian lexical context. European Journal of
Cattell, R. B. (1943a). The description of personality: Basic traits resolved into clusters.
Journal of Abnormal and Social Psychology, 38, 476–507.
Cattell, R. B. (1943b). The foundations of trait measurement. Psychological Review, 50,
559–594.
Cattell, R. B. (1945). The description of personality: Principles and findings in a factor
analysis. American Journal of Psychology, 58, 69–90.
Cattell, R. B. (1947). Confirmation and clarification of primary personality factors.
Psychometrika, 12, 197–220.
Cawley, M. J., III, Martin, J. E., & Johnson, J. J. (2000). A virtues approach to
personality. Personality and Individual Differences, 28, 997–1013.
Cheung, F. M., Van de Vijver, F. J. R., & Leong, F. T. L. (2011). Toward a new approach
to the study of personality. American Psychologist, 66, 593–603.
Church, A. T. (2000). Culture and personality: Towards an integrated cultural trait
psychology. Journal of Personality, 68, 651–703.
Church, A. T., & Katigbak, M. S. (1989). Internal, external and self/report structure of
personality in a non-western culture: An investigation of cross-language and cross-
cultural generalizability. Journal of Personality and Social Psychology, 57, 857–852.
Church, A. T., Katigbak, M. S., & Reyes, J. A. S. (1996). Toward a taxonomy of trait
adjectives in Filipino: Comparing personality lexicons across cultures. European
Church, A. T., Katigbak, M. S., & Reyes, J. A. S. (1998). Further exploration of Filipino
personality structure using the lexical approach: Do the Big Five or Big-Seven
dimensions emerge. European Journal of Personality, 12, 249–269.
Church, A. T., Reyes, J. A. S., Katigbak, M. S., & Grimm, S. D. (1997). Filipino
personality structure and the Big Five model: A lexical approach. Journal of
Coan, R. W. (1972). Measurable components of openness to experience. Journal of
Consulting and Clinical Psychology, 39, 346.
Coan, R. W. (1974). The optimal personality. New York, NY: Columbia University
Press.
Costa P. T., Busch C.M., Zonderman A. B., & McCrae R. R. (1986). Correlations of
MMPI factor scales with measures of the Five Factor model of personality. Journal of
Costa, P. T., Jr., & McCrae, R. R. (1976). Age differences in personality structure: A
cluster analytic approach. Journal of Gerontology, 21, 564–570.
Costa, P. T., Jr., & McCrae, R. R. (1985). The NEO Personality Inventory Manual.
Costa, P. T., Jr., & McCrae, R. R. (1992a). Four ways Five Factors are basic. Personality
Costa, P. T., Jr., & McCrae, R. R. (1992b). Revised NEO Personality Inventory (NEO PI-
R) and NEO Five-Factor Inventory (NEO-FFI) Professional Manual. Odessa, FL:
Dahlsgaard, K., Peterson, C., & Seligman, M. E. P. (2005). Shared virtue: The
convergence of valued human strengths across culture and history. Review of General
D’Andrade, R. G. (1965). Trait psychology and componential analysis. American
Anthropologist, 67, 215–228.
De Raad, B. (1985). Person-talk in everyday life; pragmatics of utterances about
personality. Unpublished dissertation, University of Groningen, The Netherlands.
De Raad, B. (1992). The replicability of the Big Five personality dimensions in three
word-classes of the Dutch language. European Journal of Personality, 6, 15–29.
De Raad, B. (1994). An expedition in search of a fifth universal factor: Key issues in the
lexical approach. European Journal of Personality, 8, 229–250.
personality. Göttingen, Germany: Hogrefe & Huber.
De Raad, B. (2009). Structural models of personality. In P. J. Corr & G. Matthews (Eds.),
The Cambridge handbook of personality psychology (pp. 127-147). New York, NY:
De Raad, B., & Barelds, D. P. H. (2008). A new taxonomy of Dutch personality traits
based on a comprehensive and unrestricted list of descriptors. Journal of Personality
De Raad, B., Barelds, D. P. H., Levert, E., Ostendorf, F., Mlačić, B., Di Blas, L., …
De Raad, B., Barelds, D. P. H., Timmerman, M.E., De Roover, K., Mlacic, B., & Church,
A. T. (2014). Towards a pan-cultural personality structure: Input from 11 psycho-
lexical studies. European Journal of Personality, 28, 497–510.
De Raad, B., & Ceulemans, E. (2001). The trait-dimensional scope of the characters of
Theophrastus. In R. Riemann, F. M. Spinath, & F. Ostendorf (Eds.), Personality and
temperament: Genetics, evolution, and structure (pp. 168–184). Lengerich, Germany:
Pabst Science.
De Raad, B., Di Blas, L., & Perugini, M. (1988). Two independently constructed Italian
trait taxonomies: Comparisons among Italian and between Italian and Germanic
languages. European Journal of Personality, 12, 19–41.
De Raad, B., & Doddema-Winsemius, M. (2006). De Big 5 persoonlijkheidsfactoren; een
methode voor het beschrijven van persoonlijkheidseigenschappen [The Big 5
personality factors: A method for the description of personality traits]. Amsterdam, the
Netherlands: Uitgeverij Nieuwezijds.
De Raad, B., Hendriks, A. A. J., & Hofstee, W. K. B. (1992). Towards a refined structure
of personality traits. European Journal of Personality, 6, 301–319.
De Raad, B., & Hoskens, M. (1990). Personality-descriptive nouns. European Journal of
De Raad, B., Mulder, E., Kloosterman, K., & Hofstee, W. K. B. (1988). Personality-
descriptive verbs. European Journal of Personality, 2, 81–96.
De Raad, B., & Ostendorf, F. (1996). Quantity and quality of trait-descriptive type nouns.
De Raad, B., & Peabody, D. (2005). Cross-culturally recurrent personality factors:
Analyses of three factors. European Journal of Personality, 19, 451–474.
De Raad, B., Perugini, M., Hřebíčková, M., & Szarota, P. (1998). Lingua Franca of
personality: Taxonomies and structures based on the psycholexical approach. Journal
of Cross-Cultural Psychology, 29, 212–232.
De Raad, B., Perugini, M., & Szirmák, Z. (1997). In pursuit of a cross-lingual reference
structure of personality traits: Comparisons among five languages. European Journal
De Raad, B., & Szirmák, Z. (1994). The search for the “Big Five” in a non-Indo-
European language: The Hungarian trait structure and its relationship to the EPQ and
the PTS. European Review of Applied Psychology, 44, 17–24.
De Raad, B., & Van Heck, G. L. (Eds.) (1994). The fifth of the Big Five. European
Journal of Personality, 8, Special Issue.
De Raad, B., & Van Oudenhoven, J. P. (2011). A psycholexical study of virtues in the
Dutch language, and relations between virtues and personality. European Journal of
DeYoung, C. G. (2006). Higher-order factors of the Big Five in a multi-informant sample.
Di Blas, L. (2005). Personality-relevant attribute nouns. A taxonomic study in the Italian
language. European Journal of Personality, 19, 537–557.
Di Blas, L., & Forzi, M. (1998). An alternative taxonomic study of personality-
descriptive adjectives in the Italian language. European Journal of Personality, 12, 75–
101.
Di Blas, L., & Forzi, M. (1999). Refining a descriptive structure of personality attributes
in the Italian language: The abridged Big Three circumplex structure. Journal of
Digman, J. M., & Inouye, J. (1986). Further specification of the five robust factors of
Eysenck, H. J. (1967). The biological basis of personality. Springfield, IL: Charles C
Thomas.
Eysenck, H. J. (1970). The structure of human personality (3rd ed.). London, England:
Methuen.
Eysenck, H. J. (1992). Four ways five factors are not basic. Personality and Individual
Fitzgerald, F. S. (1925). The great Gatsby. New York, NY: Charles Scribner’s.
Galton, F. (1884). Measurement of character. Fortnightly Review, 36, 179–185.
Gill, C. (1996). Personality in Greek epic, tragedy, and philosophy. Oxford, England:
Clarendon Press.
Goldberg, L. R. (1975). Toward a taxonomy of personality descriptive terms: A
description of the O.R.I taxonomy project. Oregon Research Institute Technical Report,
Vol. 15, No. 2.
Goldberg, L. R. (1981). Language and individual differences. The search for universals in
Goldberg, L. R. (1982). From ace to zombie. Some explorations in the language of
personality. In C. D. Spielberger & J. N. Butcher (Eds.), Advances in personality
assessment (Vol. 1, pp. 203–234). Hillsdale, NJ: Lawrence Erlbaum.
Goldberg, L. R. (1990). An alternative “description of personality.” The Big Five factor
Goldberg, L. R. (1992). The development of markers for the Big Five factor structure.
Goldberg, L. R. (1993a). The structure of phenotypic personality traits. American
Goldberg, L. R. (1993b). The structure of personality traits: Vertical and horizontal
aspects. In D. C. Funder, R. D. Parke, C. Tomlinson-Keasy, & K. Widaman (Eds.),
Studying lives through time: Personality and development (pp. 169–188). Washington,
DC: American Psychological Association.
Goldberg, L. R. (2006). Doing it all bass-ackwards: The development of hierarchical
factor structures from the top down. Journal of Research in Personality, 40, 347–358.
Goldberg, L. R. (2014). International Personality Item Pool: A scientific collaboratory
for the development of advanced measures of personality traits and other individual
differences (http://ipip.ori.org/). Internet Web Site. Accessed February 18, 2014.
Goldberg, L. R., & Rosolack, T. K. (1994). The Big Five structure as an integrative
framework. An empirical comparison with Eysenck’s P-E-N model. In C. F.
Halverson, G. A. Kohnstamm, & R. P. Martin (Eds.), The developing structure of
temperament and personality from infancy to adulthood (pp. 7–35). New York, NY:
Erlbaum.
Goldberg, L. R., & Somer, O. (2000). The hierarchical structure of common Turkish
person-descriptive adjectives. European Journal of Personality, 14, 497–531.
Goncharov, I. (1859; translation of 2006). Oblomov. New York, NY: Bunim &
Bannigham.
Gurtman, M. B. (1997). Studying personality traits: The circular way. In R. Plutchik & H.
R. Conte (Eds.), Circumplex models of personality and emotions (pp. 81–102).
Washington, DC: American Psychological Association.
Guttman, L. (1954). A new approach to factor analysis: The radex. In P. F. Lazarsfeld
(Ed.), Mathematical thinking in the social sciences (pp. 258–348). Glencoe, IL: Free
Press.
Hahn, D.-W., Lee, K., & Ashton, M. C. (1999). A factor analysis of the most frequently
used Korean personality trait adjectives. European Journal of Personality, 13, 261–
282.
Hampson, S. E. (1982). The construction of personality: An introduction. London,
England: Routledge & Kegan Paul.
Hampson, S. E. (1984). The social construction of personality. In H. Bonarius, G. Van
Heck, & N. Smid (Eds.), Personality psychology in Europe: Theoretical and empirical
developments. Lisse, the Netherlands: Swets & Zeitlinger.
Hendriks, A. A. J., Hofstee, W. K. B., & De Raad, B. (1998). Short behavior-descriptive
sentences as units of personality measurement. In J. Bermúdez, B. De Raad, J. De
Vries, A. M. Pérez-García, A. Sánchez-Elvira, & G. L. Van Heck (Eds.), Personality
psychology in Europe (Vol. 6, pp. 40–50). Tilburg, the Netherlands: Tilburg University
Press.
Hendriks, A. A. J., Hofstee, W. K. B., & De Raad, B. (1999). The Five-Factor Personality
Inventory (FFPI). Personality and Individual Differences, 27, 307–325.
Hendriks, A. A. J., Perugini, M., Angleitner, A., Ostendorf, F., Johnson, J. A., De Fruyt,
F., … Ruisel, I. (2003). The Five-Factor Personality Inventory: Cross-cultural
generalizability across 13 countries. European Journal of Personality, 17, 347–373.
Henss, R. (1995). From Aal to Zyniker. Personality descriptive type nouns in the German
language. European Journal of Personality, 9, 135–145.
Henss, R. (1998). Type nouns and the Five Factor model of personality description.
Hofstee, W. K. B. (2003). Structures of personality traits. In T. Millon & M. J. Lerner
(Eds.), Handbook of personality (Vol. 5, pp. 231–254). Hoboken, NJ: John Wiley.
Hofstee, W. K. B., De Raad, B., & Goldberg, L. R. (1992). Integration of the Big Five
and circumplex approaches to trait structure. Journal of Personality and Social
Hofstee, W. K. B., Kiers, H. A. L., De Raad, B., Goldberg, L. R., & Ostendorf, F. (1997).
A comparison of Big Five structures of personality traits in Dutch, English, and
German. European Journal of Personality, 11, 15–31.
Hřebíčková, M. (2007). The lexical approach to personality description in the Czech
context. Ceskoslovenska Psychologie [Czechoslovak Psychology], LI, 50–61.
Hřebíčková, M., Osecká, L., & Čermák, I. (1999). Taxonomy and structure of Czech
personality-relevant verbs. In I. Mervielde, I. Deary, F. De Fruyt, & F. Ostendorf
(Eds.), Personality psychology in Europe (Vol. 7, pp. 51–65). Tilburg, the Netherlands:
Tilburg University Press.
Imperio, S. M., Church, A. T., Katigbak, M. S., & Reyes, J. A. (2008). Lexical studies of
Filipino person descriptors: Adding personality-relevant social and physical attributes.
John, O. P. (1990). The “Big Five” factor taxonomy: Dimensions of personality in the
natural language and in questionnaires. In L.A. Pervin (Ed.), Handbook of personality:
Theory and research (pp. 66–100). New York, NY: Guilford Press.
John, O. P., Angleitner, A., & Ostendorf, F. (1988). The lexical approach to personality.
A historical review of trait taxonomic research. European Journal of Personality, 2,
171–203.
John, O. P., Hampson, S. E., & Goldberg, L. R. (1991). The basic level in personality-
trait hierarchies: Studies of trait use and accessibility in different contexts. Journal of
Kiers, H. A. L., & ten Berge, J. M. F. (1994). Hierarchical relations between methods for
simultaneous components analysis and a technique for rotation to a simple
simultaneous structure. British Journal of Mathematical and Statistical Psychology, 47,
109–126.
Klages, L. (1926). The science of character. (Translated into English 1932). London,
England: Allen & Unwin.
Kouwer, B. J. (1963). Het spel van de persoonlijkheid [The game of personality]. Utrecht,
the Netherlands: Erven J. Bijleveld.
Lee, K., & Ashton, M. C. (2004). Psychometric properties of the HEXACO Personality
Inventory. Multivariate Behavioral Research, 39, 329–358.
Loevinger, J. (1976). Ego development. San Francisco, CA: Jossey-Bass.
Mayer, J. D., Lin, S. C., & Korogodsky, M. (2011). Exploring the universality of
personality judgments: Evidence from the Great Transformation (1000 BCE–200
BCE). Review of General Psychology, 15, 65–76.
McAdams, D. P. (1994). The person: An introduction to personality psychology. Fort
Worth, TX: Harcourt Brace.
McCormick, C. C., & Goldberg, L. R. (1997). Two at a time is better than one at a time:
Exploiting the horizontal aspects of factor representations. In R. Plutchik & H. R.
Conte (Eds.), Circumplex models of personality and emotions (pp. 103–132).
McCrae, R. R. (1990). How well is Openness represented in natural languages? European
McCrae, R. R. (1994), Openness to experience: Expanding the boundaries of Factor V.
McCrae, R. R, (2010).The place of the FFM in personality psychology. Psychological
Inquiry, 21, 57–64.
McCrae, R. R., & Costa, P. T., Jr. (1985a). Updating Norman’s “Adequate taxonomy.”
Intelligence and personality dimensions in natural language and in questionnaires.
McCrae, R. R., and Costa P.T., Jr. (1985b). Comparison of EPI and psychoticism scales
with measures of the Five-Factor model of personality. Personality and Individual
McCrae, R. R., & Costa, P. T., Jr. (1987). Validation of the Five-Factor model of
personality across instruments and observers. Journal of Personality and Social
McCrae, R. R., & Costa, P. T., Jr. (1989a). Reinterpreting the Myers-Briggs type
indicator from the perspective of the Five-Factor model of personality. Journal of
McCrae, R. R., & Costa, P. T., Jr. (1989b). The structure of interpersonal traits.
Wiggins’s circumplex and the Five-Factor model. Journal of Personality and Social
McCrae, R. R., & Costa, P. T., Jr. (2008). The Five-Factor Theory of personality. In O. P.
John, R. W. Robins, & L. A. Pervin (Eds.), Handbook of personality: Theory and
research (pp. 159–181). New York, NY: Guilford Press.
McCrae, R. R., Terracciano, A., De Fruyt, F., De Bolle, M., Gelfand, M. J., Costa, P. T.,
Jr., … Yik, M. (2005). Universal features of personality traits from the observer’s
perspective: Data from 50 cultures. Journal of Personality and Social Psychology, 88,
547–561.
Milas, G., Mlačić, B., & Mikloušić, I. (2013). Construct validation of a general social
attitude scale (SAS_G). Journal of Individual Differences, 34, 203–213.
Miller, G. A. (1991). The science of words. New York, NY: Scientific American Library.
Mlačić, B. (2016). Social and reputational aspects of personality. International Journal of
Personality Psychology, 2, 15–36.
Mlačić, B., & Goldberg, L. R. (2007). An analysis of a Cross-Cultural Personality
Inventory: The IPIP Big Five Factor markers in Croatia. Journal of Personality
Mlačić, B., & Ostendorf, F. (2005). Taxonomy and structure of Croatian personality-
descriptive adjectives. European Journal of Personality, 19, 117–152.
Morales-Vives, F., De Raad, B., & Vigil-Colet, A. (2014). Psycho-lexically based virtue
factors in Spain and their relation with personality traits. The Journal of General
Psychology, 141(4), 297–325.
Mulder, B. F. (2006). Openheid voor ervaringen, lexical benaderd [Openness to
Experience, lexically approached]. Thesis, University of Groningen.
Norman, W. T. (1963). Toward an adequate taxonomy of personality attributes.
Norman, W. T. (1967). 2800 personality trait descriptors: Normative operating
characteristics for a university population. Unpublished manuscript: Department of
Psychology, University of Michigan, Ann Arbor, MI.
Osgood, C. E. (1970). Interpersonal verbs and interpersonal behavior. In J. L. Cowan
(Ed.), Studies in thought and language (pp. 133–227). Tucson, AZ: University of
Arizona Press.
Ostendorf, F. (1990). Sprache und Persönlichkeitsstruktur: Zur Validität des Fünf-
Faktoren-Modells der Persönlichkeit. Regensburg, Germany: S. Roderer Verlag.
Ostendorf, F. (1996). The structure of attitudes and its relation to the Big Five factors of
personality. Paper presented at the Symposium, Current Research on the Big Five
Factors of Personality on an International Scale, XXVI International Congress of
Psychology, Montréal, Canada, August 16–22.
Ostendorf, F., & Angleitner, A. (1994). Personality dimensions and appearance. Paper
presented at the Symposium Personality Dimensions, Seventh conference of the
European Association of Personality Psychology, Madrid, Spain, July 12-16.
Passakos, C. G., & De Raad, B. (2009). Ancient personality: Trait attributions to
characters in Homer’s Iliad. Ancient Narrative, 7, 75–95.
Peabody, D. (1987). Selecting representative trait adjectives. Journal of Personality and
Peabody, D., & De Raad, B. (2002). The substantive nature of psycholexical personality
factors: A comparison across languages. Journal of Personality and Social Psychology,
83, 983–997.
Peabody, D., & Goldberg, L. R. (1989). Some determinants of factor structures from
personality-trait descriptors. Journal of Personality and Social Psychology, 57, 552–
567.
Quevedo-Aguado, M. P., Iraegui, A., Anivarro, E. M., & Ross, P. (1996). Linguistic
descriptors of personality in the Spanish language: A first taxonomic study. European
Rotter, J. B. (1954). Social learning and clinical psychology. New York, NY: Prentice-
Hall.
Ruisel, I. (1997). Analysis of personality descriptors in the Slovak language. Studia
Psychologica, 39, 233–245.
Rümelin, G. (1890). Über die Temperamente [About temperaments]. Deutsche
Rundschau, 64, 397–412.
Ryle, G. (1949). The concept of mind. London, England: Penguin.
Saucier, G. (1997). Effects of variable selection on the factor structure of person
descriptors. Journal of Personality and Social Psychology, 73, 1296–1312.
Saucier, G. (2000). Isms and the structure of social attitudes. Journal of Personality and
Saucier, G. (2003a). An alternative multi-language structure for personality attributes.
Saucier, G. (2003b). Factor structure of English-language personality type nouns. Journal
Saucier, G. (2009). Recurrent personality dimensions in inclusive lexical studies:
Indications for a Big Six structure. Journal of Personality, 77, 1577–1614.
Saucier, G. (2010). The structure of social effects: Personality as impact on others.
Saucier, G. (2013). Isms dimensions: Toward a more comprehensive and integrative
model of belief-system components. Journal of Personality and Social Psychology,
104, 921–939.
Saucier, G., Georgiades, S., Tsaousis, I., & Goldberg, L. R. (2005). The factor structure
of Greek personality adjectives. Journal of Personality and Social Psychology, 88,
856–875.
Saucier, G., & Goldberg, L. R. (1996). The language of personality: Lexical perspectives
on the Five-Factor model. In J. S. Wiggins (Ed.), The Five-Factor model of personality.
Theoretical perspectives (pp. 21–50). New York, NY: Guilford Press.
Saucier, G., & Goldberg, L. R. (2001). Lexical studies of indigenous personality factors:
Premises, products, and prospects. Journal of Personality, 69, 847–879.
Saucier, G., Thalmayer, A. G., Payne, D. L., Carlson, R., Sanogo, L., Ole-Kotikash, L.,
… Zhou, X. (2014). A basic bivariate structure of personality attributes evident across
nine languages. Journal of Personality, 82, 1–14.
Semin, G. R., & Fiedler, K. (1988). The cognitive functions of linguistic categories in
describing persons: Social cognition language. Journal of Personality and Social
Shmelyov, A. G., & Pokhil’ko, V. I. (1993). A taxonomy-oriented study of Russian
personality-trait names. European Journal of Personality, 7, 1–17.
Shweder, R. A., & D’Andrade, R. G. (1980). The systematic distortion hypothesis. In R.
A. Shweder (Ed.), New directions of social and behavioral science, Vol. 4: Fallible
judgment in behavioral research for methodology (pp. 3–38). San Francisco, CA:
Jossey-Bass.
Singh, J. K., Misra, G., & De Raad, B. (2013). Personality structure in the trait lexicon of
Hindi, a major language spoken in India. European Journal of Personality, 27, 605–
620.
Somer, O., & Goldberg, L. R. (1999). The structure of Turkish trait-descriptive
adjectives. Journal of Personality and Social Psychology, 76, 431–450.
Strelau, J. (1983). Temperament-personality-activity. London, England: Academic Press.
Szarota, P. (1996). Taxonomy of the Polish personality-descriptive adjectives of the
highest frequency of use. Polish Psychological Bulletin, 27, 343–351.
Szirmák, Z., & De Raad, B. (1994). Taxonomy and structure of Hungarian personality
traits. European Journal of Personality, 8, 95–117.
Tellegen, A., & Waller, N. G. (1987). Re-examining basic dimensions of natural
language trait descriptors. Paper presented at the 95th Annual Convention of the
American Psychological Association, New York.
Themerson, S. (1974). Logic, labels, and flesh. London, England: Gaberbocchus.
Tupes, E. C., & Christal, R. E. (1958). Stability of personality trait rating factors obtained
under diverse conditions. USAF WADC Technical Note No. 58-61. U.S. Air Force:
Lackland Air Force Base, San Antonio, TX.
Tupes, E. C., & Christal R. E. (1961). Recurrent personality factors based on trait ratings.
USAF ASD Technical Report No. 61-97. U.S. Air Force: Lackland Air Force Base, San
Antonio, TX.
Wiggins, J. S. (1973). Personality and prediction: Principles of personality assessment.
Reading, MA: Addison-Wesley.
Wiggins, J. R. (1991). Agency and communion as conceptual coordinates for the
understanding and measurement of interpersonal behavior. In L.W. M. Grove & D.
Cicchetti (Eds.), Thinking clearly about psychology, Vol. 2: Personality and
Yu, S., Wei, L., He, W., Chai, H., Wang, D., Chen, W., & Wang, W. (2009). Description
of personality traits by Chinese adjectives: A trial on university students. Psychology of
Language and Communication, 13, 5–20.
Zhou, X., Saucier, G., Gao, D., & Liu, J. (2009). The factor structure of Chinese
personality terms. Journal of Personality, 77, 363–400.
Zuckerman, M. (1984). Sensation seeking: A comparative approach to a human trait.
Behavioral and Brain Sciences, 7, 413–471.
Zuckerman, M. (2002). Zuckerman-Kuhlman Personality Questionnaire (ZKPQ): An
alternative five-factorial model. In B. De Raad & M. Perugini (Eds.), Big Five
assessment (pp. 377–396). Goettingen, Germany: Hogrefe & Huber.
Zuckerman, M., Kuhlman, D. M., & Camac, C. (1988). What lies beyond E and N?
Factor analyses of scales believed to measure basic dimensions of personality. Journal
Factor Analytic Support for the Five Factor Model
Aidan G. C. Wright
Abstract
The Five Factor Model (FFM) has risen to prominence over the past 50 years, and
currently represents the most widely used structural model of personality
attributes. By definition, the FFM is built upon a foundation of factor-analytic
techniques. This chapter is divided into three parts. In the first, a methodological
primer is provided for those who may be less familiar with factor analytic
techniques. Second, the FFM and factor analysis are understood through a
historical review, along with updated exemplars of contemporary techniques and
applications to personality. Finally, several new directions in factor analytic
research of the FFM are reviewed, including its application to psychiatric
disorders.
Key Words: Five Factor Model, factor analysis, exploratory factor analysis,
confirmatory factor analysis, ESEM, personality structure
If determining the structure of personality traits has been the primary métier of
personality psychology through much of the twentieth century and the early part
of the twenty-first century, then factor analysis has been the primary tool of the
trade. The importance of structure in understanding personality traits and the
inventories intended to measure those traits is difficult to overstate—from
structure flows the framework that facilitates organization and comprehension of
an ever-expanding body of research. In this regard, the consensual Big Five or
Five Factor Model (FFM) is often heralded as one of the crowning achievements
of psychological science in the past century. As the predominant structural
model of personality traits, much has been written on the conceptual and
quantitative roots of the FFM. That is not to say, however, that all structural
issues in the personality trait domain have been settled. On the contrary, there
remain rapidly expanding literatures and ever-more quantitatively sophisticated
and complex studies on a variety of germane topics. Therefore, this chapter goes
beyond the historical factor analytic evidence for the FFM, and considers the
contemporary questions motivating a lively (and at times spirited) scientific
debate about the structure of personality and the role that factor analysis plays in
this discussion.1
Methodological Underpinnings
To appropriately evaluate the factor analytic evidence for the FFM, it is
necessary to keep in mind what factor analysis can and cannot do, and, in a more
nuanced way, how factor analyses perform under different conditions. As such,
this chapter starts with a nontechnical methodological review of factor analysis
(for a more thorough and technical coverage see, e.g., Fabrigar, Wegener,
MacCallum, & Strahan, 1999; Mulaik, 2009). Readers with a strong
methodological background in factor analytic techniques and latent variable
models may prefer to skip this section. The review proceeds chronologically,
starting with exploratory factor analysis (EFA), a technique in its second century
(Spearman, 1904, 1927), confirmatory factor analysis (CFA), approaching its
golden anniversary (Jöreskog, 1969), and exploratory structural equation
modeling (ESEM), still in its infancy (Asparouhov & Muthén, 2009).
Exploratory Factor Analysis

The aim of factor analytic techniques is to explain patterns of covariation
among observed or manifest (i.e., directly measured) variables using unobserved
or latent constructs. That is to say, given that responses to some stimuli (e.g.,
questionnaire items) show patterns of covariation, it is reasonable to hypothesize
that that there is an explanation for this patterning (e.g., a personality trait). In
fact, this was exactly the logic that prompted Spearman to develop factor
analysis. He had observed that those individuals who performed well on one
mental test tended to perform well on others, which gave rise to his general
theory of intelligence and the need for a quantitative method to test it. Factor
analysis was thus born. Since then, EFA has been widely applied as a technique
to determine the appropriate number of dimensions that can account for
observed patterns in a larger set of variables.
The hallmark of EFA is that the investigator does not specify the estimated
parameters and the patterning of items loading on factors, and instead these are
determined by the analysis. Figure 11.1A provides a graphic representation of
EFA. Square boxes represent observed variables, circles represent latent
variables or factors, straight arrows connecting circles and squares represent
factor loadings, arrows only pointing toward squares represent observed variable
uniqueness (i.e., variability not accounted for by the latent factors, which
includes both unique variance and error variance), and curved arrows represent
covariances/correlations. Additionally, solid lines represent model specified
parameters, whereas dashed lines represent parameters that can be specified by
the investigator. In this example there are six observed variables and two
correlated factors (i.e., an oblique model), and each of the observed variables
loads on each of the two factors. This is a basic feature of EFA, and one reason
why it is referred to as exploratory—the investigator does not assign observed
variables to factors, rather the relationship between each is estimated and the
pattern of loadings is evaluated or “interpreted” after the analysis is run.
Some have called EFA an atheoretical data analytic approach, which is a
glaring mischaracterization. Many aspects of EFA are theoretically driven. One
of the most fundamental theoretical assumptions that can be made in EFA is that
underlying dimensions account for patters of observed responses to stimuli. This
is reflected in structural diagrams that represent factor loadings as arrows
emanating from the latent factors to the observed variables.2 Furthermore, it is
frequently the case that the investigator has a hypothesis about how many factors
are needed to account for the observed variables (e.g., Spearman’s theory of
general intelligence would suggest a one-factor model). And, usually, there is a
theory about which observed variables serve as markers for the same factors.
More generally, modeling decisions should ideally be made based on substantive
theory. For instance, factors must be interpreted and labeled, and the emergence
of a factor that is uninterpretable may prompt us to select fewer factors, drop
some items, or collect more data.
EFA is, however, a very interactive technique, in the sense that several models
are often run under different conditions and compared before settling on a final
solution. Several considerations are involved in arriving at an acceptable model,
but the two primary ones include selecting the number of factors to retain and
the rotation for those factors. In each case theoretical considerations should be
the primary deciding factor, although quantitative indices have been developed
to aid in this process. As powerful desktop computing has become ubiquitous,
coarse rules of thumb such as the Kaiser–Guttman rule (i.e., retain only factors
with eigenvalues greater than 1.0) and visual tools such as Cattell’s (1966) scree
plot have given way to more quantitatively rigorous criteria such as Horn’s
(1965) parallel analysis, Velicer’s (1976) minimum average partial test, Ruscio
and Roche’s (2012) comparison data technique, and fit criteria [e.g., chi-square,
root mean square error of approximation (RMSEA)] when available based on the
estimator (e.g., maximum likelihood). Regardless of which methods are used,
and it is advisable to enlist the aid of at least one of the more rigorous
approaches, these are fallible tools that should be weighed in the decision but not
followed blindly. The investigator is still required to make careful choices based
on all pertinent information, especially theory.
Once the number of factors has been decided on, the next step generally
involves selecting a rotation for the factors. Rotation refers to adjusting the
relationship between the observed variables and latent factors through a linear
transformation. In other words, the exact pattern of factor loadings is not fixed,
and can be quantitatively adjusted to improve the interpretability of the factors.
Many rotation schemes exist (e.g., Varimax, Geomin; see Sass & Schmitt, 2010,
for a review) that attempt to maximize specific criteria (e.g., factor or variable
simplicity), with a frequent goal of achieving a simple structure (Thurstone,
1947). In a perfect simple structure each observed variable loads strongly and
exclusively on a single factor, allowing for easier interpretation. Despite the
many options available for factor rotation, the most important distinction is
whether it generates orthogonal or oblique factors. In an orthogonal rotation, the
factors are forced to be unrelated to each other, whereas in an oblique rotation
factors are allowed to correlate. As will be discussed below, choosing between
oblique and orthogonal rotation can have implications that go beyond an
individual factor analysis.
Because all items load on all factors in EFA, rotation of the factors is
arbitrary, quantitatively speaking. This is not to say that any single loading can
be adjusted to whatever the investigator wants. On the contrary, rotating a factor
causes all of its factor loadings to change based on the linear transformation.
However, there is no mathematical superiority in selecting one rotation over
another, and the decision rests on interpretability. Some have sharply criticized
EFA because of this aspect, claiming that because factor rotations are
quantitatively arbitrary, factors are therefore conceptually arbitrary and merely
conventions of human cognition. Although not a reason to disregard EFA as a
useful technique, knowledgeable factor analysts keep in mind that a factor
represents a hypothetical latent construct, and neither does it reflect a 1:1
correspondence with any particular natural phenomenon, nor can differing
interpretations for a factor be adjudicated quantitatively without external
validation.
Several other considerations require attention when conducting or evaluating a
factor analysis. Of these, the selection of variables to include is perhaps the most
important because it will constrain the resulting factor structure. In some
respects this point is so obvious as to not bear mention, but it plays out in subtle
ways. Recall that factor analysis seeks to account for patterns of covariation,
which follows directly from the included variables. If sufficient measures of a
construct are not included, it will not emerge as a factor in the analysis. For
instance, if too few measures of a construct are included (e.g., a single openness
scale), it is unlikely to emerge as a stand-alone factor, and instead these
indicators may join another factor or be orphaned with low loadings on all
factors. A reasonable but flexible heuristic is that a factor analysis should
include at least three primary indicators of a hypothesized construct. Indeed,
much of the early work on personality traits suffered from small sets of variables
that could hardly be considered comprehensive, thereby precluding any
definitive solution (Digman, 1996). On the flip side of this coin, overloading an
analysis with many variables of a very similar nature virtually guarantees that a
specific factor will emerge, even if the variables are individually conceptually
related to others in the same analysis (i.e., a grouping of variables tightly related
relatively speaking, even with marked associations with the other variables, will
stand out). As might be apparent, this can cause spurious factors to emerge that
better reflect the density of measurement as opposed to substantive distinctions
between observed variables. For instance, adding six measures of anxiety
disorders to a factor analysis of normative personality traits will likely result in
an Internalizing factor separate from neuroticism that accounts for the disorders,
even as it correlates highly with neuroticism. When the variables forming a
factor are highly redundant it has been termed a “bloated specific” factor,
denoting the excess of items used to measure a narrow piece of information.
In sum, EFA seeks to account for shared variance among observed variables
with a smaller set of reduced factors. The investigator must choose which
variables to include in the analysis, and then interpret the resulting factor
solution, deciding on how many factors to retain and how to maximize
interpretability of the loadings by selecting a rotation. Because of the high
degree of investigator decisions, EFA has been criticized as being arbitrary and
more art than science. This criticism ignores the principled nature of most
decisions and the high degree of investigator decision making in all statistical
analysis. A well-conducted factor analysis is no more subjective than any other
complex data analytic technique.
Confirmatory Factor Analysis

As the name indicates, unlike EFA, CFA was intended to serve primarily as a
hypothesis testing analytic approach. CFA shares the major conceptual basis of
EFA, in that the goal is to represent patterns of covariation among a set of
observed items with a smaller set of unobserved factors. The confirmatory
aspects are that (1) the user specifies all3 of the model parameters, and (2) the fit
(or, more specifically, the lack of fit) of the specified model to the observed data
is tested. Figure 11.1B illustrates a hypothetical typical two-factor CFA. In this
model the observed variables Y1–Y3 serve as indicators of latent factor F1 only,
and Y4–Y6 serve as indicators of F2 only. Note that one major distinction from
the EFA in Figure 11.1A is that each factor loading was specified, and not all
items load on each estimated factor. Much like the EFA model the factors are
allowed to correlate, making it an oblique model. However, there is no rotation
to choose, factors are either correlated (oblique) or uncorrelated (orthogonal).
Furthermore, each observed variable has a residual variance, reflecting unique
variability unaccounted for by the factor plus error. Finally, notice the curved
arrow between Y2 and Y5. This reflects an error covariance, indicating that there
is shared variance in items Y2 and Y5 unaccounted for by the factors that the
investigator is modeling.
Figure 11.1. (A) Exploratory factor analysis. (B) Confirmatory factor analysis. (C)
Exploratory structural equation modeling.
Were we to actually test this model, our chosen statistical package would first
optimize the values of the parameters in an effort to match the data set we were
using with some form of estimator (e.g., maximum likelihood, weighted least
squares); it would then compare the fit of the model implied covariance matrix
to the observed covariance matrix and generate goodness-of-fit indices based on
the degree of match and other criteria. It is worth noting that each model implies
a certain pattern of covariation based on its parameterization. For instance, in the
case in which there are no free error covariances, the factors must account for all
of the covariation among the observed variables. Any unaccounted for residual
covariation in the actual data will contribute to worse fit.
A detailed discussion of various fit indices goes beyond the purview of this
chapter. However, each fit index is specified quantitatively and under certain
modeling conditions it may not match the statistical test the investigator intends.
For instance, the model chi-square is often written off because it “performs
poorly in large samples.” This is inaccurate; the static is performing exactly as it
is intended to do. It is more appropriate to recognize (and state) that the chi-
square statistic tests whether the model’s implied covariance matrix fits the data
perfectly, and in large (i.e., highly powered) samples it is sensitive to very minor
sources of ill fit that are unlikely to have practical significance. In studying
personality structure we are infrequently interested in the level of precision
afforded by the chi-square statistic, but are usually interested in using large (and
therefore highly powered) samples and large variable sets that will also
contribute to a poorly fitting model as judged by the chi-square test unless large
numbers of complex factors are included. Thus, the chi-square test is often at
odds with the aims of applied personality researchers. There is nothing dubious
about selecting alternative fit indices that more closely match the desired level of
precision and account for modeling features that will be encountered in
personality structure studies. But the investigator should be making this choice
understanding the issues involved.
In CFA strict simple structure can be specified and tested by allowing each
observed variable to serve as in indicator for only one factor (e.g., as in Figure
11.1B). However, this is not a requirement, and variables may serve as indicators
for more than one factor. There are important implications for deciding whether
to make an indicator simple or complex. Recall that the estimator will attempt to
fit the model parameters to the data first. Thus if an indicator is complex,
meaning that it is influenced by more than one underlying factor, and it is
allowed to load only on a single factor, then this will result in a stronger
covariation among the factors. To provide a concrete example, consider
depression as an indicator, which is known to be associated with high
neuroticism and low extraversion (Clark & Watson, 1991). If depression is
allowed to load on neuroticism only in a model that includes other markers and
latent factors for both neuroticism and extraversion, this will result in an
increased negative correlation among neuroticism and extraversion.
Alternatively, if the depression variable is allowed to load on both, it will
decrease the latent factor correlation because the patterns of association have
now been accounted for at the level of the item loadings. As will be discussed
below, this has important implications for the way trait hierarchies are studied
and the conclusions that are drawn from them.
Another attractive feature of CFA is that it allows for principled deviation
from the assumption of conditional independence. Factor models are often
specified such that there is no covariance among the item residuals, the
assumption being that the observed variables are independent of each other once
the factors are accounted for (i.e., conditional on the factors). Although
reasonable given the goal of factor analysis, relaxing this assumption has
legitimate uses. For instance, it can be used to account for method variance
between specific item sets (e.g., items that share the same stem). However, the
unprincipled use is to be discouraged, as it can capitalize on chance in any given
data set, especially when sample size is large, and result in nonreplicable model
complexity.
Though CFA has many advantages (e.g., confirmatory nature, full control in
the ability to free and fix parameters), it is unwieldy for use as a purely
exploratory tool (i.e., when the latent structure of the data is not well understood
or is unknown), especially with large variables sets. Admittedly, many
investigators use CFA in a semiexploratory fashion in applied research (e.g.,
making modifications based on Lagrange multiplier tests, also known as
modification indices), but as confidence in the precise structure decreases, and
item set size increases, the utility of CFA for exploration declines. Yet there are
many situations in which the investigator is most interested in the data-derived
structure, or, more frequently, wants to relax the assumption of no cross-
loadings, but does not want to go through parameter by parameter testing them.
Exploratory Structural Equation Modeling

Exploratory structural equation modeling (ESEM) is a relatively recent
development in latent variable modeling (Asparouhov & Muthén, 2009), and it
allows for EFA-derived latent variables to be included within the broader SEM
framework. Another way to think about ESEM, especially as it relates to the
types of models we are discussing here, is that ESEM blends the features of EFA
(i.e., exploratory factors, range of rotations) and CFA (i.e., the ability to specify
parameters, user-specified factors, multiple group analysis) allowing for near
total flexibility in modeling. A number of considerable advantages are gained by
this innovation. These include the ability to add method factors to EFA analyses
of multiple scales from different measures, correlated residuals, and adding
parameter equalities across two scientifically interesting groups (e.g., genders,
patient versus nonpatients). Figure 11.1C provides a hypothetical example of an
ESEM model. In this diagram, in addition to two obliquely rotated EFA-defined
factors (F1 and F2), there is a third investigator specified factor (F3) that is
orthogonal to the other two. F3 could perhaps represent shared method variance
for observed variables Y1–Y3, or the assumption that they are markers for more
than one construct. Finally, the residuals for Y4 and Y6 are allowed to correlate.
In the modeling of complex personality data that has large item sets, ESEM
benefits from the efficiencies of the EFA framework, while allowing the
investigator to have control over specific study design features that are afforded
with CFA.
Similar to CFA, ESEM relies on estimation methods that ultimately result in
an implied covariance matrix that can be compared to an observed matrix in
various ways to generate goodness-of-fit indices. The fact that the EFA portion
of the structure can model a large number of potentially conceptually negligible
but statistically significant cross-loadings generally results in considerable
improvement in fit over a strict simple structure imposed by many CFAs (see
Booth & Hughes, 2014, for a summary). However, it is worth belaboring the
point that factor analytic techniques are in large part separable from the
estimation approach. Although certain estimation methods (e.g., principle factor
analysis) are reserved for EFA, estimations such as maximum likelihood and
weighted least squares can be applied to EFA, CFA, or ESEM. This
underappreciated fact often results in claims that an ESEM has been conducted,
when in reality only an EFA has been conducted with maximum likelihood-
generated estimates. Although this produces fit criteria, no additional
confirmatory work or alteration of the structure by the user has been done. As
noted above, EFA is a legitimate and very useful technique, and the objection
with labeling a maximum likelihood EFA as ESEM is related to perceived
inflation of the rigor of analysis without having done anything other than a
standard EFA. Alternatively, a maximum likelihood-based EFA could be
considered one very basic form of an ESEM, but this would seem to muddy the
methodological waters. Now armed with the contemporary thinking on factor
analysis, both historical and recent studies will be considered.
Factor Analysis of Personality Attributes
The historical development and ongoing research on the FFM are intimately
intertwined with factory analysis. The preceding review of the statistical
techniques was intended to garner a better appreciation of the model’s early fits
and starts and the challenges it is facing today.
Historical Foundations: Exploring the Structure of Personality Attributes

Domains that in hindsight are easily interpreted through the lens of the current
FFM were identified by factor analysis as far back as 1930s by investigators
such as Cattell (1933) and Guilford (Guilford & Guilford, 1934). At the time,
however, both the technical and conceptual state of affairs limited the ability of
early researchers to arrive at a replicable solution. Considering first the
technological capacities available at the time, the earliest factor analyses had to
be conducted by hand. Although calculators eventually supplanted longhand
calculations, it would be almost two decades before computers were available
for routine use. Conducting a factor analysis by hand is an incredibly time-
consuming and grueling proposition, thus limiting early efforts to a minority of
variables. Early studies were conducted with approximately 20–30 personality
variables. Contrast this with Goldberg’s (1990) definitive analysis of three data
sets of trait-descriptive terms. In the first of three studies, he factor analyzed 75
terms (more than twice that of Cattell’s 35 scales), using five different extraction
methods (e.g., principle axis, maximum likelihood, and principle components)
and two different rotations per method. Performing even one analysis with that
many terms in the early part of the twentieth century would have been
unthinkable, let alone 10, plus two additional studies with variable sets of 100
and 133, respectively. The effect of this early limitation precluded the analysis of
anything resembling a comprehensive list of variables, thus requiring more
circumscribed and idiosyncratic sets (e.g., Thurstone, 1934). This clearly
resulted in some domains being underrepresented [e.g., agreeableness in
Guildford & Guilford (1934) and conscientiousness in Thurstone (1951); see
Digman (1996) for a detailed review], and other less conspicuous influences
were likely in operation as well. Recall that the results of the EFA are driven by
the covariation of the included variables—inadequate representation of a
construct precludes the emergence of a related factor. As a result, the most
convincing and replicable solutions for this type of question will be those that
adequately sample from the full domain of personality terms. In the early part of
the twentieth century, however, it was as if investigators were attempting to
solve the personality puzzle without a full set of pieces.
Another early stumbling block beyond idiosyncrasies in early variable
selection was the view, promoted by Thurstone (1947), that “too many factors
can do no harm” (p. 509). Offered in the Old Testament of factor analysis, this
perspective is apparently responsible for Cattell’s decision to extract improbably
large numbers of factors from his datasets in the 1940s (Cattell, 1943, 1944,
1945, 1947, 1948). Drawing from Allport and Odbert’s (1936) comprehensive
list of terms, Cattell factored a set of 35 scales, and across three studies retained
12, 11, and 11 factors, respectively. Surprisingly by today’s standards, Cattell
decided to bundle all of the factors that emerged across all of his studies in his
16-factor model of personality, despite the fact that only some of these factors
replicated. Indeed, one major danger resulting from factor overextraction is that
as more and smaller factors are extracted, they are less likely to reflect
important, replicable domains, and more likely to reflect dataset-specific
variability. This is similar to the principle in CFA described above that we
should not blindly free error covariances, even if suggested by modification
indices, because it is likely capitalizing on chance variation in a specific sample.
These analyses were conducted several decades before modern quantitative
procedures were developed to inform the appropriate number of factors to retain
(e.g., Parallel Analysis, MAP Test), which would have been much more
conservative in their selection of factors. In the context of this chapter, these
early studies are worth revisiting because they demonstrate just how closely
intertwined the history of the FFM is with the history of factor analysis. Factor
analytic methods were in their infancy and their performance was not yet fully
understood, and therefore early misses are to be expected.
Over the next decade there was little advancement, aside from one notable
study that found a solution near to the Big Five (Fiske, 1949), which in fact
found a familiar four, excluding a clear agreeableness. Some have argued that
this is the provenance of the modern structure (see also the chapter by de Raad
and Mlačić). However, in the early part of the 1960s, a series of seminal studies
hammered out the structure that would ultimately come to form the FFM that is
evident today. First, Tupes and colleagues (Tupes & Christal, 1961; Tupes &
Kaplan, 1961) settled on five replicable factors in their analyses. Despite the fact
that these were mostly obscured from view by being published in U.S. Air Force
reports, several researchers (e.g., Borgatta, 1964; Norman, 1963; Smith, 1967)
did take notice and followed up with work of their own resulting in very similar
structure using different sets of scales rooted in the Allport and Odbert terms
(Norman, 1963; Smith, 1967), or as newly generated items (Borgatta, 1964). The
fertile grounds laid in the 1960s mostly lay fallow as trait psychology entered its
lost decade, ushered in by critics such as Mischel (1968).
However, the seeds had been planted, and they would bloom in the 1980s.
Interestingly, among the seminal investigations in this era was a reanalysis of
Cattell’s and Fiske’s earlier scales using more modern factor retention rules. In
this study, Digman and Takemoto-Chock (1981) found that when more stringent
criteria were employed, five replicable factors emerged across the earlier
datasets. These factors conceptually matched those found by Tupes and Christal
(1961) and Norman (1963), and are recognizable as the five factors that
comprise the FFM of today. At around the same time, Goldberg published two
chapters (1981, 1982) summarizing and distilling much of the existing factor
analytic evidence of the time, arguing that most contending models could be
organized within the framework of Norman’s adequate taxonomy (i.e.,
Norman’s five traits). According to Goldberg (1993; see also Digman, 1990), it
was based on these writings (Goldberg, 1981, 1982) that he was invited to a
conference held in 1983 by Drs. Costa and McCrae. Up to that point, Costa and
McCrae had invested their time in studying and developing a measure for a
three-factor model of personality comprised of neuroticism (N), extraversion
(E), and openness to experience (O). Based on Goldberg’s work, they adopted
agreeableness (A) and conscientiousness (C) into their model, and the NEO
Personality Inventory (NEO PI; Costa & McCrae, 1985) was published with
multiscale domains for N, E, and O, and single scale domains for A and C (see
also the chapter by Costa and McCrae). By 1992 this structure was expanded to
include an equal number of facets—six—per domain in the incomparably
popular NEO PI-Revised (NEO PI-R; Costa & McCrae, 1992). Throughout the
1980s and continuing through to today, the Costa and McCrae team produced
studies on the FFM with leporine speed. As Goldberg (1993) stated, Costa and
McCrae are the “world’s most prolific and most influential proponents of the
five factor model” (p. 30), and the same has remained true for the more than 20
years since.
Paralleling Costa and McCrae’s work, numerous other investigators
contributed to the factor analytic evidence for the FFM during the 1980s in a
variety of samples, cultures, and criteria sets (e.g., Amelang & Borkenau, 1982;
Digman, 1989; Digman & Inouye, 1986; John, 1989; McCrae, Costa, & Busch,
1986; Peabody & Goldberg, 1989; Trapnell & Wiggins, 1990). The decade
culminated with the seminal study by Goldberg (1990) mentioned above, which
showed that large sets of scales derived from the Allport and Odbert terms
resulted in highly consistent conclusions across factor analytic techniques,
rotations, and even across large and diverse sets of items (i.e., 75–133).
The early era of FFM research was based almost exclusively on EFA. Several
of the initial stumbling blocks that delayed the arrival of the five factor structure
are easily understood, in retrospect, as emerging from the state of factor analysis
and computing technology at the time. As EFA came into its own and the
methodology was better understood and its application standardized, results
became more consistent, and consensus could be achieved on the structure of
broad personality domains. One of the major take home messages of this line of
work has less to do with factor analysis per se, and more to do with good science
in general: replicate, replicate, replicate.4
Model Fit in the Modern Era: Confirming the Structure of the FFM
By the 1990s, with the FFM firmly established, researchers turned away from
explorations of basic structure and toward validation of the current model and
prediction of external variables (e.g., Ozer & Benet-Martinez, 2006). In many
respects, it could be argued that EFAs conducted and replicated in reasonably
comprehensive but different sets of trait-relevant variables (e.g., scales, items,
ratings) provide the strongest evidence for the FFM as a natural phenomenon.
Nevertheless, around this same time, CFA began gaining in popularity and was
being widely disseminated via commercial statistical packages (e.g., LISREL),
and naturally investigators sought to test the FFM structure in this more stringent
confirmatory framework.
To my knowledge, the first study to use CFA to test the FFM structure was
conducted by Borkenau and Ostendorf (1990). For the primary analyses,
Borkenau and Ostendorf (1990) included three sets of FFM scales as observed
variables in CFA models: the self-rated NEO PI, self-rated Norman scales, and
peer-rated Norman scales. In many ways, this early study was exemplary in its
application of CFA. A series of CFA models were estimated that allowed the
relevant scale scores from each measure/rater to load on one, and only one, of
five corresponding content factors, with three method factors included to
account for shared variance associated with the measure or rater. For instance,
the extraversion factor had as indicators the NEO PI extraversion, self-rated
Norman extraversion, and other-rated Norman extraversion, and all NEO PI
scales loaded on a single factor accounting for shared method variance. The
same was true for the remaining four domains and the Norman self-report and
other-report scales.
One variant of this model constrained the factor correlations to 0.0, consistent
with the theoretical view that the five factors are orthogonal or close to it, but
fared poorly in terms of fit. In contrast, an oblique model did achieve acceptable
fit, and was a significant improvement in fit (i.e., change in chi-square) over the
orthogonal model. This type of model comparison highlights the attractive
features of CFA, providing the researcher with the capacity to test interesting
theoretical questions (e.g., completely distinct versus related domains).
However, building on these results, the authors then examined a CFA model that
included the full set of lower-order markers (or facets) from each of the
measures. In contrast to the model based on domain-level scales, the expanded
model resulted in an abysmal fit leading to tempered enthusiasm by the authors
for the results as a whole.
A detailed review of this initial study is warranted, not only because of its
temporal primacy, but because many of the issues with which Borkenau and
Ostendorf were grappling have plagued the long string of CFA studies of FFM
measures since (e.g., Church & Burke; 1994; Donnellan, Oswald, Baird, &
Lucas, 2006; Gignac, Bates, & Jang, 2007; Hopwood & Donnellan, 2010; Lim
& Ployhart, 2006; McCrae et al., 1996; Parker, Bagby, & Summerfeldt, 1993;
Vassend & Skrondal, 1995, 1997). The major issue was that CFA makes highly
restrictive assumptions about the structure of personality, assuming a very strict
simple structure, with each item or scale loading on one, and only one, factor.
Although this strict simple structure is not a requirement of CFA models, it is
often how they are taught in introductory structural equation modeling courses
and presented in the literature. Furthermore, Borkenau and Ostendorf (1990)
rightly pointed out, as many others have since (e.g., DeYoung et al., 2007;
Hofstee, De Raad, & Goldberg, 1992; Hopwood & Donnellan, 2010; McCrae et
al., 1996), that markers of the FFM, be they individual items or scales, are
generally complex, meaning they often reflect more than one domain’s content
(e.g., NEO PI-R Warmth loads on both extraversion and agreeableness;
DeYoung, 2013; McCrae et al., 1996). This leaves investigators interested in
CFAs of the FFM faced with somewhat of a Catch-22. Specifically, CFA holds
the promise of providing a test of theoretical structure, or the rigorous
comparison of structure across important groups (e.g., males versus females,
Germans versus French, patients versus nonpatients), and therefore all expected
cross-loadings should be modeled. And yet, at the same time, were we to
actually consider specifying each reasonable cross-loading manually a priori, it
would require a seemingly impossible number of predictions that would likely
really be revealed only via exploratory work. For instance, take the NEO PI-R
scales; for each facet there is one clear prediction associated with the domain it
is intended to mark, but for each facet there are also four additional choices to be
made for a total of 120 possible secondary loadings—this doubles to 240 for the
60 items of the NEO-Five Factor Inventory (NEO-FFI). Alternatively, we could
allow cross-loadings post hoc based on significant modification indices; but now
the structure could hardly be considered confirmatory. This latter practice of
allowing error covariances to correlate in CFA, although perfectly allowable
quantitatively speaking (i.e., there is no necessity to make the assumption of
conditional independence), if done in a post hoc fashion to maximize fit will
generally result in many parameters that do not replicate across samples.
Borkenau and Ostendorf (1990) further noted that it was likely possible,
moving forward, to create narrow-band scales that maximized within-scale
correlations and minimized associations with other scales in such a way that
reasonable fit could be achieved in a CFA. But they also immediately discredited
this as a solution because this would be an unacceptable prioritizing of statistical
fit over the conceptual breadth of each of the five factors (i.e., a poor bandwidth
versus fidelity tradeoff). Others have suggested abandoning the enterprise of
CFA when considering the FFM, and instead reverting to EFA, and either
employing targeted rotations (i.e., Procrustes rotations; McCrae, Zonderman,
Costa, Bond, & Paunonen, 1996) or using maximum likelihood estimated EFAs
(Hopwood & Donnellan, 2010) that generate the same fit statistics as CFAs. In
each case the proponents have emphasized replicability across samples as
opposed to statistical fit between the model and the data within one sample,
paired with tests of replicability with congruence indices. By this standard, the
NEO PI-R measures (McCrae et al., 1996) and other measures of the FFM (and
non-FFM personality inventories; Hopwood & Donnellan, 2010) fare quite well.
In contrast to Borkenau and Ostendorf’s (1990) early investigation that
focused on cross-measure and rater confirmation of the theoretical structure of
the FFM, the majority of the CFA studies since have focused primarily on
whether a specific measure’s item-set or scales conform to the expected
structure. This is notable in several respects. In part this reflects the maturing
nature of the science, which has moved from unconstrained and, frankly, highly
variable criteria sets across research groups to refined and better delineated
instruments that are providing consistency of measurement across laboratories.
Accordingly, only the most extreme researchers suggested that the results of
CFA studies indicted the FFM proper (e.g., Vassend & Skrondal, 1997), and
most others limited their questions to suitability of specific measures. Yet others
have highlighted the fact that more recently developed measures, those that have
been developed in the CFA-dominated modern era, are at a disadvantage
because they will be judged against criteria that are difficult for any broadband
personality inventory to meet (Hopwood & Donnellan, 2010). This is despite the
fact that established measures developed earlier also do not meet the CFA
standards for fit, generally speaking.
Regardless of any mismatch between the overly restrictive ways in which
CFA is customarily applied, the highly sensitive nature of CFA fit indices, and
the goals of delineating the broadband structure of personality and developing
instruments to measure it, there are many attractive features of a CFA
framework that would be a shame to discard. These include the ability to test for
measurement invariance across time, groups, and possibly even different item
sets. Also, the ability to model additional factors that might influence fit, such as
method factors (e.g., Marsh, 1996; Quilty, Oakman, & Risko, 2006), or as will
be discussed more below, hierarchical structures (e.g., Digman, 1997), is a
desirable quality of CFA. An ideal method would be able to accommodate the
attractive features of EFA and CFA within a common framework.
ESEM was developed precisely to facilitate this compromise (Asparouhov &
Muthén, 2009). As noted above, ESEM allows for the estimation of exploratory
structures while also including researcher-specified constraints, paths, or factors.
Among the earliest practical applications of ESEM was the examination of the
measurement invariance of the full NEO-FFI’s 60 items (Marsh et al., 2010). In
a highly detailed exposition, Marsh and colleagues (2010) demonstrated the full
capabilities of ESEM for furthering FFM structural research by examining
invariance across gender and time and modeling additional sources of shared
method variance among items. First, the authors showed, unsurprisingly, that the
ESEM model achieved better fit than a strict simple structure CFA. Second, they
noted that NEO-FFI uses items that come from different facets to calculate
domain scores, which could lead to residual covariation among items. As such,
they allowed item residuals to freely covary if they came from the same facet.
For those unfamiliar with the NEO item sets, some facets contain highly
redundant items (e.g., Impulsiveness), making this a reasonable approach.
Allowing these items to correlate resulted in a much better fitting model. Third,
they then tested for invariance across genders (using a multigroup approach) and
time (using a longitudinal invariance approach), which required the fixing of
parameters to equality across groups or time. Although a full summary of the
results go beyond this chapter, in brief Marsh and colleagues (2010) found
support for partial measurement invariance across gender and time.
This study highlighted just how flexible the ESEM approach can be in
practice. Marsh and colleagues (2010) also noted that there has been an artificial
schism between research that has investigated the factorial structure of
personality using omnibus sets of items and inventories, and research that has
studied the measurement invariance of scales. These two literatures are not at
odds with each other, and are actually interested in the same questions.
However, the differences in methodology have kept them apart. ESEM makes
invariance testing with larger personality inventories tractable, and could lead to
improvements in measurement and the certainty of conclusions drawn in
practical research.
Another interesting feature of this study is that the authors chose to use an
oblique rotation. In part this choice made it possible to illustrate the fact that
going from a CFA with no cross-loadings, to an ESEM framework in which all
items can freely cross-load, factor intercorrelations will markedly diminish
(although they do not entirely disappear). Booth and Hughes (2014) have since
demonstrated a similar decrease in factor intercorrelations with several other
FFM measures going from a CFA to a maximum likelihood EFA framework.
These observations are not new, and in fact have long been recognized. Indeed,
McCrae and colleagues (1996) argued strongly for the use of exploratory
solutions as opposed to confirmatory solutions because the five factors are
theoretically orthogonal, and this can be specified in EFA without loss of fit.
Accordingly, McCrae and like-minded colleagues have tended to view the larger
correlations among the FFM factors obtained with CFA as being “inflated” or
even “biased upwards.”5 It is important to recognize, however, that this is a
theoretical argument, and is not based on a quantitative rationale. That is to say,
given that items and scales that form the basis of factor analytic models of
personality appear to be complex (i.e., share variance not only with other
putative markers of the same domain, but also with scales that are presumed to
be markers of other domains), factor correlations will be smaller or larger
depending on the degree of cross-loadings allowed. The complexity can either be
modeled at the level of the loadings of individual items or scales, or it can be
modeled at the level of the factor correlations. This has major theoretical
implications, both for the FFM, which is often presumed to be composed of
orthogonal factors, and because there are those who have posited theories based
on the observed patterns of factor correlations (e.g., DeYoung, 2006, 2013;
Digman, 1997). However, by reverting to EFA-derived or ESEM-derived factors
it becomes difficult, impossible in fact, to adjudicate quantitatively between
orthogonal and oblique rotations.
Therefore, investigators should be aware of the implications of choosing a
rotation if employing ESEM, just as was necessary in EFA. In general,
researchers are faced with three options, orthogonal, oblique, and target
rotations. With few exceptions (e.g., exploratory bifactor rotation; Jennrich &
Bentler, 2011), the orthogonal and oblique rotations that are generally available
are going to be calibrated, in various ways, to maximize simple structure. Given
this fact, an orthogonal rotation can be imposed, making the assumption that
factors are uncorrelated, but there is no quantitative justification to prefer this
solution to one that allows for factor correlations. More desirable, Fabrigar and
colleagues (1999) argue, is to employ oblique rotations given that if simple
structure is maximized by a solution with uncorrelated factors, this will emerge
in an oblique rotation. Alternatively, researchers may employ target rotations.
Mentioned only briefly above, this allows researchers to rotate ESEM (or EFA)
factors to a specific pattern, determined either by prior research or by a
theoretical structure. The statistical package will then try to “hit” the target
values, and will return a model with values as close as possible to those set as
the targets. This can be useful when a specific scale is intended as the primary
marker for a domain (i.e., a high target loading is specified) and other scales are
intended to be orthogonal (i.e., they would be targeted for a 0.0 loading), and
those presumed to have complex structure can be left free to load. Note that
regardless of rotation strategy, ESEM (or maximum likelihood EFA) models
will all evidence equivalent fit, and therefore the onus remains on the researcher
to select and defend a particular solution.
Our understanding of the structure of personality attributes has been deeply
influenced by factor analysis in its various forms. Early work relied on EFA,
which as a novel method was poorly understood and resulted in solutions that
fared poorly on replication. As EFA methods improved, so too did the
robustness of personality structure, ultimately arriving at the FFM by the end of
the 1980s to early 1990s. Presuming a reasonably consensual final model, a great
deal of scientific effort shifted toward validation and examining the implications
of the model. At the same time, the FFM was subjected to CFA that seemed to
threaten its foundations. However, initial concerns based on rigid applications of
CFA gave way to a more nuanced understanding of the issues involved (e.g.,
complexity of items and scales) and a recognition that the goals of developing
and testing a broadband personality structure may be at odds with the highly
sensitive nature of CFA. The recent addition of ESEM provides the necessary
compromise to broker reconciliation between the practicalities of working with
personality data and the desire to use several of the sophisticated features and
control of a confirmatory analytic framework. In many respects it would seem
that much of what factor analysis can tell us about personality has been
exhausted. Nothing could be further from the truth. On the contrary, highly
sophisticated factor analytic investigations continue to be used to address
complex questions at the cutting edge of personality science. In the next section
several of these areas with intensely active literatures will be considered.
Future Challenges for Factor Analysis in Personality Research

and the FFM
By the 1990s the FFM had spurred considerable work on personality traits as
predictors, outcomes, developmental constructs, and more. But the story does
not end there, and important structural considerations remain even today that go
beyond the FFM per se, and include the FFM in the context of a hierarchy of
traits and the utility of enlisting the FFM to serve as a broader fundamental
psychological structure.
Hierarchical Structure of Personality Traits

Although it should be clear at this point that the FFM represents a robust
solution for personality structure, it mostly reflects only one level of abstraction
in a larger structural organization of personality that goes both up, toward more
general or “meta-traits,” and down, to ever more fine-grained articulations of
personality attributes. Since their inception, factor analytic studies of personality
have assumed a hierarchical model, with the assumption being that individual
behaviors reflect the expression of attributes captured by specific descriptors,
specific descriptors combine to form narrow traits, which ultimately coalesce
into the broad domains [see, e.g., Guilford (1975) or Costa & McCrae (1995) for
a detailed discussion]. For a good example of how this has been operationalized
in a contemporary measure, consider the NEO PI-R (Costa & McCrae, 1992):
each domain is made up of six facets, which are each made up of eight items.
Thus, it could be argued that the NEO PI-R has (at least) three levels of
measurement by design.
Going up before going down, it is worth returning to the issue of the factor
intercorrelations observed among scales measuring the FFM domains. It was
noted earlier that the theoretically orthogonal domains manifested nonignorable,
and at times quite sizeable, correlations in real world data. The initial
quantitative work on this issue was conducted by Digman (1997) who examined
the higher-order structure of the five factors from 14 different studies, finding a
replicable structure that had one factor marked by neuroticism (or rather its
inverse emotional stability), agreeableness, and conscientiousness, and the other
marked by extraversion and openness. Importantly, these structures were tested
with CFA models that were generally well-fitting. Digman labeled these factors
using neutral terms, alpha and beta, respectively, although he hypothesized links
to several grand psychological theories. Since then a number of researchers have
explored these two higher-order domains in diverse datasets (e.g., Anusic,
Schimmack, Pinkus, & Lockwood, 2009). And although highly replicable, in
specific datasets they may not emerge or may evidence significant unexpected
cross-loadings. These two domains have also motivated theoretical innovation;
DeYoung (2006, 2013) has suggested the term “stability” for alpha and has
hypothesized that it reflects central serotonergic functioning, whereas he
suggested the term “plasticity” for beta, hypothesizing that it reflects
dopaminergic functioning (see also the chapter by Allen and DeYoung).
At this point, these two domains appear to be well established from a factor
analytic standpoint, and most relevant research moving forward will likely entail
further investigation of the external correlates and predictive power of these
domains (e.g., it may be that enhanced longitudinal prognostic capacity is
afforded by the broader domains relative to the five factors; are the domains
indeed reflective of functioning in the neurotransmitter systems?). However, one
aspect of this work that deserves greater factor analytic attention is the role of
the attenuation in correlations observed when cross-loadings are modeled
directly at the lower level of measurement. Significant factor correlations remain
when modeling an oblique structure with cross-loadings, but they decrease in
magnitude significantly (Booth & Hughes, 2014; Marsh et al., 2010). Testing
whether, and if so to what degree, the two higher-order factors survive in a fully
hierarchical model using second-order CFAs or ESEMs would be an important
confirmation of the robustness of these domains.
The fact that even alpha/stability and beta/plasticity correlate have led some to
hypothesize a “general factor of personality” (GFP), analogizing to the “g” of
cognitive abilities (Rushton & Irwing, 2008). This proposition has resulted in an
incredible amount of research within the past several years, often with
acrimonious commentary and critiques. Although it is difficult to guess the
motivation behind a given researcher’s decision to conduct a study and/or write a
paper with any certainty, it is hard to ignore the fact that Rushton’s writing drew
links from the GFP to distasteful explanatory theories of racial inequity in ways
that may have motivated criticism (Rushton, Bons, & Hur, 2008). Regardless of
the rationale, most of the published critiques have taken supporters of the GFP to
task on factor-analytic grounds. For instance, from a very basic perspective,
Hopwood, Wright, and Donnellan (2011), using a variety of factor analytic
techniques, showed that the GFP from different measures could hardly be
considered isomorphic, if even distally related. Pettersson and Turkheimer
(2010) demonstrated that the GFP, if representing anything substantive, likely
reflected evaluative bias. Finally, Revelle and Wilt (2013) showed that to the
extent that the GFP exists, it is of modest statistical prominence, and likely does
not warrant considerable attention. Others (e.g., Donnellan, Hopwood, &
Wright, 2012), noting problems in the CFA solutions reported (e.g.,
mismatching degrees of freedom between text and diagrams) in some research
supporting the GFP, have also shown that solutions may not replicate across
datasets. Ultimately, given the strength of the criticism, the GFP has a difficult
road ahead should the proponents seek to establish its permanence in the
hierarchy of personality traits.
Starting with five variables not much room remains to build upward, but there
is probably unlimited space to build downward. Undoubtedly the best-known
lower-order set of scales is the NEO PI-R’s 30 facets, which were designed to
meet several conceptual and statistical criteria. These included suitable factor-
analytic features, but also appropriate breadth of measurement, and sufficient
prior attention in the research literature. Other promising solutions include the
Big-Five Aspects of DeYoung, Quilty, and Peterson (2007), which reflect 10
intermediary constructs, two per domain of the FFM, that sit between the five
factors and the facets. The challenge with building downward is identifying sets
of scales that “hit” at the same level of abstraction (Guilford, 1975).
Ultimately it may be possible, given sufficient data and samples, to articulate
a broader hierarchy of personality and several levels of abstraction, with the
level of abstraction that is most conceptually or predictively useful depending on
the purpose for which it is needed. Even so, it is clear that the FFM will retain a
privileged position within any hierarchy. In the service of this aim, a technique
for establishing personality (and other) hierarchies that is based on factor
analysis, although it is not strictly factor analytic, bears mention. In what
Goldberg (2006) has affectionately yet irreverently termed the “Bass-Ackwards”
method, factor solutions of increasing complexity are estimated (e.g., one-factor,
two-factor, three-factor), the factor scores are saved, and then are correlated to
estimate the “unfolding” of a trait hierarchy. This simple but powerful technique
has been put to good use examining the hierarchical structure of various
measures, and has shown what Goldberg (1981) argued for conceptually, that
most of the enumerated trait models (i.e., the Big-Three, Big-Four, Big-Five) can
be handled by considering them as instantiations of varying levels of abstraction.
For instance, Markon Krueger, and Watson (2005), examining the joint structure
of adaptive and maladaptive personality measures, noted that at the apex sat
alpha and beta, at a level down alpha split into neuroticism and disinhibition,
next disinhibition split into agreeableness and conscientiousness (or more
accurately their inverses), and finally, at the fifth level, beta split into
extraversion and openness.
Convergence between Normal and Abnormal Personality Structure

Around the same time that personality science was beginning to gain traction
with a consensual structural model for personality attributes in the FFM, the
American Psychiatric Association (1980) made a major revision to its diagnostic
manual and system (i.e., the Diagnostic and Statistical Manual of Mental
Disorders, or DSM), and codified personality disorders (PDs) into 11 discrete,
categorical diagnoses (this has since been reduced to 10). Almost immediately,
researchers and clinicians began complaining about its lack of clinical utility and
mismatch to the phenomena it was putatively describing (Widiger & Kelso,
1983). Many stringent critiques of this system have been published over the
years (see e.g., Krueger & Eaton, 2010; Widiger & Clark, 2000; Widiger &
Trull, 2007), and therefore a detailed description of the system’s many
limitations does not bear enumeration here (see, as well, the chapter by Widiger,
Gore, Crego, Rojas, & Oltmanns). Suffice it to say, however, that the major
critiques are structural—there does not to appear to be 10 cleanly separable
disorders, nor has any credible and replicated research supported the categorical
distinction of disordered and not disordered. This is evidenced by the fact that
the personality disorder (PD) diagnoses covary within individuals at a much
higher degree than would be expected by chance, and indeed the most
commonly used diagnosis in practice has been one of “mixed” or “PD Not
Otherwise Specified” (i.e., PD is present, but does not match any of the
categories and instead reflects a blend of features; Verheul & Widiger, 2004).
Without belaboring the point, a poorly articulated structural model impedes
accurate clinical diagnosis, communication, and treatment, while also frustrating
research efforts attempting to identify etiological and maintenance mechanisms,
and develop efficient treatments.
Although the epistemological differences between psychiatry and personality
science are traditionally large, it is difficult to argue that a system developed to
map the structure and major units of personality pathology should bear no
resemblance to the empirical structure derived for basic personality. As far back
as the 1950s, Leary (1957) argued convincingly that the same system needed to
be used in the description of both to achieve harmonized scientific and clinical
endeavors. As the FFM gained prominence throughout the 1980s, a natural next
question was whether, and if so how, the five domains of basic personality
interfaced with the 10 categorical diagnoses of the DSM PDs (Widiger &
Frances, 1985). Considerable work has examined their interface, enough to
support two relatively recent meta-analyses (Samuel & Widiger, 2008a;
Saulsman & Page, 2004). However, consistent with the theme of this chapter, the
focus in this chapter will be on the use of factor analysis for answering this
question.
Wiggins and Pincus (1989) were the first to use EFA to study how the two
systems interdigitate.6 They analyzed student analogue responses to the NEO PI
domains, the Revised Interpersonal Adjective Scales—Big Five (IASR-B5;
Trapnell & Wiggins, 1990), the Minnesota Multiphasic Personality Inventory
(MMPI) PD scales of Morey, Waugh, and Blashfield (1985), and the Personality
Adjective Check List (PACL) PD scales (Strack, 1987). The pattern of loadings
on the resulting five factor structure was easily interpretable and conceptually
clear. The personality dimensions of NEO Extraversion and IASR-B5
Dominance, and the PD dimensions of MMPI Histrionic, Narcissistic, and
PACL Antisocial marked the first factor positively and MMPI and PACL
Schizoid and Avoidant scales marked the factor negatively, reflecting what
appeared to be a bipolar dimension ranging from intrusiveness/attention-seeking
to detachment/withdrawal. The second factor was most strongly marked by NEO
and IASR-B5 Neuroticism, and was additionally positively marked by MMPI
Borderline, Dependent, and Avoidant, and PACL Passive-Aggressive and
Avoidant, with smaller negative loadings from MMPI and PACL Narcissistic
and PACL Antisocial scales, thereby reflecting a modestly bipolar dimension
ranging from negative affectivity and distress to a presumed problematic lack of
concern for such. Factor three represented a dimension that ranged from
overaffiliativeness through antagonism, with positive loadings from NEO
Agreeableness, IASR-B5 Affiliation, and PACL Dependent, and negative
loadings from PACL Narcissistic and Antisocial and MMPI Paranoid. The
fourth factor reflected a dimension ranging from compulsivity to
impulsivity/indolence, with strong positive loadings from NEO PI and IASR-B5
Conscientiousness and PACL and MMPI Compulsive, and negative loadings
from MMPI Antisocial and Passive-Aggressive. Finally, NEO PI and IASR-B5
Openness and MMPI Schizotypal marked the fifth factor.
In many ways this study, although not without limitations, reflected the
seminal empirical demonstration that the FFM and pathological personality
constructs could be organized within the same structural model. Of note is that
almost all of the PD scales had strong loadings on one of the factors, and these
were easily interpretable. Also noteworthy is the fact that some scales (e.g.,
Narcissistic, Antisocial, and Dependent) differed in their loading across
instruments. For instance, MMPI Dependent was a strong marker of negative
affectivity, whereas the PACL scale of the same name marked the overly
nurturant versus antagonism domain. This type of variability in the content of
scales with the same name causes problems when used as predictors or criteria in
individual studies (e.g., Samuel & Widiger, 2008b), but can be resolved with
factor analytic studies, assuming that sufficient markers of the diverse domains
are being sampled. Finally, the emergence of the fifth factor that combined
Openness and Schizotypy presaged what has ultimately become a contentious
topic in this area of research, which will be addressed again later in this chapter.
Wiggins and Pincus’ (1989) joint EFA of PD scales and FFM scales raises the
question of what the stand-alone structure of PD would be, unbuttressed by FFM
scales. Although many individual studies have examined this over the years (see,
e.g., Wright & Zimmermann, 2015, for a review), O’Connor (2005) used EFA
techniques with target rotations to ascertain the consensus structure of the 10
DSM PD constructs pooling across many studies (see also the chapter by
O’Connor). He concluded that a four-factor structure was the best fit, with clear
resemblance to four of the FFM domains: neuroticism (e.g., borderline, avoidant,
dependent), antagonism antagonism (e.g., histrionic, narcissistic, antisocial),
extraversion/introversion (e.g., schizoid, avoidant, histrionic), and
conscientiousness (obsessive-compulsive). Thus, even without the anchors of
FFM the conceptual overlap was clear. Missing, of course, was a domain
reminiscent of the fifth factor, openness. Results such as these led some
relatively early to suggest that a four-factor, but not five factor model would be a
reasonable compromise between the FFM and the empirical structure of PD
(Widiger, 1998).
Despite this seemingly strong evidence for a model of PD that resembles the
FFM, at least in large part, there has been rigid resistance by many to replacing
the DSM’s discrete categorical model. One of the traditionally cited reasons was
that there was a lack of agreement on the structure of personality. However, it is
difficult to consider this as anything more than a contrived criticism at this point,
and each of the major alternative models is more alike than dissimilar—
especially at the primary domain level of analysis (e.g., Markon et al., 2005;
Widiger & Simonsen, 2005). Widiger and Simonsen (2005), in much the same
way Goldberg (1981) did with normal personality, reviewed 18 different models
of normal and maladaptive personality and concluded that although there are
differences in the precise make-up of the lower-order scales, all models either
contain or can be conceptually accommodated by four broad domains:
extraversion versus introversion/detachment, agreeableness versus antagonism,
emotional stability versus neuroticism/emotional dysregulation, and
constraint/conscientiousness versus disinhibition (they also identified a fifth
domain, unconventionality versus closedness to experience, but excluded it from
the primary proposal because it was not included within some prominent
models). The conceptual similarities articulated by Widiger and Simonsen
(2005) have been born out in numerous empirical factor analytic studies that
have examined these models alone (e.g., Calabrese, Rudick, Simms, & Clark,
2012; Kushner, Quilty, Tackett, & Bagby, 2011) or in combination with other
measures (e.g., Clark, Livesley, Schroeder, & Irish, 1996; Markon et al., 2005).
Many researchers have developed questionnaires/inventories that were
intended to conform (more or less) to the DSM’s structure [e.g., Personality
Diagnostic Questionnaire-4 (PDQ-4), Hyler, 1994, and Millon Clinical
Multiaxial Inventory (MCMI-III), Millon, Davis, & Millon, 1997]. However,
others have adopted a different approach, pivoting away from the DSM
constructs and instead developing measures that reflect the putative constituent
or transdiagnostic features of maladaptive/pathological personality functioning.
Examples of this approach include Clark’s Schedule for Non-Adaptive and
Adaptive Personality (SNAP; Clark, 1993), which was developed around a
theoretically triarchical temperament structure, and Livesley’s Dimensional
Assessment of Personality Pathology (DAPP; Livesley & Jackson, 2009;
Livesley, Jang, & Vernon, 1998), which was developed to match the four-factor
structure reviewed above (Livesley, Jackson, & Schroeder, 1989). Factor
analysis of each of these measures supports their intended structure, or, in the
case of the SNAP, an alternative four-factor structure is defensible (e.g.,
Calabrese et al., 2012; Pettersson et al., 2014). It is important to understand,
however, that both the SNAP and the DAPP each includes only one thought-
disordered scale out of a total of 15 and 18 scales, respectively. This makes it
very unlikely that a separate domain will emerge that captures
psychoticism/oddity/peculiarity in an EFA of these measures. Ongoing efforts to
develop measures that more explicitly are matched to the FFM are underway or
close to their conclusion. In one such effort, Widiger and his colleagues
(Widiger, Lynam, Miller, & Oltmanns, 2012) have developed individual
measures for each of the DSM PD constructs, but using FFM-based facet scales.
However, this effort has thus far emphasized a conceptual mapping of items to
lower-order constructs, and these have yet to be subjected to factor analytic work
across the different measures (see also the chapter by Widiger and colleagues).
In another effort, Simms and colleagues (2011) have been developing a
computerized adaptive test of PD (CAT-PD) that aimed to provide
comprehensive coverage of PD features, while also ensuring adequate coverage
of a pathological FFM model (see also the chapter by Simms, Williams, and
Simms). Although the measure is not yet finalized, early EFAs suggest that the
CAT-PD scales conform to expected structure when analyzed in conjunction
with normal range and pathological traits (Wright & Simms, 2014).
Leading up to the recent revision of the DSM (i.e., DSM-5) it appeared as
though there might be a shift from the categorical model to a model based on
dimensional features of PD. Indeed, this is what the DSM-5 Personality and
Personality Disorder Work Group ultimately recommended. The model they
suggested was based, in part, on five broad domains of individual differences in
personality pathology, based on 25 lower-order facets. No doubt the five broad
domains might have been expected a priori, however, these were established
based on a bottom-up process, guided by EFA.7 In brief, the Work Group
members enumerated the features they deemed necessary for the comprehensive
mapping of the PD phenotype, which resulted, after deliberations, in 37 primary
features. These were then instantiated in self-report scales and administered to
two large samples, one matched to population demographics and one reporting
previous mental health treatment, respectively, and then subjected to a variety of
factor-analytic techniques (Krueger, Derringer, Markon, Watson, & Skodol,
2012). As a result of these analyses, the initial 37 scales were reduced to 25
scales that loaded on five factors labeled negative affectivity, detachment,
antagonism, disinhibition, and psychoticism. The 25 scales and five domains
were furthered as part of the DSM-5 PD model, and also resulted in a final
instrument, the Personality Inventory for the DSM-5 (PID-5; Krueger et al.,
2012). Since then, support for the structure has been rapidly accruing via
replication in independent samples (e.g., De Fruyt et al., 2013; Wright et al.,
2012) and via independent raters (e.g., Markon, Quilty, Bagby, & Krueger,
2013; Morey, Krueger, & Skodol, 2013).
Furthermore, the five domains of the DSM-5 model would appear to offer
clear conceptual matches to the FFM. Several studies now show that when the
PID-5 domains are subjected to conjoint analyses with normative range scales
the joint structure emerges as might be predicted. De Fruyt et al. (2013) and
Thomas et al. (2013) both used EFA to show that the patterns of loadings
aligned in factors that reflected a coherent joint structure. Furthermore, Gore and
Widiger (2013) and Wright and Simms (2014) used ESEM to conduct
exploratory analyses while also accounting for method variance associated with
each of several measures. Thus, although the DSM-5 model was not developed
to match the FFM structure a priori, what has ultimately resulted is a model that
provides a clear bridge for including the FFM in the diagnostic nosology. A
further potential of this outcome is discussed below. However, although these
results are encouraging, the proposed DSM-5 model for PD was not adopted, and
its future remains uncertain at this juncture. One possibility is that research of
the type that is emerging now can be used to revise and improve the model for
inclusion in a subsequent revision.
In spite of what appear to be clear convergences related to normative and
maladaptive structures, several issues remain. Krueger and colleagues (2011)
recently summarized these as involving structure, bipolarity, hierarchy, and
range. Although there are some distinctions to be made among these four issues,
they are all variations on structural considerations and are amenable to study by
factor analytic techniques. The first two of these will be discussed.8
Structure, in the context that Krueger and colleagues (2011) meant it, refers to
the mapping of specific content onto the primary domains (i.e., likely the FFM),
and by extension the best interpretation of each domain. The reason this remains
somewhat of a challenge is that a comprehensive mapping of clinically relevant
personality domains necessarily invokes new content because basic trait models
generally do not provide adequate coverage of specific areas of impairment
(Trull, 2005). Furthermore, the relationship between pathological facets and
normal traits can be complex. For example, when items of normal range trait
measures are modified to reflect maladaptive functioning, the pattern of
covariation among domains is altered (Haigler & Widiger, 2001). It appears that
as extremity or maladaptivity is increased, content may have a tendency to be
altered as well. This is particularly evident in the specific composition of scales
related to disinhibition, constraint, and antagonism (Krueger et al., 2011). On the
one hand, disinhibition and constraint are theoretically opposite maladaptive
poles of the same dimension (i.e., conscientiousness; e.g., Samuel, 2011;
Widiger, Livesley, & Clark, 2009; Widiger & Mullins-Sweatt, 2009). However,
in different structural analyses of traits, disinhibition and constraint sometimes
emerge as opposing poles (e.g., Markon et al., 2005; Watson, Clark, &
Chmielewski, 2008) and sometimes as separate domains (e.g., De Clercq, De
Fruyt, Van Leeuwen, & Mervielde, 2006; Morey, Krueger, & Skodol, 2013). In
addition, when these domains do separate, disinhibition scales often join
antagonism scales to form a dimension that more closely resembles the
externalizing spectrum (e.g., Krueger, Markon, Patrick, Benning, & Kramer,
2007; Morey et al., 2013). Undoubtedly, measurement issues (i.e., the content of
the specific scales; e.g., Samuel & Widiger, 2010) are involved in addition to
substantive structural questions, but further research to clarify the joint structure
of normal and abnormal traits is likely warranted.
As previously alluded, an ongoing structural issue involves the fifth
personality domain. In normal range trait models there is broad support for the
domain of openness to experience/intellect (Goldberg, 1993), whereas in
maladaptive models, a dimension related to oddity, peculiarity, aberrant
thinking, or psychoticism has been suggested to capture content related to
schizotypy (Harkness & McNulty, 1994; Harkness, Finn, McNulty, & Shields,
2012; Tackett, Silberschmidt, Krueger, & Sponheim, 2008; Watson et al., 2008).
Evidence is somewhat mixed on whether these can be conceptually and
empirically integrated (e.g., Piedmont, Sherman, Sherman, Dy-Liacco, &
Williams, 2009; Watson et al., 2008). Several EFA studies find clear support for
the convergence of openness and schizotypy (e.g., De Fruyt et al., 2013; Gore &
Widiger, 2013; Markon et al., 2005; Thomas et al., 2013; Wiggins & Pincus,
1989) whereas in others the picture is murkier (e.g., Watson et al., 2008; Wright
& Simms, 2014). Returning to basic factor analytic principles may help clarify
some of the discrepant results. For instance, Watson et al. (2008) reported on
three different studies, each of which they interpreted as suggesting that
schizotypy/oddity reflected a domain outside of the FFM and did not correspond
to openness. However, although they were thorough in many respects (e.g.,
several samples and studies, different measures across studies, extraction and
comparison of several solutions), in other respects their analytic approach raises
questions about whether their conclusions can really be considered final. Most
importantly, in each of their analyses it could be argued that they oversaturated
their models with scales related to schizotypy/oddity, which may have served to
virtually guarantee that a separate factor would emerged for schizotypy/oddity.
Finally, in studies 2 and 3, scales related to schizotypy/oddity were the only
pathological scales in the analyses, raising the question of method artifact.
ESEM could fruitfully be applied here to clarify these results.
Turning to the issue of bipolarity, certain aspects of pathological personality
functioning are theorized to be extreme poles of the same dimension (Samuel,
2011; Widiger, Livesley, & Clark, 2009; Widiger & Mullins-Sweatt, 2009).
Although there is evidence to suggest that many domains operate in this way
(e.g., extraversion/detachment; Markon et al., 2005; Watson et al., 2008), some
domains, disinhibition/constraint in particular, are more variable across studies.
This can be observed in certain external correlates, in which both poles manifest
positive correlations (e.g., with obsessive–compulsive disorder, Kotov, Gamez,
Schmidt, & Watson, 2010; Wu, Clark, & Watson, 2006). Other domains, such as
antagonism/agreeableness, have specific content that is hypothesized to fall at
one end of the dimension, but instead shift domains (e.g., dependency and
attachment anxiety frequently shift to negative effectivity; Markon et al., 2005;
Wiggins & Pincus, 1989). The question of unipolarity versus bipolarity remains
an understudied issue in large part because most personality trait inventories
measure, or are keyed in the direction of, a single pole of the primary trait
domains. For instance, it is common for normal range inventories to provide
scales that tap agreeableness but not antagonism. Yet normal range and
pathological trait inventories tend to be complementary in this regard, and
therefore when studied together more of the poles receive measurement
coverage.
Relatedly, recent work by Pettersson, Turkheimer, and colleagues (2010,
2012, 2014) has revived the notion of “evaluation” (Edwards, 1957; Peabody,
1967) in personality questionnaires and their work suggests that this may be a
compelling resolution to some of the thorny issues considered here. Briefly,
evaluation is defined as the tendency to respond in the affirmative to positive
attributes, irrespective of content, and to indiscriminantly deny negative
attributes (or vice versa were the scale to be keyed toward negative evaluation).
In the study most relevant to this discussion, Pettersson and colleagues (2014)
had participants code items reflecting the DSM PD features and their opposites
for evaluativeness, transformed these into factor loadings, then compared them
with the loadings of the items on the first principle component extracted from
the same items collected from participants. The congruence coefficient between
the two was 1.0, demonstrating perfect agreement between the ratings of
evaluation and the loadings on the first principle component. The authors next
coded the SNAP items for degree of evaluation, transformed these ratings to
factor loadings for a target rotation, and rotated an EFA of the SNAP items such
that there was an evaluation factor and three unrelated content factors. The
resultant content factors demonstrated increased bipolarity (i.e., pathological
content at both poles of the dimension) relative to a solution that does not
account for item evaluativeness. Although thus far only applied to a limited set
of scales and measures, statistically isolating evaluation in order to focus on the
structure of pure content is compelling and may resolve some longstanding
concerns that are otherwise difficult to resolve factor analytically.
At the same time, isolating evaluation is not a simple solution and faces
several challenges, as any method does. For one, as with all EFAs combined
with rotation methods, the resultant factors must be interpreted, and although
some investigators may view the loadings on a first orthogonal factor as
evaluation, others may view it as reflecting personality pathology’s core
impairments (e.g., Hopwood, Malone, et al., 2011; Sharp et al., 2015). The issue
here is that the first orthogonal factor often contains impairments of seemingly
opposing content (e.g., socially inhibited and needs admiration). Impairments in
functioning across domains accord well with many theories of personality
pathology, but the evaluative perspective raises reasonable questions on
commonsense grounds. Recall that the interpretation of a dimension such as this
cannot be adjudicated within the factor analytic framework, and therefore
additional data must be gathered. Of particular importance may be temporally
sequenced data. This is because endorsing impairments of opposite content may
be justified if individuals do in fact demonstrate both, although presumably not
both at the same time (e.g., Wright, 2014). A modest amount of data exists that
suggests that personality pathology, at least of some types, predicts shifts in
content of behavior over time (Wright, 2014; Wright, Scott, Stepp, Hallquist, &
Pilkonis, 2015; Wright, Hallquist, Beeney, & Pilkonis, 2013), although more
work in this area is needed. Additionally, as Pettersson et al. (2014) note, their
analyses have been limited to self-report scales, and it is possible that evaluation
would be attenuated to a large degree with interviewer or other-report data.
Hopefully more work involving repeated sampling, different raters, and the
prediction of external criteria will emerge over the coming years.
The Next Challenge: A Metastructural Model of Personality and

Psychopathology
Investigating the overlap between the normative trait domains of the FFM and
DSM personality disorder features, with a view toward achieving structural
integration, is a natural goal. However, it may be too narrow. This is because the
relationship between the FFM traits and the personality disorders is not
privileged, and in fact the FFM traits show robust relationships with most mental
disorders (Andersen & Bienvenu, 2011; Kotov et al., 2010). It may surprise
readers to know that the strengths of association between personality traits and
clinical syndromes frequently surpass those between traits and personality
disorders. This may be due to the more clearly defined and less heterogeneous
nature of clinical syndromes relative to PDs, the low reliability of PD
assessments, or some combination of both. Regardless of the relative strength of
associations, the DSM’s clinical syndromes also appear to be moderately to
strongly related to personality traits from the FFM. Recognition of this fact
immediately raises the question of whether the structure of personality and
psychopathology writ large could be incorporated within a coherent “meta-
structural” model (see also the chapter by Bagby).
In recent years there has been a steadily growing literature on the quantitative
modeling of the structure of mental disorders. Based on the observation that
psychiatric comorbidity (i.e., diagnostic covariation) is extensive in the general
population, far outpacing chance cooccurrence (Kessler et al., 1994, 2005), and
polydiagnosis is the rule rather than the exception (Zimmerman & Mattia, 1999),
there has been an increased interest in identifying the fundamental domains of
psychopathology, in much the same way personality theorists sought to delineate
the fundamental units of personality (Krueger & Markon, 2006). Mirroring the
structural work in personality, efforts to quantitatively derive an empirical
structure of psychopathology have relied on factor analysis and related
techniques. Although this approach has been profitably applied to both child
(Achenbach, 1966; Lahey et al., 2008) and adult (Kotov et al., 2011; Krueger,
1999; Krueger & Markon, 2006) disorders, the current review is confined to
structural models of adult psychopathology.
Early investigations focused primarily on “the common mental disorders,”
which include syndromes of high population prevalence that are readily
ascertained in epidemiological samples, such as the unipolar mood disorders,
anxiety disorders, antisocial behavior, and substance abuse. Factor analyses,
usually CFA, applied to diagnoses and symptoms of the common mental
disorders resulted in what is now a well-replicated two-factor structure of
internalizing (e.g., unipolar mood disorders, anxiety disorders) and externalizing
(e.g., substance use, antisocial behavior) spectra that are robust across age, sex,
ethnicity, culture, informant type, and DSM axes (Eaton et al., 2012; Eaton,
Krueger, & Oltmanns, 2011; Forbush & Watson, 2013; Kramer, Krueger, &
Hicks, 2008; Krueger, Capsi, Moffitt, & Silva, 1998; Krueger, Chentsova-
Dutton, Markon, Goldberg, & Ormel, 2003; Lahey et al., 2008; Slade & Watson,
2006). As was discussed in the context of the factor analytic studies of
personality attributes, the particular admixture of variables analyzed constrains
the resulting structure, and early studies were understandably conservative in
their focus on common mental disorders. This approach was necessitated by the
nature of many of the early samples, which were often epidemiological, and
were coded for diagnostic categories as opposed to more fine-grained symptoms,
limiting the amount of variability for some of the rarer disorders (e.g., psychosis,
mania).
Studies have since sought out clinical samples or examined individual
symptoms in epidemiological samples to investigate an expanded structure.
There is now accumulating evidence that a thought disorder/psychosis (e.g.,
psychotic disorders, schizotypal personality disorder) spectrum is reasonably
robust across samples and criterion sets (Kotov et al., 2010; Markon, 2010; Wolf
et al., 1988; Wright, Krueger, et al., 2013). To provide more detail in the various
ways in which factor analytic techniques are being employed in this domain, one
recent study is highlighted. Wright, Krueger, and colleagues (2013) used EFA on
symptom level data from 8,841 individuals included in the 2007 Australian
National Survey of Mental Health and Well-Being, resulting in a six-factor
model, with dimensions reflecting distress, fear, obsessive–compulsive features,
alcohol abuse, drug abuse, and psychosis. Then, using CFA, models reflecting a
three-factor structure (internalizing, externalizing, and psychosis) were
compared to two different two-factor models that allowed the symptoms of
psychosis to load on internalizing or externalizing domains, respectively. The
three-factor model provided markedly better fit, supporting the three distinct but
correlated domains of internalizing, externalizing, and psychosis.
However, it is clear that spectra of internalizing, externalizing, and psychosis
cannot comprehensively account for the diverse array of pathologies of behavior
and mental functioning observed in the population. As such, several ambitious
studies have sought to expand upon this basic triarchic structure by incorporating
additional diagnoses, most notably the PDs, and have begun to uncover
additional spectra. To date only four published studies have explored the
structure of psychopathology using a broad suite of clinical syndromes and
personality disorders (Blanco et al., 2013; Kotov et al., 2011; Markon, 2010;
Røysamb et al., 2011). Although each resultant model is necessarily unique
given differences in the precise admixture of disorders (e.g., some do not include
indicators of psychosis), sampling strategy (e.g., clinical versus
epidemiological), and other features (e.g., disorder-level versus symptom-level
analyses), two additional domains appear reasonably replicable across studies.
First, Markon (2010) and Røysamb and colleagues (2011) each identified a new
spectrum they respectively termed pathological or anhedonic introversion in
EFA studies. In both cases, avoidant and dependent PDs were strong markers of
the factor, although Røysamb et al. (2011) also found that schizoid and
depressive PDs loaded strongly on the factor, which accounts for the slight
difference in conceptualization. Blanco and colleagues (2013) also found
evidence for a factor for which the strongest loadings came from avoidant and
dependent PDs and social phobia.
Second, in three studies (Blanco et al., 2013; Kotov et al., 2011; Røysamb et
al., 2011), which varied in their use of exploratory and confirmatory techniques,
a domain related to antagonism, as labeled by Kotov and colleagues, has
emerged. Again, slight differences have arisen in the makeup of this domain
across studies, although narcissistic and histrionic PDs consistently exhibit the
strongest loadings. Additional markers for this domain include obsessive–
compulsive, borderline, paranoid, and (to a lesser extent) antisocial PDs. What
these disorders share to varying degrees is an antagonistic interpersonal style
that puts afflicted individuals at odds with others. Notably, these domains of
introversion and antagonism, which emerge with the addition of PDs, each deals
with maladaptive social/interpersonal functioning, consistent with the view that
the PDs, at their core, reflect the interpersonal disorders (Benjamin, 1996; Hill,
Pilkonis, Bear, 2010; Hopwood, Wright, Ansell, & Pincus, 2013; Meyer &
Pilkonis, 2005; Pincus, 2005; Wright et al., 2012). Therefore, based on this
initial suite of studies that have included PDs in structural models of
psychopathology, the domains of introversion and antagonism appear to be good
candidates to include alongside internalizing, externalizing, and thought disorder
as broad, replicable domains of psychopathology.
Taken together, these domains bear a remarkable conceptual resemblance to
the FFM or maladaptive variants thereof (e.g., the pathological personality trait
domains included in DSM-5 Section III system of PDs as reviewed earlier). The
five domains outlined in the structural models of psychopathology are easy to
conceptually map onto the FFM: internalizing—neuroticism; externalizing—
disinhibition (or impulsivity/low conscientiousness); psychosis/thought disorder
—openness to experience/unconventionality; antagonism/low agreeableness;
pathological introversion/low extraversion. However, although intuitively
compelling, direct empirical evidence for this structural coherence is only just
emerging. In a recent study, Wright and Simms (2015) showed that the DSM-5
PD traits, clinical syndrome symptoms, and traditional DSM PD criteria, when
factored using ESEM to account for measurement effects, supported the
hypothesized structure. Nevertheless, future studies are needed that replicate this
work and combine analyses of personality traits (adaptive and maladaptive) with
DSM-defined symptoms and disorders. Were this shared structure to be
replicated, it would go a long way to providing an empirically supported basis
for the conceptualization of psychopathology, and provide a much needed bridge
between normality and psychiatric dysfunction.
Network Modeling: A Conceptual Challenge?

Recently the factor analytic basis of not only the FFM, but the structure of
personality in general (including psychopathology), has been questioned
(Cramer et al., 2010, 2012). The authors of these critiques have argued that
factor analytic techniques are inappropriate for studying the structure of
personality (and psychopathology) because they make a fundamentally incorrect
assumption—namely, that there are unobserved (i.e., latent) variables that
simultaneously “cause” observed features or behaviors. Rather, Cramer and her
colleagues suggest that individual behaviors, features, attributes, symptoms, etc.
are directly causal, and are related to each other in complex “networks.”
Accordingly, they have proposed network models as alternatives to factor
analytic techniques to understand the structure of personality. To illustrate using
examples of theirs, they might argue that the items “like to go to parties” and
“like to be around people” are not both markers of extraversion; rather, people
“like to go to parties” because they “like to be around people.” Or, for example,
the diagnostic criteria “fatigue” and “difficulty sleeping” for depression do not
both arise because of an underlying depression, but rather because a person who
has difficulty sleeping will then become fatigued. Extrapolating beyond these
simple examples, the behaviors that constitute what are termed personality and
psychopathology are not reflective of underlying entities, rather they are
personality and psychopathology, and it is in their complex associations that
personality arises as an emergent property.
On its surface, the suggestion that the field should move beyond descriptive
models of the structure of personality to understanding the microcausal and
mutually influencing inputs to a complex system is highly appealing. However,
there are grave problems with the way the network modeling approach has been
put forward that should foreclose any potential enthusiasm at this juncture.
Indeed, for each of the articles cited above there has been sharp criticism leveled
at the network approach (see commentaries associated with Cramer et al., 2010,
2012). There are currently three major problems with the network approach.
First, and most relevant here, the authors make a straw man out of factor analysis
and latent variable techniques. As described earlier in the methodological
review, one way of thinking of factor analysis is as a causal framework, with the
underlying factor directly causing the observed variables (see Figure 11.1). Also,
a frequent, but unnecessary, assumption of factor models is one of conditional
independence. Yet, also noted, factors do not necessitate such strict causal
interpretations, and can be understood at a more practical and descriptive level
(i.e., the shared variance among a group of items). Cramer and colleagues
discuss factor models as if the only way to view them is in the strictest causal
frameworks, and that these are inviolable and inflexible assumptions or
perspectives. By doing so they make assumptions few personality theorists or
applied factor analysts would make, creating a problem where there is none.
Others have made this observation as well (Belzung et al., 2010).
Second, as currently implemented, network models offer little more than
graphic representations of covariance/correlation matrices (see also Asendorf,
2012; Molenaar, 2010). In other words, the diagram that is the network model,
although visually very captivating, is defined by and directly depicts the
associations among individual items. Arguably these relationships are easier to
appreciate when visually depicted, but can quickly become overwhelming, as is
the case when the 240 items of the NEO PI-R are graphed in a network. A
corollary of this fact is that network models offer no reduction in the complexity
of the data, and as such do not reflect models in the sense of a more
parsimonious and general description of a psychological phenomenon.
Third, the descriptions of causal associations that are hypothesized to make up
personality and psychopathology constructs from the network perspective
necessarily imply temporal relationships that play out over a multitude of
(currently unknown) time scales. Yet the major demonstrations of the network
perspective employ cross-sectional data that could not possibly test the type of
causal associations that are hypothesized. Thus, there is a fundamental
disconnect between what the perspective promises and what it currently offers.
Although the current state of network models detracts little from the latent
variable perspective (including factor analysis), and offers very little in
replacement, it may be that with continued advances this perspective could
contribute greatly to personality science. For it to be successful, it will require a
greater clarity of conceptualization, better match between theory and data, and
advances in methodology to accommodate true models of interindividual
differences in intraindividual causal processes.
Conclusions
The early history of the FFM, in many respects, followed the history of factor
analysis. As factor analysis advanced, so too did the field’s understanding of the
structure of personality. From the factor analytic perspective, the evidence for
the five domains of extraversion, agreeableness, conscientiousness, neuroticism,
and openness/intellect is quite strong. Although questions have been raised about
the structural validity of the FFM, especially as instantiated in specific
instruments, these can mostly be quieted by a careful consideration of whether
the methods match the purposes of the investigation. Currently, it is
scientifically more interesting and likely to lead to greater psychological insight,
to consider in what ways the FFM may serve as an organizing framework for
understanding diverse areas of human functioning that extend beyond basic
personality, than to agonize over whether the FFM meets precise cutoffs for
confirmatory model fit. This chapter emphasized the clear convergences between
the structure of the FFM and the structure of personality pathology, and possibly
even psychopathology more generally. A full and practically useful coordination
of these systems though will require considerably more work, much of it factor
analytic. The current review was necessarily selective, prioritizing breadth of
content as opposed to depth, which has unfortunately left out the work of many
brilliant and pioneering researchers. Several new techniques that also hold
promise for future structural work in personality were not able to be covered
(i.e., Bayesian CFA, multigroup EFA) without published examples in the
literature as of yet. In spite of this, hopefully this review will stimulate others to
take up these next waves of a long history of factor analytic investigations into
the structure of personality.
Notes
1. For more detailed reviews of the historical evidence for the FFM, the reader is
directed to Digman (1996) or Goldberg (1993). At this point in the history of the
FFM it would be impossible to cover all relevant studies and applications. As a result,
many important contributions had to be neglected. Therefore this review is
necessarily selective, and this is further compounded by my choice to focus on
specific studies that provide illustrative examples as opposed to more cursory reviews
of many studies. I offer my apologies to the authors of those many excellent studies
that were not included.
2. Alternatively, a more pragmatic interpretation can be made that a factor merely
represents the shared content from a set of indicator variables, such that when
indicators covary strongly a factor can serve to parsimoniously summarize their
associations. Thus the interpretation of a factor can range from the descriptive to the
causal.
3. Most modern statistical packages that offer CFA as a technique (e.g., LISREL,
Mplus) have a number of default settings that prespecify many parameters in any
given model (e.g., error covariances are fixed at 0.0). Whether the investigator relies
on these conveniences or manually codes each parameter does not change the fact
that a decision is being made on whether to fix or free each parameter.
4. To the knowledgeable reader several important names may be conspicuously absent
from this historical narrative—names such as Eysenck and Tellegen. This is because
they generally ascribed to a theoretical model of temperament/personality domains
that was divided into three domains: Neuroticism/Negative Emotionality,
Extraversion/Positive Emotionality, and Psychoticism/Constraint. Although
undoubtedly too simplistic of a characterization of these author’s important work,
they developed measures that were intended to fit this trinity structure, and the
measures mostly did (e.g., Tellegen & Waller, 2008). As will be discussed below,
these formulations are not necessarily at odds with the FFM, and it is reasonably
possibly to coordinate both models hierarchically.
5. Note that when using sum scores of FFM domains or facets in applied research, this
is, in effect, treating each item as if it has zero cross-loadings (and a 1.0 loading on
the target scale). Thus the factor correlations observed in practice often mirror those
observed in a CFA framework.
6. Technically they used principle components analysis, which is related to, but not the
same as factor analysis. In the service of brevity, however, I will treat this work and
others like it (e.g., O’Connor, 2005) as having conducted an EFA.
7. The rationale for many of the decisions made by the Work Group was political as
opposed to scientific, and therefore was frequently driven by expediency to satisfy
competing interests. A discussion of these issues goes beyond the current chapter,
although the interested reader is directed to retrospectives by Krueger (2013), Skodol
et al. (2013), and Widiger (2013).
8. I limit my review to structure and bipolarity here, having addressed hierarchy above,
and because range, although important, generally requires the application of item
response theory techniques, which are not currently applicable to fully dimensional
scales. Nevertheless, a number of recent studies have started to examine range in
personality scales, and I direct the reader to Walton and colleagues (2008), Samuel
and colleagues (2010), and Stepp and colleagues (2012).
References
Achenbach, T. M. (1966). The classification of children’s psychiatric symptoms: a factor-
analytic study. Psychological Monographs: general and applied, 80(7), 1–37.
Allport, G. W., & Odbert, H. S. (1936). Trait-names: A psycho-lexical study.
Psychological Monographs, 47(1), i–171.
Amelang, M., & Borkenau, P. (1982). Über die faktorielle Struktur und externe Validität
einiger Fragebogen-Skalen zur Erfassung von Dimensionen der Extraversion und
emotionalen Labilität. (The factional structure and external validity of some
questionnaire scales for measurement of the dimensions of extraversion and emotional
lability.) Zeitschrift für Differentielle und Diagnostische Psychologie, 3(2), 119–145.
disorders (3rd ed.). Washington, DC: Author.
Andersen, A. M., & Bienvenu, O. J. (2011). Personality and psychopathology.
International Review of Psychiatry, 23(3), 234–247.
Anusic, I., Schimmack, U., Pinkus, R. T., & Lockwood, P. (2009). The nature and
structure of correlations among Big Five ratings: The halo-alpha-beta model. Journal
Asendorf, J. B. (2012). What do the items and their associations refer to in a network
approach to personality. European Journal of Personality, 26(4), 432–433.
Asparouhov, T. & Muthén, B. (2009). Exploratory structural equation modeling.
Structural Equation Modeling, 16, 397–438.
Belzung, C., Billette de Villemeur, E., Lemoine, M., & Camus, V. (2010). Latent
variables and the network perspective. Behavioral and Brain Sciences, 33(2–3), 150–
151.
Benjamin, L. S. (1996). Interpersonal diagnosis and treatment of personality disorders.
Blanco, C., Krueger, R. F., Hasin, D. S., Liu, S. M., Wang, S., Kerridge, B. T., … Olfson,
M. (2013). Mapping common psychiatric disorders: Structure and predictive validity in
the National Epidemiologic Survey on Alcohol and Related Conditions. JAMA
Booth, T., & Hughes, D. J. (2014). Exploratory structural equation modeling of
personality data. Assessment, 21(3), 260–271.
Borgatta, E. F. (1964). The structure of personality characteristics. Behavioral Science, 9,
8–17.
Borkenau, P., & Ostendorf, F. (1990). Comparing exploratory and confirmatory factor
analysis: A study on the 5-factor model of personality. Personality and Individual
Differences, 11(5), 515–524.
Calabrese, W. R., Rudick, M. M., Simms, L. J., & Clark, L. E. (2012). Development and
validation of big four personality scales for the schedule for nonadaptive and adaptive
personality—second edition (SNAP-2). Psychological Assessment, 24(3), 751.
Cattell, R. B. (1933). Temperament tests. II: Tests. British Journal of Psychology, 23,
308–329.
Cattell, R. B. (1943). The description of personality: Basic traits resolved into clusters.
Journal of Abnormal and Social Psychology, 38, 476–506.
Cattell, R. B. (1944). “Parallel proportional profiles” and other principles for determining
the choice of factors by rotation. Psychometrika, 9(4), 267–283.
Cattell, R. B. (1947). Confirmation and clarification of primary personality factors.
Cattell, R. B. (1948). The primary personality factors in women compared with those in
men. British Journal of Statistical Psychology, 1(2), 114–130.
Cattell, R. B. (1966). The scree test for the number of factors. Multivariate Behavioral
Research, 1, 245–276.
Church, A. T., & Burke, P. J. (1994). Exploratory and confirmatory tests of the big five
and Tellegen’s three-and four-dimensional models. Journal of Personality and Social
Psychology, 66(1), 93–114.
Clark, L. A. (1993). Schedule for Nonadaptive and Adaptive Personality (SNAP).
Minneapolis, MN: University of Minnesota Press.
Clark, L. A., Livesley, W. J., Schroeder, M. L., & Irish, S. L. (1996). Convergence of two
systems for assessing specific traits of personality disorder. Psychological Assessment,
8(3), 294–303.
Clark, L. A., & Watson, D. (1991). Tripartite model of anxiety and depression:
Psychometric evidence and taxonomic implications. Journal of Abnormal Psychology,
100(3), 316–336.
Costa, P. T., Jr., & McCrae, R. R. (1992). The Revised NEO Personality Inventory
manual. Odessa, FL: Psychological Assessment Resources.
assessment using the Revised NEO Personality Inventory. Journal of Personality
Cramer, A. O., Sluis, S., Noordhof, A., Wichers, M., Geschwind, N., Aggen, S. H., …
Borsboom, D. (2012). Dimensions of normal personality as networks in search of
equilibrium: You can’t like parties if you don’t like people. European Journal of
Personality, 26(4), 414–431.
Cramer, A. O., Waldorp, L. J., van der Maas, H. L., & Borsboom, D. (2010).
Comorbidity: A network perspective. Behavioral and Brain Sciences, 33(2–3), 137–
150.
De Clercq, B., De Fruyt, F., Van Leeuwen, K., & Mervielde, I. (2006). The structure of
maladaptive personality traits in childhood: A step toward an integrative developmental
perspective for DSM-V. Journal of Abnormal Psychology, 115(4), 639.
De Fruyt, F., De Clercq, B., De Bolle, M., Wille, B., Markon, K., & Krueger, R. F.
(2013). General and maladaptive traits in a five-factor framework for DSM-5 in a
university student sample. Assessment, 20(3), 295–307.
DeYoung, C. G. (2013). The neuromodulator of exploration: A unifying theory of the role
of dopamine in personality. Frontiers in Human Neuroscience, 7, article 762.
Digman, J. M. (1989). Five robust trait dimensions: Development, stability, and utility.
Digman, J. M. (1990). Personality structure: Emergence of the five-factor model. Annual
Review of Psychology, 41(1), 417–440.
Digman, J. M. (1996). The curious history of the five-factor model. In J. S. Wiggins
(Ed.), Theoretical foundations of the five-factor model. New York, NY: Guilford Press.
Digman, J. M., & Takemoto-Chock, N. K. (1981). Factors in the natural language of
personality: Re-analysis, comparison, and interpretation of six major studies.
Multivariate Behavioral Research, 16, 149–170.
Donnellan, M. B., Hopwood, C. J., & Wright, A. G. C. (2012). Reevaluating the evidence
for the general factor of personality in the Multidimensional Personality Questionnaire:
Concerns about Rushton and Irwing (2009). Personality and Individual Differences,
52(2), 285–289.
Donnellan, M. B., Oswald, F. L., Baird, B. M., & Lucas, R. E. (2006). The mini-IPIP
scales: Tiny-yet-effective measures of the Big Five factors of personality.
Eaton, N. R., Keyes, K. M., Krueger, R. F., Balsis, S., Skodol, A. E., Markon, K. E., …
Hasin, D. S. (2012). An invariant dimensional liability model of gender differences in
mental disorder prevalence: Evidence from a national sample. Journal of Abnormal
Eaton, N. R., Krueger, R. F., & Oltmanns, T. F. (2011). The structure and long-term
stability of the internalizing spectrum of personality and psychopathology across the
lifespan. Psychology and Aging, 26, 987–993.
Edwards, A. L. (1957). The social desirability variable in personality assessment and
research. New York, NY: Dryden.
Fabrigar, L. R., Wegener, D. T., MacCallum, R. C., & Strahan, E. J. (1999). Evaluating
the use of exploratory factor analysis in psychological research. Psychological
Methods, 4(3), 272–299.
Forbush, K. T., & Watson, D. (2013). The structure of common and uncommon mental
disorders. Psychological Medicine, 43, 97–108.
Gignac, G. E., Bates, T. C., & Jang, K. L. (2007). Implications relevant to CFA model
misfit, reliability, and the five-factor model as measured by the NEO-FFI. Personality
Goldberg, L. R. (1981). Language and individual differences: The search for universals in
Goldberg, L. R. (1982). From Ace to Zombie: Some explorations in the language of
personality. In C. D. Spielberger & J. N. Butcher (Eds.), Advances in personality
assessment (Vol. 1, pp. 203–234). Hillsdale, NJ: Erlbaum.
Goldberg, L. R. (1990). An alternative “description of personality”: The Big-Five
factor structures from the top down. Journal of Research in Personality, 40, 347–358.
Gore, W. L., & Widiger, T. A. (2013). The DSM-5 dimensional trait model and five-
factor models of personality. Journal of Abnormal Psychology, 122(3), 816–821.
Guilford, J. P. (1975). Factors and factors of personality. Psychological Bulletin, 82(5),
802.
Guilford, J. P., & Guilford, R. B. (1934). An analysis of the factors in a typical test of
introversion-extroversion. Journal of Abnormal and Social Psychology, 28, 377–399.
Haigler, E. D., & Widiger, T. A. (2001). Experimental manipulation of NEO-PI-R items.
Harkness, A. R., Finn, J. A., McNulty, J. L., & Shields, S. M. (2012). The Personality
Psychopathology—Five (PSY-5): Recent constructive replication and assessment
literature review. Psychological Assessment, 24(2), 432.
Harkness, A. R., & McNulty, J. L. (1994). The Personality Psychopathology Five (PSY-
5): Issues from the pages of a diagnostic manual instead of a dictionary. In S. Strack &
M. Lorr (Eds.), Differentiating normal and abnormal personality (pp. 291–315). New
York, NY: Springer.
Hill, J., Pilkonis, P. A., & Bear, J. (2010). Social domains, personality, and interpersonal
functioning. In L. Horowitz & S. Strack (Eds.), Handbook of interpersonal psychology:
Theory, research, assessment, and therapeutic interventions (pp. 281–296). New York,
NY: Wiley.
Hofstee, W. K., De Raad, B., & Goldberg, L. R. (1992). Integration of the big five and
63(1), 146–163.
Hopwood, C. J., & Donnellan, M. B. (2010). How should the internal structure of
personality inventories be evaluated? Personality and Social Psychology Review, 14(3),
332–346.
Hopwood, C. J., Malone, J. C., Ansell, E. B., Sanislow, C. A., Grilo, C. M., Pinto, A., …
Morey, L. C. (2011). Personality assessment in DSM-5: Empirical support for rating
severity, style, and traits. Journal of Personality Disorders, 25, 305–320.
Hopwood, C. J., Wright, A. G. C., Ansell, E. B., & Pincus, A. L. (2013). The
interpersonal core of personality pathology. Journal of Personality Disorders, 27(3),
271–295.
Hopwood, C. J., Wright, A. G. C., & Donnellan, M. B. (2011). Evaluating the evidence
for the general factor of personality across multiple inventories. Journal of Research in
Personality, 45(5), 468–478.
Horn, J. L. (1965). A rationale and test for the number of factors in factor analysis.
Hyler, S. E. (1994). PDQ-4+ personality questionnaire. New York, NY: Author.
Jennrich, R. I., & Bentler, P. M. (2011). Exploratory bi-factor analysis. Psychometrika,
76, 537–549.
John, O. P. (1989). Towards a taxonomy of personality descriptors. In D. M. Buss & N.
Cantor (Eds.), Personality psychology: Recent trends and emerging directions (pp.
261–271). New York, NY: Springer-Verlag.
Jöreskog, K. G. (1969). A general approach to confirmatory maximum likelihood factor
analysis. Psychometrika, 34(2), 183–202.
Kessler, R. C., Chiu, W. T., Demler, O., & Walters, E. E. (2005). Prevalence, severity,
and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey
Replication. Archives of General Psychiatry, 62, 617–627.
Kessler, R. C., McGonagle, K. A., Zhao, S., Nelson, C. B., Hughes, M., Eshleman, S.,
Wittchen, H. U., & Kendler, K. S. (1994). Lifetime and 12-month prevalence of DSM-
III-R psychiatric disorders in the United States: results from the National Comorbidity
Survey. Archives of General Psychiatry, 51, 8–19.
Bulletin, 136(5), 768–821.
Kotov, R., Ruggero, C. J., Krueger, R. F., Watson, D., Yuan, Q., & Zimmerman, M.
(2011). New dimensions in the quantitative classification of mental illness. Archives of
General Psychiatry, 68(10), 1003.
Kramer, M. D., Krueger, R. F., & Hicks, B. M. (2008). The role of internalizing and
externalizing liability factors in accounting for gender differences in the prevalence of
common psychopathological syndromes. Psychological Medicine, 38, 51–61.
Krueger, R. F. (1999). The structure of common mental disorders. Archives of General
Psychiatry, 56(10), 921.
Krueger, R. F. (2013). Personality disorders are the vanguard of the post-DSM-5.0 era.
Personality Disorders: Theory, Research, and Treatment, 4(4), 355–362.
Krueger, R. F., Capsi, A., Moffitt, T. E., & Silva, P. A. (1998). The structure and stability
of common mental disorders (DSM-III-R): A longitudinal-epidemiological study.
Journal of Abnormal Psychology, 107, 216–227.
Krueger, R. F., Chentsova-Dutton, Y. E., Markon, K. E., Goldberg, D. P., & Ormel, J.
(2003). A cross-cultural study of the structure of comorbidity among common
psychopathological syndromes in the general health care setting. Journal of Abnormal
Krueger, R. F., Derringer, J., Markon, K. E., Watson, D., & Skodol, A. E. (2012). Initial
construction of a maladaptive personality trait model and inventory for DSM-5.
Krueger R. F., & Eaton, N. R. (2010). Personality traits and the classification of mental
disorders. Personality Disorders: Theory, Research, and Treatment, 1(2), 97–118.
Krueger, R. F., Eaton, N. R., Clark, L. A., Watson, D., Markon, K. E., Derringer, J., …
Livesley, W. J. (2011). Deriving an empirical structure of personality pathology for
DSM-5. Journal of Personality Disorders, 25(2), 170–191.
Krueger, R. F., & Markon, K. E. (2006). Reinterpreting comorbidity: A model-based
approach to understanding and classifying psychopathology. Annual Review of Clinical
Krueger, R. F., Markon, K. E., Patrick, C. J., Benning, S. D., & Kramer, M. D. (2007).
Linking antisocial behavior, substance use, and personality: An integrative quantitative
model of the adult externalizing spectrum. Journal of Abnormal Psychology, 116(4),
645.
Kushner, S. C., Quilty, L. C., Tackett, J. L., & Bagby, R. M. (2011). The hierarchical
structure of the Dimensional Assessment of Personality Pathology (DAPP-BQ).
Journal of Personality Disorders, 25(4), 504–516.
Lahey, B. B., Rathouz, P. J., van Hulle, C., Urbano, R. C., Krueger, R. F., Applegate, B.,
… Chapman, D. A. (2008). Testing structural models of DSM-IV symptoms of
common forms of child and adolescent psychopathology. Journal of Abnormal Child
Leary, T. (1957). Interpersonal diagnosis of personality. New York, NY: Ronald Press.
Lim, B. C., & Ployhart, R. E. (2006). Assessing the Convergent and Discriminant
Validity of Goldberg’s International Personality Item Pool A Multitrait-Multimethod
Examination. Organizational Research Methods, 9(1), 29–54.
Livesley, W. J., & Jackson, D. N. (2009). Manual for the dimensional assessment of
personality pathology—basic questionnaire. Port Huron, MI: Sigma Press.
Livesley, W. J., Jackson, D. N., & Schroeder, M. L. (1989). A study of the factorial
structure of personality pathology. Journal of Personality Disorders, 3, 292–306.
Livesley, W. J., Jang, K. L., & Vernon, P. A. (1998). The phenotypic and genetic
structure of traits delineating personality disorder. Archives of General Psychiatry, 55,
941–948.
Markon, K. E. (2010). Modeling psychopathology structure: A symptom-level analysis of
Axis I and II disorders. Psychological Medicine, 40, 273–288.
Markon, K. E., Quilty, L. C., Bagby, R. M., & Krueger, R. F. (2013). The development
and psychometric properties of an informant-report form of the Personality Inventory
for DSM-5 (PID-5). Assessment, 20(3), 370–383.
Marsh, H. W. (1996). Positive and negative global self-esteem: A substantively
meaningful distinction or artifactors? Journal of Personality and Social Psychology,
70(4), 810–819.
Marsh, H. W., Lüdtke, O., Muthén, B., Asparouhov, T., Morin, A. J., Trautwein, U., &
Nagengast, B. (2010). A new look at the big five factor structure through exploratory
structural equation modeling. Psychological Assessment, 22(3), 471–491.
McCrae, R. R., Costa, P. T., Jr., & Busch, C. M. (1986). Evaluating comprehensiveness in
personality systems: The California Q-Set and the five-factor model. Journal of
Meyer, B., & Pilkonis, P. A. (2005). An attachment model of personality disorders. In M.
F. Lenzenweger & J. F. Clarkin (Eds.), Major theories of personality disorder (2nd ed.,
Millon, T., Davis, R., & Millon, C. (1997). MCMI-III manual (2nd ed.). Minneapolis,
MN: National Computer Systems.
Mischel, W. (1968). Personality and assessment. New York, NY: Wiley.
Molenaar, P. C. (2010). Latent variable models are network models. Behavioral and
Brain Sciences, 33(2–3), 166.
Morey, L. C., Krueger, R. F., & Skodol, A. E. (2013). The hierarchical structure of
clinician ratings of proposed DSM-5 pathological personality traits. Journal of
Morey, L. C., Waugh, M. H., & Blashfield, R. K. (1985). MMPI scales for DSM-III
personality disorders: Their derivation and correlates. Journal of Personality
Assessment, 49(3), 245–251.
Mulaik, S. A. (2009). Foundations of factor analysis (2nd ed.). Boca Raton, FL: CRC
press.
Replicated factor structure in peer nomination personality ratings. The Journal of
Abnormal and Social Psychology, 66(6), 574–583.
personality disorders. Journal of Clinical Psychology, 61(3), 323–345.
Parker, J. D., Bagby, R. M., & Summerfeldt, L. J. (1993). Confirmatory factor analysis of
the Revised NEO Personality Inventory. Personality and Individual Differences, 15(4),
463–466.
Peabody, D. (1967). Trait inferences: Evaluative and descriptive aspects. Journal of
Personality and Social Psychology Monographs, 7 (Whole No. 644).
Peabody, D., & Goldberg, L. R. (1989). Some determinants of factor structures from
personality-trait descriptors. Journal of Personality and Social Psychology, 57, 552–
567.
Pettersson, E., Mendle, J., Turkheimer, E., Horn, E. E., Ford, D. C., Simms, L. J., &
Clark, L. A. (2014). Do maladaptive behaviors exist at one or both ends of personality
traits? Psychological Assessment, 26, 433–446.
Pettersson, E., & Turkheimer, E. (2010). Item selection, evaluation, and simple structure
in personality data. Journal of Research in Personality, 44(4), 407–420.
Pettersson, E., Turkheimer, E., Horn, E. E., & Menatti, A. R. (2012). The general factor
of personality and evaluation. European Journal of Personality, 26(3), 292–302.
(2009). Using the five-factor model to identify a new personality disorder domain: The
case for experiential permeability. Journal of Personality and Social Psychology,
96(6), 1245.
Pincus, A. L. (2005). A contemporary integrative interpersonal theory of personality
disorders. In J. Clarkin & M. Lenzenweger (Eds.), Major theories of personality
disorder (2nd ed., pp. 282–331). New York, NY: Guilford Press.
Quilty, L. C., Oakman, J. M., & Risko, E. (2006). Correlates of the Rosenberg self-
esteem scale method effects. Structural Equation Modeling, 13(1), 99–117.
Revelle, W., & Wilt, J. (2013). The general factor of personality: A general critique.
Røysamb, E., Kendler, K. S., Tambs, K., Orstavik, R. E., Neale, M. C., Aggen, S. H., …
Reichborn-Kjennerud, T. (2011). The joint structure of DSM-IV axis I and axis II
disorders. Journal of Abnormal Psychology, 120, 198–209.
Ruscio, J., & Roche, B. (2012). Determining the number of factors to retain in an
exploratory factor analysis using comparison data of known factorial structure.
Rushton, J. P., Bons, T. A., & Hur, Y.-M. (2008). The genetics and evolution of a general
factor of personality. Journal of Research in Personality, 42, 1173–1185.
Rushton, J. P., & Irwing, P. (2008). A general factor of personality (GFP) from two meta-
analyses of the big five: Digman (1997) and Mount, Barrick, Scullen, and rounds
(2005). Personality and Individual Differences, 45, 679–683.
Samuel, D. B. (2011). Assessing personality in the DSM–5: The utility of bipolar
constructs. Journal of Personality Assessment, 93(4), 390–397.
Samuel, D. B., Simms, L. J., Clark, L. A., Livesley, W. J., & Widiger, T. A. (2010). An
item response theory integration of normal and abnormal personality scales.
Personality Disorders: Theory, Research, and Treatment, 1(1), 5.
Samuel, D. B., & Widiger, T. A. (2008a). A meta-analytic review of the relationships
between the five-factor model and DSM-IV-TR personality disorders: A facet level
analysis. Clinical Psychology Review, 28(8), 1326.
Samuel, D. B., & Widiger, T. A. (2008b). Convergence of narcissism measures from the
perspective of general personality functioning. Assessment, 15, 364–374.
Samuel, D. B., & Widiger, T. A. (2010). A comparison of obsessive–compulsive
personality disorder scales. Journal of Personality Assessment, 92(3), 232–240.
Sass, D. A., & Schmitt, T. A. (2010). A comparative investigation of rotation criteria
within exploratory factor analysis. Multivariate Behavioral Research, 45, 73–103.
Saulsman, L. M., & Page, A. C. (2004). The five-factor model and personality disorder
empirical literature: A meta-analytic review. Clinical Psychology Review, 23(8), 1055–
1085.
Sharp, C., Wright, A. G. C., Fowler, J. C., Freuh, C., Allen, J. G., Oldham, J., & Clark, L.
A. (2015). The structure of personality pathology: Both general (“g”) and specific (“s”)
factors? Journal of Abnormal Psychology, 124, 387–396.
CAT–PD Project. Journal of Personality Assessment, 93(4), 380–389.
Slade, T., & Watson, D. (2006). The structure of common DSM-IV and ICD-10 mental
disorder in the Australian general population. Psychological Medicine, 36, 1593–1600.
Smith, G. M. (1967). Usefulness of peer ratings of personality in educational research.
Educational and Psychological Measurement, 27, 967–984.
Spearman, C. (1904). General intelligence, objectively determined and measured.
American Journal of Psychology, 15, 201–293.
Spearman, C. (1927). The abilities of man. New York, NY: Macmillan.
Stepp, S. D., Yu, L., Miller, J. D., Hallquist, M. N., Trull, T. J., and Pilkonis, P. A.
(2012). Integrating competing dimensional models of personality. Personality
Strack, S. (1987). Development and validation of an adjective checklist to assess the
Millon personality types in a normal population. Journal of Personality Assessment,
51, 572–587.
dimensional model of personality disorder: Incorporating DSM cluster A
characteristics. Journal of Abnormal Psychology, 117(2), 454–459.
Development of the Multidimensional Personality Questionnaire. The SAGE Handbook
of Personality Theory and Assessment, 2, 261–292.
Thomas, K. M., Yalch, M. M., Krueger, R. F., Wright, A. G. C., Markon, K. E., &
Hopwood, C. J. (2013). The convergent structure of DSM-5 personality trait facets and
the five-factor model (FFM) trait domains. Assessment, 20(3), 308–311.
Thurstone, L. L. (1934). The vectors of mind. Psychological Review, 41, 1–32.
Thurstone, L. L. (1947). Multiple factor analysis. Chicago, IL: University of Chicago
Press.
Thurstone, L. L. (1951). The dimensions of temperament. Psychometrika, 16(1), 11–20.
Trapnell, P. D., & Wiggins, J. S. (1990). Extension of the Interpersonal Adjective Scales
to include the Big Five dimensions of personality. Journal of Personality and Social
Trull, T. J. (2005). Dimensional models of personality disorder: Coverage and cutoffs.
Journal of Personality Disorders, 19(3), 262–282.
Tupes, E. C, & Christal, R. E. (1961). Recurrent personality factors based on trait ratings
(Technical Report No. ASD-TR-61-97). Lackland Air Force Base, Lackland, TX: U.S.
Air Force.
Tupes, E. R., & Kaplan, M. (1961). Similarity of factors underlying peer ratings of
socially acceptable, socially unacceptable, and bipolar personality traits (Technical
Report No. ASD-TN-61-48). Personnel Laboratory, Aeronautical Systems Division,
Air Force Systems Command. Lackland Air Force Base, Lackland, TX: U.S. Air Force.
Vassend, O., & Skrondal, A. (1995). Factor analytic studies of the NEO Personality
Inventory and the five-factor model: The problem of high structural complexity and
conceptual indeterminacy. Personality and Individual Differences, 19(2), 135–147.
Vassend, O., & Skrondal, A. (1997). Validation of the NEO Personality Inventory and the
five-factor model. Can findings from exploratory and confirmatory factor analysis be
reconciled? European Journal of Personality, 11(2), 147–166.
Velicer, W. F. (1976). Determining the number of components from the matrix of partial
correlations. Psychometrika, 41, 321–327.
Verheul, R. & Widiger, T. A. (2004). A meta-analysis of the prevalence and usage of the
Personality Disorder Not Otherwise Specified (PDNOS) diagnosis. Journal of
Personality Disorders, 18, 309–319.
relevance to psychopathology: II. Further articulation of a comprehensive unified trait.
Widiger, T. A. (1998). Four out of five ain’t bad. Archives of General Psychiatry, 55,
865–866.
Widiger, T. A. (2013). A postmortem and future look at the personality disorders in
DSM-5. Personality Disorders: Theory, Research, and Treatment, 4(4), 382–387.
Widiger, T. A., & Clark, L. A. (2000). Toward DSM–V and the classification of
psychopathology. Psychological Bulletin, 126, 946–963.
Widiger, T. A., & Frances, A. (1985). The DSM-III personality disorders: Perspectives
from psychology. Archives of General Psychiatry, 42(6), 615–623.
Widiger, T. A., & Kelso, K. (1983). Psychodiagnosis of axis II. Clinical Psychology
Review, 3(4), 491–510.
Widiger, T. A., Livesley, W. J., & Clark, L. A. (2009). An integrative dimensional
classification of personality disorder. Psychological Assessment, 21(3), 243.
Widiger, T. A., Lynam, D. R., Miller, J. D., & Oltmanns, T. F. (2012). Measures to assess
maladaptive variants of the Five-Factor model. Journal of Personality Assessment,
94(5), 450–455.
Widiger, T. A., & Mullins-Sweatt, S. N. (2009). Five-factor model of personality
disorder: A proposal for DSM-V. Annual Review of Clinical Psychology, 5, 197–220.
Widiger, T. A., & Simonsen, E. (2005). Alternative dimensional models of personality
disorder: Finding a common ground. Journal of Personality Disorders, 19(2), 110–130.
disorder: Shifting to a dimensional model. American Psychologist, 62(2), 71–83.
Wiggins, J. S., & Pincus, A. L. (1989). Conceptions of personality disorders and
dimensions of personality. Psychological Assessment: A Journal of Consulting and
Clinical Psychology, 1(4), 305.
Wolf, A. W., Schubert, D. S. P., Patterson, M. B., Grande, T. P., Brocco, K. J., &
Pendleton, L. (1988). Associations among major psychiatric diagnoses. Journal of
Consulting and Clinical Psychology, 56, 292–294.
Wright, A. G. C. (2014). Integrating trait and process based conceptualizations of
pathological narcissism in the DSM-5 era. In A. Besser (Ed.), Handbook of psychology
of narcissism: Diverse perspectives. Hauppauge, NY: Nova Science Publishers.
Wright, A. G. C., Hallquist, M. N., Beeney, J. E., & Pilkonis, P. A. (2013). Borderline
personality pathology and the stability of interpersonal problems. Journal of Abnormal
Psychology, 122(4), 1094–1100.
Wright, A. G. C., Krueger, R. F., Hobbs, M. J., Markon, K. E., Eaton, N. R., & Slade, T.
(2013). The structure of psychopathology: Toward an expanded quantitative empirical
model. Journal of Abnormal Psychology, 122(1), 281–294.
Wright, A. G. C., Scott, L. N., Stepp, S. D., Hallquist, M. N., & Pilkonis, P. A. (2015).
Personality pathology and interpersonal problem stability. Journal of Personality
Disorder (in press).
Conjoint analyses of the CAT-PD, PID-5, and NEO-PI-3 trait models. Personality
Wright, A. G. C. & Simms, L. J. (2015). A metastructural model of mental disorders and
pathological personality traits. Psychological Medicine, 45(11), 2309–2319.
Wright, A. G. C., Thomas, K. M., Hopwood, C. J., Markon, K. E., Pincus, A. L., &
Krueger, R. F. (2012). The hierarchical structure of DSM-5 pathological personality
traits. Journal of Abnormal Psychology, 121(4), 951–957.
Wright, A. G. C., & Zimmermann, J. (2015). At the nexus of science and practice:
Answering basic clinical questions in personality disorder assessment and diagnosis
with quantitative modeling techniques. In S. Huprich (Ed.), Personality disorders:
Assessment, diagnosis, and research. Washington, DC: American Psychological
Association.
Wu, K. D., Clark, L. A., & Watson, D. (2006). Relations between obsessive–compulsive
disorder and personality: Beyond Axis I–Axis II comorbidity. Journal of Anxiety
Disorders, 20(6), 695–717.
Zimmerman, M., & Mattia, J. I. (1999). Psychiatric diagnosis in clinical practice: is
comorbidity being missed? Comprehensive Psychiatry, 40(3), 182–191.
Childhood Personality and Temperament
Sarah S. W. De Pauw
Abstract
This chapter discusses the blossoming research on childhood personality,
addressing salient questions on its measurement, its foundational structure, and its
convergence with child temperament. First, the discovery and contemporary
measurement of FFM-antecedents in young age is reviewed, highlighting
promises as well as pitfalls of the various approaches available today. Second,
divergences between child and adult taxonomies are delineated, as accumulating
research suggests that the structure of child personality is not identical to the
established FFM-structure in adults. Finally, the issue of temperament-personality
convergence is considered. Whereas influential narrative reviews proposed that
temperament and personality appear to be “more alike than different,” empirical
research suggests that a simple hierarchical mapping fails to capture the
complexity of these relationships. We invite students of behavioral individuality
in childhood and adolescence to take into account the salient differences between
the various personality approaches, child and adult personality taxonomies, and
child temperament and personality “vocabularies.”
Key Words: personality, temperament, childhood, adolescence, development,
assessment, taxonomy
It is almost impossible to overestimate the impact the Five Factor Model

(FFM) has had on the study of adult personality in the past four decades. The
pursuit to unravel the basic architecture of adult traits has provided the
personality field with structure and clarity, which has greatly accelerated the
discipline’s progress. Most trait researchers now agree that personality is
hierarchically structured, with higher-order domains subsuming narrower traits.
Most scholars also agree that the FFM (or the Big Five) constitutes a unique,
comprehensive level of this personality hierarchy (John, Naumann, & Soto,
2008; Markon, Krueger, & Watson, 2005).
One major strength of the FFM is its attested comprehensiveness to describe
adaptive individual differences in adults. To date, a considerable number of
studies have documented the impressive power of the FFM to link many
competing adult personality systems under its “five factor umbrella” (Halverson
et al., 2003; John et al., 2008; McCrae & Costa, 1996). This well-documented
convergence between the FFM and alternative personality models has
significantly contributed to the establishment of the FFM as a unifying
framework that allows organization and integration of the wide ocean of adult
trait descriptors into one single reference structure. This integrative character of
the FFM has also provided an excellent framework for generating new
hypotheses and exciting research on how adult personality impacts a wide array
of life outcomes (e.g., Roberts, Kuncel, Shiner, Caspi, & Goldberg, 2007).
These major advances in the domain of adult personality have sparked a
growing interest in uncovering FFM antecedents in younger age groups. There is
now-particularly among personality psychologists- a wide acceptance of the
notion that the FFM can be used as a valid framework to describe behavioral
individuality in adolescents and even in children, at least from the preschool
years on (John, Caspi, Robins, Moffitt, & Stouthamer-Loeber, 1994; Mervielde,
De Clercq, De Fruyt, & Van Leeuwen, 2005; Shiner & Caspi, 2003; Shiner &
DeYoung, 2013; Soto & John, 2014; Soto, John, Gosling, & Potter, 2008;
Tackett, Kushner, De Fruyt, & Mervielde, 2013). However, unlike in adulthood,
in which the FFM reigns as the “sovereign framework” for trait assessment in
both research and practice, the study of childhood personality still remains “a
work in progress” with important lingering questions on the measurement,
structure, and conceptualization, of youngsters’ behavioral traits.
First, questions remain on how to measure FFM personality in youth.
Notwithstanding that one of the most seminal papers on the “Little Five,”
describing FFM antecedents in adolescent boys (John et al., 1994), has already
celebrated its twentieth birthday, researchers and practitioners still have a need
for clear recommendations on which measures to use to assess a child’s
personality (Tackett et al., 2013). In the first section of this chapter, we critically
review the discovery history of FFM antecedents in childhood. Along the way,
the most important contemporary personality instruments for youth are
highlighted, focusing on their strengths and limitations, and their convergence
with the adult FFM framework.
Second, throughout this review, we discuss accumulating evidence that the
structure underlying child personality is less equivalent to adult personality than
was originally expected on the basis of the FFM. Recent empirical and
conceptual analyses no longer radically assert that five dimensions are enough to
capture youth’s traits. Studies increasingly identify Activity as a major sixth
dimension to consider (De Pauw, Mervielde, & Van Leeuwen, 2009; Shiner &
DeYoung, 2013; Soto & John, 2014). Gradually, other divergences between
child and adult personality taxonomies become visible and these are highlighted
in the second part of this chapter. In addition, we formulate some thoughts on
salient themes that need further consideration to advance the budding child
personality discipline.
Finally, essential questions remain about the conceptual convergence between
childhood personality and temperament, which is the historical “sovereign
framework” for trait description in children and adolescents. In the past decade,
some influential, yet primarily personality-inspired, scholars have made a
compelling case that temperament and personality systems describing children’s
and adolescents’ traits can be considered to be “more alike than different,” at
least from the preschool years on (Caspi, Roberts, & Shiner, 2005; Caspi &
Shiner, 2006, 2008; De Pauw & Mervielde, 2010; Mervielde et al., 2005; Shiner,
1998; Shiner & Caspi, 2003; Shiner & DeYoung, 2013). Moreover, these
authors have proposed—mainly based on narrative analyses—that the FFM
provides a comprehensive framework to capture the conceptual dimensions
underlying both personality and temperament traits. The third part of this chapter
discusses the growing evidence on temperament–personality convergence and
focuses on what we can learn from the temperament perspective. Finally, we
propose a detailed and nuanced taxonomy describing how the wide range of
temperament and personality traits that are currently assessed in youngsters can
be integrated.
While discussing the blossoming child personality field, we will point out
some vital issues that need further empirical or conceptual consideration. It is
our hope that these issues will inspire fruitful dialogues between scholars from
different disciplines, including personality psychology and developmental
psychology. Ideally, these discussions should go beyond disputes about which
framework (i.e., personality or temperament) will prove superior to
conceptualize childhood traits. What both disciplines (i.e., personality
psychology and developmental psychology) need, is a more universal and cross-
disciplinary approach to individual differences among youth (Caspi & Shiner,
2006; De Pauw & Mervielde, 2010). A consensus on a more comprehensive
conceptual trait system would help to better clarify the wide heterogeneity in
pathways that children and adolescents follow throughout their development.
Ideally, such a trait system should be tied closely to individual differences in
adults in order to facilitate theories relating individual trait differences to
development across the entire lifespan. As will become evident throughout this
chapter, a flexible adaptation of the FFM is a good candidate to provide the
conceptual underpinnings of such a comprehensive taxonomy for childhood
traits.
Discovery and Measurement of Child Personality

How Shall We Speak of Personality Differences in Children and Adolescents?
Child temperament yes, but personality? Historically, the conceptualization
and measurement of individual differences among infants, toddlers, preschool
children, and school-age children are based on temperament models.
Temperament has been traditionally distinguished from personality because it
refers to individual differences that appear from birth onward, remain relatively
stable across the lifespan, and presumably have a strong genetic or
neurobiological basis (Goldsmith et al., 1987; Mervielde et al., 2005). In
addition, temperament has been conceived as the “attentional, activational, and
attentional core” of personality around which personality develops under the
influence of the environment (Rothbart & Bates, 1998, 2006). According to this
view, personality is wider in scope than temperament and refers to consistent
patterns of thoughts, feelings, and actions, including skills, values, defenses,
beliefs, morals, and social cognitions (Rothbart & Bates, 1998, 2006).
Based on this perspective, the study of individual differences among children
and adolescents has been—for many years—the uniquely reserved territory of
developmental psychologists, who conceptualized trait differences in terms of
temperament since the early 1960s (Rothbart, 2012; Thomas, Chess, Birch,
Hertzig, & Korn, 1963). In contrast, personality psychologists primarily devoted
their attention to the structure and measurement of individual differences in
adulthood. For several decades, the literatures on temperament and personality
were separate worlds. Researchers created divergent systems based on widely
varying constructs and dimensions, seeking little opportunities for cross-
fertilization.
This trend changed, however, in the late 1980s to early 1990s. In this period,
described as the “electrifying” early years of FFM research (Goldberg, 1993),
studies on adult personality impressively documented how the “Big Five”
dimensions emerged across a wide variety of ages, languages, and cultures (John
et al., 2008). These great successes in the adult literature inspired multiple
research groups from all over the world to inquire about the applicability of the
FFM in younger age groups, such as adolescents or even children. At the time,
examining the FFM in children was considered an audacious endeavor. By
launching this quest for FFM antecedents in developmental research, historical
borders between the temperament and personality territories were trespassed and
the traditional interpretations of both established constructs began to erode.
The discovery of “the Little Five” in adolescents and children. The
pioneering efforts of the multiple research groups independently searching for
FFM personality antecedents can be considered as very successful. There is now
rich and ample evidence that five factor-like dimensions can indeed be retrieved
in adolescents and even in children. In Tables 12.1 and 12.2, we provide an
overview of the most important actors in research on the “discovery of FFM
antecedents”—historically referred to as the “Little Five” (John et al., 1994)—in
younger age groups. Both tables show that the research groups searching for the
“Little Five” derived their conclusions based on a wide variety of informants,
strategies, and methods of assessment (Goldberg, 2001a; John et al., 1994;
Mervielde & De Fruyt, 1999; Shiner & Caspi, 2003). Regarding informants, five
factor-like dimensions have now been found across ratings provided by parents
and teachers (Table 12.1) as well as across peer-reports and self-reports (Table
12.2). Regarding strategies, five factor-like dimensions emerged in studies using
questionnaires, adjective lists, and rating scales, as well as in studies using peer
nomination procedures and studies using generic and natural language
descriptions of children and their parents. Regarding methods of assessment, five
factor-like dimensions have emerged in studies using Q-sorts, trait adjective
lists, behavioral statements, bipolar scales, and even combinations of rating
scales and child interviews. Many of these studies documented that the “Little
Five” retrieved in childhood or adolescence are theoretically—and in some
cases, even empirically (e.g., De Fruyt, Mervielde, Hoekstra, & Rolland, 2000)
—related to the Big Five dimensions established in adults. Moreover, many of
these studies have provided compelling evidence that Little Five dimensions,
retrieved in youth, are associated (both cross-sectionally and longitudinally) with
a wide range of theoretically meaningful external correlates (e.g., De Pauw et al.,
2009; Graziano, Jensen-Campbell, & Finch, 1997; John et al., 1994; Markey,
Markey, Tinsley, & Ericksen, 2002).
How Shall We Measure Personality in Children and Adolescents?
The summary above gives an affirmative answer to the central question of
whether the FFM can be used for the description of individual differences among
adolescents, and even children. Yet, in this chapter, we aim to illuminate a
number of unresolved issues in current FFM approaches to childhood
personality. One important issue is that--unlike in adulthood--where there is a
wide array of well-validated FFM instruments available (in all sorts and lengths),
a comparably clear set of instruments for child assessment is still missing. There
are not many well-validated FFM instruments for use in applied settings.
Moreover, there is relatively little critical discussion on which instrument should
be preferred for which purposes (Tackett et al., 2013). As a consequence, the
search for a psychometrically sound (and preferably short) FFM instrument
remains a major obstacle for researchers and clinicians interested in addressing
children’s personalities in their studies or practices today.
Table 12.1 Prominent FFM Instruments for Child and Adolescent Personality
Assessment, Relying on Caregiver Informants
Strategy Key Number Age of Informant Method of Facets Dimensions
Reference of Items Target Assessment
for
Application
in Youth
I.a. Top-down adult-oriented inventories not modified
Bipolar markers Mervielde et 25 4–12 Teacher Bipolar markers – N, E, I, A, C
(Goldberg, 1992) al. (1995)
Bipolar markers Graziano et al. 50 10–14 Teacher Bipolar markers – N, E, I, A, C
(Goldberg, 1992) (1997)
Unipolar markers Gerris et al. 30 12+ Parents Adjectives – N, E,
(Goldberg, 1992) (1998) Resourcefulness,
A, C
NEO Five Factor Parker and 60 11–14 Parent Parent Questionnaire – N, E, O, A, C
Inventory (Costa & Stumpf (1998) 10–13
McCrae, 1992) Markey et al.
(2002)
NEO-Personality Martin and 240 9–15 Parent Questionnaire 30 N, E, O, A, C
Inventory-R Friedman
(Costa & (2000)
McCrae, 1992)
I.b. Top-down adult inventories explicitly modified to fit youth
Experimental Big Ehrler et al. 9–18 Teachers Ratings – N, E, O, A, C
Five trait markers (1999)
Big Five Barbaranelli et 65 3–14 Parent, Questionnaire – Emotional
Questionnaire— al. (2003) teacher Instability,
Children Intellect/
Openness,
Energy/E, A, C
Big Five Inventory John et al. 20 3–20 Parents Questionnaire – N, E, O,
Marker scales (2008)
M5-PS Grist et al. 90 3–5 Teacher Questionnaire – N, E, O,
(2012)
II.a. Generic descriptions of child development
Digman teachers’ Goldberg 39 6–12 Teacher Adjectives (11) Emotional
general ratings (2001a) Stability, E, I,
Friendliness, C
Project Shiner and >115 10 Multi- Multi-report – N, E, O, A, C
Competence Masten informant (interviews,
personality (2012) composite questionnaires)
indices (general (parent,
interview, rating teacher,
scales, child)
Devereaux
inventory)
II.b. Regrouping of items of other theoretical models
Common Language John et al. 48 3–15 Parent Q-sort FFM – N, E, O, A, C
Version of the (1994) scales
California Child Q-
sort (Caspi et al.,
1992)
John et al. (1994) 100 3–15 Parent Q-sort – Anxious
Distress,
Sociability,
O, A, C,
Irritability,
Activity
Van Lieshout and 100 3–17 Parent, Q-sort – Emotional
Haselager (1994) teacher Stability, E,
O,a A, C,
Activity,
Dependency
Soto and John 100 3–20 Parent Questionnaire – N, E, O,a A,
(2014) C, Activity
School Behavior Resing et al. 52 4–12 Teacher Questionnaire – Emotional
Checklist (Zaal, (1999) Stability, E, A,
1978) Attitude toward
School Work
Multidimensional Tackett et 34 11 Parent Questionnaire – Negative
Personality al. (2008) Emotionality,
Ratings (Tellegen Positive
& Waller, 2008) Emotionality,
Absorption/
Openness, A,
Constraint
III. Bottom-up based on free descriptions lexical research
Inventory of Halverson et 108 2–15 Parent, Questionnaire 15 N, E, I,
Childhood al. (2003) teacher A/Manageability,
Individual Deal et al. C
Differences (2007)
Inventory of Deal et al. 50 2–15 Parent, Questionnaire - N, E, I,
Childhood (2007) teacher A/Manageability,
Individual C
Differences, short
version
Hierarchical Mervielde 144 6–15 Parent, Questionnaire 18 Emotional
Personality et al. (2009) teacher Stability, E,
Inventory for Imagination,
Children Benevolence,
C
Note. We abbreviated the FFM domain names if the original reports explicitly use the Big Five (NEIAC) or
FFM (NEOAC) nomenclature. N, Neuroticism; E, Extraversion; O, Openness-to-experience; I, Intellect; A,
Agreeableness; C, Conscientiousness.
a Weaker psychometric properties are reported for this dimension.
Table 12.2 Prominent FFM Instruments for Child and Adolescent Personality
Assessment, Relying on Youth As Informant
Strategy Key Number Age of Informant Method of Facets Dimensions
Reference of Items Target Assessment
for
Application
in Youth
I.a. Top-down adult-oriented inventories not modified
Bipolar markers Graziano et al. 50 10–14 Self Bipolar markers – N, E, I, A, C
(Goldberg, 1992) (1997)
Bipolar markers Gerris et al. 30 12+ Self Unipolar – N, E, I, A, C
(Goldberg, 1992) (1998) markers
Selected set of Scholte et al. 25 13–15 Self Questionnaire – Emotional
bipolar markers (1997) Stability, E, O/I,
A, C
Selected set of Scholte et al. 20 13–15 Peers Peer – Aggression–
unipolar markers (1997) nominations Inattentiveness,
Achievement–
Withdrawal, Self-
confidence,
Sociability,
Emotionality–
Nervousness
Selected set of Mervielde and 25 8–12 Peers Peer – Emotional
bipolar markers De Fruyt nominations Stability–
(2000) Extraversion,
Agreeableness,
Intellect–
Conscientiousness
NEO-Personality De Fruyt et al. 240 12–17 self Questionnaire 30 N, E, O, A, C
Inventory-R (2000)
(Costa &
McCrae, 1989)
NEO-Five Factor Markey et al. 60 10–13 Self Questionnaire – N, E, O, A, C
Inventory (Costa (2002)
& McCrae,
1989)
NEO-Personality McCrae et al. 240 12+ Self Questionnaire 30 N, E, O, A, C
Inventory-3 (2005)
(McCrae et al.,
2005)
Big Five Soto et al. 40 10+ Self Questionnaire – N, E, I, A, C
Inventory (John (2008)
& Srivastava,
1999)
Five Factor Hendriks et 100 12+ Self Questionnaire – Emotional
Personality al. (2008) Stability, E,
Inventory Autonomya, A,
(Hendriks, C
1997)
I.b. Top-down adult inventories explicitly modified to fit youth
Adolescent Lounsbury et 55 11–18 Self Questionnaire – Emotional
Personality Style al. (2003) Stability, E, O, A,
Inventory C
Five Factor McGhee et al. 75 9–18 Self Questionnaire 30 Emotional
Personality (2007) Regulation, E, O,
Inventory– A, C
Children
Big Five Barbaranelli 65 8–13 Self Questionnaire – Emotional
Questionnaire– et al. (2003) Instability,
Children Energy/E, I/O,
A, C
II. Generic descriptions of child development
Common- van Lieshout 100 11–14 Self, peers Q-sort – Emotional
Language and Haselager Stability, E,
version of the (1994) Activity, A, C (no
California Child O)
Q-sort (Caspi et
al., 1992)
Adjective Check Parker and 103 13–15 Self Adjectives – N,a E,a O,a A, C
List (Gough & Stumpf (1998)
Heilbrun, 1983)
Berkeley Measelle et 60 6–7 Self Puppet – N, E, O,a A, C
Puppet al. (2005) interview
Interview
III. Bottom-up
Inventory of Halverson et 108, 40 12–17 Self Questionnaire 15 N, E, O, A, C
individual al. (2003)
differences Deal et al.
(2007)
Inventory of Deal et al. 40 12–17 Self Questionnaire – N, E, O, A, C
individual (2007)
differences–short
Hierarchical Mervielde 144 12–17 Self Questionnaire 18 Emotional
Personality et al. (2009) Stability, E,
Inventory for De Fruyt et Imagination,
Children al. (2000) Benevolence, C
Note. We abbreviated the FFM domain names if the original reports explicitly use the Big Five (NEIAC) or
FFM (NEOAC) nomenclature. N, Neuroticism; E, Extraversion; O, Openness-to-Experience; I, Intellect; A,
Agreeableness; C, Conscientiousness.
a Weaker psychometric properties are reported for this dimension.
Three approaches to delineate FFM antecedents in children and

adolescents. In considering the history of research on child personality (and
childhood traits in general), it is crucial to distinguish research that relies upon
theory-driven construction methods from research that does not depart from such
a priori theoretical assumptions. Obviously, research using bottom-up strategies
to delineate the structure of childhood traits has the potential to provide stronger
and less biased support for the notion that the FFM supplies a valid framework
for describing childhood traits. In the search for the “Little Five” in nonadult
groups, three approaches have been used, all of which vary in their adoption of a
top-down or bottom-up rationale. Each of these approaches has successfully
retrieved FFM-like dimensions in caregiver reports (see Table 12.1) and, to a
lesser degree, in self-ratings and peer-ratings of children (see Table 12.2).
Top-down: Downward adapting adult-oriented measures. The first strategy
is clearly top-down, departing from the a priori assumption that the FFM will
provide valid trait descriptions in nonadult age groups. This approach sees no
problems in stretching the FFM model and its adult-based measures by applying
it to younger groups. When interpreting results of this type of research, it must
be kept in mind that results are strongly driven by the a priori hypothesis that the
FFM structure is replicable in nonadults. As such, there is a risk of confirmation
bias and a potential lack of opportunities for falsification of the hypothesis. On
the other hand, this approach has the major advantage that it makes it possible to
use the same scales across different age periods. This is a great asset for scholars
wanting to make comparisons across developmental age groups (McCrae, Costa,
& Martin, 2005).
Roughly, this top-down methodology has two versions. In the first version,
FFM instruments developed for adults are administered without major
modifications to younger age groups. In the second version, instruments that
have a clear link to FFM instruments developed for adults are modified
extensively to warrant that the item content suits the behavioral repertoire of
younger groups.
The oldest examples of the first version of top-down research (i.e., using adult
measures in nonadult groups without major modifications) are based upon FFM
adjective markers, stemming from the psycholexical research tradition in adults
(Goldberg, 1990). These adjective lists have now been widely applied as rating
forms for teachers and parents to describe children even as young as age 4.
Factor analyses on these commonly short lists (e.g., 25 to 50 adjectives)
generally support a five-dimensional structure underlying caregiver reports in
childhood and young adolescence, closely resembling the Big Five structure in
adults (Branje, Van Aken, Van Lieshout, & Mathijssen, 2003; Graziano et al.,
1997; Klimstra, Hale, Raaijmakers, Branje, & Meeus, 2009; Mervielde, Buyst,
& De Fruyt, 1995; Vermulst & Gerris, 2005). In teacher ratings of 4- and 5-year-
old children, however, research has had difficulties in retrieving separate
Intellect and Conscientiousness dimensions (Mervielde et al., 1995).
Because adjective lists are less complex and time-consuming than full
questionnaire statements, they have also been a preferred medium in studies
exploring how (pre)adolescents perceive their own and (their peers’)
personalities (Klimstra et al., 2009; Mervielde & De Fruyt, 2000; Scholte, Van
Aken, & van Lieshout, 1997). Such research has replicated clear FFM structures
underlying self-ratings in adolescents from 12 years and older (Klimstra et al.,
2009; Scholte et al., 1997). The structure of young adolescents’ personality
ratings of their peers on Big Five adjective lists in peer nomination designs,
however, appears less differentiated (e.g., Mervielde & De Fruyt, 2000; Scholte
et al., 1997). In a study critically examining peer nomination procedures,
Mervielde and De Fruyt (2000) retrieved a three-factor structure in children’s
peer nomination scores on bipolar personality adjectives: The first factor
included a combination of Extraversion and Emotional Stability, the second
factor represented a combination of Conscientiousness and Intellect, and the
third factor clearly represented Agreeableness. This lower differentiation in
children’s perceived personality evaluations might reflect more limited cognitive
capacities in judging others, but could also denote methodological problems
associated with the selected nomination scales or peer nomination procedure
(Mervielde & De Fruyt, 2000).
Next to adjective lists, it is now fairly common to administer adult
questionnaires without major item-phrasing modifications to nonadult groups.
This approach is particularly popular in obtaining self-ratings in older
adolescents, but is also sporadically used to obtain caregiver reports of
youngsters’ personality (see, e.g., Markey et al., 2002; Martin & Friedman,
2000; Parker & Stumpf, 1998). In the past 15 years, several studies have
successfully demonstrated that (pre)adolescents are capable of providing reliable
and valid judgments of their own personality on common adult-oriented FFM
measures. Replications of the FFM structure in older children and adolescents
are now available for self-report versions of the NEO inventories (De Fruyt et
al., 2009; McCrae et al., 2002; Parker & Stumpf, 1998), the Big Five Inventory
(BFI; Denissen, Geenen, Van Aken, Gosling, & Potter, 2008; Soto et al., 2011),
and the Five Factor Personality Inventory (FFPI; Hendriks, Kuyper, Offringa, &
Van der Werf, 2008; Szirmak, 2005), among others. Taking into account the
critique that some of the NEO Personality Inventory-Revised (NEO PI-R) items
are too difficult for use in adolescents as well as in adults with lower educational
levels, 38 of 240 original NEO PI-R items have recently been reformulated in
the NEO’s third revision. In the past years, this NEO-PI-3 and its shorter
counterpart, NEO-FFI-3, have extensively been validated as a self-report
instrument in adolescence, across multiple cultures and languages (e.g., De Fruyt
et al., 2009). The original U.S. manual officially proposes the NEO-set of
questionnaires for acquiring personality self-ratings in adolescents aged 12 years
and beyond. Norms are also available, but only for a broad reference group, aged
12 to 20 years. Other NEO manuals such as the Dutch version (Hoekstra & De
Fruyt, 2014) set the age limit more conservatively to 16 years. Shorter FFM
instruments such as the BFI (44 items) and FFPI (100 items) have now been
validated for self-report in even younger age groups. Studies have shown that the
presupposed FFM structures can be replicated in self-reports as early as age 10
years, even though internal consistencies and differentiation among FFM
domains tend to improve with age across late childhood and adolescence
(Hendriks et al., 2008; Soto et al., 2008, 2011).
A second, alternative version of the top-down approach is to adapt existing
adult-oriented FFM measures, explicitly modifying item phrasing and rating
instructions to make them more suitable for child or adolescent assessment. This
approach draws content for the prespecified FFM dimensions from two major
sources. First, adult FFM item pools [such as the trait adjectives of Goldberg
(1999), the International Personality Item Pool of Goldberg (2001b), or the NEO
items of Costa and McCrae (1989)] are screened for personality descriptions that
are thought to also apply to children’s behavioral repertoire. Second, these
prestructured FFM dimensions are further supplemented by newly written child-
or adolescent-oriented items (for instance, suggested by a panel of
developmental experts such as classroom teachers) to attain a more age-
appropriate coverage of the FFM domains. Notable examples of this approach
are the Adolescent Personality Style Inventory which was developed specifically
for adolescent self-report (Lounsbury et al., 2003), the Five Factor Personality
Inventory for Children and its experimental precursor versions (Ehrler, Evans, &
McGhee, 1999; McGhee, Ehrler, & Buckhalt, 2007), and the Big Five Inventory
for Children, which is one of the few questionnaires available today that is
simultaneously validated across self-, parent-, and teacher-ratings (Barbaranelli,
Caprara, Rabasca, & Pastorelli, 2003).
Whereas the nonadapted adult measures usually target adolescence and rely
on self-ratings, these downward extensions generally focus on middle childhood
(age 9 years and beyond) and rely on caregiver-reports as well as self-reports.
Recently, researchers adopted a similar approach to construct a caregiver
instrument to address FFM personality in preschool age (Grist, Socha, &
McCord, 2012). The M5-PS-inventories (a 90-item and 35-item version) were
developed on the basis of a long list of traits that the authors derived from
Goldberg’s International Personality Item Pool (2001b) as they considered these
traits to be relevant for preschoolers. In a second step, these items were rewritten
to fit preschoolers, using expert panels of kindergarten teachers to validate or
supplement these descriptions. Studies with the M5-PS replicated four of the
FFM factors in teacher ratings of preschoolers. Only the Openness scale
performed poorly in terms of internal consistency and convergent validity (Grist
et al., 2012).
TOP-DOWN AND/OR BOTTOM-UP: DERIVING FFM PROXIES FROM

EXISTING ASSESSMENT MEASURES
The second strategy derives FFM marker scales from existing instruments that
specifically target child development (rather than adult development) but were
originally developed outside the FFM sphere of influence. This strategy also
comes in two versions. A first version is based upon a-theoretical lists of generic
descriptions of youth’s development. A second version derives FFM scores from
measures that were originally operationalizing other theoretical models. To date,
research has made a strong case that the dimensions underpinning these rich,
child-focused item pools show striking similarities to the FFM domains in
adults, based upon analyses inspired by both the top-down and bottom-up
rationale.
The best known and most influential example of the use of a-theoretical
generic lists of child development can be found in the work of John Digman.
Digman was among the first to address the complexity of childhood personality
while searching for the basic structure underlying personality in middle
childhood. From 1959 to 1967, Digman asked 88 teachers to make judgments of
personality attributes of 2,572 pupils aged 7 to 13 years. He did so in a series of
six data collections in Hawaii. These lists were composed of 36 to 63 personality
attributes, phrased as a single word or short phrase (e.g., Energetic: active; full
of pep; vigorous; movements are quick, darting). In some samples, these
attributes were accompanied by a definition developed beforehand from focus
groups in which teachers were asked to provide typical examples of classroom
behaviors relating to that concept. In Digman’s first analyses of these data (in an
era in which computerized data-analytical applications were still in their
infancy), he suggested that seven or eight factors would underlie these child
personality attribute ratings. Later, Digman and Takemoto-Chock (1981),
Digman and Inouye (1986), and Digman and Shmelyov (1996) carried out other
studies based on similar lists of teacher ratings. They finally reported five
recurring factors, across multiple cultures, that showed remarkable similarity to
the Big Five dimensions retrieved in adulthood.
In 2001, Lew Goldberg thoroughly reanalyzed the original Hawaiian datasets
of Digman, which are so unique because these trait descriptors were chosen long
before the FFM became the “sovereign” personality framework in adults. Using
bottom-up-inspired data-analyses (principal components analysis and
exploratory factor analysis), Goldberg (2001a) concluded that in all six
Hawaiian samples, a clear and robust five factor structure emerged, and, that no
other domains than the Big Five factors surfaced from these different datasets.
This research provides compelling evidence for the validity of the Big Five
model for teacher-based personality assessment in middle childhood.
Other scholars have also used the FFM as a taxonomic framework to organize
generic descriptions collected long before the FFM became the dominant
personality paradigm, enabling the study of child personality in and across
archival datasets. For example, Shiner (2000) derived child personality markers
similar to four of the FFM dimensions (not Neuroticism) using multi-informant
and multi-method descriptions of children’s competences in various
developmental task domains that were collected in the Project Competence
longitudinal study (Masten & Tellegen, 2012). These four traits, labeled
Agreeableness, Mastery Motivation (akin to Openness/Intellect), Academic
Conscientiousness, and Surgent Engagement (akin to Extraversion), were
highlighted as important contributors to resilience and adjustment across time.
Recently, Shiner and Masten (2012) reorganized the item groupings of these
original child personality scales to attain even more clearly aligned, reliable, and
valid FFM scales. These top-down modifications produced purer Extraversion
and Openness proxies, in addition to the already established scales for
Agreeableness and Conscientiousness. Shiner and Masten (2012) were also able
to construct a proxy scale for child Neuroticism by selecting items from a
generic behavior questionnaire rated by the parents. This reanalysis for all five
FFM domains revealed that individuals who were rated around 10 years of age
with more favorable traits on four of the FFM domains (except Extraversion)
had the most favorable developmental outcomes in both emerging (around 20
years) and young adulthood (around 30 years of age).
Research based upon this second approach (deriving FFM-proxies from
existing, a-theoretical developmental questionnaires) has strongly supported the
validity of the FFM as a model to describe individual differences in children. In
addition, this approach has had tremendous value in establishing the relevance of
child personality, its longitudinal consistency, and its impact on important life
domains, even though the original measurements were not operationalizing
FFM-based child personality. Long-term longitudinal studies based on this
approach have uncovered the vital importance of child personality for
physiological dysfunction, health, or mortality risk (e.g., Hampson & Goldberg,
2006; Kern, Hampson, Goldberg, & Friedman, 2014) and for psychosocial
(mal)adjustment and resilience (Masten & Tellegen, 2012; Shiner & Masten,
2012), even across long stretches of time.
This second approach also includes the regrouping of items based upon
alternative theoretical models. The California Child Q-set (CCQ; Block &
Block, 1980) and its Common Language adaptation (Caspi et al., 1992) is the
most notorious example of this strategy and can be considered as the most
popular instrument in the history of child personality assessment. In contrast to
the a-theoretical lists that offer a more broad perspective on child development,
the (Common) CCQ sort was developed within the clear theoretical model
postulated by Jack Block and Jeanne Block (1980). This model distinguishes two
major dimensions. The first dimension, ego-control, refers to a child’s tendency
to contain rather than to express emotional and motivational impulses. The
second dimension, ego-resiliency, refers to a child’s tendency to respond flexibly
rather than rigidly to changing situational demands or particularly stressful
situations. The CCQ is composed of 100 behavioral statements (to be sorted in a
fixed Q-sort distribution) that target the description of personality, cognitive, and
social attributes of children in this ego-control/ego-resiliency framework.
John et al. (1994) were the first to evaluate whether the FFM domains and/or
additional factors could be replicated in the structure of Common CCQ ratings.
They analyzed data from the Pittsburgh Youth Longitudinal Study, focusing on
early adolescent boys (12 years) and using maternal ratings. They investigated
this issue from both a top-down and bottom-up approach. First, a lucid top-down
strategy was used: The authors aimed to construct rationally developed FFM
scales for use with children and adolescents. To this end, they had two coryphaei
of FFM research in adults (Oliver John and Robert McCrae) independently
select those CCQ items that they judged to be clearly related to the FFM in
adults. These five FFM scales were composed of 48 CCQ items and showed
good to excellent validity (even though Openness-to-Experience had the smallest
number of items and showed the lowest reliability).
All FFM scales were found to be associated meaningfully with a wide array of
external criteria. This study was introduced as “discovering the Little Five” in
young adolescents while providing a first validation of its nomological network.
To date, these top-down-derived FFM CCQ scales have been widely adopted.
Research based on these scales has strongly contributed to the growing evidence
that childhood personality is a major predictor of cognitive, academic,
behavioral, as well as mental health outcomes, both concurrently (e.g., John et
al., 1994; Van Lieshout & Haselager, 1994) and longitudinally (e.g., Abe, 2005;
Abe & Izard, 1999; Asendorpf & Van Aken, 2003; Gjerde & Cardilla, 2009;
Lamb, Chuang, Wessels, Broberg, & Hwang, 2002).
Interestingly, John et al. (1994) also adopted a bottom-up perspective to
address the broader issue of adolescent personality structure by performing
principal component analyses on the entire 100 CCQ item set. Notably, they
recovered a variant of their proposed Big Five, but they also identified two
additional factors. Three factors were found that differed little from the top-
down proposed FFM scales: Agreeableness, Conscientiousness, and Openness-
to-Experience. Interestingly, the two other FFM domains were defined more
narrowly, and were labeled Sociability (instead of Extraversion) and Anxious
Distress (instead of Neuroticism). In addition, two complementary factors were
recovered, labeled Activity and Irritability. Sociability was defined by elements
such as sociability and expressiveness, whereas energy, social presence, and
activity level (traits typically loading extraversion in adults) loaded the sixth
factor, Activity. Anxious distress represented a more limited range of negative
affects (e.g., anxiety, nervous worry, guilt, low self-esteem) than typically found
on adult Neuroticism factors. The Irritability factor was defined by more
inappropriate and immature negative affect regulation as reflected in whining,
crying, tantrums, excessive sensitivity to teasing, frustration, and irritability.
Hence, these bottom-up analyses suggest that both the Extraversion and
Neuroticism constructs as found in adults might be more differentiated in
youngsters, hinting that Extraversion splits out in two distinct traits (Sociability
and Activity) whereas Neuroticism is represented in two distinct aspects of
negative affect (Anxious versus Irritable Distress). John et al. (1994)
recommended the further study and replication of these two additional factors in
other samples and instruments because they thought it was unlikely that these
two factors were error factors, instrument-specific factors, or just facets of the
five classic FFM dimensions. They considered both factors as good candidates to
represent relatively independent dimensions in young adolescence or even
childhood.
In the same year, Van Lieshout and Haselager (1994) performed similar
principal component analyses on the Dutch translation of the CCQ set using
parent as well as teacher ratings of 720 children and adolescents. In this broader
age group (3 to 16 years) these authors also recovered a variant of the FFM and
two additional factors, which they interpreted as Activity (strongly similar to the
one reported by John et al., 1994) and Dependency (not found by John et al.,
1994), a trait that reflects emotional overreliance on significant others (e.g.,
eager to please, manipulative). The separate Irritability trait was not found in the
factor structures across age groups. Yet, Irritability-related CCQ items primarily
loaded on the Agreeableness factor. The factor Neuroticism (labeled Emotional
Stability) was primarily indicated by anxious distress and low self-worth. Van
Lieshout and Haselager (1994) also provided unique analyses of the
multidimensional factor structure of self-ratings and peer-ratings (made by best
friends) of 158 early adolescents. Compared to adult person descriptions,
however, adolescent peer-descriptions and self-descriptions fitted the FFM less
clearly. Both in self-reports and best friends’ descriptions, the strongest support
was found for the factors Agreeableness, Emotional Stability, and
Conscientiousness. In best friends’ person descriptions, a combined Openness–
Activity factor was found, but a clear Extraversion factor was absent. In self-
ratings, traces of Extraversion were found to be present (especially Sociability
and Social Attractiveness versus Social Withdrawal) in addition to a separate
Activity factor. Notably, a clear Openness factor was not found.
Even though this work on the CCQ is a prominent example of the approach
that derives FFM proxies from questionnaires operationalizing alternative
theoretical models, other examples are available. For example, Resing,
Bleichrodt, and Dekker (1999) used principal component analyses to evaluate
the multidimensional structure underlying the School Behavior Checklist of Zaal
(1978), an instrument targeting child adjustment in the field of educational
practice. They recovered a four-dimensional structure, which they interpreted in
terms of four of the FFM dimensions (Extraversion, Attitude toward School
Work, Agreeableness, and Emotional Stability). Interestingly, a separate
Openness component could not be established in this study.
In the past decade, the approach of deriving FFM proxies of alternative
theoretical models has even expanded its scope beyond instruments targeting
child or adolescent development. For example, studies have now successfully
developed FFM markers from Tellegen’s originally adult-oriented
Multidimensional Personality Questionnaire (Tellegen & Waller, 2008) for use
with children and adolescents (e.g., Cukrowicz, Taylor, Schatschneider, &
Iacono, 2006; Tackett, Krueger, Iacono, & McGue, 2008). These proxies are
labeled as Negative and Positive Emotionality (akin to Neuroticism and
Extraversion), Agreeableness, Constraint (akin to Conscientiousness), and
Absorption (akin to Openness-to-Experience).
Bottom-up: Constructing new taxonomies from free description research.
The two approaches discussed so far have been of utmost importance in the
process of validating FFM dimensions in nonadult groups. The top-down
strategies (e.g., the rational FFM scale development, downward FFM
adaptations) have provided the most unambiguous replications of the FFM in
nonadult groups. Other strategies, also including elements of a bottom-up
approach, have yielded more problems in identifying all five domains (Openness
in particular) or have identified more than five domains. Moreover, from Tables
12.1 and 12.2, it can be inferred that although all retrieved traits can be
organized easily within the FFM framework, not all of these retrieved factors
may be qualitatively similar. In this regard, the range of diverse labels chosen by
scholars interpreting their factor structures and describing the resulting scales
should be noted: for example, how similar are scales such as Openness,
Resourcefulness, Absorption?
As there is no guarantee that semantically similar labels actually share the
same trait variance, further empirical validation is needed to avoid the “jingle-
jangle” fallacy (Block, 1995). This phenomenon reminds trait psychologists that
they sometimes study the same trait under different names (jingle) or use the
same label to describe different traits (jangle). Unfortunately, surprisingly few
studies have addressed the empirical relationships between multiple coexisting
child personality measures, so that the empirical convergence remains for a large
part unknown (some exceptions can be noted: Barbaranelli et al., 2003; De Fruyt
et al., 2000; Parker & Stumpf, 1998; Soto & John, 2014; Tackett et al., 2013).
There are other problems with these first two approaches too. Most
importantly, both approaches do not provide any warranty that the personality of
children or adolescents is assessed optimally (De Fruyt et al., 2000). As the first
approach imposes the structure of the adult FFM to nonadult groups and as the
second approach derives scores from scales not designed for FFM assessment,
both approaches run the risk that other factors, beyond the FFM, are important
for the assessment of individual differences in childhood and adolescence.
Therefore, a third approach aims to start from a conceptually blank slate and
aims to develop new taxonomies of childhood traits from the bottom-up, based
upon careful observation of the wide range of behavioral differences manifested
by children and adolescents.
Examples of such pure bottom-up approaches are studies relying on free-
response techniques. Research has found evidence that the FFM can be used to
organize and classify a substantial portion of children’s free descriptions of their
own and peers’ personalities as elicited by projective personality tests (Donahue,
1994). More impactful, however, has been the research on parental and teachers’
free descriptions. Regarding teacher descriptions, Mervielde (1994) showed that
more than 60% of 3,265 personal constructs, generated by 226 teachers, can be
clearly classified within the FFM framework. Regarding parent descriptions, a
large international collaborative research project was initiated by Kohnstamm,
Halverson, Mervielde, and Havill (1998). This project assembled an immense
database of parents’ free descriptions of their children in and across seven
countries, which were subsequently coded on the basis of an intuitive, rational
coding system (yet not entirely unaffected by FFM premises). A key finding was
that—across countries—70% to 80% of descriptors can be easily classified
within the first five categories of the lexicon. Markedly, these first five
categories show strong correspondence to the FFM in adults. The research
consortium noted, however, that this finding alone is insufficient to fully
establish the FFM’s validity to describe childhood individual differences
because these results are still contingent on a prestructured classification process
(Kohnstamm et al., 1998; Mervielde et al., 2005).
Hence, more work is required to illuminate the underpinnings of the
personality dimensions relevant for a specific age group. In the wake of the
international classification study, two research groups undertook the endeavor to
thoroughly structure and classify the raw material of free parental descriptions,
collected for children from multiple age groups. In doing so they used a bottom-
up approach. This method is exceptional as it provides a unique occasion to
disentangle the lower versus higher levels of child personality taxonomies. In
adulthood, similar studies of the natural, culture-specific language to describe
personality have greatly helped to clarify the taxonomical framework of
personality (John et al., 2008). These time- and labor-intensive projects resulted
in two childhood trait taxonomies, from which two influential personality
instruments were developed: the Hierarchical Personality Inventory for Children
(HiPIC; Mervielde, De Fruyt, & De Clercq, 2009) and the Inventory for
Childhood Individual Differences (Deal, Halverson, Martin, Victor, & Baker,
2007; Halverson et al., 2003).
For the construction of the HiPIC, more than 9,000 Flemish parental
descriptions were collected and organized into 100 homogeneous clusters that
covered three age groups: 5 to 7, 8 to 10, and 11 to 13 years. Notably, for each of
these age groups, principal component analyses at the item level clearly tended
to cluster items according to FFM-like dimensions. Additional analyses were
done to explore the facet structure within each of these five dimensions in the
three age-specific item sets. In each age-specific set, 18 highly reliable facets
were found. Given the substantial overlap in content across these three groups,
the three age-specific item sets were merged into one item set measuring
personality in primary school children (Mervielde & De Fruyt, 1999; Mervielde
et al., 2005).
The 18 HiPIC facets primarily load on five broadband dimensions, labeled as
Conscientiousness, Benevolence, Extraversion, Imagination, and Emotional
Stability. These five are clearly related, yet not identical, to adult FFM domains.
Conscientiousness (grouping the facets Achievement Motivation, Concentration,
Perseverance, and Orderliness) and Extraversion (grouping the facets Shyness,
Optimism, and Expressiveness, but also Energy—which did not acquire the
status of a separate factor in this study) can be considered as closest to their adult
counterparts. Benevolence appears broader in content than adult Agreeableness:
the facets Egocentrism and Irritability have the highest loading on this factor
[reminiscent of the Irritable Distress factor of John et al. (1994)], in addition to
Dominance (a typical Extraversion facet in adults), Compliance, and Altruism
(reflecting the purest Agreeableness content). Imagination is more limited than
adult Openness because it groups only Intellect, Curiosity, and Creativity.
Finally, the smallest HiPIC component is Emotional Stability, grouping an
Anxiety facet and one measuring Low Self-Confidence [content reminiscent of
the Anxious Distress factor of John et al. (1994)].
Although the HiPIC was originally targeting primary school age, its factor
structure has now been recovered in parent ratings of preschool children (De
Pauw et al., 2009) as well as in parent- and self-ratings of adolescents (e.g., van
den Akker, Dekovic, Asscher, & Prinzie, 2014). Moreover, a joint principal-
component analysis of the HiPIC and NEO PI-R (Costa & McCrae, 1992)
largely confirmed the proposed correspondence of HiPIC and NEO PI-R
relations using adolescent self-reports (De Fruyt et al., 2000). Only two HiPIC
facets, Dominance and Irritability, as rated by adolescents aged 12 to 17 years,
were situated more at the intersection of Benevolence–Extraversion and
Benevolence–Neuroticism, respectively, than documented in parent-report data
(where these two facets uniquely load on Benevolence). This finding suggests
that these two traits might migrate from childhood to adulthood, taking a more
intermediate position in adolescent personality taxonomies (De Fruyt et al.,
2000).
Even though the construction of the Inventory for Childhood Individual
Differences (ICID) is related to the HiPIC, it followed a more theory-driven
process because the classification of descriptors was closely aligned with the
intuitive rational FFM-coding system that was developed for the international
collaboration project (Kohnstamm et al., 1998). The ICID is based upon free
parental descriptions collected from a U.S. sample and items imported from
similar instruments constructed in the Netherlands, Greece, and China
(Halverson et al., 2003; Mervielde et al., 2005). The initial questionnaire resulted
in 144 items that matched the distribution of the FFM phrases in the parental
lexicon. The factor structure of this original ICID globally resembled the adult
Big Five, but also showed some deviations, such as a restricted Intellect factor
(which was labeled as Openness but includes only one facet, Intellect), a
broadened Extraversion factor (with primary loadings of the facets Openness and
Considerate), as well as a Neuroticism factor primarily indicated by Shyness
(Halverson et al., 2003; Mervielde et al., 2005). In 2007, the ICID was reduced
and refined into a 108-item instrument (Deal et al., 2007) assessing 15 mid-level
scales that show a purer factoring into the FFM. Still, five facet scales (Positive
Emotions, Negative Affect, Compliant, Distractible, and Intellect) are assigned
to more than one FFM domain.
In accordance with the goals and results of the international collaborative
project, the ICID domains are labeled with exact FFM nomenclature:
Neuroticism (comprising Fearful, Negative Affect, and Distractible as facets),
Extraversion (grouping Sociability, Shy, Activity Level, and Positive Emotions),
Openness (capturing Intellect and Openness), Agreeableness (consisting of the
facets Considerate, Compliant, Positive Emotions, Antagonism, Strong-Willed,
and Negative Affect), and Conscientiousness (consisting of the facets Organized,
Achievement Oriented, Distractible, Compliant, and Intellect).
Are We There Yet? Strengths and Weaknesses of the Three Approaches

Each of the three discussed approaches clearly has its own assets and pitfalls.
The greatest advantage of the first, top-down, approach is that it enables
youngsters’ traits to be described according to the same reference structure as in
adults. This is a major asset for researchers interested in personality
development, because it allows them to perform both cross-sectional and
longitudinal comparisons across age (De Fruyt et al., 2009; McCrae et al., 2005;
Soto et al., 2008). Using a single set of constructs (perhaps with some items
adapted to age-specific population characteristics) across development, from
early childhood to late adulthood, might be the ultimate dream for charting
lifespan development in personality (John et al., 1994). However, there are
important caveats in simply imposing the adult FFM structure onto younger age
groups. For example, how far can we go in lowering the age limit of adult-
oriented FFM measures? Can we be sure that these item sets, derived from the
complex world of adults, are appropriate to describe children’s behavioral
reality? Finding structural invariance of adult measures in nonadult groups is not
sufficient to assert that these measures are also the “best” assessment of child
personality. Moreover, imposing an adult structure on nonadult groups may
obscure important age-specific features. Hence, this line of research should be
preferably complemented with more bottom-up approaches and by research
rigorously screening other developmental psychological constructs (De Fruyt et
al., 2000; Kohnstamm et al., 1998; Soto et al., 2008).
The second approach, deriving FFM proxies from alternative models,
provides strong evidence that FFM-like dimensions underlie the rich and
heterogeneous descriptions of children’s behavioral differences. However, these
scales can only be considered as proxies for the FFM as they are restricted by the
theoretical framework of the original instrument or by the a priori selection of
the researcher. In the CCQ, for example, content related to Agreeableness is
overrepresented whereas content related to Openness-to-Experience is
predominantly restricted to intellectual and imaginative tendencies (Gjerde &
Cardilla, 2009; Soto & John, 2014). Hence, the derived scales do not necessarily
reflect child characteristics that are perceived as salient by observers at different
ages. Moreover, it is difficult to deduce whether retrieved personality factor
structures actually reflect child features or simply result from instrument-specific
idiosyncrasies. Furthermore, other important traits beyond Big Five may be
relevant in younger age groups and are risked to be overlooked by this method.
The third approach is a remarkable illustration that the development of five
factor personality theory in childhood is not restricted to theory-driven
construction methods and hence cannot merely be considered as an extension of
adult personality theory. As evidenced by the construction histories of the HiPIC
and, to lesser extent, the ICID, this bottom-up approach relies in a unique way
on an age-specific informant-based approach to instrument construction.
Consequently, this approach may be considered to provide the most
comprehensive description of individual differences in childhood (Soto & John,
2014; Tackett et al., 2013). It also yields the most compelling evidence for the
validity of the FFM in childhood. Moreover, both the HiPIC and ICID
taxonomies are exceptional in contemporary child personality assessment
because they include lower-level facets. In adults, the hierarchical framework of
the FFM is well-established, with each FFM dimension subsuming several, more
specific lower-order facets. Even though same-dimension facets are substantially
correlated (around .4 in adults), correlations are not nearly perfect. Thus, lower-
order facets provide unique information and reveal distinctive, or even
idiosyncratic, developmental changes (Caspi et al., 2005; Soto et al., 2011).
Likewise, recent research with the HiPIC and ICID facets provides crucial
information on the development (e.g., Deal, Halverson, Havill, & Martin, 2005;
de Haan, De Pauw, Van den Akker, & Prinzie, under review) and differential
predictive impact (e.g., Becht, Prinzie, Dekovic, Van den Akker, & Shiner,
2015; Prinzie, van Harten, Dekovic, van den Akker, & Shiner, 2014) of the more
specific, concrete, lower-order facets.
Yet this third approach also has drawbacks. Researchers interested in
addressing childhood personality with this approach have no other choice but to
use different scales at different ages, which hampers comparisons in
developmental studies wanting to adopt a lifespan perspective (but note that
studies have found adequate properties for young adults’ HiPIC and ICID self-
and other-ratings; Deal et al., 2005; Van den Akker et al., 2014). Moreover, even
though HiPIC as well as ICID shows excellent reliability and validity when used
in adolescence (e.g., De Fruyt et al., 2000; Tackett et al., 2013; van den Akker et
al., 2014), both taxonomies were originally developed to represent the structure
of parental descriptions of young and preadolescent children (HiPIC: 5–13 years;
ICID: 3–13 years). To our knowledge, a similar attempt to chart the personality
of adolescents from a bottom-up approach has not been undertaken, so that it is
not entirely clear whether there are other salient facets typifying adolescents’
personality beyond those described in the HiPIC or ICID. For example, it is
feasible that given its prominence in the adolescent developmental literature, a
trait such as “Self-Worth” (Harter, Waters, & Whitesell, 1998) would play a
much more pronounced role in bottom-up adolescent personality taxonomies
(De Fruyt et al., 2000). From a more practical point of view, the length of the
HiPIC and ICID measures is sometimes considered a barrier. For the ICID, a
shorter, validated version is already available, whereas for the HiPIC, a shorter
version for research purposes is currently under construction (De Fruyt &
Mervielde, 2013).
Child versus Adult Personality Taxonomies

Toward a Better Understanding of Child Personality Traits in Their Own
Right
Even though the quest for the most optimal child personality instrument is not
settled yet, our review reveals that researchers and practitioners today can
choose from a wide array of measures to address FFM(-like) traits in childhood
or adolescence. More intriguing, however, is the more fundamental discussion
about the basic architecture of children’s and adolescent’s personality. In this
section, the issue of child personality structure is addressed more deeply by
describing the apparent differences between child and adult personality
taxonomies, based on the research reviewed so far.
Are there traits above and beyond the FFM? Our state-of-the-art review of
child personality measures summarizes compelling evidence that childhood
personality models, developed in the tradition of the FFM model in adults
(McCrae & Costa, 1987), can validly and reliably describe the structure of
individual trait differences of adolescents, preadolescents, primary school
children, and even preschool children (e.g., De Pauw et al., 2009; Grist et al.,
2012). These findings are vitally important as they counter the classic allegation
that “temperament” would be the childhood precursor of later developing
“personality.” Instead, these findings emphasize that childhood traits, from
preschool age onward, can be rightfully considered as personality (Caspi &
Shiner, 2006; De Pauw & Mervielde, 2010; Shiner & DeYoung, 2013).
On the other hand, the contention that the FFM provides the “best”
comprehensive model to describe individual differences in children and
adolescents has gradually been toned down in recent years. Even though all five
FFM dimensions have clear relevance to capture behavioral individuality in
youth, scholars should attend more to the developmental nuances of personality
structure across childhood and adolescence (De Pauw et al., 2009; Soto & John,
2014; Tackett et al., 2013). From our review, it becomes evident that the
strongest and purest replications of the FFM structure are typically found in
studies in (pre)adolescence (starting from approximately age 10 years) that adopt
top-down approaches, with clearly prestructured FFM measures. Yet, even some
of these top-down studies have had difficulties in retrieving all five of the FFM
dimensions (Openness-to-Experience in particular). Moreover, the bottom-up-
driven approaches emphasize that the basic structure underlying childhood traits
looks less similar to the adult structure, particularly when considering the
younger age groups (i.e., primary school and preschool age).
Therefore, we concur with Tackett and colleagues’ recent call that “the study
of child personality should move forward to gain a better understanding of child
personality traits in its own right, rather than assuming child personality traits
are merely identical to those measured in adults” (Tackett et al., 2013, p. 739).
Slowly but steadily, the outlines of the differences between adult and child
personality taxonomies become more visible. First, there is increasing evidence
that “Activity” should be regarded as an important additional factor in youth.
Second, evidence suggests that there are developmental nuances in the content of
the other five classic FFM domains.
FROM LITTLE FIVE TO LITTLE SIX? THE CASE OF ACTIVITY

Recently, Soto and John (2014) published a groundbreaking paper in which
they profoundly reevaluate the multidimensional structure of parental CCQ
ratings from an explicit bottom-up perspective. They applied Goldberg’s (2006)
“bass-ackwards” method to analyze the foundational hierarchical structure of
parental CCQ ratings of 16,000 children, adolescents, and young adults. These
ratings were sampled from a large online platform in which parents completed a
slightly adapted Common CCQ in a questionnaire format (instead of the fixed Q-
sort). The authors sampled 500 males and 500 females for each individual year
from age 3 to 17 years, and from a broader young adult group aged 18 to 20
years, hence enabling the cross-sectional comparison of trait structure across 16
age groups. Goldberg’s (2006) “bass-ackwards” method makes it possible to
examine the hierarchical structure based on the interrelations between principal-
component scores from rotated solutions. Beginning with one dimension, each
next analysis extracts and rotates an extra dimension. The correlations among the
component scores from adjacent levels are then interpreted as path coefficients
in a hierarchical structure.
This study is unique because it provides very detailed and comprehensive
insights in the hierarchical structure of personality traits from early childhood to
early adulthood. Its findings emphasize that the personality trait structure of
youths is clearly hierarchically organized, but that significant similarities as well
as differences between youth and adult personality hierarchies should be noted.
Most importantly, Soto and John conclude that, from middle childhood onward
(age 6 years and beyond), children’s personality hierarchy rests on a foundation
of six, rather than five, dimensions. Next to traits interpreted as Extraversion,
Agreeableness, Conscientiousness, Neuroticism, and Openness-to-Experience,
Activity emerged systematically as a major factor across the structural models.
Accordingly, the authors proposed that these “Little Six” are to be preferred
above the “Little Five” when speaking about child and adolescent traits. These
Little Six proved to be more replicable than the Big Five across childhood and
adolescence. However, Soto and John (2014) noted that the Big Five structure
became more replicable with age and was clearly recovered in the late
adolescence and early adult age groups (ages 15 to 18–20 years). In the 18–20
year age groups, the Big Five was even more plainly recovered than the Little
Six. Activity, in particular, was less apparent in the emerging adult group, and
content related to this trait was subsumed by Extraversion as well as
Conscientiousness in the Big Five model. These findings imply that during the
transition to adulthood, the personality hierarchical foundation may shift from a
Little Six structure to a Big Five structure (Soto & John, 2014).
Importantly, Soto and John (2014) also reported that fewer and simpler
structures replicated more easily in the early childhood samples (ages 3 to 5
years). Only three structures were replicated in these young age groups: a one-
dimensional structure (labeled Adjustment), a two-dimensional structure
(discerning Benevolence from Emotionality), and an alternative five-
dimensional structure. In this latter structure, versions of Agreeableness,
Neuroticism, Extraversion, and Conscientiousness are retrieved, in addition to a
dimension that combines aspects of Activity and Openness-to-Experience. Soto
and John (2014) interpret this finding as preliminary support for the theoretical
hypothesis of Caspi et al. (2005) that Openness-to-Experience may begin to
develop early in life, but due to ongoing cognitive development, its conceptual
core may shift from exploration of the physical world in early childhood to
intellectual curiosity and imagination by middle childhood and early
adolescence. Hence, openness in young children would not only be defined by
cognitive characteristics (e.g., imagination, creativity), but also by aspects of
motor activity and physical exploration.
Even though this study’s results are confined by their cross-sectional nature
and the use of only parent reports on a particular measure (CCQ), it is a
landmark for the child personality discipline, as it gently challenges the bold
proposals of earlier times (e.g., Caspi et al., 2005; De Pauw & Mervielde, 2010;
Digman, 1994; Mervielde et al., 2005) that a strict version of the FFM is the
“best” model to describe child and adolescent trait differences. Nevertheless,
more studies, particularly in younger age groups and based on longitudinal
designs, are needed to further substantiate the suggestions of Soto and John
(2014) that personality structure would follow a curvilinear developmental
trajectory, becoming more complex from early to middle childhood (when the
Little Six would emerge), and then less complex from adolescence to adulthood
(when they merge into the Big Five).
SIMILAR, NOT IDENTICAL: OUTLINING THE DIFFERENCES BETWEEN

ADULT AND CHILD TAXONOMIES
In addition to the growing recognition of Activity as a sixth fundamental
factor (Shiner & DeYoung, 2013; Soto & John, 2014), marked progress has been
made in uncovering differences between child and adult personality taxonomies
in the past decades. Here we summarize the most salient differences that have
been identified for the classic FFM domains, primarily in studies adopting the
bottom-up rationale.
The case of Extraversion/Sociability. The growing evidence that Activity
represents an independent dimension from at least middle childhood to middle
adolescence implies that Extraversion represents more narrow content in young
age (primarily capturing content related to Sociability) and that the meaning of
this trait gradually expands when getting older. The age-specific analyses of
Soto and John (2014), for example, illustrate that the meaning of Activity shifts
dramatically when middle childhood groups are compared to older adolescent
groups. Whereas during primary school age, Activity is indicated mainly by
physical aspects such as energy level and motor activity, at older ages this factor
includes more aspects related to psychological agency, such as motivation and
competitive drive. This shift may reflect a broader developmental process by
which Activity gradually merges with aspects such as sociability to form a
broader Extraversion component (e.g., De Pauw & Mervielde, 2010; De Pauw et
al., 2009; Eaton, 1994) but also by which Activity merges with achievement–
motivation aspects to form a broader Conscientiousness component (Shiner &
DeYoung, 2013; Soto & John, 2014).
Studies also demonstrated other developmental differences between child and
adult Extraversion. For example, in all bottom-up analyses, there is little
evidence for a separate Honesty–Humility trait in children, even though this is a
very prominent trait in adults (Soto & John, 2014). Another difference relates to
Assertiveness, or its counterparts in child taxonomies, Dominance (HiPIC) or
Strong-Willed (ICID). Whereas this is a salient facet of Extraversion in adults,
the HiPIC and ICID taxonomies allocate this content to the broader
Agreeableness/Benevolence factor. Clearly, more developmental research on
these diverse trait aspects is needed.
The case of Agreeableness/Benevolence. Bottom-up analyses suggest that the
childhood analogue of Agreeableness (i.e., Benevolence) captures a much
broader content than its adult counterpart. Specifically, this factor combines
content related to parents’ views of manageability of a child (i.e., low irritability,
low egocentrism–Neuroticism facets in adults—and low assertiveness—an
Extraversion facet in adults) with more typical content related to Agreeableness
such as altruism, compassion, and cooperativeness (De Pauw et al., 2009;
Mervielde et al., 2009; Tackett et al., 2008, 2012). A label such as Benevolence
might better reflect this heterogeneity in content than the classic label
Agreeableness (Mervielde et al., 2009).
It is remarkable that parental free description-based taxonomies primarily
identify negatively poled markers of this domain, reflecting
Disagreeableness/Antagonism. This is evidenced by HiPIC facets such as
Dominance, Egocentrism, and Irritability or ICID facets such as Strong-Willed
and Antagonism. In children, Agreeableness/Benevolence is also represented by
aspects of Agreeable Compliance, a feature less present in the typical FFM
models in adults (i.e., Compliance in HiPIC and Obedience in ICID). All these
traits primarily reflect salient differences in how well children can self-regulate
in the service of maintaining positive relationships with significant others and
how well they can inhibit their hostile and aggressive impulses (Shiner &
DeYoung, 2013). Notably, these bottom-up-derived taxonomies identify less
descriptors of individual differences in empathic emotional reactions to others’
emotions, concerns, and desires. Only one HiPIC facet (Altruism) and one ICID
facet (Considerate) reflect more prosocial tendencies, which are so central in
adult models of Agreeableness. It is interesting that in free descriptions research,
parents use more lexicon to denote the quality of the social interactions of their
children (are they obedient or stubborn, irritable or indulgent) than to describe
their children’s empathic responses to others.
In contrast to free description research, FFM-oriented analyses of existing,
generic instruments for child development have identified much more “pure”
markers of Agreeableness in children and adolescents. The analyses of both the
CCQ scales and the Digman scales emphasize that traits such as warmth,
kindness, considerate, helpfulness, empathy, friendliness, and trustworthiness
can readily be recognized, even in very young children (Goldberg, 2001a; Soto
& John, 2014). The inclusion of such pure “Agreeableness” content in the
bottom-up constructed inventories may have important added value for a more
comprehensive assessment of this childhood trait.
Interestingly, developmental psychologists have traditionally paid much
attention to identifying childhood markers of prosocial and aggressive
tendencies with relevance to the Agreeableness/Benevolence trait (Graziano &
Eisenberg, 1997; Knafo & Israel, 2012). Yet this line of developmental research
has been evolving in relative isolation of the work of other developmental
psychologists on temperament. Temperament theorists have typically not
considered Agreeableness-related traits (e.g., Aggression, Considerateness,
Empathy) in their models, perhaps because they have regarded these traits as
being products of socialization in conjunction with other temperamental
tendencies, rather than temperament per se (Shiner & DeYoung, 2013). Yet there
are multiple reasons to consider these traits relevant to
Agreeableness/Benevolence as being fundamentally like other temperament
traits. There is emerging evidence that such Agreeableness-related traits are
discernible early in life and show moderate stability throughout early childhood.
For example, 1-year-old children already vary in their display of physical
aggression and the average display of such aggressiveness peaks around the age
of 3 years (Tremblay et al., 2004). By preschool age, children show major
differences in physical aggression as well as in relational aggression and these
two types of aggression are strongly related in young children (Côté,
Vaillancourt, Barker, Nagin, & Tremblay, 2007). In addition, even children as
young as 14 months exhibit moderately stable and situationally consistent
differences in empathy and moderately stable tendencies toward prosocial
behaviors by the age of 3 years (Knafo, Zahn-Waxler, Van Hulle, Robinson, &
Rhee, 2008). Moreover, as discussed before, research using bottom-up
approaches has provided strong support that Agreeableness/Benevolence
constitutes a basic trait dimension, even from very early in development.
Developmental work on childhood analogs of Agreeableness/Benevolence
raises further interesting research suggestions. For example, developmental
scholars have emphasized that Antagonism/Aggressiveness and Prosocial
tendencies should not be considered to be opposite ends of a single dimension,
even though they tend to covary strongly and negatively throughout childhood
(Graziano & Eisenberg, 1997; Knafo & Israel, 2012; Shiner, 1998). In future
child personality taxonomies, more work can be done to unravel the potential
distinctiveness of these traits. Including these different aspects in a single
dimension might run the risk that the differential patterns of these traits are being
obscured (Shiner, 1998).
The case of neuroticism/emotional stability. Commensurate with the broader
content of Agreeableness/Benevolence, a more restricted Neuroticism factor in
childhood is suggested by bottom-up research. This trait tends to be indicated
primarily by content related to anxious distress and lower feelings of self-worth.
Given the key importance of Neuroticism in the development of mental health
and psychopathology, it is remarkable that parental free description research
derived only a few descriptors covering this domain (e.g., HiPIC Anxiety and
Self-Confidence, ICID Fearful and Negative Affect). A developmental
psychopathological perspective may provide a much wider lexicon to describing
nuances of Emotional Distress/Stability in youngsters. For example, it seems
relevant to differentiate Fearfulness (related to specific stimuli) from a more
general Anxiety disposition (De Pauw & Mervielde, 2010) or to differentiate
between more mood colors within a broader negative affect category (e.g.,
discomfort, sadness, misery). Adult Neuroticism is covered by a more diverse
range of features including self-consciousness and social anxiety, vulnerability
to distress, feelings of depression, sadness, despondency and loneliness,
impulsiveness (childhood analogs of this trait tend to load on
Conscientiousness), and angry-hostility (childhood analogs of this trait tend to
load on Benevolence). It is remarkable that similar traits have not been retrieved
from parental free description research, even though some of these markers have
been retrieved in factor analyses of the generic developmental questionnaires.
Hence, future research on childhood personality structure might look for a more
comprehensive conceptualization of this trait at both the positive and negative
poles. Finally, it is noteworthy that many FFM studies prefer the label Emotional
(In)stability instead of Neuroticism in describing this trait in youngsters,
possibly because this label has a less pejorative connotation to use in youth.
The case of Conscientiousness. Both top-down and bottom-up approaches
emphasize that Conscientiousness is a very robust childhood personality factor
covering basic tendencies such as impulse control, effortful attention, or task
persistence as well as individual differences in orderliness, dependability, and
the motivation to strive for high standards and to pursue goals over time in a
determined manner. This broader content of a Conscientiousness trait has been
documented in both top-down and bottom-up approaches and has been identified
as early as preschool age. It is likely, however, that these tendencies become
even more salient to other observers as children are faced with the demands for
more sophisticated work in primary school (Shiner & DeYoung, 2013).
Although this broad content of the Conscientiousness factor in childhood
meshes well with the adult Big Five domain, the social aspect of
Conscientiousness, as manifested in facets such as Dutifulness or Responsibility,
is not recovered in bottom-up taxonomic work in childhood (De Pauw et al.,
2009; Soto & John, 2014). Given the growing demands from society on
productivity and accountability during adolescence, it seems plausible that traits
such as dutifulness and responsibility become even more salient throughout
adolescence. Hence, they are not included in child taxonomies, but they could be
retrieved by future bottom-up endeavors entangling specific taxonomies
underlying adolescents’ personality. Evaluating further differences between
child and adult Conscientiousness, some researchers have noted that in
childhood, traits such as attentional and impulse control appear to be more
central in children than in adults (De Pauw et al., 2009; Soto & John, 2014).
The case of Openness-to-Experience/Intellect. The fifth FFM trait remains
the most heterogeneous FFM dimension of adult personality (John et al., 2008).
The longstanding debate about the best conceptualization of this trait as either
“Openness-to-Experience” or “Intellect” has significantly marked the early
decades of FFM or Big Five research. The questionnaire-based FFM tradition
preferred the term “Openness,” whereas the psycholexical Big Five tradition
preferred the term “Intellect” (see also the chapter by De Raad and Mlačić). This
historical differential preference for Openness versus Intellect is also reflected in
the nomenclature presented in Tables 12.1 and 12.2.
Also in childhood, Openness/Intellect remains the most controversial factor
(Gjerde & Cardilla, 2009; Herzhoff & Tackett, 2012; Mervielde, De Fruyt, &
Jarmuz, 1998). A first controversy relates to when this Openness/Intellect
dimension in children first emerges. In this regard, some scholars have argued
that Openness may not be a meaningful personality dimension prior to
adolescence (Lamb et al., 2002). Lamb et al. (2002) analyzed CCQ ratings from
a longitudinal study of 102 Swedish children, followed from 2.5 to 15 years of
age (across five waves), relying on the top-down FFM CCQ scales proposed by
John et al. (1994). Their study showed that the Openness-to-Experience scale
was adequately reliable in the parent Q-sorts of adolescents, but did not reach
adequate reliabilities when children were younger. In our review above, we have
pointed out that both in top-down and bottom-up approaches, several studies
reported low reliability for Openness/Intellect scales. Moreover, in some studies,
this trait does not emerge as a separate factor in instruments’ structural analyses
(e.g., Resing et al., 1999). In addition, particularly at younger ages, studies have
found that content related to Intellect tends to group together with content related
to Conscientiousness (Mervielde et al., 1995, 1998).
Still, it should be stated that a majority of studies within the top-down
approach have retrieved psychometrically sound scales for Openness or Intellect,
both in caregiver and self-reports (see Tables 12.1 and 12.2). In addition,
bottom-up strategies have yielded strong evidence that Openness or Intellect
content does emerge as a separate trait in component structures of generic
developmental questionnaires. In analyses of the Digman scales, this domain
groups traits such as imaginative, aesthetically sensitive, and quick to learn,
whereas in the CCQ, this domain groups traits such as creative, active fantasy
life, interesting and arresting child, responds to humor, or resourcefulness in
initiating activities. Strong evidence also comes from free description research
showing that parents and teachers (from many countries) spontaneously and
frequently use words referring to the Openness/Intellect domain when asked to
describe their children or pupils (Halverson et al., 2003; Kohnstamm et al., 1998;
Mervielde et al., 1995, 1998). Again, it can be stated that there might be more
problems in discerning Openness/Intellect as a separate trait in preschoolers.
Even though the youngest children in this free response research were 3 years of
age (Halverson et al., 2003), there remains ambiguity about the point in early
childhood at which this trait emerges (De Pauw et al., 2009; Herzhoff & Tackett,
2012; Shiner & DeYoung, 2013). More developmental research about the early
manifestations of Openness-related traits is certainly required.
A second controversy relates to how this Openness/Intellect trait is best
conceptualized in childhood. Even though parental free descriptions provide
strong support for the existence of such a dimension, its content is clearly more
restricted than in adult FFM models. In adulthood, the classic debate about the
terms Openness versus Intellect has been largely resolved by the recognition that
both Openness and Intellect describe distinct yet equally central aspects of this
FFM domain, with Openness capturing perceptual and aesthetic interests and
with Intellect capturing intellectual interests (Shiner & DeYoung, 2013; see also
the chapter by Sutin). Both tendencies are—to some degree—present in the NEO
PI-R Openness scale of Costa and McCrae (1992), which distinguishes Intellect-
capturing facets Open to Ideas (e.g., intellectually curious and open to new
ideas) and Open to Values (e.g., reexamination of traditional social, religious,
and political values) from more Openness-capturing facets Open to Fantasy
(e.g., vivid imagination and fantasy life), Open to Feelings (e.g., receptiveness to
inner emotional states and valuing emotional experience), and Open to Actions
(e.g., trying new activities, visiting new places, trying new foods).
In contrast, the childhood analogs of the Openness/Intellect dimensions are
clearly more restricted in terms of content. In the ICID, the Openness domain is
heavily defined by intellectual tendencies, even though it is labeled as Openness.
The HiPIC does allow for a differentiation between intellectual (such as quick to
learn, clever, insightful, curious) and imaginative tendencies (such as
imaginative, creative, and aesthetically sensitive) within the Imagination trait.
Significantly, the joint factor analysis of NEO PI-R and HiPIC facet scales in
adolescents’ self-reports (De Fruyt et al., 2000) revealed a clearly identifiable
factor indicated by both the NEO Openness and HiPIC Imagination scales. Yet
although the HiPIC Intellect scale had a primary loading on this joint factor, the
correlations with the NEO PI-R Openness facets were small to moderate. This
finding suggests that adult personality measures such as the NEO PI-R do not
cover the Intellect domain as thoroughly as childhood personality measures such
as the HiPIC. In addition, this finding suggests that in bottom-up approaches,
features of intellect are more determining of child personality structure than of
adult structure (De Fruyt et al., 2000; Tackett et al., 2012).
More work can be done to broaden the scope of this Openness/Intellect trait.
Some researchers have already suggested additional markers of childhood
Openness/Intellect that go beyond the ones identified in bottom-up approaches.
For example, children high on this trait might be enthusiastically involved in
extracurricular activities, eager to take on creative and intellectual work, and
show imaginativeness and resourcefulness in play, confidence, and adaptability
in the face of uncertainty (Abe, 2005; Goldberg, 2001a; Shiner & DeYoung,
2013). As we will discuss in the third part of this chapter, it has also been
suggested that perceptual sensitivity, a trait delineated by the temperament
tradition, might be considered as a childhood analog of the aesthetical and
perceptual interests of adult Openness (De Pauw et al., 2009; Herzhoff &
Tackett, 2012).
Multiple studies have documented meaningful associations between childhood
Openness/Intellect traits and a wide array of psychosocial outcomes, including
the development of mental health problems (Herzhoff & Tackett, 2012; John et
al., 1994), competencies (Herzhoff & Tackett, 2012), and individuals’ values,
political beliefs, or academic and creative achievements (Caspi & Shiner, 2006;
Shiner & DeYoung, 2013). This research emphasizes the importance of this trait
in youth. Nevertheless, more research is certainly needed on how to evaluate this
trait in a more comprehensive (i.e., broader than intellective tendencies) and
developmentally appropriate manner (Herzhoff & Tackett, 2012). The additional
markers of childhood analogs mentioned before provide interesting candidates to
attain a more comprehensive coverage of this fascinating childhood trait.
Are We There Yet? Further Challenges for Research on Child Personality

The research reviewed in this second part clearly illustrates the marked
progress that has been made in unraveling the underlying structure of child and
adolescent personality. Yet, as noted throughout this review, this field of
research clearly remains a “work in progress.” Most importantly, the higher-
order objective of “delineating the most comprehensive structure of individual
differences, at each point in development” (Caspi & Shiner, 2006) cannot be
considered as fully attained. In this section, a number of lacunae and unresolved
issues that require further empirical and conceptual attention are examined.
Personality in babies? How early can we tell? Very little is known about
when the Little Six dimensions first emerge in development and about what the
developmental precursors of these traits are. Therefore, it is not clear from what
point in development we can rightfully speak of youngsters’ traits as personality.
Put in other words, what are the lower-age boundaries of FFM measures in early
childhood?
A number of studies have now suggested that FFM-like dimensions can
already be retrieved in parent ratings of children as young as age 2 or 3 years,
relying upon parental CCQ ratings (Abe, 2005; Abe & Izard, 1999; Asendorpf &
Denissen, 2006; Lamb et al., 2002) and HiPIC or ICID ratings (de Haan et al.,
under review; Halverson et al., 2003; Tackett et al., 2012; Zupancic, 2004;
Zupancic, Podlesek, & Kavcic, 2006). These studies generally support the notion
that content related to Activity, Extraversion/Sociability,
Benevolence/Agreeableness, Emotional Stability, Conscientiousness, and even
Openness/Imagination is already salient in the minds of informants who judge
toddlers prior to the preschool years. However, just as there are problems with
top-down approaches stretching adult measures for use in adolescents and
children, there are problems in the downward stretching of preschool- or middle
childhood-appropriate measures to infants and toddlers.
Most importantly, such approaches do not address the question of the extent to
which Little Five or Little Six structures fully and adequately represent
behavioral individuality in these young children. Therefore, more work is needed
to resolve the question how individual differences are best captured in toddlers
and infants. One straightforward approach to disentangle infants’ and toddlers’
trait structure is to scrutinize relevant trait concepts from the broader
developmental literature in the youngest age groups. As the cradle of the most
prominent temperament models can be found in infancy (Buss & Plomin, 1975;
Rothbart, 1981; Thomas & Chess, 1977; Thomas et al., 1963), a simultaneous
evaluation of multiple temperament instruments developed for assessment in the
first year would be a valuable starting point to delineate the major traits in these
youngest groups (De Pauw et al., 2009; Mervielde & De Pauw, 2012). To search
for traits beyond those already part of traditional temperament models, in
addition, one could collect natural language descriptors of infants and toddlers.
There is one small-scaled Slovenian study that has already analyzed free parental
descriptions of infants and toddlers (Zupancic, 2004). This study applied the
coding scheme as used in the international study of Kohnstamm and colleagues
(1998) to classify parents’ descriptions of infants/toddlers (mean age 15 months)
and preschool children (mean age 50 months). Interestingly, over 80% of
parental responses of both groups were coded within the FFM, suggesting that
the FFM taxonomy might be useful for describing even younger children than
previously thought.
As we discussed in the review for older children, the complementary
combination of a top-down approach (items constructed by theorists to assess
potentially relevant traits in children) and a bottom-up approach (researchers
collected descriptors frequently used by caregivers or researchers analyze fine-
grained observations of child behaviors) may have the greatest potential to
uncover the most comprehensive trait taxonomies. We advocate a similar
approach for research on younger children. In addition, empirical studies, ideally
including side-by-side multiple age-appropriate trait measures from multiple
temperament and personality frameworks, are needed to further validate these
conceptually derived comprehensive sets of traits in early childhood (Caspi &
Shiner, 2006; De Pauw et al., 2009). Given the considerable and rapid
developmental changes in the first years of life, such empirical research requires
measurements at multiple short time intervals, for example, using cohorts of
children at 3, 6, 9, 12, 15, and 18 months. Repeated follow-up studies of these
samples will make it possible to detect the time in development at which
personality-like traits emerge and merge with temperament-based precursors.
Such longitudinal work may also help to better understand how trait differences
in preverbal children unfold into the personality characteristics that can be
readily identified in preschoolers. This line of research has much to gain from an
open-minded collaboration between developmental and personality
psychologists.
How will we capture change and continuity in personality structure
development? In recent years, the improved measurement of children’s
personality has given an impetus to the study of normative personality
development. Multiple studies have now examined mean-level change and
differential stability of FFM personality traits prior to adult age, particularly
from middle childhood to early adulthood (De Fruyt et al., 2006; de Haan et al.,
under review; Klimstra et al., 2009; McCrae et al., 2002; Roberts, Walton, &
Viechtbauer, 2006; van den Akker et al., 2014). Denissen, van Aken, Penke, and
Wood (2013) recently performed a meta-analysis of those studies reporting
mean-level age differences in child personality FFM traits, thereby charting
differences across early and late adolescence (from 10 to 20 years). Their meta-
analysis summarized 14 studies (reporting data on 20 samples, including 16 with
a longitudinal design) targeting mean-level differences across this time period.
Both caregiver and self-ratings were included in the study, but type of informant
was not studied as a moderator. Intriguingly, this meta-analysis found that three
personality factors (Neuroticism, Extraversion, and Agreeableness) showed
substantial continuity from middle childhood to late adolescence. For
Conscientiousness, however, there was evidence of a linear mean-level shape of
effect sizes, suggesting that this trait tends to decrease in early adolescence but
then increases later in adolescence. In addition, Openness-to-Experience showed
both a linear and quadratic trend, decreasing rather steeply in early adolescence
and increasing relatively less steeply later in adolescence.
Recent work on both large cross-sectional (Soto, 2015; Soto et al., 2008,
2011) and longitudinal (de Haan et al., under review; van den Akker et al., 2014)
samples further replicated these interesting patterns of Conscientiousness and
Openness-to-Experience throughout the transition from late childhood into
adolescence. Interestingly, these studies also suggest a similar curvilinear age
pattern for Agreeableness/Benevolence: This trait tends to increase from middle
to late childhood, but temporarily declines from pre- to mid-adolescence, and
increases again thereafter. Van den Akker and colleagues (2014) cross-validated
these temporary declines in HiPIC Benevolence, Conscientiousness, and
Imagination across both mother and child self-ratings of personality. They also
demonstrated a temporary decline in Emotional Stability in early adolescence
followed by an increase thereafter, but this latter trend was only found in the
mother reports. Together, these results are now interpreted as evidence for the
“disruption hypothesis” (Soto, 2015), alternatively labeled as “a temporary
defiance of the maturity principle” (van den Akker et al., 2014). The “maturity
principle” refers to the observation in many adult studies that personality
characteristics become gradually more adaptive and desirable. The discussed
research hence suggests that early adolescence (10–15 years) is an exception to
this maturity principle. During this period, characterized by substantial changes
in biological (puberty), psychological (separation–individuation), and social
(increased importance of peer relationships) development, personality
development shows temporary declines in psychosocial maturity, at least in two
or three of the FFM traits (Soto, 2015; van den Akker et al., 2014).
Supplementing this work on big traits, de Haan et al. (under review)
conducted one of the first long-term longitudinal studies tracing lower-order
HiPIC personality facets across two samples, spanning an age range from young
childhood to young adulthood. This study provides further evidence that
distinctive developmental changes are most apparent in the transition from late
childhood to mid-adolescence, particularly for Benevolence and
Conscientiousness facets. Moreover, this study emphasizes that even same-
dimension facets show idiosyncratic changes, indicating that examining
personality development exclusively on the level of higher-order traits might
obscure distinctive changes of the more specific, concrete, lower-order facets.
To summarize, recent research reveals important personality changes
throughout childhood and adolescence that require further empirical attention
and understanding. In addition, it should be noted that even though these
changes are salient, they are limited in terms of absolute size. Hence, this
research also provides support to the conception that childhood personality traits
show both continuity and change (Asendorpf & Denissen, 2006; Denissen et al.,
2013). Thus, future research faces the challenge of better understanding the
origins and reasons of stability and robustness in personality across the lifespan
on the one hand, as well as the multiple aspects of change in personality
development on the other.
This research endeavor will benefit from an integration of all retrieved
findings on developmental continuity and change into one single hierarchical
taxonomy, such as the Big Five or Little Six. Such a taxonomy can serve as a
steering framework to “orient developmental research about the origins and
sequelae of personality differences across the life course” (John et al., 1994, p.
161). Ideally, this taxonomy will include relevant traits for each age group,
which can then be compared across the different stages in development. Using
such a taxonomy will enable researchers to chart homotypical and heterotypical
paths of personality development and to study how different phenotypes are
linked to one another across time (Caspi et al., 2005; Caspi & Shiner, 2006). To
trace the developing structure of personality from infancy to old age, such a
taxonomy should be necessarily conceptualized as a dynamic evolving system.
For example, a trait conceived as a facet at a younger age might evolve into a
more encompassing behavioral trait later in life, so that it becomes a domain
rather than a facet. Conversely, a trait that subsumes broader content in
childhood may become a more specific personality feature when children get
older and may become a more crystallized, stand-alone trait at older ages.
Methodological issues: The need for more varied methods and
informants. To close this second part of the chapter, we briefly highlight some
methodological problems still plaguing research on childhood personality.
Informant troubles. First, the overreliance on questionnaires is a well-known
concern for adult personality research that also applies to child personality
research. In adults, personality structure is primarily derived from self-report
questionnaires. In younger populations, and particularly in children, self-report
measures are generally not the first option, given the nature of language and
cognitive development in these age groups. Hence, the personality structure of
children and adolescents is derived almost by default from ratings by parents or
teachers. Given this overreliance on caregiver reports critics may wonder
whether the Big Five/Little Six structure identified so consistently across studies
exists mainly or only in the mind of adult perceivers. Therefore, it is imperative
to unravel when and how children’s perceptions of themselves and significant
others develop. This is important not only from a developmental point of view,
but also to gain further insight in the developmental roots of the FFM and its
generalizability across the lifespan (Mervielde & De Fruyt, 2000).
In Table 12.2, we summarized the evidence retrieving FFM-like dimensions
based upon children and adolescents’ self-reports and peer-reports. This
summary shows that children above age 10 can reliably and validly provide self-
reports on their personality characteristics. Yet when we consider the factor-
analytic evidence from more bottom-up strategies (e.g., of CCQ sorts or
adjective lists), multiple studies have reported deviant factor structures emerging
from adolescents’ self-reports and peer-reports (e.g., Mervielde & De Fruyt,
2000; Scholte et al., 1997; Van Lieshout & Haselager, 1994). Hence, the
question when the Big Five/Little Six structure emerges in youngsters’ self-
reports cannot be considered as fully answered.
In this regard, there is one interesting study suggesting that the FFM structure
might already show up in self-descriptions of personality in early to middle
childhood. Measelle, John, Ablow, Cowan, and Cowan (2005) found that young
children aged 5 to 7 years could already provide meaningful, longitudinally
stable, and reliable self-descriptions on top-down-derived FFM dimensions
when they were administered an age-appropriate puppet interview (Berkeley
Puppet Interview). Interestingly, the authors noticed that these Big Five self-
perceptions became more differentiated across a 2-year interval. By age 5 years,
there was substantial convergence between the children’s self-reports and adult
ratings on Extraversion, Agreeableness, and Conscientiousness. By age 6 and 7
years, convergence between raters was shown for Neuroticism and Openness-to-
Experience, respectively (Measelle et al., 2005). However, more research is
needed to replicate these findings and to determine from what age children’s
self-reports can be used to describe their own personalities. One other study, for
example, has been less successful in determining reliable and valid FFM-like
factors from a similar puppet interview in 4- and 5-year-old children’s self-
reports (Roth, Dadds, & McAloon, 2004).
Validation across raters and instruments. As noted throughout the discussion
of measurement, very few studies have simultaneously included teacher-ratings,
parent-ratings, and self-ratings of child personality measures [but for exceptions
see Barbaranelli et al. (2003) and van den Akker, Dekovic, & Prinzie (2010)].
As a consequence, validity and replicability across raters have not been
examined systematically for most contemporary child personality measures. A
further crucial validation issue pertains to the empirical convergence across these
wide sets of currently available instruments. In contrast with the well-
documented relations among common FFM measures in adulthood (e.g., John et
al., 2008; Markon et al., 2005), surprisingly few studies have included and
compared multiple coexisting measures for child personality. As an exception,
Tackett and colleagues (2013) recently provided a rigorous evaluation of the
empirical congruence between two well-established bottom-up child personality
measures, that is, the HiPIC (English translation version) and the short version of
the ICID(-S). This study demonstrated clear convergence between the higher-
order domains of both instruments within the same sample, with the highest
correlations between analogous traits in each measure. Notably, this analysis
also documented some asymmetries in content between the two measures. Most
outstandingly, unique variance in ICID-S Openness positively predicted HiPIC
Extraversion, whereas HiPIC Imagination negatively predicted ICID-S
Extraversion. In addition, some marked divergences in predictive validity of
these measures were noted. For example, this study identified HiPIC
Benevolence as being a strong negative predictor of internalizing problems but
with ICID-S Agreeableness being a strong but positive predictor of internalizing
problems. This positive connection between ICID-S Agreeableness and
internalizing is surprising and requires further consideration. According to the
authors, the positive link with internalizing might reflect the “nonantagonistic”
content captured by the ICID trait. In this way, this positive relation would
corroborate the evidence in adult groups that extreme elevations in
Agreeableness (e.g., being overly accommodating, self-sacrificing, or self-
effacing) are a risk factor for internalizing symptoms (Widiger, 2011). This
effect could also reflect the more heterogeneous content of this trait as ICID-S
Agreeableness includes the facets Considerate, Compliant, Antagonism, and
Strong-Willed, as well as the subscales Positive Emotions and Negative Affect
(Deal et al., 2007). Alternatively, this phenomenon may be caused by the fact
that different facets within ICID-S Agreeableness and HiPIC Benevolence may
manifest differential and idiosyncratic predictive patterns. These differential
effects are then obscured if only the higher-order dimensions are considered.
This notion is supported by recent longitudinal evidence on the HiPIC facets.
This study revealed that two facet scales of Benevolence, Irritability and
Altruism, positively predicted pathways of elevated anxious and depressive
symptoms. Moreover, the facet Compliance positively predicted pathways of
elevated anxiety (Prinzie et al., 2014).
Taken together, this discussion emphasizes that there is no guarantee that
semantically similar labels actually share the same trait variance (i.e., the jingle-
jangle fallacy mentioned before), so that researchers, particularly in the study of
childhood individual differences, should be extremely cautious to judge traits of
different frameworks only by their labels (De Pauw et al., 2009). Moreover, this
discussion strongly emphasizes the need for research that thoroughly addresses
the differential patterns of the lower-order facets across development (e.g., Becht
et al., 2015; de Haan et al., under review; Prinzie et al., 2014).
Validation across cultures. The cross-cultural generalizability of personality
structure in childhood also remains a vital yet understudied topic in child
personality research (Caspi et al., 2005; Tackett et al., 2012). As an exception,
Tackett and colleagues (2012) have evaluated hierarchical personality structures
based upon parent ratings of the ICID-long version across five countries and four
age groups. Based upon the “bass-ackwards method” of Goldberg (2006), this
study provided strong evidence that the hypothesized FFM-like structure of the
ICID could be replicated across all countries and was already salient in the
youngest age groups. Interestingly, some of the differences between child and
adult taxonomies discussed before (i.e., the prominence of antagonism in
Agreeableness and the covariation of Conscientiousness-Intellect in childhood)
were found to replicate across countries. Yet more work is welcomed to further
investigate cross-cultural differences in emergent personality in youth, potential
cultural differences in informants and methods, and in the role that culture plays
in shaping youngsters’ personality throughout development (Tackett et al.,
2012).
Moving beyond questionnaires. Finally, there is a clear need to go beyond the
classic approach of administering questionnaires in order to establish child
personality structure by more varied methods and reporters. Future research
should aim to validate the questionnaire ratings of child personality with other
methodological approaches, such as behavioral tasks or observational measures.
Such behavioral observation research (for example, coding thin slices of
behavioral observations) has been successfully carried out in adult FFM research
(Borkenau, Mauer, Riemann, Spinath, & Angleitner, 2004; Borkenau, Riemann,
Angleitner, & Spinath, 2001) but is only very limited in the study of child
personality (for exceptions see Markey, Markey, & Tinsley, 2004; Tackett,
Herzhoff, Kushner, & Rule, 2015). In further pursuing this line of work,
inspiration can be gathered from temperament research (Shiner & DeYoung,
2013) in which there is a rich tradition of relying not only on questionnaires but
also on other methods, such as home observation systems, to code naturalistic
observations of children (Buckley, Klein, Durbin, Hayden, & Moerk, 2002) and
laboratory tasks that elicit specific contexts in which children’s behaviors can be
observed and coded (Goldsmith & Rothbart, 1999; Majdandzic & van den
Boom, 2007).
Lessons from Research on Temperament

Toward a More Comprehensive Taxonomy of Childhood Traits
In the conclusion of the preceding section, we argued that research on
children’s personality structure may benefit from the methodological diversity
typical for research in the temperament tradition. In addition, a more essential
question can be raised: How does childhood personality relate—in substantive
terms—to this historical “sovereign framework” for trait description in youth?
For too many years, research on child personality has been proceeding in relative
isolation from research on child temperament. As a consequence, our current
understanding of the convergence between adult and child personality is greater
than our understanding of the convergence between child temperament and child
personality (De Pauw & Mervielde, 2010; Tackett, 2006). In this final part of the
chapter, we address the understudied topic of temperament–personality
convergence. Integrating all reviewed information on childhood personality and
temperament, we finally propose a more comprehensive taxonomy of trait
differences in childhood.
The stretching of temperament and personality concepts. There is a
historical reason why temperament and personality research are not better
integrated. Temperament researchers have primarily focused on individual
differences among infants, toddlers, and preschoolers (Buss & Plomin, 1984;
Rothbart, 1981; Rothbart & Ahadi, 1994; Thomas et al., 1963) whereas
personality psychologists primarily targeted traits in adults (John et al., 2008).
To document the continuity of temperament in older age groups, theorists have
stretched their constructs to suit school-aged children (Hegvik, McDevitt, &
Carey, 1982; Simonds & Rothbart, 2004), adolescents (Capaldi & Rothbart,
1992), and even adults (Evans & Rothbart, 2007; Thomas, Mittelman, Chess,
Korn, & Cohen, 1982). Notably, this operation entails the reverse movement of
the one we discussed for top-down child personality research, where adult-
oriented personality models have been adapted downward to document the
relevance in younger ages. The risk of both stretching operations is that they
may have narrowed the scope and representativeness of the sampled behavioral
repertoires. More specifically, these stretching operations have presumably
narrowed the representativeness of the sampled behavioral repertoires of
temperament for older groups and of personality for younger groups.
Nevertheless, these stretching operations have resulted in the current availability
of coexisting temperament and personality measures that can be used by
researchers and practitioners in almost all age groups (except in infants and
toddlers). As we aim to compile a more comprehensive trait taxonomy based
upon both the child personality and temperament tradition, we first discuss the
major trait content embedded in the most prominent temperament models.
What’s in temperament? Historically, temperament has been studied
intensively from the 1960s, primarily by developmental psychologists. However,
after more than 60 years of discussion, there is still no clear consensus on a
single model, definition, or on the main dimensions underlying temperamental
variation of children (Goldsmith et al., 1987; Mervielde & De Pauw, 2012).
Here, we briefly sketch the content of three temperament models that have
generated the most theoretical and methodological contributions. These are the
models of Thomas and Chess et al. (1963, 1977), Buss and Plomin (1975, 1984),
and Rothbart (1981, 2012).
Thomas and Chess. In their landmark New York Longitudinal Study,
Alexander Thomas and Stella Chess followed the development of 138 babies
from white middle-class families and 95 infants from economically
disadvantaged families, using extensive observations, psychiatric assessments,
and in-depth interviews with caregivers (Thomas et al., 1963; Thomas & Chess,
1977). They became intrigued by the marked differences that could already be
observed very early in development. They first described these differences as
“primary, innate response patterns.” Notably, these differences were not easily
explained by the prevailing behavioristic or psychoanalytical paradigms of that
time. As such, their work launched the scientific study of early apparent,
relatively stable, internal individual differences, under the label of temperament.
According to Thomas and Chess (1977), temperament refers to stylistic (e.g.,
how intense does an infant cry) rather than motivational (e.g., why does an infant
cry) or content (e.g., what does an infant do while crying) aspects of behavior.
From inductive content analysis of interviews with 22 parents of infants, Thomas
and Chess derived nine “behavioral styles” that they considered relevant for the
child’s later psychosocial adjustment. These traits were (1) Activity level, (2)
Rhythmicity of biological functions, (3) Approach or withdrawal from new
stimuli, (4) Adaptability to new situations, (5) Threshold of sensory
responsiveness, (6) Intensity of emotional reactions regardless of their positive
or negative quality, (7) General quality of mood (e.g., being either pleasant or
irritable), (8) Distractibility (e.g., the capacity for external stimuli to interrupt a
child’s behavior), and (9) Task persistence or the degree to which a child can
sustain attention.
These traits were first identified in infants, but were later also assumed to be
fundamental in toddlers, preschoolers, and primary school age. Even though the
same nine dimensions are presumed to exist across the lifespan, behavioral styles
theory emphasizes that these nine behavioral styles are manifested in different
behaviors throughout development, analogous to the growing behavioral
repertoire of the child. Hence, this tradition emphasizes the need for age-specific
item sets for assessing the nine behavioral styles. Notably, the assumed basic
traits are identically labeled from infancy to late childhood (even though they are
assessed with different items), except for the trait Rhythmicity, which was
described as Predictability/Quality of Organization in middle childhood (Hegvik
et al., 1982). In this age group, this trait is broader than rhythmicity in biological
functions (e.g., sleeping and eating) but taps more overall predictability as well
as task performance.
Even though Thomas and Chess’s nine-dimensional conceptualization is still
well represented in past and present research, there appears little empirical
support for the proposed nine-dimensional structure. A dozen item-level factor
analyses of behavioral styles instruments were reviewed by Martin, Wisenbaker,
and Huttunen (1994) and Presley and Martin (1994). These authors suggested
that four rather than nine factors are sufficient to categorize the items. These
dimensions were labeled Irritable Distress, Social Inhibition, Activity, and
Attention.
Buss and Plomin. Whereas Thomas and Chess initially focused on traits
appearing in infancy, Arnold Buss and Robert Plomin (1975, 1984) chose to
focus on traits that show relative continuity from infancy into adulthood. As
such, they framed temperament as being the developmental precursor of later
developing personality. They proposed five inclusion criteria, specifying that a
trait can be considered truly as “temperament” only if it is substantially
heritable, apparent in phylogenetic relatives, evolutionary adaptive, and if it is
relatively stable during childhood and retained into adulthood. From these
criteria, Buss and Plomin (1975) proposed that only four traits can be considered
as truly temperamental in nature. These traits were Emotionality (i.e., intensity
of emotion, first undifferentiated distress but later focusing on both fear and
anger), Activity (i.e., quantity of motor activity), Sociability (i.e., closeness to
others), and Impulsivity (i.e., quickness versus inhibition). This latter dimension
was later dropped because the evidence for the heritability of impulsivity was
mixed. Also, in the latest version of their model, Buss and Plomin (1984) make a
theoretical distinction between the traits Sociability (i.e., the preference for
interaction with others) and Shyness (i.e., feeling uncomfortable in the
interaction with unfamiliar others). In empirical studies, however, it remains
difficult to distinguish these two traits through factor analysis (De Pauw et al.,
2009; Gasman et al., 2002; Mervielde & De Pauw, 2012). The work of Buss and
Plomin (1984) is unique as it points researchers to the importance of
understanding trait consistency across multiple age groups. Moreover, even
though their model is criticized as painting a too narrow picture of temperament,
they have identified some of the most important traits appearing across
temperament models (Shiner & DeYoung, 2013).
Rothbart. Mary Rothbart’s (1981, 2012) model is without a doubt the most
popular model in contemporary temperament research. Inspired by the great
discoveries in psychobiology in the 1970s and 1980s, she extended the original
temperament construct of Thomas and Chess to also encompass emotional,
motivational, and attentional processes. Rothbart (1981, 2012) defined
temperament as “constitutional differences in reactivity and self-regulation,”
with “constitutional” referring to the biological make-up of a person. In
Rothbart’s model, temperamental differences are determined by the
responsiveness of underlying neural systems. Reactivity then refers to the extent
to which these neural systems are aroused. These systems are primarily systems
responsible for emotional (affect), physical (active), and orienting (attention)
responses. Self-regulation, then, refers to the processes that make it possible to
control and modulate this reactivity. These differences in reactivity and self-
regulation are considered to be relatively consistent across situations and time,
but their expressions can be modified by heredity, maturation, and experience.
Hence, according to Rothbart’s theory, new temperament traits may emerge over
time, for example, in response to new challenging developmental tasks or as a
result of physiological maturation in a child.
Even though this model has strong psychobiological underpinnings, its default
assessment method has also been caregiver questionnaires. Initially, Rothbart
developed her model to describe temperament in the first year of life. In the past
two decades, the model was conceptually and methodologically extended to
older age groups, first to preschool and young school children (Rothbart &
Ahadi, 1994) and then to adolescents (Capaldi & Rothbart, 1992), and only
recently, to toddlers (Putnam, Gartstein, & Rothbart, 2006) and adults (Evans &
Rothbart, 2007). It can be noted that these extensions were primarily developed
as part of doctoral dissertations. The principal aim of these studies was to link
temperamental differences to other important developmental constructs (e.g.,
puberty, reward sensitivity). As such, Rothbart and her students were primarily
interested in identifying temperamental indicators of salient neurological
systems in specific age groups, rather than in constructing comprehensive
temperament taxonomies across these age groups. Consequently, the multiple
questionnaires developed for the multiple age groups show only limited
consideration of traits that were already found in younger or older age groups.
As a result, Rothbart’s instruments vary markedly in the number and content of
proposed lower-order traits (Putnam, Ellis, & Rothbart, 2001). This large variety
of lower-order temperament scales is a major impediment for longitudinal
research with this model because it is not clear whether changes in temperament
across age groups reflect true developmental changes or only differences in
measurement (Mervielde & De Pauw, 2012). Moreover, given the nonsystematic
and primarily theory-driven construction of Rothbart’s measures, it is hard to
establish the comprehensiveness of the proposed taxonomies for each age group.
Nevertheless, factor analyses of Rothbart and colleagues’ measures yield
strong support for the conception that three trait dimensions underlie
temperament ratings from infancy through late childhood (Rothbart & Bates,
2006; Rothbart, 2012). These dimensions are labeled Negative Affect, Surgency,
and Effortful Control. Negative Affect refers to children’s tendencies toward a
wide range of negative emotions such as sadness, fear, anxiety, and difficulty to
be soothed after high arousal, but also encompasses irritability and frustration.
Surgency taps into children’s levels of energy and physical activity, and rapid
approach but also expressions of positive emotions, pleasure and excitement in
social interaction. Effortful Control refers to individual differences in focusing
and sustaining attention, and inhibiting dominant responses in favor of a
subdominant response. Whereas these differences in self-control typically
become most apparent in the second and third year of life (Kochanska &
Knaack, 2003), individual differences in focusing and sustaining attention can
already be detected in infants. This precursor of Effortful control is labeled
Orienting Regulation. This trait shows substantial cross-time correlations with
Effortful Control assessed in toddler and preschool age (Putnam et al., 2001;
Putnam, Rothbart, & Gartstein, 2008).
Notably, in more recent conceptualizations of her framework, Rothbart has
suggested that the broader Negative Affect-trait might be decomposed into
Anxious Distress versus Irritable Distress to accommodate the empirical
observation that Negative Affect strongly relates to both internalizing and
externalizing problems (Rothbart & Posner, 2006). Interestingly, this distinction
mirrors the distinction made in bottom-up child personality taxonomies between
Anxious and Irritable Distress (John et al., 1994). Another notable recent
evolution in Rothbart’s model pertains to the content of Effortful Control. In
Rothbart’s taxonomies for toddlers, preschoolers, and school children (but not in
adolescents), this trait encompasses pleasure in low-intensity activities (e.g.,
enjoys looking at picture books, enjoys sitting in the sun), and sensitivity to
perceptual experiences (e.g., notices the smoothness or roughness of objects she
or he touches). Recent research (De Pauw et al., 2009; Herzhoff & Tackett,
2012) now suggests that these scales form a separate Sensitivity trait, distinct
from the more attention-focused trait of Effortful Control. Delineating
Sensitivity as a distinct childhood trait would hence result in a better alignment
of Rothbart’s child and adult temperament taxonomies. In the recently developed
taxonomy for adults, similar Sensitivity traits are subsumed by a fourth higher-
order domain, which is labeled Orienting Sensitivity (Evans & Rothbart, 2007).
Finally, Rothbart’s taxonomies for adolescents also propose another higher-order
trait, which relates to the need to spend time with significant others; this trait is
labeled Need for Affiliation.
Are We There Yet? What Is the Basic Structure of Temperament?

As noted, all three “classic” temperament traditions had their cradle in infant
research and were later stretched to older age groups. Moreover, all three models
have their original roots in the conceptualization of Thomas and Chess (1977).
Both Buss and Plomin (1975) and Rothbart (1981) departed from Thomas and
Chess’s original framework, modifying and extending it with their specific
emphases about the nature of temperament. The resulting taxonomies hence can
be considered as the product of primarily top-down and theory-driven
construction methods (Halverson et al., 2003). This implies that it is uncertain
whether important domains were excluded because these theorists were not
interested in them, or that dimensions were included because the theorists wrote
many items to measure what they believed to be important (Halverson et al.,
2003).
Therefore, it is not surprising that there is insufficient agreement on the
dimensional structure among the different models of temperament (Halverson et
al., 2003; Mervielde & De Pauw, 2012). This deficient empirical validation of
the theoretical underpinnings can be considered a major weakness of
temperament research, which has resulted in an uncontrolled growth of disparate
measures capturing all sorts of temperament traits. In this regard, the dearth of
empirical studies that link the various temperament models and accompanying
instruments is notorious (De Pauw et al., 2009; Mervielde & De Pauw, 2012).
As a consequence of this lack of empirical evidence, scholars interested in the
basic structure of temperament are forced to turn to narrative, semantic analyses
to condense the multitude of proposed temperament traits into a higher-order
structure. Such reviews (e.g., Rothbart & Bates, 2006; Shiner & Caspi, 2003)
now generally agree that child temperament can be summarized by at least three
higher-order traits: Surgency/Positive Emotionality, Negative
Emotionality/Affect, and Effortful Control/Constraint (notice that these labels
are primarily Rothbart’s terminology). Other reviews (De Pauw & Mervielde,
2010; Mervielde et al., 2005) have also followed the suggestion of Buss and
Plomin (1975) to separate Activity from Sociability/Extraversion. These
narrative reviews have had tremendous value in advancing the integration of
temperament findings across studies and models. Yet these narrative reviews
have most strongly relied on the more conveniently arranged set of theoretically
proposed dimensions instead of on the host of empirically derived age-specific
temperament factors across studies (De Pauw et al., 2009; Mervielde & De
Pauw, 2012). To empirically delineate the basic structure of child temperament
definitely remains a key priority for future research.
How Shall We Conceive the Linkages between Child Personality and

Temperament?
The narratively identified traits in child temperament show striking
similarities to the higher-order traits of child personality. Therefore, the most
crucial questions remain: How should we conceive the linkages between child
temperament and child personality? And, how shall we speak of individual trait
differences in youth? Do we need separate temperament and personality
frameworks? These key questions have fueled some fiery conference and email
discussions in the past decades, with temperament adherents making strong calls
to perpetuate the distinction and with (mainly adult) personality adherents
pleading to simply conceptualize all child trait differences as personality. Here,
we review some important arguments and recent evolutions in this debate.
Temperament is not the biologically based precursor of personality. Even
though temperament scholars have had their disputes concerning the nature and
number of dimensions, there is a general consensus that temperament forms the
biologically based foundation of later-developing personality (Buss & Plomin,
1984; Goldsmith et al., 1987; Rothbart & Bates, 2006). In some temperament
models, this relation is seen as a sequence, with temperamental traits being the
early building blocks of substrates that develop into more complex personality
traits over time (Cicchetti, 1990). The environment is assumed to interact with
temperament to produce personality (Kagan, Snidman, Zentner, & Peterson,
1999). Another viewpoint holds that temperament is a lifelong, yet distinct,
component of personality. This position is, for example, asserted by Rothbart
and Ahadi (1994), who emphasize that the “personality domain contains much
more than temperament,” such as patterns of habitual behavioral skills, content
of individual thoughts, values, needs, and goals, and perceptions of self, others,
and events.
Definitely, personality can be conceived as a broader concept than
temperament (Shiner & DeYoung, 2013). For instance, McAdams and Pals
(2006) distinguish three different levels within the study of personality: basic
traits, characteristic adaptations, and personal narratives. First, the basic traits
are the relative consistencies in behavior, emotions, and cognitions across
contexts and across time, and have—at least in adults—the Big Five as the
underlying structure. Second, the characteristic adaptations are a wide range of
motivational, social-cognitive, and developmental adaptations that are specific to
a particular time, place, or role. These are, for example, the goals and values that
an individual holds regarding academic performance or romantic relationships.
Finally, from adolescence on, humans begin to form personal narratives, which
help them to make sense of their identities and selves over time. These narratives
are totally unique to each person but can be studied empirically in terms of their
common features across individuals. Yet, even though personality can and
should be studied well beyond this level, the study of basic traits remains the
most prevalent approach in personality psychology (and all research reviewed in
this chapter).
At this level of basic traits, temperament and personality are hard to
distinguish empirically (for a review see McCrae et al., 2000; Shiner &
DeYoung, 2013). Arguments against this historical divide stem from behavioral
genetic studies (both temperament and personality traits arise from the complex
interplay of genes and experience, including personality-only traits such as
Openness-to-Experience and Agreeableness), ethological studies (animals
display proxies of the most important temperament traits in childhood and of all
key personality traits in adulthood), as well as longitudinal studies (both
temperament and personality show substantial stability and change across time).
Other salient arguments arise from the reviewed evidence in the first two parts of
this chapter, documenting that individual differences manifested by children
beyond age 3 (and perhaps younger) can be rightfully considered as
“personality.” Moreover, the strong conceptual similarities across childhood
temperament and personality further indicate that the distinction between these
constructs is more the historical result of research traditions than a substantive
matter.
The FFM seems relevant, yet not sufficient to embrace both domains Based
on the evidence reviewed here, prominent scholars now agree that temperament
and personality systems have many traits in common (Rothbart, 2012; Rothbart
& Bates, 2006; Shiner, 1998), and that temperament and personality increasingly
appear to be “more alike than different” (Caspi et al., 2005; Caspi & Shiner,
2006, 2008; De Pauw & Mervielde, 2010; Mervielde et al., 2005; Shiner, 1998;
Shiner & Caspi, 2003; Shiner & DeYoung, 2013). In a number of seminal, yet
primarily narrative, reviews (e.g., Caspi et al., 2005; De Pauw & Mervielde,
2010; Mervielde et al., 2005), it has been argued that the Five Factor Model, as
retrieved in adults, appears to embrace the content embedded in both trait
systems. These reviews consider the FFM dimensions as overarching dimensions
in a preliminary joint taxonomy capturing both temperament and personality. In
this taxonomy, it is hypothesized that the FFM dimensions of Neuroticism,
Extraversion, and Conscientiousness represent the three temperament domains
of Negative Emotionality, Surgency/Positive Emotionality, and Effortful
Control/Constraint, respectively. In this joint view, the child personality system
is considered the most comprehensive because it also includes Agreeableness
and Openness-to-Experience, traits that have not been recognized as major
temperament dimensions. Notably, the latest version of this taxonomy (Shiner &
DeYoung, 2013) also incorporates the recent emphasis on Activity in child
personality as well as temperament research. This version no longer proposes the
FFM, but the Little Six as higher-order underpinnings of a joint taxonomy.
These reviews and their proposed taxonomies have had great significance in
improving communication among researchers using different trait concepts, and
in integrating research findings across different research traditions and
disciplines. Yet we emphasize that these taxonomies too were necessarily based
on narrative analyses and conceptual comparisons, because empirical studies on
temperament and personality have been essentially lacking.
More empirical work is needed to disentangle temperament–personality
relations. In recent years, such empirical research relating temperament and
personality has become available slowly but steadily. A handful of studies have
been conducted in adults (Angleitner & Ostendorf, 1994; Evans & Rothbart,
2007; McCrae & Costa, 1985) and, more recently, in children (De Pauw &
Mervielde, 2011; De Pauw et al., 2009; De Pauw, Mervielde, Van Leeuwen, &
De Clercq, 2011; Deal et al., 2007; Grist et al., 2012; Halverson et al., 2003;
Tackett et al., 2013). In both adults and in children, consistent support has now
been found for the expected associations of temperamental Negative Affect,
Surgency/Positive Emotions, and Effortful Control/Constraint with FFM-based
scores on Neuroticism, Extraversion, and Conscientiousness, respectively.
However, in all the above identified studies, it is consistently noted that the
mapping of temperament on FFM personality is less clean as would be expected
by the semantic similarity of scale labels. There appears to be no perfect one-to-
one correspondence between the expected connections, so that a simple
hierarchical mapping fails to capture the complexity of these interrelations.
These observations indicate that temperament and personality cannot be
considered as totally redundant systems of child trait differences. Instead, these
findings imply that temperament and child personality scales capture unique as
well as overlapping variance (De Pauw et al., 2009, 2011; Tackett et al., 2013).
Strikingly, these reports also suggest that content related to the FFM traits
Agreeableness/Benevolence and even Openness-to-Experience/Imagination is
not entirely missing in temperament models. Studies have remarkably converged
in demonstrating substantial connections between Agreeableness/Benevolence
and negative Affect/Emotionality traits (e.g., De Pauw & Mervielde, 2011; De
Pauw et al., 2009, 2011; Evans & Rothbart, 2007; Tackett et al., 2013) and
between Agreeableness/Benevolence and temperament scales capturing
Impulsivity, Activity, and Effortful Control (e.g., De Pauw & Mervielde, 2011;
De Pauw et al., 2009, 2011; Deal et al., 2007). In addition, multiple studies have
now also identified substantial correlations between Openness-to-
Experience/Intellect scales and Rothbart’s Effortful Control scale (De Pauw et
al., 2009; Evans & Rothbart, 2007; Grist et al., 2012; Herzhoff & Tackett, 2012;
Tackett et al., 2013).
One study (De Pauw et al., 2009) went beyond simple correlational
convergence. In a sample of 4- to 5-year-old children, we carried out a joint
principal component analysis of HiPIC personality facets as well as facets from
the three described temperament measures. We retrieved six dimensions rather
than the supposed FFM dimensions. These were labeled Activity, Sociability,
Emotionality, Disagreeableness, Conscientiousness, and a sixth unexpected
factor, which was labeled Sensitivity. This latter factor is defined by specific
content from Rothbart’s childhood Effortful Control scale. As already noted in
the appraisal of Rothbart’s model, a similar Sensitivity trait has been identified
in her adult temperament taxonomy. Also, as noted in the research suggestions to
recover more childhood analogs of Openness-to-Experience/Intellect, research
has now repeatedly shown substantial connections between Sensitivity-related
content and Openness-to-Experience-related content in both adults (Evans &
Rothbart, 2007) and children (De Pauw et al., 2009; Herzhoff & Tackett, 2012).
Yet more work is definitely needed to further establish whether these two traits
should be integrated into a single dimension in childhood. In this regard, in the
joint component analysis in preschoolers, all HiPIC Imagination facets loaded on
a broader Conscientiousness factor. Interestingly, the separate HiPIC principal-
component analysis clearly differentiated between the two domains. Hence, it is
possible that the imbalance of imagination-related content across temperament
and personality languages has precluded the Imagination trait from surfacing as
a major independent dimension in this study (De Pauw et al., 2009).
To summarize, the empirical relationships between temperament and
personality traits have not been addressed comprehensively (De Pauw &
Mervielde, 2010; De Pauw et al., 2009; Tackett et al., 2013). The substantial
empirical links between temperament and personality traits suggest that both
systems are indeed “more alike than different” (Caspi et al., 2005, p. 454), but
that they are less redundant than expected on the basis of the semantic
similarities of scale labels.
Temperament and Personality as Complementary Languages of Trait

Differences
So, how shall we speak about individual differences in children and
adolescents: as temperament or as personality or as temperament and as
personality? Both temperament and personality taxonomies can be considered as
reliable and valid “languages” to describe childhood traits (De Pauw &
Mervielde, 2011; De Pauw et al., 2009, 2011). These languages are clearly
related, yet also complement each other as measures of individual differences in
childhood and adolescence. How much unique and shared “vocabulary” these
two languages have in describing behavioral individuality across developmental
age groups remains a critical question for future empirical research.
Based on their historical backgrounds, the temperament language presumably
provides the richest lexicon for the younger age groups, whereas the personality
language presumably provides the richest lexicon for older age groups. The
reviewed evidence encourages the continuous evaluation of both similarities and
differences between the temperament and personality measures in further
studies. Also, it seems vitally important that researchers interested in integrating
study findings across temperament and personality frameworks take into account
these salient differences between temperament and personality “vocabulary” and
avoid an assimilation of findings based solely on semantic similarity.
With these cautionary notes in mind, we have drafted a new version of a
preliminary joint taxonomy of temperament and personality traits (presented in
Table 12.3). In this taxonomy, we attempted to integrate in a nuanced fashion
many reviewed convergences between adult and child personality taxonomies on
the one hand, and between child personality and temperament taxonomies on the
other. For example, at the top row, the higher-order structure differentiates
between seven potential basic traits. With this proposal for a differentiated
structure, we encourage researchers to further inquire the separateness of
Activity from Sociability (see the comments on child Extraversion), of Openness
/Imagination from Conscientiousness, and of Sensitivity from Openness
/Imagination.
In addition, we encourage researchers to take a developmental approach to
further unravel the basic structure underlying youth’s traits. The way the
taxonomy is displayed in Table 12.3 may mistakenly suggest that the
temperament/personality structure is static and can be applied invariantly across
childhood and adolescence. We hope that future research will yield more
insights in the rapid developmental changes that determine children’s personality
development. In this regard, more empirical work on the lower-order structure of
childhood traits is definitely needed. The middle row of Table 12.3 summarizes
some of the most important groupings that are identified throughout the
temperament and personality languages. These facet-level traits can be
considered as capturing an important part of the mid-level of a joint childhood
trait taxonomy. Moreover, we have underlined the facet traits (primarily
stemming from temperament) that are well-established to be readily observable
in preverbal children (Caspi & Shiner, 2006; Rothbart & Bates, 2006; Shiner,
1998).
A final goal of this taxonomy is to encourage future joint factor analytic work
in order to enable decisions on the location of “ambiguous” traits (Caspi &
Shiner, 2006; Rothbart & Bates, 2006). These traits appear to show coherent
conceptual or empirical linkages to more than one higher-order trait across the
different temperament and/or personality languages. Some of the most salient
“conceptual blends” are summarized in the bottom row of the taxonomy.
Summary
This chapter discussed state-of-the-art insights in child and adolescent
personality, addressing salient questions on its measurement, its foundational
structure, and its convergence with child temperament. First, we discussed the
discovery and contemporary measurement of personality in youth, highlighting
pitfalls as well as promises of the various approaches. Second, we pleaded for
the study of child personality traits in its own right, rather than assuming that
child personality taxonomies are identical to those established in adults. Finally,
we reviewed how the relationships between child temperament and child
personality are more messy than one would hope and noted that a simple
hierarchical mapping fails to capture the complexity of these relationships.
Instead, we advocated to consider temperament and personality to be “more
alike than different” trait systems but also as systems that adequately
complement each other as “languages” of individual differences. Researchers
from both personality and developmental psychology are invited to take into
account the salient differences between the various child personality approaches,
child and adult personality taxonomies, and child temperament and personality
“vocabularies.” In addition, researchers are invited to avoid assimilation of trait
findings in childhood based solely on the semantic similarity of labels.
Acknowledgments
This manuscript is dedicated to the memory of Ivan Mervielde (1947-2011)
and his pioneering research identifying developmental antecedents of the Five
Factor-Model in childhood and adolescence.
Table 12.3 A Common Taxonomy of Temperament and Personality in Children and

Adolescents
Domain Activity Sociability Emotionality Benevolence Conscientiousness
Level
Abbreviation ACT SOC EMO BEN CON
Alternative Activity Positive Negative Agreeableness, Task Persistence,
labels level, Emotionality, Emotionality, Friendliness, Constraint,
Surgency Surgency, Emotional Love, Effortful Control
Extraversion Stability®, Hostility®
Neuroticism
Facet level Activity level Positive Anxious distress Antagonism® Attention control,
(energy, emotions (joy, (fear, anxiety) Concentration
physical pleasure) Affective distress Egocentrism® Activation control,
restlessness) Sociability (sadness, Willfulness® Perseverance
High Need for discomfort, Compliance Orderliness
Intensity Affiliation soothability) Self- Empathy, Achievement
Pleasure Expressiveness confidence Altruism motivation
Optimism Prosocial
tendencies
Warmth, Trust
Conceptual SOC + ACT SOC® + EMO + BEN® BEN® + CON® + BEN® +
blends EMO ACT
SOC®
Talkativeness Social Irritable distress Dominance, Impulsivity®,
Inhibition, (anger/frustration) Assertiveness Inhibitory control
Withdrawal, Manageability®
Shyness
Note. Traits denoted with ® should be interpreted as reversed coded. Traits marked in italics are primarily
derived from personality literature. Traits that are underlined are well-established as recognizable in
preverbal children.
a Research suggests that Openness-to-Experience scales are strongly related to Conscientiousness in
preschool children.
b Some research suggests that Sensitivity might be subsumed by childhood Openness-to-Experience, but
this should be further validated empirically.
References
Abe, J. A. A. (2005). The predictive validity of the Five-Factor Model of personality with
preschool age children: A nine year follow-up study. Journal of Research in
Personality, 39(4), 423–442.
Abe, J. A. A., & Izard, C. E. (1999). A longitudinal study of emotion expression and
personality relations in early development. Journal of Personality and Social
Psychology, 77(3), 566–577.
Angleitner, A., & Ostendorf, F. (1994). Temperament and the Big Five factors of
Asendorpf, J. B., & Denissen, J. J. A. (2006). Predictive validity of personality types
versus personality dimensions from early childhood to adulthood: Implications for the
distinction between core and surface traits. Merrill-Palmer Quarterly-Journal of
Developmental Psychology, 52(3), 486–513.
Asendorpf, J. B., & Van Aken, M. A. (2003). Validity of Big Five personality judgments
in childhood: A 9 year longitudinal study. European Journal of Personality, 17(1), 1–
17.
Barbaranelli, C., Caprara, G. V., Rabasca, A., & Pastorelli, C. (2003). A questionnaire for
measuring the Big Five in late childhood. Personality and Individual Differences,
34(4), 645–664.
Becht, A. I., Prinzie, P., Deković, M., Van den Akker, A. L., & Shiner, R. L. (2015).
Child personality facets and overreactive parenting as predictors of aggression and
rule-breaking trajectories from childhood to adolescence. Development and
Psychopathology, 1–15.
Block, J. (1995). A contrarian view of the 5-factor approach to personality description.
Block, J., & Block, J. H. (1980). The California Child Q-set. Palo Alto, CA: Consulting
Borkenau, P., Mauer, N., Riemann, R., Spinath, F. M., & Angleitner, A. (2004). Thin
slices of behavior as cues of personality and intelligence. Journal of Personality and
Social Psychology, 86(4), 599.
Borkenau, P., Riemann, R., Angleitner, A., & Spinath, F. M. (2001). Genetic and
environmental influences on observed personality: Evidence from the German
Observational Study of Adult Twins. Journal of Personality and Social Psychology,
80(4), 655.
Branje, S. J., Van Aken, M. A., Van Lieshout, C. F., & Mathijssen, J. J. (2003).
Personality judgments in adolescents’ families: The perceiver, the target, their
relationship, and the family. Journal of Personality, 71(1), 49–81.
Buckley, M. E., Klein, D. N., Durbin, C. E., Hayden, E. P., & Moerk, K. C. (2002).
Development and validation of a q-sort procedure to assess temperament and behavior
in preschool-age children. Journal of Clinical Child and Adolescent Psychology, 31(4),
525–539.
Buss, A. H., & Plomin, R. (1975). A temperament theory of personality development.
New York, NY: Wiley.
Buss, A. H., & Plomin, R. (1984). Temperament: Early developing personality traits.
Capaldi, D. M., & Rothbart, M. K. (1992). Development and validation of an early
adolescent temperament measure. Journal of Early Adolescence, 12(2), 153–173.
Caspi, A., Block, J., Block, J. H., Klopp, B., Lynam, D., Moffitt, T. E., &
Stouthamerloeber, M. (1992). A “common-language” version of the California Child
Q-set for personality assessment. Psychological Assessment, 4(4), 512–523.
Caspi, A., & Shiner, R. L. (2006). Personality development. In W. Damon, R. Lerner, &
N. Eisenberg (Eds.), Handbook of child psychology: Vol. 3. Social, emotional, and
personality development (6th ed., pp. 300–364). New York, NY: Wiley.
Caspi, A., & Shiner, R. L. (2008). Temperament and personality. In M. Rutter, D. Bishop,
S. Pine, S. Scott, J. Stevenson, E. Taylor, & A. Thapar (Eds.), Rutter’s child and
adolescent psychiatry (5th ed., pp. 182–199). London, England: Blackwell.
Cicchetti, D. (1990). Perspectives on the interface between normal and atypical
development. Development and Psychopathology, 2, 329–333.
Costa, P. T., & McCrae, R. R. (1989). The NEO-PI/NEO-FFI manual supplement.
Costa, P. T., & McCrae, R. R. (1992). Professional manual: Revised NEO Personality
Inventory (NEO-PI-R) and NEO Five-Factor-Inventory (NEO-FFI). Odessa, FL:
Costa, P. T., & McCrae, R. R. (1992). Professional manual: NEO inventories: NEO
Personality Inventory-3 (NEO-PI-3), Revised NEO Personality Inventory (NEO-PI-
RTM), and NEO Five-Factor Inventory-3 (NEO-FFI-3). Odessa, FL: Psychological
Côté, S. M., Vaillancourt, T., Barker, E. D., Nagin, D., & Tremblay, R. E. (2007). The
joint development of physical and indirect aggression: Predictors of continuity and
change during childhood. Development and Psychopathology, 19(01), 37–55.
Cukrowicz, K. C., Taylor, J., Schatschneider, C., & Iacono, W. G. (2006). Personality
differences in children and adolescents with attention-deficit/hyperactivity disorder,
conduct disorder, and controls. Journal of Child Psychology and Psychiatry, 47(2),
151–159.
De Fruyt, F., Bartels, M., Van Leeuwen, K. G., De Clercq, B., Decuyper, M., &
De Fruyt, F., De Bolle, M., McCrae, R. R., Terracciano, A., Costa, P. T., Jr., Ahn, C.-Y.,
… Shimonaka, Y. (2009). Assessing the universal structure of personality in early
adolescence: The NEO-PI-R and NEO-PI-3 in 24 cultures. Assessment, 16(3), 301–
311.
De Fruyt, F., & Mervielde, I. (2013). Hierarchical Personality Inventory for Children—
Experimental short version. Ghent University: Unpublished questionnaire.
De Fruyt, F., Mervielde, I., Hoekstra, H. A., & Rolland, J. P. (2000). Assessing
adolescents’ personality with the NEO PI-R. Assessment, 7(4), 329–345.
de Haan, A. D., De Pauw, S. S. W., Van den Akker, A. L., & Prinzie, P. (under review).
The only constant is change: Long-term development of lower-order Big Five facets.
Deal, J. E., Halverson, C. F., Havill, V. L., & Martin, R. P. (2005). Temperament factors
as longitudinal predictors of young adult personality. Merrill-Palmer Quarterly-
Journal of Developmental Psychology, 51(3), 315–334.
Deal, J. E., Halverson, C. F., Martin, R. P., Victor, J., & Baker, S. (2007). The Inventory
of Children’s Individual Differences: Development and validation of a short version.
Denissen, J. J. A., Geenen, R., Van Aken, M. A. G., Gosling, S. D., & Potter, J. (2008).
Development and validation of a Dutch translation of the Big Five Inventory (BFI).
Denissen, J. J. A., van Aken, M. A. G., Penke, L., & Wood, D. (2013). Self-regulation
underlies temperament and personality: An integrative developmental framework.
Child Development Perspectives, 7(4), 255–260.
De Pauw, S. S. W., & Mervielde, I. (2010). Temperament, personality and developmental
traits. Child Psychiatry & Human Development, 41(3), 313–329.
De Pauw, S. S. W., & Mervielde, I. (2011). The role of temperament and personality in
problem behaviors of children with ADHD. Journal of Abnormal Child Psychology,
39(2), 277–291.
De Pauw, S. S. W., Mervielde, I., & Van Leeuwen, K. G. (2009). How are traits related to
problem behavior in preschoolers? Similarities and contrasts between temperament and
personality. Journal of Abnormal Child Psychology, 37(3), 309–325.
De Pauw, S. S. W., Mervielde, I., Van Leeuwen, K. G., & De Clercq, B. J. (2011). How
temperament and personality contribute to the maladjustment of children with autism.
Journal of Autism and Developmental Disorders, 41(2), 196–212.
Digman, J. M. (1994). Child personality and temperament: Does the five-factor model
embrace both domains? In C. F. Halverson, G. A. Kohnstamm, & R. P. Martin (Eds.),
The developing structure of temperament and personality from infancy to adulthood
(pp. 323–338). Hillsdale, NJ: Erlbaum.
Digman, J. M., & Inouye, J. (1986). Further specification of the 5 robust factors of
personality. Journal of Personality and Social Psychology, 50(1), 116–123.
Digman, J. M., & Shmelyov, A. G. (1996). The structure of temperament and personality
in Russian children. Journal of Personality and Social Psychology, 71(2), 341.
Digman, J. M., & Takemotochock, N. K. (1981). Factors in the natural-language of
personality—Re-analysis, comparison, and interpretation of 6 major studies.
Multivariate Behavioral Research, 16(2), 149–170.
Donahue, E. M. (1994). Do children use the Big Five, too? Content and structural form in
personality description. Journal of Personality, 62(1), 45-66.
Eaton, W. (1994). Temperament, development, and the five-factor model: Lessons from
activity level. In C. F. Halverson, G. A. Kohnstamm, & R. P. Martin (Eds.), The
Ehrler, D. J., Evans, J. G., & McGhee, R. L. (1999). Extending big-five theory into
childhood: A preliminary investigation into the relationship between big-five
personality traits and behavior problems in children. Psychology in the Schools, 36(6),
451–458.
Evans, D., & Rothbart, M. K. (2007). Developing a model for adult temperament. Journal
of Research in Personality, 41(4), 868–888.
Gasman, I., Purper-Ouakil, D., Michel, G., Mouren-Simeoni, M. C., Bouvard, M., Perez-
Diaz, F., & Jouvent, R. (2002). Cross-cultural assessment of childhood temperament—
A confirmatory factor analysis of the French Emotionality Activity and Sociability
(EAS) Questionnaire. European Child & Adolescent Psychiatry, 11(3), 101–107.
Gerris, J. R. M., Houtmans, M. J. M., Kwaaitaal-Roosen, E. M. G., Schipper, J. C.,
Vermulst, A. A., & Janssens, J. M. A. M. (1998). Parents, adolescents and young
adults in Dutch families: a longitudinal study. Nijmegen: Institute of Family Studies,
University of Nijmegen.
Gjerde, P. F., & Cardilla, K. (2009). Developmental implications of openness to
experience in preschool children: Gender differences in young adulthood.
Goldberg, L. R. (1990). An alternative description of personality—The Big-5 factor
Psychological assessment, 4(1), 26.
measuring the lower-level facets of several five-factor models. In I. Mervielde, I.
Goldberg, L. R. (2001a). Analyses of Digman’s child-personality data: Derivation of big-
five factor scores from each of six samples. Journal of Personality, 69(5), 709–743.
Goldberg, L. R. (2001b). International Personality Item Pool. A scientific collaboratory
for the development of advanced measures of personality traits and other individual
differences. Internet Web Site, http://ipip.ori.org/.
factor structures from the top down. Journal of Research in Personality, 40(4), 347–
358.
Goldsmith, H. H., Buss, A. H., Plomin, R., Rothbart, M. K., Thomas, A., Chess, S., …
McCall, R. B. (1987). Round-table—What is temperament 4 approaches. Child
Development, 58(2), 505–529.
Goldsmith, H. H., & Rothbart, M. K. (1999). Laboratory temperament assessment battery
—Locomotor version 31. Eugene, OR: University of Oregon, Department of
Psychiatry.
Gough, H. G., and Heilbrun, A. B., Jr. (1983).Adjective Check List manual. Palo Alto,
CA: Consulting Psychologists Press.
Graziano, W. G., Jensen-Campbell, L., & Finch, J. F. (1997). The self as a mediator
between personality and adjustment. Journal of Personality and Social Psychology, 73,
392–404.
Grist, C. L., Socha, A., & McCord, D. M. (2012). The M5-PS-35: A Five-Factor
Personality questionnaire for preschool children. Journal of Personality Assessment,
94(3), 287–295.
Halverson, C. F., Havill, V. L., Deal, J., Baker, S. R., Victor, J. B., Pavlopoulos, V., …
Wen, L. (2003). Personality structure as derived from parental ratings of free
descriptions of children: The inventory of child individual differences. Journal of
Personality, 71(6), 995–1026.
Hampson, S. E., & Goldberg, L. R. (2006). A first large cohort study of personality trait
stability over the 40 years between elementary school and midlife. Journal of
Harter, S., Waters, P., & Whitesell, N. R. (1998). Relational self-worth: Differences in
perceived worth as a person across interpersonal contexts among adolescents. Child
Development, 69(3), 756–766.
Hegvik, R. L., McDevitt, S. C., & Carey, W. B. (1982). The Middle Childhood
Temperament Questionnaire. Journal of Developmental and Behavioral Pediatrics,
3(4), 197–200.
Hendriks, A. A. J. (1997). The construction of the Five-Factor Personality Inventory
(FFPI). Unpublished dissertation. Groningen, The Netherlands: University of
Groningen.
Hendriks, A. A. J., Kuyper, H., Offringa, G. J., & Van der Werf, M. P. (2008). Assessing
young adolescents’ personality with the five-factor personality inventory. Assessment,
15, 304–316.
Herzhoff, K., & Tackett, J. L. (2012). Establishing construct validity for Openness-to-
Experience in middle childhood: Contributions from personality and temperament.
Hoekstra, H. A., & De Fruyt, F. (2014). Dutch manual of the NEO-PI-3 en NEO-FFI-3
questionnaires. Amsterdam, Netherlands: Hogrefe Uitgevers.
John, O. P., Caspi, A., Robins, R. W., Moffitt, T. E., & Stouthamer-Loeber, M. (1994).
The Little 5—Exploring the nomological network of the 5-factor model of personality
in adolescent boys. Child Development, 65(1), 160–178.
John, O. P., & Srivastava, S. (1999). The Big Five trait taxonomy: History, measurement,
and theoretical perspectives. In L. A. Pervin & O. P. John (Eds.), Handbook of
personality: Theory and research (2nd ed., pp. 102-138). New York, NY: Guilford
Press.
Five trait taxonomy: History, measurement, and conceptual issues. In O. P. John, R. W.
Robins, & L. A. Pervin (Eds.), Handbook of personality: Theory and research (3rd ed.,
Kagan, J., Snidman, N., Zentner, M., & Peterson, E. (1999). Infant temperament and
anxious symptoms in school age children. Development and Psychopathology, 11(2),
209–224.
Kern, M. L., Hampson, S. E., Goldberg, L. R., & Friedman, H. S. (2014). Integrating
prospective longitudinal data: Modeling personality and health in the Terman Life
Cycle and Hawaii Longitudinal Studies. Developmental Psychology, 50(5), 1390.
Klimstra, T. A., Hale, W. W., Raaijmakers, A. W., Branje, S. J. T., & Meeus, W. H. J.
(2009). Maturation of personality in adolescence. Journal of Personality and Social
Psychology, 96(4), 898–912.
Knafo, A., & Israel, S. (2012). Empathy, prosocial behavior, and other aspects of
kindness. In M. Zentner & R. L. Shiner (Eds.), Handbook of temperament. New York,
NY: Guilford Press.
Knafo, A., Zahn-Waxler, C., Van Hulle, C., Robinson, J. L., & Rhee, S. H. (2008). The
developmental origins of a disposition toward empathy: Genetic and environmental
contributions. Emotion, 8(6), 737.
Kochanska, G., & Knaack, A. (2003). Effortful control as a personality characteristic of
young children: Antecedents, correlates, and consequences. Journal of Personality,
71(6), 1087–1112.
Kohnstamm, G. A., Halverson, C. F., Mervielde, I., & Havill, V. L. (Eds.). (1998).
Parental descriptions of child personality: Developmental antecedents of the Big Five?
Mahwah, NJ: Lawrence Erlbaum Associates.
Lamb, M. E., Chuang, S. S., Wessels, H., Broberg, A. G., & Hwang, C. P. (2002).
Emergence and construct validation of the big five factors in early childhood: A
longitudinal analysis of their ontogeny in Sweden. Child Development, 73(5), 1517–
1524.
Lounsbury, J. W., Tatum, H., Gibson, L. W., Park, S.-H., Sundstrom, E. D., Hamrick, F.
L., & Wilburn, D. (2003). The development of a Big Five adolescent personality
inventory. Journal of Psychoeducational Assessment, 21(2), 111–133.
Majdandzic, M., & van den Boom, D. C. (2007). Multimethod longitudinal assessment of
temperament in early childhood. Journal of Personality, 75(1), 121–167.
Markey, P. M., Markey, C. N., & Tinsley, B. J. (2004). Children’s behavioral
manifestations of the Five-Factor Model of personality. Personality and Social
Psychology Bulletin, 30(4), 423–432.
Markey, P. M., Markey, C. N., Tinsley, B. J., & Ericksen, A. J. (2002). A preliminary
validation of preadolescents’ self-reports using the Five-Factor Model of personality.
Martin, L. R., & Friedman, H. S. (2000). Comparing personality scales across time: An
illustrative study of validity and consistency in life-span archival data. Journal of
Martin, R. P., Wisenbaker, J., & Huttunen, M. (1994). Review of factor analytic studies
of temperament measures based on the Thomas-Chess structural model: Implications
for the Big Five. In C. F. Halverson, G. A. Kohnstamm, & R. P. Martin (Eds.), The
Masten, A. S., & Tellegen, A. (2012). Resilience in developmental psychopathology:
Contributions of the project competence longitudinal study. Development and
Psychopathology, 24(02), 345–361.
McAdams, D. P., & Pals, J. L. (2006). A new big five—Fundamental principles for an
integrative science of personality. American Psychologist, 61(3), 204–217.
McCrae, R. R., & Costa, P. T. (1985). Openness to experience. In R. Hogan & W. H.
Johnson (Eds.), Perspectives in personality (Vol. 1, pp. 145–172). Greenwich, CT: JAI
Press.
McCrae, R. R., & Costa, P. T. (1987). Validation of the 5-Factor Model of personality
across instruments and observers. Journal of Personality and Social Psychology, 52(1),
81–90.
McCrae, R. R., & Costa, P. T. (1996). Toward a new generation of personality theories:
Theoretical contexts for the five-factor model. In J. Wiggins (Ed.), The five-factor
model of personality: Theoretical perspectives (pp. 51–87). New York, NY: Guilford.
McCrae, R. R., Costa, P. T., & Martin, T. A. (2005). The NEO-PI-3: A more readable
revised NEO Personality Inventory. Journal of Personality Assessment, 84(3), 261–
270.
McCrae, R. R., Costa, P. T., Ostendorf, F., Angleitner, A., Hrebickova, M., Avia, M. D.,
… Smith, P. B. (2000). Nature over nurture: Temperament, personality, and life span
development. Journal of Personality and Social Psychology, 78(1), 173–186.
Mervielde, I. (2002). Personality trait development from age 12 to age 18:
McGhee, R. L., Ehrler, D. J., & Buckhalt, J. A. (2007). Five-Factor Personality Inventory
—Children. Austin, TX: PRO-ED.
Measelle, J. R., John, O. P., Ablow, J. C., Cowan, P. A., & Cowan, C. P. (2005). Can
children provide coherent, stable, and valid self-reports on the Big Five dimensions? A
longitudinal study from ages 5 to 7. Journal of Personality and Social Psychology,
89(1), 90–106.
Mervielde, I. (1994). A Five-Factor Model classification of teachers’ constructs on
individual differences among children ages 4 to 12. In C. F. Halverson, G. A.
Kohnstamm, & R. P. Martin (Eds.), The developing structure of temperament and
personality from infancy to adulthood (pp. 323–338). Hillsdale, NJ: Erlbaum.
Mervielde, I., Buyst, V., & De Fruyt, F. (1995). The validity of the Big-5 as a model for
teachers ratings of individual-differences among children aged 4–12 years. Personality
Mervielde, I., De Clercq, B., De Fruyt, F., & Van Leeuwen, K. (2005). Temperament,
personality, and developmental psychopathology as childhood antecedents of
personality disorders. Journal of Personality Disorders, 19(2), 171–201.
Mervielde, I., & De Fruyt, F. (1999). Construction of the Hierarchical Personality
Inventory for Children (HiPIC). In I. Mervielde, I. Deary, F. De Fruyt, & F. Ostendorf
(Eds.), Personality psychology in Europe (pp. 107–127). Tilburg, Netherlands: Tilburg
University Press.
Mervielde, I., & De Fruyt, F. (2000). The Big Five personality factors as a model for the
structure of children’s peer nominations. European Journal of Personality, 14(2), 91–
106.
Mervielde, I., De Fruyt, F., & De Clercq, B. J. (2009). Manual of the Hierarchical
Personality Inventory for Children. Amsterdam, Netherlands: Hogrefe.
Mervielde, I., De Fruyt, F., & Jarmuz, S. (1998). Linking openness and intellect in
childhood and adulthood. In G. A. Kohnstamm, C. F. Halverson, I. Mervielde, & V. L.
Havill (Eds.), Parental descriptions of child personality: Developmental antecedents of
the big five? (pp. 105–142). Mahwah, NJ: Erlbaum.
Mervielde, I., & De Pauw, S. S. W. (2012). Models of child temperament. In M. Zentner
& R. L. Shiner (Eds.), Handbook of temperament. New York, NY: Guilford Press.
Parker, W. D., & Stumpf, H. (1998). A validation of the five-factor model of personality
in academically talented youth across observers and instruments. Personality and
Presley, R., & Martin, R. P. (1994). Toward a structure of preschool temperament—
Factor structure of the Temperament Assessment Battery for Children. Journal of
Personality, 62(3), 415–448.
Prinzie, P., van Harten, L. V., Dekovic, M., van den Akker, A. L., & Shiner, R. L. (2014).
Developmental trajectories of anxious and depressive problems during the transition
from childhood to adolescence: Personality × Parenting interactions. Development and
Psychopathology, 26(4), 1077–1092.
Putnam, S. P., Ellis, L. K., & Rothbart, M. K. (2001). The structure of temperament from
infancy through adolescence. In A. Eliasz & A. Angleitner (Eds.), Advances and
proceedings in research on temperament (pp. 165–182). Lengerich, Germany: Pabst
Scientist Publisher.
Putnam, S. P., Gartstein, M. A., & Rothbart, M. K. (2006). Measurement of fine-grained
aspects of toddler temperament: The Early Childhood Behavior Questionnaire. Infant
Behavior & Development, 29(3), 386–401.
Putnam, S. P., Rothbart, M. K., & Gartstein, M. A. (2008). Homotypic and heterotypic
continuity of fine-grained temperament during infancy, toddlerhood, and early
childhood. Infant and Child Development, 17(4), 387–405.
Resing, W. C. M., Bleichrodt, N., & Dekker, P. H. (1999). Measuring personality traits in
the classroom. European Journal of Personality, 13(6), 493–509.
Roberts, B. W., Kuncel, N. R., Shiner, R. L., Caspi, A., & Goldberg, L. R. (2007). The
Roth, J. H., Dadds, M. R., & McAloon, J. (2004). Evaluation of a Puppet Interview to
measure young children’s self-reports of temperament. Behaviour Change, 21(1), 37–
56.
Rothbart, M. K. (1981). Measurement of temperament in infancy. Child Development,
52(2), 569–578.
Rothbart, M. K. (2012). Advances in temperament: History, concepts, and measures. In
M. Zentner & R. L. Shiner (Eds.), Handbook of temperament. New York, NY: Guilford
Press.
Rothbart, M. K., & Ahadi, S. A. (1994). Temperament and the development of
personality. Journal of Abnormal Psychology, 103(1), 55–66.
Rothbart, M. K., & Bates, J. E. (1998). Temperament. In W. Damon & N. Eisenberg
(Eds.), Handbook of child psychology: Vol. 3. Social, emotional, and personality
development (5th ed., pp. 105–176). New York, NY: Wiley.
Rothbart, M. K., & Bates, J. E. (2006). Temperament. In W. Damon, R. Lerner, & N.
Eisenberg (Eds.), Handbook of child psychology: Vol. 3. Social, emotional, and
personality development (6th ed., pp. 99–166). New York, NY: Wiley.
Rothbart, M. K., & Posner, M. I. (2006). Temperament, attention, and developmental
psychopathology. In D. Cicchetti & D. J. Cohen (Eds.), Developmental
psychopathology: Vol. 2. Developmental neuroscience (2nd ed., pp. 465–501). New
York, NY: Wiley.
Scholte, R. H. J., Van Aken, M. A. G., & van Lieshout, C. F. M. (1997). Adolescent
personality factors in self-ratings and peer nominations and their prediction of peer
acceptance and peer rejection. Journal of Personality Assessment, 69, 534–554.
Shiner, R. L. (1998). How shall we speak of children’s personalities in middle childhood?
A preliminary taxonomy. Psychological Bulletin, 124(3), 308–332.
Psychiatry and Allied Disciplines, 44(1), 2–32.
Shiner, R. L., & DeYoung, C. G. (2013). The structure of temperament and personality
traits: A developmental perspective. In P. Zelazo (Ed.), Oxford handbook of
developmental psychology. New York, NY: Oxford University Press.
Shiner, R. L., & Masten, A. S. (2012). Childhood personality as a harbinger of
competence and resilience in adulthood. Development and Psychopathology, 24(2),
507–528.
Simonds, J., & Rothbart, M. K. (2004, October). The Temperament in Middle Childhood
Questionnaire (TMCQ): A computerized self-report instrument for ages 7–10. Paper
presented at the Occasional Temperament Conference, Athens, GA.
Soto, C. J. (2015). The little six personality dimensions from early childhood to early
adulthood: Mean-level age and gender differences in parents’ reports. Journal of
Personality. doi:10.1111/jopy.12168
personality traits from early childhood to early adulthood. Journal of Personality,
82(3), 182–199.
Soto, C. J., John, O. P., Gosling, S. D., & Potter, J. (2008). The developmental
psychometrics of Big Five self-reports: Acquiescence, factor structure, coherence, and
differentiation from ages 10 to 20. Journal of Personality and Social Psychology,
94(4), 718–737.
Szirmak, Z. (2005). The Big Five Model of Personality and Primary Prevention in
Adolescence. Berlin, Germany: Free University Berlin. Unpublished doctoral
dissertation.
in children and adolescents. Clinical Psychology Review, 26(5), 584–599.
Tackett, J. L., Herzhoff, K., Kushner, S. C., & Rule, N. (2015). Thin Slices of Child
Personality: Perceptual, Situational, and Behavioral Contributions. Journal of
Personality and Social Psychology. [Epub ahead of print].
Tackett, J. L., Krueger, R. F., Iacono, W. G., & McGue, M. (2008). Personality in middle
childhood: A hierarchical structure and longitudinal connections with personality in
late adolescence. Journal of Research in Personality, 42(6), 1456–1462.
Tackett, J. L., Kushner, S. C., De Fruyt, F., & Mervielde, I. (2013). Delineating
temperament, and behavioral problems. Assessment, 20(6), 738–751.
Tackett, J. L., Slobodskaya, H. R., Mar, R. A., Deal, J., Halverson, C. F., Jr., Baker, S. R.,
… Besevegis, E. (2012). The hierarchical structure of childhood personality in five
Personality, 80(4), 847–879.
Development of the Multidimensional Personality Questionnaire. In G. J. Boyle, G.
Matthews, & D. H. Saklofske (Eds.), The Sage handbook of personality theory and
assessment (Vol. 2).Thousand Oaks, CA: Sage.
Thomas, A., & Chess, S. (1977). Temperament and development. New York, NY:
Brunner/Mazel.
Thomas, A., Chess, S., Birch, H. G., Hertzig, M. E., & Korn, S. (1963). Behavioral
individuality in early childhood. New York, NY: New York University Press.
Thomas, A., Mittelman, M., Chess, S., Korn, S. J., & Cohen, J. (1982). A temperament
questionnaire for early adult life. Educational and Psychological Measurement, 42(2),
593–600.
Tremblay, R. E., Nagin, D. S., Séguin, J. R., Zoccolillo, M., Zelazo, P. D., Boivin, M., …
Japel, C. (2004). Physical aggression during early childhood: Trajectories and
predictors. Pediatrics, 114(1), e43–e50.
van den Akker, A. L., Dekovic, M., Asscher, J., & Prinzie, P. (2014). Mean-level
personality development across childhood and adolescence: A temporary defiance of
the maturity principle and bidirectional associations with parenting. Journal of
van den Akker, A. L., Dekovic, M., & Prinzie, P. (2010). Transitioning to adolescence:
How changes in child personality and overreactive parenting predict adolescent
adjustment problems. Development and Psychopathology, 22(1), 151–163.
Van Lieshout, C. F., & Haselager, G. (1994). The Big Five personality factors in Q-sort
descriptions of children and adolescents. In C. F. Halverson, G. A. Kohnstamm, & R.
P. Martin (Eds.), The developing structure of temperament and personality from
infancy to adulthood (pp. 293–318). Hillsdale, NJ: Erlbaum.
Vermulst, A. A., & Gerris, J. R. M. (2005). Manual of the Quick Big Five Personality
Inventory. Leeuwarden, Netherlands: LDC.
Widiger, T. A. (2011). Integrating normal and abnormal personality structure: A proposal
for DSM-V. Journal of Personality Disorders, 25(3), 338–363.
Zaal, J. N. (1978). Social-emotional behavior in the classroom. Doctoral dissertation,
Wolters-Noordhoff, Groningen, The Netherlands.
Zupancic, M. (2004). Parental free descriptions of child personality: Applicability of The
Five-Factor Model taxonomy from infancy through pre-school years. Studia
Psychologica, 46(2), 145–162.
Zupancic, M., Podlesek, A., & Kavcic, T. (2006). Personality types as derived from
parental reports on 3-year-olds. European Journal of Personality, 20(4), 285–303
Animal Personality
Alexander Weiss and Marieke C. Gartner
Abstract
Animal personality has been studied for decades, and a recent renaissance in the
field has revealed links to health and life outcomes that echo those found in
humans. Some of this research is tied to the Five Factor Model—the predominant
model of human personality—which informs animal personality research as well,
and allows for comparative work that points to evolutionary pathways that
delineate phylogenetic continuity. From personality facets and traits to factors,
this work has implications for human and nonhuman animal genetics, life history
strategies, survival, and well-being, as well as development and social
relationships. Working together, scientists from a variety of fields who study
personality can hope to puzzle out causality, use personality as a tool for health,
and simply define personality, across species, and therefore evolutionary time.
Key Words: animal, behavioral syndrome, coping styles, evolution, genetics,
health, personality, phylogeny, temperament, well-being
Animal personality research is a major endeavor in fields ranging from

comparative psychology to evolutionary biology to the agricultural and
veterinary sciences. As a consequence, there are multiple approaches to studying
personality in animals, some being rooted in older research traditions and others
that are just being developed. Although much has been written about what can
be learned about animal personality by understanding more about human
personality, including the Five Factor Model (Digman, 1990), we will show that
the converse is also true.
A Brief History
Early Origins
The observations that animals are consistent in how they behave, how they
react to the world, and even how they perceive the world probably originated
long before modern psychology or biology. In the scientific sense, the study of
animal personality has a pedigree that extends at least as far back as to Charles
Darwin’s The Expression of the Emotions in Man and Animals (Darwin,
1872/1998).
More recent origins of the idea that animals have personality goes back to
Pavlov, who described four types of personality profiles or, to use his
terminology, “nervous system types” (excitable, lively, calm, and inhibited;
Pavlov, 1941) in dogs (Canis familiaris). According to Pavlov, these personality
types dictated how dogs learned—for example, how rapidly dogs were able to
extinguish a conditioned response. According to Locurto (2007), these ideas
influenced Eysenck’s first model of human personality, which was built on two
axes, one ranging from neurotic to emotionally stable and the other ranging from
introversion to extraversion.
Early studies on nonhuman primate personality began in the 1930s and 1940s
at the Yale Laboratories of Primate Biology. In speeches that Yerkes gave (e.g.,
1939), he noted the objections that some had to his research but he and those
with whom he worked not only observed individual differences in the
chimpanzees (Pan troglodytes) they studied they also presented their work with
no doubt that personality existed in these animals. Crawford (1938) noted that
just a few days with chimpanzees revealed individual differences among them.
To test this idea, she had people who worked with laboratory chimpanzees rate
them on 22 behaviors and traits via a survey. She found good interrater and
retest reliabilities for the items, and based on intercorrelations of the items,
suggested two group factors. The first was composed of intelligence, motor skill,
confidence in the observer, and desire to please; the second was composed of
friendliness with strangers, cheerfulness, and noisiness. Timidity with strangers,
although a single item, was negatively related to almost all other items.
Yerkes later observed in a speech that “… there is no question about the
reality of chimpanzee mind, individuality, personality” (1939, p. 97). Working
with laboratory chimpanzees, he noted that they generally have two types of
personality: commanding and obedient, but also discussed how some
chimpanzees are immediately trusting, whereas others are not, as well as patient,
tractable, docile, suggestive, gentle, friendly, or their opposites. Similarly, Hebb
(1949) noted that chimpanzee personality is similar to that of humans in
complexity. He tested this idea with behavioral reactions of chimpanzees to both
humans and inanimate objects. He found test–retest reliability over 8 years, and
individual variation in the behavior of chimpanzees toward familiar caretakers, a
“timid” unfamiliar human, a “bold” unfamiliar human, and inanimate objects.
By the 1960s, however, the idea of personality in nonhuman animals was
losing favor in the scientific community, and especially in ethology and
psychology. At the same time, Jane Goodall traveled to what is now Tanzania’s
Gombe National Park to study chimpanzees. While there, she named the animals
she was observing—including David Graybeard, the first animal observed to
make and use tools, and the first to let her observe him—and attributed
personalities to them (as well as acknowledging that they had minds and
feelings; Goodall, 1986), using words such as “affectionate” and “supportive.”
The Decline
However, despite such promising beginnings, with research being led by the
most prominent researchers of the day, the study of animal personality soon
thereafter suffered a steep decline. Although there were, to our knowledge, only
a few direct attacks on animal personality research (see, e.g., Goodall, 1990), it
suffered from multiple shifts in the science of human and animal behavior.
Most prominent among these shifts was a change in the view of what should
constitute the scientific method for the study of behavior. In psychology this
change was exemplified by the rise of the radical behaviorists, who argued that
the focus of research should be on observable behavior and how the
consequences of behavior and the cues that signal these consequences change
behavior (Skinner, 1964). Psychologists also did not view individuals’ inner
worlds, that is, their thoughts and emotions, as being valid subjects of study
(Skinner, 1964). As such, although radical behaviorists did not argue against the
existence of personality or against studying it, they viewed personality as
behavioral habits that reflected each individual’s learning history and not
individual differences in traits rooted in the organism’s biology (Skinner, 1964).
For those interested in studying animal behavior, there was no refuge to be
found in the study of ethology, either. Although the early ethologists disagreed
with the radical behaviorists on many fronts, most notably on whether we should
study behaviors that are learned or behaviors that naturally occur, they agreed
that only observable behavior merited scientific study (Tinbergen, 2005).1
Another development hindering the study of animal personality was the
decline in trait-oriented human personality research that followed the publication
of Walter Mischel’s Personality and Assessment (1968). In his book, Mischel
drew on social learning theory and early cognitive psychology, but he also
reevaluated the evidence for personality that had been collected in previous
studies, such as the association between individuals’ personalities and how they
behave within groups (Mischel, 1968, p. 79). Based on his findings and a review
of the field, he concluded that personality stability was an illusion, resulting
from cognitive biases, such as projection or the tendency of individuals to see
others only in certain situations that elicit those behaviors, and to other artifacts.
Mischel also pointed out that the correlations between personality and actual
behavior such as the aforementioned behavior in groups rarely exceeded .3 and
were thus small in comparison to the power of situations to predict behavior.
Mischel therefore argued that it would be more fruitful to study the dynamics of
behaviors and their elicitation by situations and cues. Mischel’s more recent
work exemplifies this by operationalizing personality as individual differences in
the response of individuals across multiple situations (Shoda & Mischel, 1996).
One development with a mixed effect on the study of animal personality was
the emergence of cognitive ethology in the 1970s. Donald Griffin (1978), a
founder of the field of animal cognition, put forward the view that consciousness
was not exclusive to humans, and could be scientifically studied. On the other
hand, some prominent researchers in animal cognition were (and still are) of the
opinion that we should not use cognition as understood in humans as a basis for
making predictions about animal cognition or in interpreting animal behavior
(e.g., Shettleworth, 2012). Moreover, these researchers labeled Darwin’s (1871)
view that the differences between humans and animals were continuous as
“anthropomorphic” or “anthropocentric” (e.g., Shettleworth, 2012). They argued
that when explaining the seemingly intelligent behavior of animals, we should
adhere to Morgan’s Canon (1894) and should use the simplest explanation
possible.
A recent example illustrates this divide in views on how to interpret animal
behavior. Santino is a male chimpanzee housed in the Furuvik Zoo in Gävle,
Sweden (Osvath, 2009). He was known for throwing stones and pieces of
concrete at visitors for at least a decade. Further investigation revealed that his
“missiles” were not merely objects that had come to hand, but that he collected
and stored the stones and pieces of concrete in strategic locations, and had even
shaped the concrete. This behavior took place prior to the zoo’s opening hours
and so there was a delay of several hours between the gathering and throwing of
stones and concrete. These and other behaviors led to the conclusion that
Santino’s behavior was premeditated. Others, however, argued that because the
caches of stones were not observed until Santino’s behavior caught the attention
of the keepers, it is unclear whether they had been gathered for some other
reason (Shettleworth, 2010, 2012; Suddendorf & Corballis, 2010). Of course this
is a possibility, though it is worth noting that neither Shettleworth nor
Suddendorf and Corballis commented on the fact that Santino had broken off
and shaped the pieces of concrete that he added to his arsenal.
The Renaissance
The 1970s also marked the start of new beginnings in the study of personality
in nonhuman animals. This period was originally focused on studies of
nonhuman primates, but grew to encompass other species. It was also marked by
the use of psychometric and multivariate methods now commonly used in
human personality research. The use of these methods likely helped further
legitimize the study of animal personality (Joan Stevenson-Hinde, personal
communication, May 25, 2014). The studies of this period were chiefly focused
on personality or behaviors related to it in baboons (Papio anubis; Buirski,
Kellerman, Plutchik, Weininger, & Buirski, 1973), chimpanzees (Buirski,
Plutchik, & Kellerman, 1978), and rhesus macaques (Macaca mulatta;
Stevenson-Hinde & Zunz, 1978; Stevenson-Hinde, Stillwell-Barnes, & Zunz,
1980a, 1980b). Wild baboons in Kenya were rated on the 12-item Emotions
Profile Index (Buirski et al., 1973). The interrater reliabilities all exceeded .7,
and individual differences were found among the troop. The authors also found
that more submissive animals were rated as more social and affectionate, as well
as more fearful, than dominant animals. This work was replicated in wild
chimpanzees (Buirski et al., 1978), and again the ratings were reliable.
Individual differences were found, as were sex differences (females were more
timid, depressed, and trusting than males, whereas males were more aggressive,
sociable, and impulsive). The authors also found a positive relationship between
dominance rank and aggressiveness. Finally, they compared profile results
among baboons, chimpanzees, and humans, and found very high correlations
among males, with all groups showing high scores in gregariousness and trust,
and lower scores on depression, aggression, and distrust.
Stevenson-Hinde and Zunz (1978) and Stevenson-Hinde et al., (1980a, 1980b)
carried out what are now considered seminal works on nonhuman primate
personality. The first stage of this research (the 1978 study) involved developing
what has come to be known as the Madingley Questionnaire (Stevenson-Hinde
& Hinde, 2011),2 which was composed of “behaviorally-defined adjectives”
based on behaviors they observed in rhesus macaques and on Sheldon’s (1942)
measure of human temperament. They then asked individuals with considerable
experience in observing and recording the behaviors of these monkeys to rate
each monkey using this questionnaire at three time points. They used a principal
components analysis to interpret their results for each set of ratings. In the first
phase, the 19 items deemed to have sufficient interrater reliabilities yielded the
same two components, Confident to Fearful and Active to Slow. In the second
phase, 23 items had sufficient interrater reliabilities. Principal components
analysis of these items found two components similar to those identified in the
earlier waves and a third component, Sociable to Solitary.
In addition to identifying these components, they identified age, sex, and
maternal effects. Males scored higher on the first two factors, whereas females
scored higher on the third. Over the study’s 4 years, Confident to Fearful scores
remained stable, whereas the remaining two factors were stable only in adults.
For example, young males with an adverse experience within the first 8 months
of life were more Active than other males, but just as Confident and/or Sociable.
In terms of maternal effects, Confident and Sociable mothers had Confident and
Sociable infants, respectively. Active mothers had infants that were Fearful.
Finally, first-time mothers showed stability in their scores for Active from
prepregnancy to postpregnancy.
Feaver, Mendl, and Bateson (1986) used a method similar to this early primate
work to assess personality in domestic cats (Felis silvestris catus). Using
behavioral observations and a questionnaire based on the Madingley
Questionnaire but altered to suit cat behavior, the authors found three personality
factors: Alert, Sociable, and Equable (as all the cats were female, sex effects
were not investigated).
Personality work in domestic dogs began in the 1930s, and slowly began
proliferating (Jones & Gosling, 2005). Rating methods, however, did not begin
until the 1980s (Serpell, 1983), and became more common in the 1990s and
beyond. Previous work utilized test batteries and observational tests. Work in
dogs has been breed or work related most commonly.
The Age of Discovery

The field of animal personality began to truly emerge in the 1990s with two
studies that set the path for the research that has continued to date. In the first
study, Gold and Maple (1994) assessed nearly 300 gorillas (Gorilla gorilla)
housed in zoos using a modified version of the Madingley Questionnaire. Their
study identified personality factors they labeled Dominant, Extroverted, Fearful,
and Understanding, which suggested that some personality dimensions in great
apes were human-like. The second study was conducted by King and Figueredo
(1997), who asked zoo keepers and volunteers to rate 100 chimpanzees on the
Chimpanzee Personality Questionnaire. This questionnaire consisted of 43
adjectives, of which 41 were sampled from each of the domains in Goldberg’s
characterization of the Five Factor Model (1990). Adjectives were adapted for
use in chimpanzees by pairing each with one to three sentences that set them in
the context of behavior (King & Figueredo, 1997). Factor analysis of the 41
reliable items revealed six factors. The first and largest factor was unlike
domains found in human studies of personality and contained a mix of items
denoting assertiveness, independence, low fear, and aggression. It resembled the
rhesus macaque dimension Confidence (Stevenson-Hinde & Zunz, 1978) and the
gorilla dimension Dominant (Gold & Maple, 1994). This chimpanzee dimension
was thus labeled Dominance. The remaining five factors, labeled Surgency,
Dependability, Agreeableness, Emotionality, and Openness, resembled the Five
Factor Model domains of Extraversion, Conscientiousness, Agreeableness,
Neuroticism, and Openness, respectively.3 Finally, during this period,
personality began to be studied in nonmammals in natural populations, to
understand its evolutionary implications. For example, the shy–bold continuum
was viewed as a product of natural selection, with implications for survival
strategies, frequency-dependent selection, and phenotypic plasticity (Wilson,
Clark, Coleman, & Dearstyne, 1994).
King and Figueredo’s 1997 study was soon followed by two other articles. In
the first Gosling (1998) asked raters to assess the personality traits of spotted
hyenas (Crocuta crocuta). Traits were sampled from the Big Five mini-markers
(Saucier, 1994, cited in Gosling, 1998), traits used in prior studies of animal
personality, and traits devised to capture aspects of hyena behavior that were not
captured by the other traits. Principal components analysis of the 44 reliable
items yielded dimensions labeled Assertiveness, Excitability, Human-Directed
Agreeableness, Sociability, and Curiosity. In the second article Capitanio (1999)
used a variant of the Madingley Questionnaire. In addition to identifying three
rhesus macaques dimensions similar to those described by Stevenson-Hinde et
al. (1978, 1980a), he took up the gauntlet (John P. Capitanio, personal
communication, June 1, 2014) laid down by Mischel (1968) and showed that the
ratings predicted behavior over long periods of time and in situations other than
those in which personality was assessed.
Thus, by the late 1990s, enough work had been done on personality in a
variety of species for a review to be carried out. Gosling and John (1999) found
19 studies of personality factors in 12 species, and found, based on the Five
Factor Model, that Extraversion, Neuroticism, and Agreeableness generalized
the most across those species, with Openness being identified in fewer species
and Conscientiousness being identified only in chimpanzees.
This was followed by Gosling’s (2001) review and meta-analysis that showed
strong evidence for the interrater and retest reliabilities of animal personality
measures and also for their association with behavior. This review agreed with
the 1999 review by Gosling and John with respect to which personality
dimensions appeared most commonly and highlighted the tendency for animal
personality studies to identify dimensions related to dominance and/or activity.
One interesting difference between human and nonhuman animal personality is
the common occurrence of a Dominance factor in the latter, and its absence in
the former. Recent work has shown that although there is no independent
Dominance factor in humans, there are Dominance-related traits, as well as a
dimension measured by the Psychopathic Personality Inventory (Lilienfeld &
Andrews, 1996) called Fearless Dominance. This dimension is composed of
traits such as social dominance, charm, physical fearlessness, and immunity to
anxiety (Lilienfeld et al., 2012). This dimension is associated with healthy
psychological adjustment (e.g., Benning, Patrick, Hicks, Blonigen, & Krueger,
2003; Patrick, Edens, Poythress, Lilienfeld, & Benning, 2006). In terms of the
Five Factor Model, higher levels of Extraversion and Openness and lower levels
of Neuroticism and Agreeableness predict Fearless Dominance (Ross, Benning,
Patrick, Thompson, & Thurston, 2009).
The Modern Era

Establishing reliability and validity in nonhuman animal personality
assessment and determining personality structure are still continuing in a variety
of species that have not had the same focus as primates (e.g., in felids, Gartner,
Powell, & Weiss, 2014). In addition, a new focus in personality research aims to
explain it in evolutionary and genetic terms (e.g., Adams, King, & Weiss, 2012;
Suomi, 2006; Weiss, King, & Figueredo, 2000), and to link it to health outcomes
(e.g., Capitanio, Mendoza, & Baroncelli, 1999; Weiss, Gartner, Gold, &
Stoinski, 2013). Behavioral ecologists also address evolutionary issues, but with
a different focus, using the traits that comprise personality to show that it can
affect species distribution, species adaptability, and speciation rates (e.g., Réale,
Reader, Sol, McDougall, & Dingemanse, 2007; Sih, Bell, Johnson, & Ziemba,
2004). Reviews of this type of work showed that personality affects fitness in a
wide range of wild and captive species (Dingemanse & Réale, 2005; Smith &
Blumstein, 2008).
The Five Factor Model’s Contribution to Animal Personality

Research
Despite the rebirth of animal personality research, the scars from the previous
assaults remain. This is particularly the case when it comes to
anthropomorphism, which Jane Goodall referred to as the “cardinal sin” in
ethology (1990, p. 13). Although there are some researchers who make efforts to
avoid even the appearance of the possibility of anthropomorphism, others tend to
make due with purchasing indulgences before or confessing and engaging in acts
of contrition after committing the sin. The latter cases are recognizable by
Introduction and Discussion sections in which much ink is spilled to justify the
use of methods common to the human literature such as questionnaire ratings
(e.g., Weiss, Adams, Widdig, & Gerald, 2011). Whether these tendencies are
some ingrained Western tradition that predates the modern scientific era is hard
to say, but it is an unfortunate state of affairs. The knowledge gained in
comparative psychology, evolutionary biology, anthropology, and any number
of similar fields by adopting Darwin’s (1871) gradualist view of a continuity of
species has been great. Not surprisingly, questions, approaches, and findings
from human personality research and the Five Factor Model have contributed to
the development of animal personality research.
Identifying Species-Typical Structures

Perhaps the most important idea that found its way from studies of human
personality and the Five Factor Model has been the importance of studying
personality structure, that is, testing whether the Five Factor Model or Big Five
is a human universal (McCrae & Costa, 1997), is a linguistic-cultural artifact,
and is found only in Anglo-Germanic language groups (Saucier & Goldberg,
2001), or whether fewer dimensions (e.g., Eysenck, 1970), more dimensions
(e.g., Lee, Ogunfowora, & Ashton, 2005), or a different configuration of traits
(e.g., Zuckerman, Kuhlman, Thornquist, & Kiers, 1991) best characterize human
personality differences (see also the chapters by Allik and Realo, by Costa and
McCrae, and by De Raad and Mlačić).
The importance of studying personality structure as a species universal in
animals came to the fore with King and Figueredo’s (1997) study of chimpanzee
personality. The finding that the similarities and differences between humans
and their close phylogenetic cousins revealed differences in degree, that is,
chimpanzees possessed five human-like domains, and differences in kind,
namely the identification of a chimpanzee-specific Dominance domain, allowed
for the idea that the evolutionary continuity of personality could be studied.
Further research took a cue from the studies that showed that the Five Factor
Model exists in cultures very different from that in which it was identified
(McCrae, Terracciano, & 78 Members of the Personality Profiles of Cultures
Project, 2005). These studies sought to determine whether King and Figueredo’s
(1997) findings were robust across different samples or, putting it another way,
whether the personality structure that they identified was a “chimpanzee
universal.” To this end, researchers used the same questionnaire to study
personality structure among chimpanzees living in a naturalistic sanctuary in the
Republic of the Congo (King, Weiss, & Farmer, 2005) and in Yerkes National
Primate Research Center, a laboratory setting (Weiss, King, & Hopkins, 2007).
They also used an expanded version of the questionnaire to study chimpanzees
living in zoos, research institutes, and a sanctuary located in Japan (Weiss et al.,
2009). These studies yielded similar results. Targeted orthogonal Procrustes
rotation (McCrae, Zonderman, Costa, Bond, & Paunonen, 1996) yielded overall
congruence coefficients of .89, .90, and .85, respectively, indicating that the
whole structure replicated, high congruence coefficients indicating that
Dominance, Extraversion, Conscientiousness, and Agreeableness replicated, and
markedly lower congruence coefficients for Neuroticism and Openness that were
not consistent with the replicability of these dimensions. In short, the personality
structure identified by King and Figueredo (1997) was, for the most part, not
likely an artifact of the zoo environment, the English language, or culturally
based expectations on the part of raters. As Neuroticism or Openness was
defined by only three and two questions, respectively, and so would be less
stable, the question of whether one or both of these domains are a feature of
chimpanzee personality remains.
Of course it is possible that features of the questionnaire, such as the
adjectives chosen and how descriptor sentences were worded, may be wholly or
partly responsible for the high degree of consistency in chimpanzee personality
structure. Excluding this possibility requires studying personality using this same
instrument in species that differ enough in their evolutionary origins, in their
socioecology, and in other features (Gosling, 2001; Gosling & Graybeal, 2007).
With respect to nonhuman primates, this was tested in a study of orangutans
using an augmented version of the questionnaire used in King and Figueredo’s
study (1997) that included five new items that attempted to tap into the
Neuroticism and Openness domains (Weiss, King, & Perkins, 2006). Orangutans
shared a common ancestor with humans and chimpanzees approximately 15
million years ago (Purvis, 1995). Crucially, orangutans differ from chimpanzees
and humans in that they are semisolitary (Galdikas, 1985; MacKinnon, 1974).
As such, their personality might be expected to differ from humans and
chimpanzees, two very social species. This is what Weiss et al. (2006) found.
Orangutan personality was characterized by Extraversion, Neuroticism, and
Agreeableness domains that resembled those of chimpanzees and humans, a
Dominance domain, which somewhat resembled that of chimpanzees, though it
mainly included traits indicative of low Agreeableness, such as “bullying,” and a
domain, “Intellect,” which combined traits that tapped Openness, such as
“intelligent,” and those related to Conscientiousness, such as “independent.”
A later study by Weiss, Adams, Widdig, et al. (2011) investigated rhesus
macaques, which shared a common ancestor with humans and great apes
approximately 25 to 30 million years ago (Purvis 1995). Although rhesus
macaques, like chimpanzees and humans, are a highly social species, their
societies are differently arranged; females remain in the group in which they
were born and arrange themselves in hierarchies based around female lines,
whereas males leave the group into which they were born just prior to puberty
(Colvin, 1986; Gouzoules & Gouzoules, 1987; Manson, 1995; Melnick & Pearl,
1987). This study used an augmented version of the instrument used to rate
orangutans (Weiss et al., 2006) that included six additional items related to
Conscientiousness and Openness. The study found six personality dimensions.
Two domains, Confidence and Anxiety, represented emotional reactions to
social and nonsocial stimuli, respectively. There was also a Dominance domain
similar to the chimpanzee and orangutan Dominance domains (King &
Figueredo, 1997; Weiss et al., 2006), and the social aspect of Extraversion and
traits related to Agreeableness were found in a domain labeled Friendliness.
Finally, there were separate domains related to Activity and Openness.
Recently, a similar study investigated personality structure in brown capuchin
monkeys (Sapajus apella; Morton et al., 2013), which share a very distant
common ancestor with the macaques, the great apes, and humans some 40
million years ago (Purvis, 1995). Of particular interest was the question of
whether brown capuchin monkey personality would resemble that of
chimpanzees, since brown capuchin monkeys, like other capuchin monkeys,
behaviorally resemble chimpanzees in many respects, including their fission–
fusion social structure and the creation and use of tools (Fragaszy, Visalberghi,
& Fedigan, 2004). Brown capuchin monkey personality structure resembled that
of chimpanzees as it included a domain similar to Dominance labeled
Assertiveness, as well as Neuroticism and Openness domains. It also included a
narrow variant of Conscientiousness labeled Attentiveness (Morton et al., 2013).
The only difference between brown capuchin monkey and chimpanzee
personality structure was that items related to Extraversion and Agreeableness
loaded on a single dimension, labeled Sociability (Morton et al., 2013).
These findings are not unique to this family of questionnaires or studies of
nonhuman primates. Similar findings were described in studies that relied on
other animal personality questionnaires, including the Madingley Questionnaire
and a blend of the Madingley Questionnaire and King and Figueredo’s
questionnaire developed by Konečná et al. (2008). Studies using the Madingley
Questionnaire identified different structures in rhesus macaques (Capitanio,
1999; Stevenson-Hinde & Zunz, 1978), gorillas (Gold & Maple, 1994),
chimpanzees (Murray, 2011), bonobos (Pan paniscus; Murray, 1995), elephants
(Loxodonta africana; Lee & Moss, 2012), and other species. Likewise, studies
using the questionnaire devised by Konečná and her colleagues identified
different structures in two species of free-ranging Old World monkeys,
Hanuman langurs (Semnopithicus entellus; Konečná et al., 2008) and Barbary
macaques (Macaca sylvanus; Konečná, Weiss, Lhota, & Wallner, 2012).
Similarly, using a questionnaire developed for felids that was also based on
the Five Factor Model, the personality of domestic cats, Scottish wildcats (Felis
silvestris grampia), clouded leopards (Neofelis nebulosa), snow leopards
(Panthera uncia), and African lions (Panthera leo) was assessed and compared
(Gartner et al., 2014). This study, which revealed differences among species in
both Panthera and Felidae, from solitary to semisolitary to social species, also
found a striking similarity across these species, with most having a Dominance
domain, a Neuroticism domain, and an Impulsivity domain (Scottish wildcats
had a Self-Control factor and an Agreeableness factor that were related to others’
Impulsiveness domains; clouded leopards had a mixed
Dominance/Impulsiveness domain and an Agreeableness/Openness domain that
were related to others’ Impulsiveness domains; and snow leopards had a mixed
Impulsiveness/Openness domain that was related to others’ Impulsiveness
domains). This suggests that the evolution of personality in felids may be
continuing on the trait level, or possibly among the Dominance domains, which
differed across species.
Clearly, studies of animal personality are able to identify robust structures that
characterize a given species. More importantly, these studies are able to detect
different structures where they would be expected.
Understanding and Interpreting Personality Traits

The Five Factor Model has also benefited animal personality research by
illustrating how to understand animal personality traits, that is, by working out
what theoretical constructs personality dimensions represent and what
predictions can be made about these dimensions. As Gosling and John (1999)
noted, this is important as there are numerous ways to measure human and
animal personality, each with a different name, and more may even be developed
in the future. A similar problem beset studies of human personality, and
therefore early research on the Five Factor Model points the way forward. One
key to it becoming the dominant model of personality is illustrated in the
Revised NEO Personality Inventory manual (Costa & McCrae, 1992), chapters
in this volume, and countless articles; scales for many different instruments are
highly correlated with the Five Factor Model domains and facets (see also the
chapter by O’Connor). In short, the Five Factor Model largely subsumes other
scales.
Some early animal personality research conducted before the widespread
acceptance and adoption of the Five Factor Model either applied the same scales
and structures across species, as was the case in studies using the Emotions
Profile Index (Martau, Caine, & Candland, 1985), or referred to existing human
personality models, such as Eysenck and Eysenck’s (1964), when interpreted the
resulting personality dimensions (Chamove, Eysenck, & Harlow, 1972).
However, these approaches, perhaps because of the widespread fear of
anthropomorphism, were rare. King and Figueredo (1997) were perhaps the first
to explicitly link the composition of the personality factors they discovered to
the human Five Factor Model, namely by indicating from which of the Five
Factor domains each item was sampled (see their Table 1), and using this as a
means to label factors. In other words, although they did not impose the Five
Factor Model on chimpanzees, they linked similar chimpanzee factors to their
human counterparts based on the phylogenetic, behavioral, cognitive, and
physiological similarities between the species.
Not many would argue that the Five Factor Model sensu stricto should be the
personality model for any species other than humans, let alone higher groupings,
such as taxa. However, the importance to animal personality research of finding
theoretical constructs (domains) that explain variation across a given species or
taxa and leads to predictions cannot be overstated. This was acknowledged and
cemented in Gosling’s reviews that used the Five Factor Model to help interpret
findings from previous studies (Gosling, 2001; Gosling & John, 1999). This
point was also not lost on the authors of a review and discussion of animal
personality research in behavioral ecology. Specifically, Réale et al. (2007)
linked behavioral tests used to assess personality in nonhuman animals to the
“temperament traits” of Aggressiveness, Sociability, Boldness, Activity, and
Exploration (see their Figure 2), and stressed the importance of validating these
measures (see also Carter, Feeney, Marshall, Cowlishaw, & Heinsohn, 2012).
Two studies illustrate the importance of seeing traits as measures of
theoretical constructs. Boldness is a label behavioral ecologists have given a trait
in a range of species, including humans, that captures the degree to which
individuals are risk prone as opposed to risk averse (Wilson et al., 1994).
Boldness is often measured via behavioral tests (Réale et al., 2007), such as how
likely an individual will be caught in a baited trap. Although seemingly simple
and straightforward, Bergvall, Schäpers, Kjellander, and Weiss (2011) showed
that behavioral tests for boldness in fallow deer (Dama dama), such as how long
it took subjects to enter an arena that contained a novel object, were associated
with a wide range of rated traits capturing aspects of the Five Factor Model
domains of low Neuroticism and high Extraversion, Conscientiousness, and
Openness. In a study of wild chacma baboons (Papio ursinus), Carter, Marshall,
Heinsohn, and Cowlishaw (2012) compared multiple behavioral tests assumed to
measure boldness, that is, responses to a model puff adder and an unfamiliar
food item. Contrary to their predictions, they found that alarm calls were
associated with more time spent inspecting the model puff adder and that
responses to the model puff adder that would be interpreted as “boldness” were
not associated with responses to the novel food stimuli that would be interpreted
as “boldness.” In short, studies of animal personality would benefit from
personality models and the ability to examine associations between different
measures and these models.
Facets
The examination of personality at the level of facets has not yet taken hold in
much of animal personality research. This is despite the fact that there is no
reason not to believe that animal personality, like human personality (Costa &
McCrae, 1995), is hierarchically organized (King & Weiss, 2011). This view is
consistent with classic ethological perspectives on how behavior is organized
(Eibl-Eibesfeldt, 1970) and views of personality by modern behavioral
ecologists (Araya-Ajoy & Dingemanse, 2014; Réale et al., 2007).
Human personality research has benefited hugely from the study of facets, for
example, as they make it possible to better understand why some personality
domains are related to health outcomes. Specifically, studies that examine the
association between personality and health have revealed that among the five
human domains, Conscientiousness stands out as a robust predictor of health
behaviors and outcomes, such as cigarette smoking, obesity, inflammatory
markers, and premature mortality (Deary, Weiss, & Batty, 2010). Studies that
have examined these associations at the facet level show that, for example, the
tendency for more conscientious individuals to be higher in self-discipline is
what underlies the association of Conscientiousness with longer life (Weiss &
Costa, 2005) and with taking better care of one’s health (Terracciano & Costa,
2004). Studies of human personality development have also benefited from an
examination of facets: although there are overall modest changes in mean levels
of personality domains over time, there is some variability in trajectories at the
facet level (Terracciano, McCrae, Brant, & Costa, 2005).
One deliberate study of animal personality facets was conducted by King,
Weiss, and Sisco (2008), who examined sex and age differences in chimpanzee
personality to determine whether the overall patterns matched those of humans.
As part of their study, along with looking at the domain level, they divided, a
priori, Extraversion and Conscientiousness into two facets each. The
Extraversion facets were labeled Gregariousness and Activity, as they were
similar in composition to the like-named NEO PI-R facets (Costa & McCrae,
1992). The Conscientiousness facets were labeled Tameness and Predictability.
The former described individuals who did not display aggressive behavior,
whereas the latter described individuals who were reliable and consistent in their
behavior. These facets showed different associations with sex and age (King et
al., 2008). Whereas Extraversion was lower in older age groups and did not
differ between males and females, males were higher in Activity than females,
and the age effects for Gregariousness, but not Activity, were stronger in females
than males. Distinct patterns of sex and age differences also characterized the
Conscientiousness facets. Males were lower in Conscientiousness, and although
age effects suggested an increase, the effect was not significant. However,
although the sex and age differences were similar for both facets of
Conscientiousness, the age effects indicated that older individuals were higher in
both facets.
Examining personality at the level of facets may also help researchers
understand the evolutionary mechanisms responsible for maintaining heritable
personality variation. Considerable research has shown that personality in
humans and animals is related to survival and reproductive success and we
would therefore expect there to be no additive genetic variation underlying
personality (Dingemanse & Réale, 2005; Nettle & Penke, 2010; Smith &
Blumstein, 2008). However, that is not what is found in humans (Bouchard &
Loehlin, 2001) or animals (van Oers, de Jong, van Noordwijk, Kempenaers, &
Drent, 2005). One proposed explanation for this maintained additive genetic
variation is that the fitness benefits of personality traits vary across different
environments, a process that is also known as balancing selection by
environmental heterogeneity (Penke, Denissen, & Miller, 2007). Another (not
incompatible) proposed mechanism is that each personality trait comes with
fitness costs and benefits (see Table 1 in Nettle, 2006).
Although much of the additive genetic variance within each human
personality domain is shared in common (McCrae, Jang, Livesley, Riemann, &
Angleitner, 2001; Pilia et al., 2006; Yamagata et al., 2006) and would therefore
react to selection in the same way, there is some evidence that the specific or
residual variance of the Five Factor Model facets is heritable (Jang, McCrae,
Angleitner, Riemann, & Livesley, 1998; see also the chapter by Jarnecke and
South). Thus, another mechanism that may act to maintain additive genetic
variation, alone or in combination with these other mechanisms, is heterogeneity
within personality domains. This is consistent with the findings noted above that
health consequences such as length of life appear to be related to specific facets,
such as Self-Discipline. Moreover, it is easy to see how some facets of
Extraversion, such as Warmth, or of Agreeableness, such as Trust, may be more
strongly associated with mating success than their Activity or Compliance
facets, respectively. Similarly, the personality facets of a given domain may
differ in terms of their respective fitness costs or benefits. Turning again to
Extraversion, although its Excitement-Seeking facet may lead to injury or early
mortality due to misadventure, its Activity facet may lead to health promotion
behaviors such as regular exercise, and, consequently, longer life (see also the
chapter by Wilt and Revelle).
Going Beyond Behavior
One less appreciated way in which the Five Factor Model has and can
continue to contribute to animal personality research has been in the work
showing that personality domains and facets are more than “stand-ins” for
behavior, but are biologically based basic tendencies (McCrae & Costa, 1999).
This led researchers to examine the associations between personality and
outcomes beyond simple behaviors (Deary et al., 2010; Ozer & Benet-Martínez,
2006; Roberts, Kuncel, Shiner, Caspi, & Goldberg, 2007). For example, as noted
briefly before, personality has implications for health, including immune
function, stress, morbidity, mortality, and well-being (see also the chapter by
Kern and Friedman).
Immune function. In nonhuman primates, research with rhesus (Capitanio et
al., 1999, 2008) and long-tailed macaques (Macaca fascicularis; Kaplan et al.,
1991) shows that the personality factor Sociability has a protective role in
relation to the immune system, similar to what is found in humans (Ironson,
O’Cleirigh, Schneiderman, Weiss, & Costa, 2008). Rhesus monkeys higher in
Sociability who are exposed to social stress experience an increase in antibodies
in response to inoculation with simian immunodeficiency virus (SIV), whereas
those lower in Sociability had a decrease in antibodies. In addition to the
negative effects of the virus and the resulting acquired immune deficiency
syndrome (AIDS), an inability to fight an infection following SIV infection can
negatively affect longevity (Capitanio et al., 2008).
Similarly, cynomolgus monkeys (Macaca fascicularis) both high in affiliation
and low in aggression exposed to social stress show a healthier immune
response, with a greater proliferation of lymphocytes (a type of white blood cell)
in response to stimulation with mitogens (a chemical substance that causes cells
to divide; this is a common way of testing immune function), and greater natural
killer (NK) cytolytic activity (cells that kill virally infected cells; Kaplan et al.,
1991; see also Cohen, Kaplan, Cunnick, Manuck, & Rabin, 1992).
In pigtailed (Macaca nemestrina) and bonnet (Macaca radiata) macaques, the
personality factor Emotional Reactivity in response to a stressor may
compromise immunity. In one experiment, monkeys were separated from their
mothers; those that vocalized more on the first day of separation had less
proliferation of white blood cells 2 weeks later in response to mitogens
(Laudenslager, Held, Boccia, Reite, & Cohen, 1990). Similar results were found
in rhesus monkeys (Laudenslager, Rasmussen, Berman, Suomi, & Berger,
1993).
Stress. Activation of the hypothalamic–pituitary–adrenal (HPA) axis is
related to stress (Tsigos & Chrousos, 2002), disease susceptibility, and also to
personality. In humans, Neuroticism can influence increases in anxiety and the
tendency to expose oneself to stressors (Bolger & Shilling, 1991). Similarly,
Wielebnowski, Fletchall, Carlstead, Busso, and Brown (2002) found that
clouded leopards rated as more fearful/tense, and who self injured, paced, slept,
and hid more often, had increased overall, base, and peak fecal corticoid
concentrations, indicating chronic stress.
A relationship between personality and stress is found in nonhuman primates
as well. Rhesus macaques rated as higher in Excitability had lower basal cortisol
concentrations, and those rated as higher in Confidence had higher cortisol
concentrations (Capitanio, Mendoza, & Bentson, 2004). Brown capuchins
showed positive and negative correlations between personality traits strong and
submissive, respectively, and baseline cortisol, as well as correlations between
apprehensive, fearful, insecure, and tense and confident, curious, effective, and
opportunistic, respectively, and peak cortisol (Byrne & Suomi, 2002).
There is evidence that personality interacts with behavioral reactions to stress
as well. Solitary, irritable, and aggressive Diana monkeys (Cercopithecus diana)
demonstrate increased abnormal behavior during high visitor density, whereas
active, playful, and excitable monkeys show an increase in species-typical
behaviors, including play (Barlow, Caldwell, & Lee, 2007). Sapolsky (1994)
found that certain behavioral styles were associated with lower basal cortisol
concentrations, including those that allow the animal to differentiate between
threatening and neutral situations, those in which the animal is the initiator of
aggression, those that dictate how the animal behaves after either winning or
losing a battle with a rival in which winners display affiliative behavior and
losers display displacement behavior, and those in which the preceding three
traits are correlated.
Similarly, a review of the relationship between personality and cortisol in
birds found that for those species studied [great tits (Parus major), Japanese
quail (Coturnix coturnix japonica), and hens (Gallus gallus domesticus)], those
with proactive personalities had lower corticosterone stress responses to stimuli
than those with reactive personalities (Cockrem, 2007).
Morbidity and mortality. Immune function and stress are not the only way
that personality affects morbidity. In humans, high Conscientiousness and
Extraversion and low Neuroticism are associated with a reduced risk for mental
disorders such as depression, panic attacks, generalized anxiety disorder, and
substance abuse; high Conscientiousness is also associated with a reduced risk
for physical disorders such as diabetes, high blood pressure, and sciatica
(Goodwin & Friedman, 2006).
A link has also been found between personality and disease contraction in
domestic cats. Natoli et al. (2005) analyzed temperament, social rank, and
prevalence of feline immunodeficiency virus (FIV), which is transmitted by
biting (Fromont, Artois, Langlais, Courchamp, & Pontier, 1997), in three cat
colonies. The most aggressive and affiliative (proactive) males who marked
(spraying or rubbing cheeks) frequently had the highest social rank and the most
reproductive success, but were also more likely to be infected with FIV. The
opposite result is found in domestic cats with feline leukemia virus (FeLV),
which is transmitted mainly during affiliative interactions including licking and
grooming (Fromont et al., 1997). More aggressive cats, then, have lower levels
of FeLV, whereas socially active cats have higher levels (Fromont et al., 1997).
This link between personality and morbidity can be found in other species as
well. Female Sprague–Dawley rats (Rattus norvegicus) labeled as neophobic had
significantly more risk of developing spontaneous mammary and pituitary
tumors than neophilic females, and therefore increased mortality—they died 6
months earlier (Cavigelli, Yee, & McClintock, 2006). In addition, aggressive
wild-type rats are more susceptible to experimental autoimmune
encephalomyelitis (an animal model of brain inflammation) than nonaggressive
rats (Kavelaars, Heijnen, Tennekes, Bruggink, & Koolhaas, 1999). This is not
surprising, as aggression bears similarities to toxic characteristics of type A
personality in humans, which is characterized by, among other things,
antagonistic hostility (related to lower Agreeableness and moderately related to
higher Neuroticism: Dembroski & Costa, 1987), and which may play some role
in coronary heart disease (Booth-Kewley & Friedman, 1987).
Personality has also been shown to predict mortality in a variety of species,
including humans; both Conscientiousness and Extraversion seem to be
protective and influence longevity, whereas Neuroticism has an unclear effect
(Roberts et al., 2007). Male neophobic Sprague–Dawley rats are 60% more
likely to die at any point in time than neophilic rats, with a 20% decrease in
overall lifespan (despite dying of the same causes; Cavigelli & McClintock,
2003). In gorillas, Extraversion predicts longer survival, regardless of
demographics such as age, sex, or husbandry practices (Weiss et al., 2013).
Well-Being. Personality is also one of the strongest and most consistent
predictors of well-being in humans (DeNeve & Cooper, 1998; Steel, Schmidt, &
Shultz, 2008) and nonhuman primates (e.g., chimpanzees; King & Landau,
2003), especially in relation to Extraversion and Neuroticism, where the former
is related to positive affect and the latter is related to negative affect. In
chimpanzees, King and Landau (2003) found that subjective well-being was
related to higher Dominance, Extraversion, and Conscientiousness. Weiss et al.
(2009) replicated those results, but also found a positive relationship between
subjective well-being and Agreeableness and Openness and a negative
relationship with Neuroticism. In addition, orangutan personality—specifically
Extraversion, Agreeableness, and low Neuroticism—is also related to subjective
well-being (Weiss et al., 2006). Finally, a similar relationship is found in rhesus
macaques, where higher Confidence and Friendliness and lower Anxiety are
related to subjective well-being, concurrently and prospectively (Weiss, Adams,
Widdig, et al., 2011).
Similarly, in clouded leopards, snow leopards, and African lions, Neuroticism
is negatively related to well-being; in lions, Impulsiveness is also negatively
related to well-being (Gartner et al., 2014). In clouded leopards, well-being is
positively related to Agreeableness/Openness (Gartner et al., 2014), whereas is
Scottish wildcats it is positively related to Self-Control, a factor with similarities
to Conscientiousness (Gartner & Weiss, 2013).
These relationships are important because subjective well-being is associated
with longer life in humans (Diener & Chan, 2011) and orangutans (Weiss,
Adams, & King, 2011). As such, well-being may be a good marker for health
outcomes, as personality may be influencing health via subjective well-being.
For example, in long-tailed macaques, depression, a facet of the negative affect
aspect of subjective well-being, mirrors that in humans in terms of physiology,
neurobiology, and behavior, including increased cardiovascular disease risk,
increased mortality, and more, and is subject to individual differences in terms of
response to environmental challenges (Willard & Shively, 2012).
Learning and social relationships. Personality reaches beyond health and
well-being outcomes, with implications for both learning and social
relationships. In humans, higher Conscientiousness (Busato, Prins, Elshout, &
Hamaker, 2000), Openness, and Extraversion (Duff, Boyle, Dunleavy, &
Ferguson, 2004) are associated with academic success, whereas Neuroticism and
Agreeableness are associated with less academic success. As mentioned earlier,
Pavlov observed differences between excitable and inhibited dogs and between
lively and calm dogs in learning (Pavlov, 1941). For example, Excitable dogs
were able to learn quickly in response to both strong and weak stimuli, but were
slower in learning tasks that required switching between the two. Lively dogs
were best at associative learning in general, whereas Calm dogs learned
consistently but slowly. Finally, Inhibited dogs were slow to learn. More recent
work has shown that personality is associated with other types of learning as
well. In fallow deer, those animals who were high on a component called
Flexibility were better learners in reinforcement tasks (Bergvall et al., 2011),
whereas capuchin monkeys that rated high on the personality factor Openness
and low on Assertiveness performed better on cognitive tasks (Morton, Lee, &
Buchanan-Smith, 2013).
Social behavior is also affected by personality. In humans, similarity in
Extraversion, Openness, and Agreeableness is related to friendship, and this
extends to chimpanzees as well, in terms of Sociability and Boldness (Massen &
Koski, 2014). Similarly, social rank in Barbary macaques is affected by
personality: Confidence is associated with higher future rank (Konečná et al.,
2012). In terms of social learning, both bolder and more anxious baboons
showed greater improvement in task solving after watching a demonstrator
(Carter, Marshall, Heinsohn, & Cowlishaw, 2014). Slow-to-explore male great
tits have shorter latencies when accompanied by a faster companion, whereas
fast males show no effect from companion birds (van Oers, Klunder, & Drent,
2005).
Animal personality research’s contribution to the Five Factor

Model
Unlike many areas of psychology, such as behavior genetics (e.g., Scott &
Fuller, 1965), personality psychology developed largely independently from
animal research. This is unfortunate, particularly as the fields that grew from or
with animal research can arguably be said to have benefited from the experience.
What, then, might research on the Five Factor Model gain from the animal
personality work that has been conducted and that may be conducted?
Phylogeny and Ultimate Explanations

One overarching contribution that animal personality can make involves
offering new means to test hypotheses about personality evolution. Although
attempting to understand human personality evolution is important and an active
area of research (see Nettle, 2006), as pointed out by Gosling and Graybeal
(2007), researchers who ignore phylogeny may do so at their peril, in that they
may make unwarranted assumptions about the environment and timing during
which a particular personality dimension evolved. To be consistent with Gosling
and Graybeal’s choice of sociability as an example, let us consider Extraversion.
If individual differences in Extraversion describe differences in adaptive
strategies within social contexts, then we would expect to find it only in social
species. Thus, a test of this hypothesis would be to search for this trait in solitary
or semisolitary species. As is clear from the previous study of orangutan
personality (Weiss et al., 2006), this prediction is not supported.
Comparing species, then, offers a means of conducting strong inferential tests
(Platt, 1964). These tests make it possible to rule out hypotheses concerning
when domains and facets of the Five Factor Model evolved and what their
functions are (Gosling & Graybeal, 2007). To see how this works, it is only
necessary examine Figure 13.1, a cladogram created to try and capture the
results of the then existing work on personality in humans, chimpanzees,
orangutans, and rhesus macaques (Weiss, Adams, Widdig, et al., 2011). It
highlights, for example, the likelihood that Conscientiousness evolved before or
after the common ancestor of chimpanzees split from that of gorillas. A study of
gorilla personality, then, would resolve when this would have happened.
Moreover, as gorilla social structure differs from that of humans, chimpanzees,
and brown capuchin monkeys, being composed of unimale harems (Robbins,
1995, 1996), finding Conscientiousness in gorillas would rule out the hypothesis
that Conscientiousness evolved in species with human-like social structures. A
preliminary report on captive gorilla personality (Schaefer & Steklis, 2014)
suggests that Conscientiousness did evolve in the common ancestor of
chimpanzees and gorillas and that the complex social structures that characterize
humans, chimpanzees, and brown capuchin monkeys did not lead to its
emergence. The other hypotheses are equally testable and there is no doubt that
this picture will change as more data are collected.
The benefits of grounding studies on human personality evolution within
results from animal research are clear. For example, recent work in this vein in
modern populations was conducted by Figueredo, Vásquez, Brumbach, and
Schneider (2007) who investigated links between Five Factor Model domains
and psychometric measures of life history strategies. They found evidence that
personality profiles characterized by low Neuroticism and high Extraversion,
Openness to Experience, Agreeableness, and Conscientiousness were associated
with the adoption of a slower life history strategy, that is, a tendency to delay
reproduction and to invest more in individual offspring. Likewise, a study of
rural villages in Senegal (Alvergne, Jokela, & Lummaa, 2010) showed that in
women, higher Neuroticism was associated with having more offspring and that
among women in lower social classes, higher Neuroticism was also associated
with lower child body mass. This study also found that men who scored higher
on Extraversion were in higher social classes, were more likely to be
polygynous, and had more children. Finally, a recent study of another traditional
society, the Tsimane of Bolivia, revealed associations between personality and
fitness, as measured by fertility and health, though in women these associations
were not consistent across regions (Gurven, von Rueden, Stieglitz, Kaplan, &
Rodriguez, 2014).
Figure 13.1. Cladogram of the hypothesized patterns of personality evolution in the

parvorder Catarrhini. Personality structures are described as a combination of “basic” or
“blended” dimensions, for example, Friendliness in rhesus macaques is a blend of
Altruism and Sociability. The evolutionary transitions are interpreted as the integration or
disintegration of these dimensions, shown by horizontal arrows between groups of
dimensions. The possible transition points, indicated by the dashed lines, are placed
according to phylogenetic parsimony. This picture is likely to change as more species are
assessed. The structure of the genus Pan is represented by chimpanzees and the family
Cercopithecidae by rhesus macaques. [From Weiss, A., Adams, M. J., Widdig, A., &
Gerald, M. S. (2011). Rhesus macaques (Macaca mulatta) as living fossils of hominoid
personality and subjective well-being. Journal of Comparative Psychology, 125, 72–83.
Figure by the authors, licensed under a Creative Commons Attribution Unported License
and published under the terms of this license.]
Clearing the Air

The universality of the Five Factor Model. Along with contributing to
understanding the evolutionary bases of personality domains and the
maintenance of additive genetic variation, research on animal personality can
inform fundamental debates about the Five Factor Model. Take, for example, the
question of whether the Five Factor Model is a human universal (McCrae &
Costa, 1997) or is limited to so-called WEIRD (Western, Educated,
Industrialized, Rich, and Democratic) people (Henrich, Heine, & Norenzayan,
2010). Cross-cultural studies of personality using an “etic” approach, in which a
questionnaire is translated into multiple languages, have found that the Five
Factor Model can mostly be identified even in non-Western and/or less-
developed nations (e.g., McCrae et al., 2005; see also the chapter by Allik and
Realo). On the other hand, cross-cultural studies using an “emic” approach, in
which questionnaires are devised based on personality descriptors from within
that culture, do find cross-cultural differences in personality structure (Saucier &
Goldberg, 2001). They suggest that differences between findings from the emic
approach and those from the etic approach indicate that the etic approach
constrains the structure that can emerge in samples (Saucier & Goldberg, 2001;
see also the chapter by De Raad and Mlačić).4
Research on animal personality decreases the possibility that finding cross-
cultural similarities in structure is constrained when the same questionnaire is
applied across cultures. As noted earlier, studies of closely related animal
species, for example, felids, using the same survey tool, shows evidence for
species differences in personality (Gartner et al., 2014). Thus, human personality
structure appears to largely an expression of the human genome, although the
mostly etic study of the Tsimane (Gurven, von Rueden, Massenkoff, Kaplan, &
Lero Vie, 2013) suggests that personality structure may also be a reflection of
more modern social groups and environments.
Personality change. Animal personality research can also inform studies of
personality development. In adulthood, as humans age, their personality changes
in ways reflecting greater maturity, a pattern that includes declines in
Neuroticism and Extraversion, increases in Agreeableness and
Conscientiousness, and increases in Openness followed by declines (Roberts,
Wood, & Caspi, 2008). There has been much debate over whether these trends
reflect biological development (McCrae & Costa, 2003) or whether they are
driven by social roles taken on by individuals, such as employment or marriage
(Roberts, Wood, & Smith, 2005). Each side draws on several findings to support
its position (see the chapter by Terracciano for more details). Those who favor
Five Factor Theory point out that the amount of change is modest, is similar
across different cultures, and is underpinned by shared genetic effects (McCrae
& Costa, 2003). Those who favor Social-Investment Theory point out that within
and across cultures, these shifts in personality are contemporaneous with
transitions in life, such as leaving home, beginning work, and beginning a
family, and that behavioral genetic studies find evidence for nonshared
environmental influences (Roberts et al., 2005). However, the comparable trends
across cultures could just reflect the influence of experiences and roles that are
part of being humans. Similarly, differences within and between cultures in
terms of the rate of maturation could reflect genetic variation that, for example,
makes some individuals’ personalities develop more rapidly, thus preparing
them for these roles sooner.
McCrae et al. (2000) noted that developmental trends in the behavioral traits
of primates resemble those found in humans. A study by King et al. (2008)
explored this further by directly comparing human and chimpanzee personality
development and found that the pattern of age differences of the Five Factor
Model domains was mostly conserved in chimpanzees. They also found that,
correcting for the fact that chimpanzees do not live as long as humans, the
magnitudes of age differences were similar in these two species (King et al.,
2008). The main thing setting apart chimpanzee personality development from
that of their human cousins is that male chimpanzees, but not male humans,
show a slight, albeit significant, difference in some trends. In particular, as noted
earlier, compared to females, males maintained higher Activity scores into older
age and the association between age and Agreeableness was lower in males. In
addition, the chimpanzee Dominance factor, while showing an increase and then
a decline across age groups in females, showed an increase and plateau in males.
As King et al. (2008) note, these findings support Five Factor Theory. However,
they also noted that the sex difference in age-related trends suggests that
selection has acted upon the way in which personality develops. In particular, the
pattern of male chimpanzee personality development would be beneficial to
species in which intragroup aggression among males is relatively high, such as it
is in chimpanzees as opposed to humans (Wrangham, Wilson, & Muller, 2006,
cited in King et al., 2008).
Unraveling causality. One notable contribution made by animal personality
research has been with respect to explaining why human (and animal)
personality traits are associated with life outcomes. Take as an example the
association between personality and subjective well-being. In a 2002 study,
Weiss, King, and Enns showed that in addition to being heritable in
chimpanzees, subjective well-being shared nearly all of its genetic variance with
Dominance, that is, all of the personality and subjective well-being variance
could be accounted for by common genetic influences; they named this
phenomenon “covitality” (p. 1147). Since then, studies of personality and well-
being in humans (Hahn, Johnson, & Spinath, 2013; Keyes, Kendler, Myers, &
Martin, 2015; Weiss, Bates, & Luciano, 2008) and orangutans (Adams et al.,
2012) have also shown that the associations between personality domains and
well-being are mediated by common genetic influences.
Whereas personality is linked to a variety of life chances and outcomes,
causality is not always apparent in humans as well as in nonhuman animals. For
example, although a modest relationship between Type A personality and
coronary heart disease has been found (Booth-Kewley & Friedman, 1987),
methodological and other factors may play a role in this association, making it
difficult to know what is causing what (Friedman & Booth-Kewley, 1988;
Matthews, 1988). Later work described the components of Type A personality
that were likely to affect this type of disease, namely antagonistic hostility,
which is related to lower Agreeableness and higher Neuroticism (Dembroski,
MacDougall, Costa, & Grandits, 1989). The question is, how exactly is health
influenced by personality? One hypothesis is that health behaviors act as
mediators between personality and its links to morbidity and mortality (Deary et
al., 2010). For example, Conscientiousness is related to health-promoting
behaviors (e.g., exercise) and to avoiding health-harming behaviors (e.g., fast
driving). High Neuroticism and low Agreeableness have also been shown to be
related to health-harming behaviors (Terracciano & Costa, 2004).
Conscientiousness is also linked to better management of disease; for example,
high Conscientiousness was associated with better adherence to medication in
end-stage renal disease patients (Christensen & Smith, 1995). However, these
mediators do not wholly account for the relationship between personality and
mortality, and other factors such as the patient–health care practitioner
relationship, demographic factors, socioeconomic status, other health behaviors,
cognitive abilities, or other personality dimensions may be playing a role, and
should be further explored (Deary et al., 2010).
Future Directions
Where in the study of animal (and even human) personality does this exciting
field next point? We think the most important directions to pursue are those that
will lead to the growth of both animal and human personality. At the most basic
level, to fully understand personality requires obtaining comparable personality
measures, whether they are behavioral observations and/or tests, ratings, or some
new method to be developed, on multiple traits and an even broader range of
species. This will make it possible to test, for example, the extent to which
personality tracks phylogeny (King & Weiss, 2011) and is an evolutionary
character (Araya-Ajoy & Dingemanse, 2014). This work could help understand
not just the evolutionary bases of personality variation, but also the genetics of
the coupling and uncoupling of traits over evolutionary time.
Second, much like the research on personality and well-being cited here,
studies of animal personality and outcomes related to welfare, psychopathology,
and psychological needs could be used to better understand these associations in
humans and could be informed by research on humans, too. In short,
understanding whether personality is an actual causal factor or whether these
associations are mediated by third variables, such as common genes, would lead
to the ability to either identify individuals at risk, intervene to reduce risk, or
both.
There is much more that those who study human and animal personality can
learn from one another. With personality now being studied across a large
variety of fields, conversations involving these subject areas and
interdisciplinary research can only increase our knowledge of personality, its
relationships and evolutionary history, and its potential to affect life outcomes
for both humans and animals. We look forward to the discussion.
Acknowledgments
We greatly appreciate the thoughts and time that Joan Stevenson-Hinde, Sam
Gosling, Lars Penke, and John Capitanio contributed prior to and during the
drafting of this chapter.
Notes
1. Curiously, whereas Tinbergen is often cited by those who advocate a behavioral
approach to studying animal personality [e.g., Uher, J. (2008). Comparative personality
research: Methodological approaches. European Journal of Personality, 22, 475–496], Skinner is
not.
2. The Madingley Questionnaire and the 54-item Hominoid Personality Questionnaire
are freely available at http://extras.springer.com/2011/978-1-4614-0175-9.
3. To minimize the number of different terms in the literature, these labels were later
replaced with the more common labels for the five domains.
4. The similarity between emic and etic approaches to studying cultures and the
different approaches available to study animal personality have been noted [ Gosling,
S. D., and John, O. P. (1998, May). Personality dimensions in dogs, cats, and hyenas. Paper presented
at the annual meeting of the American Psychological Society, Washington, DC; Uher, J. (2008).
Comparative personality research: Methodological approaches. European Journal of Personality, 22,
475–496].
References
Adams, M. J., King, J. E., & Weiss, A. (2012). The majority of genetic variation in
orangutan personality and subjective-well being is nonadditive. Behavior Genetics, 42,
675–686.
Alvergne, A., Jokela, M., & Lummaa, V. (2010). Personality and reproductive success in
a high-fertility human population. Proceedings of the National Academy of Sciences of
the United States of America, 107, 11745–11750.
Araya-Ajoy, Y. G., & Dingemanse, N. J. (2014). Characterizing behavioural ‘characters’:
An evolutionary framework. Proceedings of the Royal Society B-Biological Sciences,
281(1776), 2013–2645.
Barlow, C. J. C., Caldwell, C. A., & Lee, P. C. (2007). Individual differences and
reactions to visitors in zoo-housed Diana monkeys. In S. Dow & F. Clark (Eds.),
Proceedings of the 8th annual symposium on zoo research (pp. 131–142). London,
England: British and Irish Association of Zoos and Aquariums.
Benning, S. D., Patrick, C. J., Hicks, M. M., Blonigen, D. M., & Krueger, R. F. (2003).
Factor structure of the psychopathic personality inventory: Validity and implications
for clinical assessment. Psychological Assessment, 15, 340–350.
Bergvall, U. A., Schäpers, A., Kjellander, P., & Weiss, A. (2011). Personality and
foraging decisions in fallow deer, Dama dama. Animal Behaviour, 81, 101–112.
Bolger, N., & Shilling, E. A. (1991). Personality and the problems of everyday life: The
role of Neuroticism in exposure and reactivity to daily stressors. Journal of
Booth-Kewley, S., & Friedman, H. S. (1987). Psychological predictors of heart disease: A
quantitative review. Psychological Bulletin, 101, 343–362.
Bouchard, T. J., Jr., & Loehlin, J. C. (2001). Genes, evolution, and personality. Behavior
Genetics, 31, 243–273.
Buirski, P., Plutchik, R., & Kellerman, H. (1978). Sex differences, dominance, and
personality in the chimpanzee. Animal Behaviour, 26, 123–129.
Buirski, P., Kellerman, H., Plutchik, R., Weininger, R., & Buirski, N. (1973). A field
study of emotions, dominance, and social behavior in a group of baboons (Papio
anubis). Primates, 14, 67–78.
Busato, V. V., Prins, F. J., Elshout J. J., & Hamaker, C. (2000). Intellectual ability,
learning style, personality, achievement motivation and academic success of
psychology students in higher education. Personality and Individual Differences, 29,
1057–1068.
Byrne, G., & Suomi, S. J. (2002). Cortisol reactivity and its relation to homecage
behavior and personality ratings in tufted capuchin (Cebus apella) juveniles from birth
to six years of age. Psychoneuroendocrinology, 27, 139–154.
Capitanio, J. P. (1999). Personality dimensions in adult male rhesus macaques: Prediction
of behaviors across time and situation. American Journal of Primatology, 47, 299–320.
Capitanio, J. P., Mendoza, S. P., & Baroncelli, S. (1999). The relationship of personality
dimensions in adult male rhesus macaques to progression of simian immunodeficiency
virus disease. Brain Behavior and Immunity, 13, 138–154.
Capitanio, J. P., Mendoza, S. P., & Bentson, K. L. (2004). Personality characteristics and
basal cortisol concentrations in adult male rhesus macaques (Macaca mulatta).
Psychoneuroendocrinology, 29, 1300–1308.
Capitanio, J. P., Abel, K., Mendoza, S. P., Blozis, S. A., McChesney, M. B., Cole, S. W.,
& Mason, W. A. (2008). Personality and serotonin transporter genotype interact with
social context to affect immunity and viral set-point in simian immunodeficiency virus
disease. Brain, Behavior, and Immunity, 22, 676–689.
Carter, A. J., Marshall, H. H., Heinsohn, R., & Cowlishaw, G. (2012). How not to
measure boldness: Demonstration of a jingle fallacy in a wild social primate. Animal
Behaviour, 84, 603–609.
Carter, A. J., Feeney, W., Marshall, H. H., Cowlishaw, G., & Heinsohn, R. (2012).
Animal personality: What are behavioural ecologists measuring? Biological Reviews,
88, 465–475.
Carter, A. J., Marshall, H. H., Heinsohn, R., & Cowlishaw, G. (2014). Personality
predicts the propensity for social learning in a wild primate. PeerJ, 2, e283.
Cavigelli, S. A., & McClintock, M. K. (2003). Fear of novelty in infant rats predicts adult
corticosterone dynamics and an early death. Proceedings of the National Academy of
Sciences of the United States of America, 100, 16131–16136.
Cavigelli, S. A., Yee, J. R., & McClintock, M. K. (2006). Infant temperament predicts life
span in female rats that develop spontaneous tumors. Hormones and Behavior, 50,
454–462.
Chamove, A. S., Eysenck, H. J., & Harlow, H. F. (1972). Personality in monkeys: Factor
analyses of rhesus social behavior. Quarterly Journal of Experimental Psychology, 24,
496–504.
Christensen, A. J., & Smith, T. W. (1995). Personality and patient adherence: Correlates
of the five-factor model in renal dialysis. Journal of Behavioral Medicine, 18, 305–
313.
Cockrem, J. F. (2007). Stress, corticosterone responses and avian personalities. Journal of
Ornithology, 148, S169–S178.
Cohen, S., Kaplan, J. R., Cunnick, J. E., Manuck, S. B., & Rabin, B. S. (1992). Chronic
social stress, affiliation, and cellular immune response in nonhuman primates.
Colvin, J. D. (1986). Proximate causes of male emigration at puberty in rhesus monkeys.
In R. G. Rawlins & M. J. Kessler (Eds.), The Cayo Santiago macaques (pp. 131–157).
Albany, NY: SUNY Press.
R) and NEO Five-Factor Inventory (NEO-FFI) professional manual. Odessa, FL:
Crawford, M. P. (1938). A behavior rating scale for young chimpanzees. Journal of
Darwin, C. (1871). The descent of man and selection in relation to sex. London, England:
John Murray.
Darwin, C. (1872/1998). The expression of the emotions in man and animals. Oxford,
England: Oxford University Press.
Deary, I. J., Weiss, A., & Batty, G. D. (2010). Intelligence and personality as predictors
of illness and death: How researchers in differential psychology and chronic disease
epidemiology are collaborating to understand and address health inequalities.
Psychological Science in the Public Interest, 11, 53–79.
Dembroski, T. M., & Costa, P. T., Jr. (1987). Coronary prone behavior: Components of
the type A pattern and hostility. Journal of Personality, 55, 211–235.
Dembroski, T. M., MacDougall, J. M., Costa, P. T., Jr., & Grandits, G. A. (1989).
Components of hostility as predictors of sudden death and myocardial infarction in the
Multiple Risk Factor Intervention Trial. Psychosomatic Medicine, 51, 514–522.
DeNeve, K. M., & Cooper, H. (1998). The happy personality: A meta-analysis of 137
personality traits and subjective well-being. Psychological Bulletin, 124, 197–229.
Diener, E., & Chan, M. Y. (2011). Happy people live longer: Subjective well-being
contributes to health and longevity. Applied Psychology: Health and Well-Being, 3, 1–
43.
Digman, J. M. (1990). Personality structure: Emergence of the Five-Factor Model.
Annual Review of Psychology, 41, 417–440.
Dingemanse, N. J., & Réale, D. (2005). Natural selection and animal personality.
Behaviour, 142, 1159–1184.
Duff, A., Boyle, E., Dunleavy, K., & Ferguson, J. (2004). The relationship between
personality, approach to learning and academic performance. Personality and
Eibl-Eibesfeldt, I. (1970). Ethology: The biology of behavior (trans. E. Klinghammer).
New York, NY: Holt, Rinehart & Winston.
Eysenck, H. J. (1970). The biological basis of personality (2nd ed.). Springfield, IL:
Charles C Thomas.
Eysenck, H. J., & Eysenck, S. B. G. (1964). Manual of the Eysenck Personality Inventory.
London, England: University of London Press.
Feaver, J., Mendl, M., & Bateson, P. (1986). A method for rating the individual
distinctiveness of domestic cats. Animal Behavior, 34, 1016–1025.
Figueredo, A. J., Vásquez, G., Brumbach, B. H., & Schneider, S. M. R. (2007). The K-
factor, covitality, and personality: A psychometric test of life history theory. Human
Nature, 18, 47–73.
Fragaszy, D. M., Visalberghi, E., & Fedigan, L. M. (2004). The complete capuchin: The
biology of the genus Cebus. New York, NY: Cambridge University Press.
Friedman, H. S., & Booth-Kewley, S. (1988). Validity of the type A construct: A reprise.
Fromont, E., Artois, M., Langlais, M., Courchamp, F., & Pontier, D. (1997). Modelling
the feline leukemia virus (FeLV) in natural populations of cats (Felis catus).
Theoretical Population Biology, 52, 60–70.
Galdikas, B. M. F. (1985). Orangutan sociality at Tanjung-Puting. American Journal of
Primatology, 9, 101–119.
Gartner, M. C., Powell, D. M., & Weiss, A. (2014). Personality structure in the domestic
cat (Felis silvestris catus), Scottish wildcat (Felis silvestris grampia), clouded leopard
(Neofelis nebulosa), snow leopard (Panthera uncia), and African lion (Panthera leo):
A comparative study. Journal of Comparative Psychology, 128(4), 414–426.
Gartner, M. C., & Weiss, A. (2013). Scottish wildcat (Felis silvestris grampia) personality
and subjective well-being: Implications for captive management. Applied Animal
Behaviour Science, 147, 261–267.
Gold, K. C., & Maple, T. L. (1994). Personality assessment in the gorilla and its utility as
a management tool. Zoo Biology, 13, 509–522.
Goldberg, L. R. (1990). An alternative ‘description of personality’: The Big-Five Factor
Goodall, J. (1986). The chimpanzees of Gombe: Patterns of behavior. Cambridge, MA:
Belknap Press of Harvard University.
Goodall, J. (1990). Through a window: 30 years of observing the Gombe chimpanzees.
London, England: Weidenfeld & Nicolson.
Goodwin, R. D., & Friedman, H. S. (2006). Health status and the five-factor personality
traits in a nationally representative sample. Journal of Health Psychology, 11, 643–654.
Gosling, S. D. (1998). Personality dimensions in spotted hyenas (Crocuta crocuta).
Journal of Comparative Psychology, 112(2), 107–118.
Gosling, S. D. (2001). From mice to men: What can we learn about personality from
animal research? Psychological Bulletin, 127, 45–86.
Gosling, S. D., & Graybeal, A. (2007). Tree thinking: A new paradigm for integrating
comparative data in psychology. The Journal of General Psychology, 134, 259–277.
Gosling, S. D., & John, O. P. (1998). Personality dimensions in dogs, cats, and hyenas.
Annual meeting of the American Psychological Society, Washington, DC.
cross-species review. Current Directions in Psychological Science, 8, 69–75.
Gouzoules, S., & Gouzoules, H. (1987). Kinship. In B. B. Smuts, et al. (Eds.), Primate
societies (pp. 299–305). Chicago, IL: University of Chicago Press.
Griffin, D. R. (1978). Prospects for a cognitive ethology. Behavioral and Brain Sciences,
4, 527–538.
universal is the Big Five? Testing the Five-Factor Model of personality variation
among forager-farmers in the Bolivian Amazon. Journal of Personality and Social
Gurven, M., von Rueden, C., Stieglitz, J., Kaplan, H., & Rodriguez, D. E. (2014). The
evolutionary fitness of personality traits in a small-scale subsistence society. Evolution
and Human Behavior, 35, 17–25.
Hahn, E., Johnson, W., & Spinath, F. M. (2013). Beyond the heritability of life
satisfaction—The roles of personality and twin-specific influences. Journal of
Hebb, D. O. (1949). Temperament in chimpanzees: I. Method of analysis. Journal of
Comparative and Physiological Psychology, 42, 192–206.
Henrich, J., Heine, S. J., & Norenzayan, A. (2010). The weirdest people in the world?
Behavioral & Brain Sciences, 33, 61–83.
Ironson, G. H., O’Cleirigh, C., Schneiderman, N., Weiss, A., & Costa, P. T., Jr. (2008).
Personality and HIV disease progression: The role of NEO-PI-R Openness,
Extraversion, and profiles of engagement. Psychosomatic Medicine, 70, 245–253.
1556–1565.
Jones, A. C., & Gosling, S. D. (2005). Temperament and personality in dogs (Canis
familiaris): A review and evaluation of past research. Applied Animal Behaviour
Science, 95, 1–53.
Kaplan, J. R., Heise, E. R., Manuck, S. B., Shively, C. A., Cohen, S., Rabin, B. S., &
Kasprowicz, A. L. (1991). The relationship of agonistic and affiliative behavior
patterns to cellular immune function among cynomolgus monkeys (Macaca
fascicularis) living in unstable social-groups. American Journal of Primatology, 25,
157–173.
Kavelaars, A., Heijnen, C. J., Tennekes, R., Bruggink, J. E., & Koolhaas, J. M. (1999).
Individual behavioral characteristics of wild-type rats predict susceptibility to
experimental autoimmune encephalomyelitis. Brain, Behavior, and Immunity, 13, 279–
286.
Keyes, C. L. M., Kendler, K. S., Myers, J. M., & Martin, C. C. (2015). The genetic
overlap and distinctiveness of flourishing and the Big Five personality traits. Journal of
Happiness Studies, 16(3), 655–668.
King, J. E., & Figueredo, A. J. (1997). The Five-Factor Model plus Dominance in
chimpanzee personality. Journal of Research in Personality, 31, 257–271.
King, J. E., & Landau, V. I. (2003). Can chimpanzee (Pan troglodytes) happiness be
estimated by human raters? Journal of Research in Personality, 37, 1–15.
King, J. E., & Weiss, A. (2011). Personality from the perspective of a primatologist. In A.
Weiss, J. E. King, & L. Murray (Eds.), Personality and temperament in nonhuman
primates (Developments in primatology: Progress and prospects) (pp. 77–99). New
York, NY: Springer.
King, J. E., Weiss, A., & Farmer, K. H. (2005). A chimpanzee (Pan troglodytes) analogue
of cross-national generalization of personality structure: Zoological parks and an
African sanctuary. Journal of Personality, 73, 389–410.
King, J. E., Weiss, A., & Sisco, M. M. (2008). Aping humans: Age and sex effects in
chimpanzee (Pan troglodytes) and human (Homo sapiens) personality. Journal of
Konečná, M., Lhota, S., Weiss, A., Urbánek, T., Adamová, T., & Pluháček, J. (2008).
Personality in free-ranging Hanuman langur (Semnopithecus entellus) males:
Subjective ratings and recorded behavior. Journal of Comparative Psychology, 122,
379–389.
Konečná, M., Weiss, A., Lhota, S., & Wallner, B. (2012). Personality in Barbary
macaques (Macaca sylvanus): Temporal stability and social rank. Journal of Research
in Personality, 46, 581–590.
Laudenslager, M. L., Held, P. E., Boccia, M. L., Reite, M. L., & Cohen, J. J. (1990).
Behavioral and immunological consequences of brief mother-infant separation: A
species comparison. Developmental Psychobiology, 23, 247–264.
Laudenslager, M. L., Rasmussen, K. L., Berman, C. M., Suomi, S. J., & Berger, C. B.
(1993). Specific antibody levels in free-ranging rhesus monkeys: Relationships to
plasma hormones, cardiac parameters, and early behavior. Developmental
Psychobiology, 26, 407–420.
Lee, K., Ogunfowora, B., & Ashton, M. C. (2005). Personality traits beyond the Big Five:
Are they within the HEXACO space? Journal of Personality, 73, 1437–1463.
Lee, P. C., & Moss, C. J. (2012). Wild female African elephants (Loxodonta africana)
exhibit personality traits of leadership and social integration. Journal of Comparative
Lilienfeld, S. O., & Andrews, B. P. (1996). Development and preliminary validation of a
self-report measure of psychopathic personality traits in noncriminal population.
Lilienfeld, S. O., Waldman, I. D., Landfield, K., Watts, A. L., Rubenzer, S., &
Faschingbauer, T. R. (2012). Fearless dominance and the US presidency: Implications
of psychopathic personality traits for successful and unsuccessful political leadership.
Locurto, C. (2007). Individual differences and animal personality. Comparative
Cognition and Behavior Reviews, 2, 67–78.
MacKinnon, J. (1974). In search of the red ape. New York, NY: Holt, Rinehart, &
Winston.
Manson, J. H. (1995). Do female rhesus macaques choose novel males? American
Journal of Primatology, 37, 285–296.
Martau, P. A., Caine, N. G., & Candland, D. K. (1985). Reliability of the Emotions
Profile Index, Primate Form, with Papio hamadryas, Macaca fuscata, and two Saimiri
species. Primates, 26, 501–505.
Massen, J. J. M., & Koski, S. E. (2014). Chimps of a feather sit together: Chimpanzee
friendships are based on homophily in personality. Evolution and Human Behavior, 35,
1–8.
Matthews, K. A. (1988). Coronary heart disease and type A behaviors: Update on and
alternative to the Booth-Kewley and Friedman (1987) quantitative review.
McCrae, R. R., & Costa, P. T., Jr. (1999). A Five-Factor Theory of personality. In L.
Pervin & O. P. John (Eds.), Handbook of personality (2 ed., pp. 159–181). New York,
NY: Guilford Press.
McCrae, R. R., & Costa, P. T., Jr. (2003). Personality in adulthood: A Five-Factor
Theory perspective. New York, NY: Guilford Press.
McCrae, R. R., Costa, P. T., Jr., Ostendorf, F., Angleitner, A., Hřebíčková, M., Avia, M.
D., … Smith, P. B. (2000). Nature over nurture: Temperament, personality, and life
span development. Journal of Personality and Social Psychology, 78, 173–186.
covariation of personality traits. Journal of Personality, 69, 511–535.
Melnick, D. J., & Pearl, M. C. (1987). Cercopithecines in multimale groups: Genetic
diversity and population structure. In B. B. Smuts, et al. (Eds.), Primate societies (pp.
121–134). Chicago, IL: University of Chicago Press).
Mischel, W. (1968). Personality and assessment. New York, NY: John Wiley & Sons.
Morgan, C. L. (1894). An introduction to comparative psychology. London, England: W.
Scott.
Morton, F. B., Lee, P. C., & Buchanan-Smith, H. M. (2013). Taking personality selection
bias seriously in animal cognition research: A case study in capuchin monkeys
(Sapajus apella). Animal Cognition, 16, 677–684.
Morton, F. B., Lee, P. C., Buchanan-Smith, H. M., Brosnan, S. F., Thierry, B., Paukner,
A., … Weiss, A. (2013). Personality structure in brown capuchin monkeys (Sapajus
apella): Comparisons with chimpanzees (Pan troglodytes), orangutans (Pongo spp.),
and rhesus macaques (Macaca mulatta). Journal of Comparative Psychology, 127,
282–298.
Murray, L. (2011). Predicting primate behavior from personality ratings. In A. Weiss, J.
E. King, & L. Murray (Eds.), Personality and temperament in nonhuman primates (pp.
129–167). New York, NY: Springer.
Murray, L. E. (1995). Personality and individual differences in captive great apes.
Cambridge, England: University of Cambridge.
Natoli, E., Say, L., Cafazzo, S., Bonanni, R., Schmid, M., & Pontier, D. (2005). Bold
attitude makes male urban feral domestic cats more vulnerable to feline
immunodeficiency virus. Neuroscience and Biobehavioral Reviews, 29, 151–157.
Nettle, D., & Penke, L. (2010). Personality: Bridging the literatures from human
psychology and behavioural ecology. Philosophical Transactions of the Royal Society
of London B Biological Sciences, 365, 4043–4050.
Osvath, M. (2009). Spontaneous planning for future stone throwing by a male
chimpanzee. Current Biology, 19, R190–R191.
Patrick, C. J., Edens, J. F., Poythress, N. G., Lilienfeld, S. O., & Benning, S. D. (2006).
Construct validity of the Psychopathic Personality Inventory two-factor model with
offenders. Psychological Assessment, 18, 204–208.
Pavlov, I. P. (1941). Lectures on conditioned reflexes. Volume two: Conditioned reflexes
and psychiatry (trans. W. H. Gantt). London, England: Lawrence and Wishart Limited.
Penke, L., Denissen, J. J. A., & Miller, G. F. (2007). The evolutionary genetics of
Pilia, G., Chen, W. M., Scuteri, A., Orrú, M., Albai, G., Dei, M., … Schlessinger, D.
(2006). Heritability of cardiovascular and personality traits in 6,148 Sardinians. PLoS
Genetics, 2, e132.
Platt, J. R. (1964). Strong inference. Science, 146, 347–353.
Purvis, A. (1995). A composite estimate of primate phylogeny. Philosophical
Transactions of the Royal Society of London, Series B: Biological Sciences, 348, 405–
421.
Réale, D., Reader, S. M., Sol, D., McDougall, P. T., & Dingemanse, N. J. (2007).
Integrating animal temperament within ecology and evolution. Biological Reviews, 82,
291–318.
Robbins, M. M. (1995). A demographic analysis of male life history and social structure
of mountain gorillas. Behaviour, 132, 21–47.
Robbins, M. M. (1996). Male-male interactions in heterosexual and all-male wild
mountain gorilla groups. Ethology, 102, 942–965.
Roberts, B. W., Wood, D., & Caspi, A. (2008). The development of personality traits in
adulthood. In O. P. John (Ed.), Handbook of personality: Theory and research (3rd ed.,
Roberts, B. W., Wood, D., & Smith, J. L. (2005). Evaluating Five Factor Theory and
social investment perspectives on personality trait development. Journal of Research in
Ross, S. R., Benning, S. D., Patrick, C. J., Thompson, A., & Thurston, A. (2009). Factors
of the psychopathic personality inventory criterion-related validity and relationship to
the BIS/BAS and five-factor models of personality. Assessment, 16, 71–87.
Sapolsky, R. M. (1994). Individual differences and the stress response. Seminars in
Neuroscience, 6, 261–269.
markers. Journal of Personality Assessment, 63, 506–516.
Saucier, G., & Goldberg, L. R. (2001). Lexical studies of indigenous personality factors:
Premises, products, and prospects. Journal of Personality, 69, 847–879.
Schaefer, S. A., & Steklis, H. D. (2014). Personality and subjective well-being in captive
male western lowland gorillas living in bachelor groups. American Journal of
Primatology, 76(9), 879–889.
Scott, J. P., & Fuller, J. L. (1965). Genetics of the social behavior of the dog. Chicago, IL:
University of Chicago Press.
Serpell, J. A. (1983). The personality of the dog and its influence on the pet-owner bond.
In A. H. Katcher & A. M. Beck (Eds.), New perspectives on our lives with companion
animals (pp. 57–63). Philadelphia, PA: University of Pennsylvania Press.
Sheldon, W. H. (1942). The varieties of temperament: A psychology of constitutional
differences. New York, NY: Harper & Brothers.
Shettleworth, S. J. (2010). Clever animals and killjoy explanations in comparative
psychology. Trends in Cognitive Sciences, 14, 477–481.
Shettleworth, S. J. (2012). Darwin, Tinbergen, and the evolution of comparative
cognition. In J. Vonk & T. K. Shackelford (Eds.), The Oxford handbook of comparative
evolutionary psychology (pp. 529–546). New York, NY: Oxford University Press.
Shoda, Y., & Mischel, W. (1996). Toward a unified, intra-individual dynamic conception
of personality. Journal of Research in Personality, 30, 414–428.
Sih, A., Bell, A. M., Johnson, J. C., & Ziemba, R. E. (2004). Behavioral syndromes: An
integrative overview. Quarterly Review of Biology, 79, 241–277.
Skinner, B. F. (1964). Science and human behavior (11 ed.). New York, NY: The
Macmillan Company.
Smith, B. R., & Blumstein, D. T. (2008). Fitness consequences of personality: A meta-
analysis. Behavioral Ecology, 19, 448–455.
Stevenson-Hinde, J., & Hinde, C. A. (2011). Individual characteristics: Weaving
psychological and ethological approaches. In A. Weiss, J. E. King, & L. Murray (Eds.),
Personality and temperament in nonhuman primates (Developments in primatology:
Progress and prospects) (pp. 3–14). New York, NY: Springer.
Stevenson-Hinde, J., Stillwell-Barnes, R., & Zunz, M. (1980a). Subjective assessment of
rhesus monkeys over four successive years. Primates, 21, 66–82.
Stevenson-Hinde, J., Stillwell-Barnes, R., & Zunz, M. (1980b). Individual differences in
young rhesus monkeys: Consistency and change. Primates, 21, 498–509.
Stevenson-Hinde, J., & Zunz, M. (1978). Subjective assessment of individual rhesus
monkeys. Primates, 19, 473–482.
Suddendorf, T., & Corballis, M. C. (2010). Behavioural evidence for mental time travel in
nonhuman animals. Behavioural Brain Research, 215, 292–298.
Suomi, S. J. (2006). Risk, resilience, and gene x environment interactions in rhesus
monkeys. Annals of the New York Academy of Sciences, 1094, 52–62.
Terracciano, A., & Costa, P. T., Jr. (2004). Smoking and the Five-Factor Model of
personality. Addiction, 99, 472–481.
Aging. Psychology and Aging, 20, 493–506.
Tinbergen, N. (2005). On aims and methods of ethology (Reprinted from Zeitschrift fur
Tierpsychologie, 20, 410, 1963). Animal Biology, 55, 297–321.
Tsigos, C., & Chrousos, G. P. (2002). Hypothalamic-pituitary-adrenal axis,
neuroendocrine factors and stress. Journal of Psychosomatic Research, 53, 865–871.
Uher, J. (2008). Comparative personality research: Methodological approaches. European
van Oers, K., Klunder, M., & Drent, P. J. (2005). Context dependence of personalities:
Risk-taking behavior in a social and a nonsocial situation. Behavioral Ecology, 16,
716–723.
van Oers, K., de Jong, G., van Noordwijk, A. J., Kempenaers, B., & Drent, P. J. (2005).
Contribution of genetics to animal personalities: A review of case studies. Behaviour,
142, 1191–1212.
Weiss, A., Adams, M. J., & King, J. E. (2011). Happy orang-utans live longer lives.
Biology Letters, 7, 872–874.
Weiss, A., Adams, M. J., Widdig, A., & Gerald, M. S. (2011). Rhesus macaques (Macaca
mulatta) as living fossils of hominoid personality and subjective well-being. Journal of
Weiss, A., Bates, T. C., & Luciano, M. (2008). Happiness is a personal(ity) thing: The
genetics of personality and well-being in a representative sample. Psychological
Science, 19, 205–210.
Weiss, A., & Costa, P. T., Jr. (2005). Domain and facet personality predictors of all-cause
mortality among Medicare patients aged 65 to 100. Psychosomatic Medicine, 67(5),
724–733.
Weiss, A., King, J. E., & Enns, R. M. (2002). Subjective well-being is heritable and
genetically correlated with dominance in chimpanzees (Pan troglodytes). Journal of
Weiss, A., King, J. E., & Figueredo, A. J. (2000). The heritability of personality factors in
chimpanzees (Pan troglodytes). Behavior Genetics, 30, 213–221.
Weiss, A., King, J. E., & Hopkins, W. D. (2007). A cross-setting study of chimpanzee
(Pan troglodytes) personality structure and development: Zoological parks and Yerkes
National Primate Research Center. American Journal of Primatology, 69, 1264–1277.
Weiss, A., King, J. E., & Perkins, L. (2006). Personality and subjective well-being in
orangutans (Pongo pygmaeus and Pongo abelii). Journal of Personality and Social
Weiss, A., Inoue-Murayama, M., Hong, K. W., Inoue, E., Udono, T., Ochiai, T., … King,
J. E. (2009). Assessing chimpanzee personality and subjective well-being in Japan.
American Journal of Primatology, 71, 283–292.
Weiss, A., Gartner, M. C., Gold, K. C., & Stoinski, T. S. (2013). Extraversion predicts
longer survival in gorillas: An 18-year longitudinal study. Proceedings of the Royal
Society B-Biological Sciences, 280, 20122231.
Wielebnowski, N. C., Fletchall, N., Carlstead, K., Busso, J. M., & Brown, J. L. (2002).
Noninvasive assessment of adrenal activity associated with husbandry and behavioral
factors in the North American clouded leopard population. Zoo Biology, 21, 77–98.
Willard, S. L., & Shively, C. A. (2012). Modeling depression in adult female cynomolgus
monkeys (Macaca fascicularis). American Journal of Primatology, 74, 528–542.
Wilson, D. S., Clark, A. B., Coleman, K., & Dearstyne, T. (1994). Shyness and boldness
in humans and other animals. Trends in Ecology and Evolution, 9, 442–446.
Wrangham, R. W., Wilson, M. L., & Muller, M. N. (2006). Comparative rates of violence
in chimpanzees and humans. Primates, 47, 14–26.
Yerkes, R. M. (1939). The life history and personality of the chimpanzee. The American
Naturalist, 73, 97–112.
Zuckerman, M., Kuhlman, D. M., Thornquist, M., & Kiers, H. (1991). Five (or three)
robust questionnaire scale factors of personality without culture. Personality and
Behavior and Molecular Genetics of the Five Factor
Model
Amber M. Jarnecke and Susan C. South
Abstract
Behavior and molecular genetics informs knowledge of the etiology, structure,
and development of the Five Factor Model (FFM) of personality. Behavior
genetics uses quantitative modeling to parse the relative influence of nature and
nurture on phenotypes that vary within the population. Behavior genetics research
on the FFM has demonstrated that each domain has a heritability (proportion of
variation due to genetic influences) of 40–50%. Molecular genetic methods
attempt to identify specific genetic mechanisms associated with personality
variation. To date, findings from molecular genetics are tentative, with significant
results failing to replicate and accounting for only a small percentage of the
variance. However, newer techniques hold promise for finding the “missing
heritability” of FFM and related personality domains. This chapter presents an
overview of commonly used behavior and molecular genetic techniques, reviews
the work that has been done on the FFM domains and facets, and offers a
perspective for future directions.
Key Words: behavior genetics, molecular genetics, genes, environment, twin,
family
Personality is a construct that taps into how people tend to feel, think, and
behave. The Five Factor Model (FFM) of personality proposes that personality
can be captured according to a person’s standing on five domains: neuroticism,
extraversion, openness, agreeableness, and conscientiousness. Each of the five
higher-order domains are further subdivided into six facets (e.g., for
extraversion, they are warmth, gregariousness, assertiveness, activity,
excitement-seeking, and positive emotions). There is variability between
individuals on each of these domains and facets. For example, some individuals
may be more warm and assertive whereas others are cold and submissive. It is
the unique combination of a person’s standing across the facets and domains that
describes his or her personality. It is reasonable to ask, though, why some people
are more or less likely to be extraverted and outgoing, open to new experiences
and ideas, or prone to experiencing negative emotions. To determine why a
person’s personality manifests in a particular way we can look to both genes and
environment for an explanation.
The study of behavior and molecular genetics can help us examine where
personality traits come from, how they develop, and how they change over time.
Behavior genetics methods provide estimates about the comparative influence of
genes and environment on a given personality trait (domain or facet) in the
population. The behavior genetics field has firmly established that personality
stems from both genetic and environmental influences; that is, both nature and
nurture are at play in the origin of personality. Although behavior genetics
methods demonstrate that genes play a role in personality, they cannot provide
any information on which genes are exerting their influence on personality.
From there it is necessary to turn to the methods of molecular genetics. Over the
past few decades, the growing field of personality molecular genetics has
attempted to identify measured genes that contribute to the variance in
personality traits.
In this chapter, behavioral genetics and molecular genetics research on FFM
personality domains and facets are reviewed. A broad overview of the methods
will be presented, applications and limitations of this work in personality will be
addressed, and recommendations for future work will be provided.
Behavior Genetics of Personality

Univariate Twin Studies
This chapter begins with an overview of the rationale and techniques of the twin
method, the most commonly used behavior genetics method, before moving on
to a review of what has been found for the FFM domains and facets and related
personality traits.
Twin methodology. Behavior genetics relies on genetically informative
family data—and a majority of the behavior genetics work on personality has
relied on twin samples. Twins provide a natural experiment, offering the
opportunity to unravel the comparative influence of genes and environment on a
given phenotype; a phenotype is a variable, like personality, that varies
continuously within the population. If a variable has little or no variance in the
population, then it makes no sense to try to explain its variance (e.g., by and
large the entire human population has two legs, in the absence of any rare
genetic mutation). Personality traits do vary continuously throughout the
population, and it is this variation from person to person that we are trying to
understand. That is, we are trying to explain why some individuals are very
extraverted, enjoying the companionship of others on a near-constant basis,
whereas others are more introverted and prefer being alone. The source of that
variation can logically be located in either genetic differences between
individuals in the population or differences in the environment which they
encounter. Twins who grow up together in the same household have known
degrees of genetic relatedness and shared family experiences, making them
invaluable for understanding personality variation. There are, of course, other
types of genetically informative family data that are useful for understanding
genetic and environmental influences on personality (we discuss the powerful
adoption design below), but for many pragmatic reasons the twin study is the
most widely used.
Identical, or monozygotic (MZ), twins come from one fertilized egg that is
split in two in utero and they share 100% of their genes. Fraternal, or dizygotic
(DZ) twins, are the product of two eggs fertilized in utero at the same time. DZ
twins share approximately 50% of their genes and, other than being born at
roughly the same time and sharing a uterus, and are no more likely to be similar
than any two nontwin siblings. DZ twins can be either same-sex (boy–boy, girl–
girl) or opposite-sex (boy–girl) pairs; the latter are necessary for some types of
modeling that investigate sex differences in genetic and environmental
influences. To collect sample sizes that are large enough to have adequate power
for biometric modeling, many researchers have established databases of twins
based on publicly available birth records. In the United States, these twin
databases are often regionally located, although the twin sample from the
Midlife in the United States Study (MIDUS; see Kendler, Thornton, Gilman, &
Kessler, 2000) is unique in being a nationwide sample of adult twins. An
important consideration for twin databases is how zygosity is determined. For
many twin samples, researchers will use a screener that asks questions such as
“How often are you and your sibling mistaken for each other?” More recently,
researchers have been able to confirm zygosity using genetic screening.
Comparing the correspondence between MZ twins and DZ twins (all twins
having been reared together, see below for twins reared apart) allows us to
determine the extent to which genes and environment explain phenotypic
variation.
To first determine the genetic and environmental variance of a trait, the
correlations between twin pairs on a phenotype can be estimated and the
magnitude of the correlation for MZ twins is compared to the correlation for DZ
twins. A rough estimate of heritability, or the proportion of variance in a variable
due to genetic differences between people, is calculated using the formula
2•(rMZ–rDZ). For instance, if we collected data on the NEO Personality
Inventory-Revised (NEO PI-R; Costa & McCrae, 1992) from a sample of MZ
and DZ twins, and the correlation between MZ twins for Neuroticism was .65
and the correlation between DZ twins was .40, then heritability would equal
2(.65–.40) = .50. That is, approximately 50% of the variance in NEO PI-R
Neuroticism would be explained by genetic differences between people in the
sample. In general, when the MZ correlation is larger than the DZ correlation the
presence of genetic effects on a trait is inferred. If the MZ correlation is greater
than twice the DZ correlation, the presence of dominant (nonadditive) genetic
effects is indicated.
Formal biometric modeling of twin data can be accomplished through the use
of structural modeling software that compares the covariance, or similarity,
within MZ and DZ twin pairs on a phenotype, providing estimates for the
proportion of varianceexplained by genetic and environmental influences. The
univariate biometric twin model shown in Figure 14.1 decomposes the total
variance of one phenotype, P, into three components that account for the total
variance in the population: additive genetic influences, abbreviated A in the
figure, common or shared environmental influences, abbreviated C, and
nonshared or unique environmental influences, abbreviated E (Plomin, DeFries,
McClearn, & McGuffin, 2008). In this model, some parameters are fixed to
certain values, which are derived from known relationships about twins reared
together. Correlations between the A latent variables are set to 1.0 for MZ twins
and 0.5 for DZ twins to account for the genetic relatedness of twins in a pair; the
correlation between C latent variables is set to 1.0 to account for twins sharing
the same common environment.
Figure 14.1. Univariate ACE model for the decomposition of a phenotype, P, into
additive genetic (A), shared environmental (C), and nonshared environmental (E). Both
members of the twin pair (P Twin 1 and P Twin 2) are shown. Correlations between latent
genetic factors are set to 1.0 for MZ twins and 0.5 for DZ twins. Correlations between
latent common environmental factors are set to 1.0 for MZ and DZ twins. Total
phenotypic variance is calculated by summing the squared paths leading to it: a2 + c2 +
e2.
Heritability (abbreviated h2) is the genetic variance A over the total variance
in the phenotype or trait (A + C + E). It is important to note that heritability does
not indicate how much of one person’s genes influence his or her behavior. It is
a population parameter that measures the amount of genetic variance over total
trait variance (including genetic, shared, and nonshared environmental
variation). Thus, if it is found that a personality trait has a heritability estimate of
40%, this means that genetic variation between people in that given population
accounts for 40% of the trait’s total variance. It would not be correct to infer that
genes account for 40% of the personality trait in any one individual. Heritability
estimates from twin models assume that genetic influences are additive,
suggesting that the total genetic variance can be explained by many genes with
small effect sizes located at different positions (or loci) on the genome.
However, it is possible that nonadditive genetic influences also contribute to
phenotypic variation. There may be effects from a dominant gene or genetic
contributions may interact with one another to influence personality. In fact,
there is research supporting the presence of nonadditive genetic effects on
personality (e.g., Keller, Coventry, Heath, & Martin, 2005). Estimates of
nonadditive genetic effects (usually abbreviated D in biometric models) can be
included in the univariate model shown in Figure 14.1, but because of model
constraints it is necessary to remove one of the other sources of variance, usually
C.
Another source of influence on the variation of a phenotype is the shared or
common environment (C). The shared environmental component of variance
reflects the degree to which twins in a pair are alike because they grew up under
the same roof. Aspects of the shared environment may include socioeconomic
status (SES), neighborhood, similar peer groups, and/or comparable interactions
with parents. Just as heritability is a proportion, the proportion of variance due to
shared environment can be calculated [C/(A+C+E)], and is often abbreviated c2.
The last source of variance examined by biometric models is from the nonshared
or unique environment (referred to as E, raw variance, or e2, for proportion of
variance). The nonshared environmental component reflects the amount that
twins are different from one another even though they share the same genetic
material and grew up in the same household. The nonshared environment may
include traumatic events, different peer groups, events in utero, and unique
interactions with parents. Measurement error is also accounted for in the
nonshared environmental estimate. Therefore, nonshared environmental
influences may be inflated if there is bias or imprecision in measurement.
Limitations and assumptions of the twin method. Differentiating
environmental components that might fall under the shared environment versus
the nonshared environment may be difficult to identify. For example, having
similar peer groups may be thought of as a shared environmental influence that
makes siblings more similar to one another. However, a peer group may act as a
nonshared environmental influence if one twin’s experience with the peer group
is different or is perceived as different from his or her co-twin. Differences
between twins on this experience would cause them to be less similar to one
another and would be accounted for within the nonshared environmental
component of variance.
The twin method also assumes that twins are, in general, just like other
nontwin singletons. Some may question whether twins have unique personalities
just by virtue of being a twin. Johnson and colleagues (2002) examined this issue
using data on the Multidimensional Personality Questionnaire (MPQ; Tellegen,
1982) from the Minnesota Twin Registry, and found that the only difference
between twins and singletons was greater Social Closeness in twins. Twin
modeling is also based on the assumption that DZ twins share, on average, 50%
of their differentiating genes. If assortative mating occurs, that is, the twins’
parents are correlated on the phenotype of interest (e.g., perhaps they selected
one another as a mate on the basis of sharing common personality traits), this
could lead to greater than 50% sharing of genes. This would result in a higher
DZ correlation than if there were no assortative mating, thus resulting in lower
heritability estimates. Twin researchers should be careful to specifically model
any assortative mating, but in general most studies of personality have found
little evidence of similarity or assortative mating for personality (e.g., Watson et
al., 2004).
Biometric modeling with twin data also relies on the equal environments
assumption (EEA). We infer from twin data that if there are more similarities
between MZ pairs raised in the same environment than DZ pairs, genetic
influences are contributing to this similarity. Some have questioned, however, if
MZ pairs are more alike than DZ twin pairs because their parents treated them in
the same way (e.g., dressed them alike, treated them alike) than the parents of
DZ twins. In this case, genetic influences would be overestimated. Although this
is a concern regarding biometric models, there is evidence to support the EEA.
In the cases in which the environment does seem to have an effect of making
MZ twins more alike, research suggests that this does not impact phenotypic
correlations for personality (Loehlin & Nichols, 1976; Scarr & Carter-Saltzman,
1979).
Further support for the EEA is found in studies of twins reared apart. A
unique and small group, these are twins who were separated usually at a very
young age, were raised by different families, and often do not even know of the
other’s existence. It might be expected that heritability estimates for twins reared
apart would be lower for twins raised in different environments, but studies
generally find that twins reared apart show heritability estimates similar to those
shown in studies examining twins reared in the same environment. For example,
in a study of reared-together twins and twins reared apart, correlations for MZ
twins reared apart were very similar to correlations for MZ twins reared together
on the MPQ scales (Tellegen et al., 1988). In addition, little support for influence
from the shared environment was found for 2 of the 14 MPQ scales. Only the
Positive Emotionality and Social Closeness scales showed a significant influence
from the shared environment. This suggests that rearing environment may have
less of an influence on the development of personality previously thought.
Findings from univariate twin studies. With regard to FFM research, it is
consistently found that both nature and nurture contribute to personality traits.
The heritability of all five higher-order domains (neuroticism, extraversion,
openness, agreeableness, and conscientiousness) ranges from 40% to 50% and
the majority of the remaining variance is ascribed to nonshared environmental
factors (Bouchard & Loehlin, 2001). Studies examining the heritability of FFM
domains have sampled participants from the United States, Canada, and
Germany and have used variations of Costa and McCrae’s (1992) measures,
including in particular the NEO PI-R and the NEO-Five Factor Inventory (NEO-
FFI) in addition to other Big Five or FFM measures. Many of these studies have
found evidence for nonadditive genetic variance in one or more traits; however,
the domains that show nonadditive influences vary between studies. This might
be due to differences in measurement or population.
Heritability estimates have also been derived for FFM facets that lie within the
higher-order domains. As it turns out, there is some variability in heritability
estimates across facets within a domain (Jang, Livesley, & Vernon, 1996). One
study using the NEO PI-R in a sample of Canadian adolescents and adults found
that facets of Neuroticism showed the greatest consistency in genetic (26–44%)
and environmental influences (56–74%), with the exception of Angry Hostility,
for which nonadditive genetic influences were found (d2 = .33). Facets
belonging to the other four FFM domains presented greater variability in their
etiology. Not surprisingly, additive genetic effects were found for some facets in
each domain (e.g., Gregariousness from Extraversion, Fantasy from Openness,
Altruism from Agreeableness). Additionally, nonadditive genetic effects were
also found for other facets in extraversion, openness, and agreeableness (e.g.,
Warmth from Extraversion, Aesthetics from Openness, and Compliance from
Agreeableness). For a few facets there was no evidence of significant additive or
nonadditive genetic influences, with all variance accounted for by shared and
nonshared environmental influences: Feelings from the Openness domain,
Modesty under Agreeableness, and Order, Self-Discipline, and Deliberation
under Conscientiousness.
The studies reviewed above did not directly address the possibility of sex
differences in estimates of genetic and environmental influences. Many behavior
genetics studies control for effects of sex or gender (McGue & Bouchard, 1984),
but a few in the personality literature have specifically examined sex differences.
There are different types of sex differences that can be examined; qualitative sex
differences exist when there are different genes operating for men and women,
whereas quantitative sex differences occur when the genes are the same but the
magnitude of genetic influences is different across sex. One study that drew its
sample from twins from the United States and their families, as well as twins
from Finland and Australia, found that the broad sense heritability (a
combination of additive and nonadditive genetic influences) of extraversion was
roughly the same for males and females (.47 and .52 for males and females,
respectively) but heritability estimates of neuroticism were higher for females
(.51) than males (.38; Eaves, Heath, Neale, Hewitt, & Martin, 1998). Another
study that assessed personality using the MPQ did not find sex differences in the
higher order factors of Negative Emotionality (similar to FFM neuroticism),
Positive Emotionality (similar to FFM extraversion), or Constraint (a blend of
FFM agreeableness and conscientiousness) but did find different heritability
estimates for the lower-order scales of Alienation (25% for men, 16% for
women), Control (2% for men, 20% for women), and Absorption (11% for men,
29% for women; Finkel & McGue, 1997).
For most univariate ACE models of personality, estimates of the shared
environment are typically close to 0. This suggests that the family in which you
are raised has little or no impact on personality variation beyond the influences
from genes, leading some to conclude that the families have much less influence
on development than other contextual factors (e.g., peer groups; see Harris,
1998). There are numerous rebuttals to this argument from outside the field of
behavior genetics (see Collins, Maccoby, Steinberg, Hetherington, & Bornstein,
2000). In addition, a few twin studies have uncovered shared environmental
influence on certain personality traits. For example, Krueger, Hicks, and McGue
(2001) found support for the effects of the shared environment on the personality
trait of altruism. It is also possible that many of the FFM twin studies that do not
find evidence of shared environment are hampered by assessment method. In a
study that examined NEO-FFI personality traits rated by unacquainted observers,
influence from the shared environment was found (Borkenau, Riemann,
Angleitner, & Spinath, 2001). In addition, it is possible that estimates of the
shared environment hover around 0 because environmental factors have an
interactive influence on personality. For example, researchers have reported
higher estimates of C among individuals at the extreme ends of the parent–child
relationship (an example of G×E that is addressed further later in this chapter;
Krueger, South, Johnson, & Iacono, 2008). Standard biometric models would
capture this source of influence in genetic or nonshared environmental estimates.
However, newer biometric moderation models are capable of finding shared
environmental influences at extreme ends of specific environments. These types
of models will be discussed in greater detail later in the chapter.
Behavior genetic research using twin samples to examine personality
generally, and the FFM specifically, is limited in some regard. Most of the
research has used adult twin samples and self-report methods. The reliance on
these samples and methods may bias estimates. When self-reports are
supplemented by observer or peer reports, estimates from genetic influences
increase (e.g. Borkenau et al., 2001; Riemann, Angleitner, & Strelau, 1997;
Wolf, Angleitner, Spinath, Riemann, & Strelau, 2004). More recently, behavior
genetics work on personality has included child and adolescent samples, rather
than just adult samples, which have self- and observer-reported personality
traits; however, thus far this work has not incorporated the FFM personality
domains. Interestingly, the behavior genetic studies on personality in children
have demonstrated the presence of shared environmental influences. In a sample
of 9- and 10-year-old children, substantial heritability estimates were found for
the Self-directedness and Harm Avoidance subscales of the Junior Character and
Temperament Inventory (Cloninger, Svrakic, & Przybeck, 1993); shared
environmental estimates were found for the Novelty Seeking and
Cooperativeness subscales (Isen, Baker, Raine, & Bezdjian, 2009). In another
study examining inhibitory control (IC) in toddlers, 58% of the variation was
attributed to genetic influences, 26% to shared environmental influences, and
16% to nonshared environmental influences for parent ratings; estimates were
38% genetic and 62% nonshared environment using observer ratings. These
studies provide some evidence that developmental period and rater may
differentially influence estimates for the genetic and environmental effects on
personality; however, these estimates need to be replicated in future work and
extended to measures that assess the FFM domains and facets.
Multivariate and Longitudinal Twin Studies

In the twin studies discussed thus far, univariate biometric models were used
to obtain estimates of the influence of genes and environment on FFM domains
and facets considered individually. This type of model can be extended to
examine the etiology of multiple personality traits or other phenotypes, which is
referred to as multivariate biometric modeling. These types of models are useful
in examining the structure and development of personality, as discussed below.
Multivariate twin methodology. Multivariate biometric modeling in
personality research can be used to decompose the covariance between two or
more personality traits or the covariance between a personality trait and another
phenotype (e.g., a putative environmental risk factor). The model estimates the
genetic and environmental influences unique to each phenotype as well as
genetic and environmental variance common between phenotypes. The common
variance presents as a correlation. For example, the genetic correlation
represents the amount of genetic variance Phenotype A shares in common with
Phenotype B. The respective environmental correlations represent the degree to
which environmental influences on Phenotype A are shared with environmental
influences on Phenotype B.
Examining the overlap between genetic and environmental influences on
various personality facets and domains has the potential to inform our
understanding of personality structure. In addition, exploring the overlap in
personality traits with a longitudinal design may help us understand personality
stability. Finding that nonshared environmental correlations among personality
traits are large across time points, for example, would indicate that the stability
of personality is largely influenced by environmental experiences unique to a
twin in a twin pair.
There are three types of multivariate models that can be used to examine
overlapping sources of variance. The simplest model employs a Cholesky
decomposition, which parses the genetic and environmental variance shared
between two or more phenotypes. This model is shown in Figure 14.2; for
simplicity, one-half of the twin pair is displayed. In this model, one set of ACE
influences (A1, C1, E1) is estimated for all the phenotypes and a second set of
uncorrelated ACE influences (A2, C2, E2) is estimated for only the downstream
variable. This model estimates the degree to which etiologic influences are
shared between multiple phenotypes but does not impose any a priori constraints
on why or how these influences might be shared; it is up to the researcher to
determine the order of entry of the variables into the path model a priori. It is
also possible to remove parameters, with the ultimate goal of obtaining the most
parsimonious model to explain the relationships among the variables. This type
of model would be useful, for example, to examine the shared genetic and
environmental influences between an FFM domain and a putatively related
environmental variable.
Figure 14.2. Path model for a Cholesky decomposition of variance into additive genetic
(A), shared environmental (C), and nonshared environmental (E) sources. Shown for one
twin in a pair.
Figure 14.3. An independent pathways model for the genetic and environmental
decomposition of variance into common latent genetic (A), shared environmental (C), and
nonshared environmental (E) sources as well as ACE estimates specific to each
phenotype. Shown for one twin in a pair.
The independent pathways (IP) model (Figure 14.3) builds on the Cholesky
decomposition by positing that there are common latent genetic and
environmental factors with direct effects on the phenotypes and genetic and
environmental influences specific to each phenotype. As such, it is a stricter
model than the Cholesky decomposition. In this model, one set of ACE factors is
estimated for all phenotypes. Additionally, unique ACE components are derived
for each phenotype in the model to estimate genetic and environmental
influences that are not common to the other phenotypes.
The common pathways (CP) model is the third multivariate model. It imposes
the greatest amount of structure on the etiologic influences contributing to a set
of phenotypes (Figure 14.4). This model is most similar to a phenotypic factor
analysis. It provides a single latent construct, P, that accounts for the covariance
among multiple phenotypes that is decomposed into genetic and environmental
effects (AP, CP, EP). It also includes the ACE influences unique to each
indicator. If an FFM domain is etiologically coherent, the CP model should fit
the data better than the IP model or Cholesky decomposition.
Findings from multivariate twin modeling: environmental correlates. The
univariate twin model was vital in showing that personality variation was due to
genetics and nonshared environment. However, the great challenge over the past
few decades has been to determine the source of that nonshared variation. The
“gloomy prospect” suggests that finding any aspect of the nonshared
environment will be difficult if not impossible, because these environmental
experiences are so idiosyncratic (Turkheimer, 2000; Turkheimer & Waldron,
2000). Examining the overlap in the sources of variance between a personality
trait and another variable has implications for understanding nonshared
environmental factors that contribute to nearly 50% of the variance in
personality. For example, the Cholesky decomposition can include a personality
trait and a second environmental variable to estimate genetic and environmental
influences shared between the two. This may help explain why personality is
phenotypically related to a certain environmental variable.
Figure 14.4. A common pathways model with a single latent construct, P, decomposed
into additive genetic (A), shared environmental (C), and nonshared environmental (E)
sources of variance. Shown for one twin in a pair.
If a moderate to large genetic correlation (rA) was found, this would indicate
that a personality trait and environmental variable are associated because
common genetic influences contribute to both outcomes. Many variables of
interest that are putatively environmental (e.g., parenting, marital satisfaction)
are found to be heritable (Kendler & Baker, 2007), and some researchers are
interested in examining the genetic and environmental associations between
these environmental variables and personality. For instance, the FFM personality
domains of neuroticism, agreeableness, and conscientiousness share some
common genetic influences with parent–child relationship quality (Riemann,
Kandler, & Bleidorn, 2012), and wives’ personality traits of optimism and
aggression share common genetic influences with their reports of marital
satisfaction as well as their husbands’ reports of marital satisfaction (Spotts et
al., 2005). In another sample of adult twins, both self-report and peer-report
ratings of NEO PI-R personality domains shared common genetic influences
with retrospective reports of rearing environment (Kandler, Riemann, &
Kämpfe, 2009); estimates for unique environmental influences shared between
personality and rearing environment were small. This finding suggests that
unique environmental experiences manifesting in the rearing environment have
only a small to negligible influence on adult personality. In addition, this study
found that genetic correlations were greater for self-reported, versus peer-
reported, personality, indicating that genetic influences may affect an
individual’s perception of past environments and self.
Multivariate twin modeling: personality structure. Findings from
multivariate biometric models have also informed the etiologic structure of
personality. Just as factor analysis has been used to examine the phenotypic
structure of personality, multivariate biometric modeling can determine whether
the etiologic structure of the FFM domains mirrors that of the phenotypic.
Studies examining the etiologic structure of personality thus far have suggested
that the structure is complex. For example, in a combined sample of Canadian
and German adult twins, Jang and colleagues (2002) found that lower-order
facets of the NEO PI-R did not load as anticipated onto five genetic factors. For
each FFM higher-order personality domain, two genetic and two nonshared
environmental liabilities were found. This suggests that the higher-order FFM
domains may not be truly coherent constructs.
Another study examined the etiologic structure of Big Five personality
domains in the MIDUS study, which includes a nationwide sample of adult
American twins who completed a measure of trait adjectives taken from existing
inventories. The authors found that common factors accounted for the variance
in extraversion and neuroticism (Johnson & Krueger, 2004). This suggests that
these two FFM domains were unitary latent personality constructs. The same
study found that the IP model provided the best fit for openness and
conscientiousness and the Cholesky provided the best fit for agreeableness.
These findings suggest that the openness, conscientiousness, and agreeableness
domains may have an etiologic structure that is not as coherent as the
extraversion and neuroticism domains.
Finally, another study found support for a robust etiologic structure of the
FFM. This study utilized a cross-cultural sample of twins from Canada,
Germany, and Japan and factor analyzed the genetic and environmental
correlations among NEO PI-R facets (Yamagata et al., 2006). This procedure
yielded five genetically robust domains that mirrored the FFM higher-order
domains. This pattern was consistent across cultures, supporting the idea that the
FFM is universal.
The findings from these studies appear somewhat inconsistent with one
another. To explain this, it is possible that the FFM fails to reflect the nature of
personality. Nevertheless, the FFM does present more unity in its etiologic and
phenotypic structures than other models of personality, such as the Temperament
and Character Inventory (TCI; Cloninger, Svrakic, & Przybeck, 1993) for which
the genetic and phenotypic structures of the inventory yielded little agreement
(Ando et al., 2004). An alternative interpretation is that personality is
hierarchical and each facet-level trait of the FFM has its own genetic and
environmental influences (Jang, McCrae, Angleitner, Riemann, & Livesley,
1998). If this is the case, current measures of personality may fail to capture
etiologically coherent personality constructs. In the future, findings from
biometric modeling research may help inform personality inventories to produce
more etiologically robust personality domains.
Findings from multivariate twin modeling: development and stability. As
mentioned above, multivariate models can be extended to examine genetic and
environmental variance on personality over time. This type of modeling allows
us to estimate the relative influence of genes and environment on personality
stability at different points in time. Many studies using longitudinal biometric
models have found that the same genetic influences contribute to personality
traits at different ages. For example, one study examining withdrawn behavior in
children aged 3 to 12 years found that a substantial proportion of variation was
due to genetic effects at all ages, and that these genetic effects contributed to the
stability of withdrawal across time (Hoekstra, Bartels, Hudziak, Van
Beijsterveldt, & Boomsma, 2008). This investigation also found that shared
environmental effects, though modest, explained the stability of withdrawal in
girls but not boys. Estimates of nonshared environmental influences increased
over time but the nonshared environmental correlation decreased, suggesting that
different effects of the nonshared environment played a role at different ages.
Another study examining the stability of MPQ traits across two waves of data
during late adulthood found that large genetic and nonshared environmental
correlations explained the high stability in traits over time (Johnson, McGue, &
Krueger, 2005).
More recently, researchers have extended these developmental models by
using biometric latent growth curve models. In these models, the latent slope
(change) and intercept (initial level) factors are estimated for at least three waves
of personality data, and then the factors are decomposed into genetic and
environmental components. A recent study examined MPQ higher-order
personality traits across three waves, from late adolescence into adulthood. The
authors found that changes in traits were greater from the first wave to the
second than from the second to the third. In addition, it was found that genetic
and nonshared environmental influences both contributed to personality change
(e.g., the slope factor) (Hopwood et al., 2011). Genetic influences contributed to
change and the stability of trait levels over time and nonshared environmental
influences had an effect on changes in traits. Another recent study used this
approach, looking at change and stability of NEO PI-R traits in a sample of
German adults across three waves, each 5 years apart (Bleidorn, Kandler,
Riemann, Spinath, & Angleitner, 2009). This study found that genetic and
nonshared environmental influences each contributed to personality stability and
change; however, the etiology of personality change differed by domain or facet.
Strong genetic effects contributed to the change in neuroticism,
conscientiousness, and agreeableness; however, nonshared environmental
influences contributed to most of the change in extraversion and openness. In
addition, the genetic and environmental influences on a given facet tended to
vary compared to other facets within the same domain.
Findings from multivariate twin modeling: relationships with
psychopathology and well-being. Multivariate biometric models are also useful
for examining the shared etiology between personality and other relevant
outcomes, such as psychopathology or physical health. One study looked at the
association between subjective well-being and Big Five/FFM domains in the
MIDUS twin sample. The authors found that an IP model with one additive
genetic factor accounted for the covariance between subjective well-being and
the five personality domains, suggesting that the genetic variance on happiness
could be accounted for completely by the genetic variance on the Big Five/FFM
domains (Weiss, Bates, & Luciano, 2008). In other research, multivariate models
have examined the overlap between FFM and related traits and different forms
of mental illness (e.g., Agrawal, Jacobson, Prescott, & Kendler, 2004; Kendler,
Gatz, Gardner, & Pedersen, 2006). Most recently, researchers have incorporated
personality dimensions into latent domains of internalizing (e.g., mood
disorders, anxiety disorders) and externalizing (e.g., substance dependence,
antisocial personality disorder) psychopathology; for instance, neuroticism fits
well into the internalizing spectrum (e.g., Hettema, Neale, Myers, Prescott, &
Kendler, 2006; South & Krueger, 2008).
Biometric Moderation Models

In the past 10 years, another form of biometric model has gained prominence
in the field. Biometric moderation models have become popular for their ability
to specifically model gene × environment interaction (G×E). This refers to the
idea that genetic influences on a trait may vary depending on the presence or
absence of an environmental moderator variable, often thought of as a risk or
protective factor. For example, an individual may inherit a genetic predisposition
to express a particular personality trait, but this trait may only manifest given a
particular environment. Traditional univariate biometric models provide a
heritability estimate that is constant and population specific, but G×E can be
tested in twins using biometric moderation models. These models allow the
etiologic influences on a phenotype to differ by a person’s standing in a given
environment. One study utilizing these models found that the heritability of
positive and negative emotionality varied as a function of parent–child
relationship quality in a sample of adolescents (Krueger, South, Johnson, &
Iacono, 2008). For instance, genetic influences on positive emotionality
decreased (72% to 34%), shared environmental influences increased (1% to
46%), and nonshared environmental influences decreased (27% to 20%) from
low to high parent–child conflict. From low to high levels of parental
involvement, genetic influences on negative emotionality increased (31% to
59%), whereas shared environmental (36% to 21%) and nonshared
environmental (33% to 19%) influences decreased. Another study found that
etiologic influences on emotional instability (akin to neuroticism) varied by level
of maternal indulgence and family conflict (Jang, Dick, Wolf, Livesley, & Paris,
2005). Heritability estimates differed as a function of family variables as did
shared environmental variance. Considering the lack of shared environment
found in many twin studies, this introduces the possibility that shared
environmental effects may be found only in the most extreme environments.
Biometric moderation models may not always explain variations in
personality, however. Several studies have failed to find that environmental
variables moderate the etiologic influences on personality; for example, one
study found that the genetic and environmental influences on neuroticism did not
vary by level of family environment (Kendler, Aggen, Jacobson, & Neale,
2003). Another possibility is that personality may act as the moderator; for
instance, South and Krueger (2014) found that the genetic influences on alcohol
use problems varied as a function of level of FFM/Big Five conscientiousness,
with greater heritability of alcohol problems found at higher levels of
conscientiousness. There is much room for work in this area, as biometric
moderation models of personality and related variables (e.g., relationships,
health) are relatively underutilized compared to other phenotypes (e.g., alcohol
use disorders; see Young-Wolff, Enoch, & Prescott, 2011).
Adoption Studies and Extended Family Designs

Most of the behavior genetics work on personality has utilized twin samples to
decompose the variance attributable to genetic and environmental effects.
Families with one or more adopted children make up another genetically
informative sample that may provide information about the relative influence of
genes and environment on a phenotype like personality. Adopted,
nonbiologically related children share 100% of their rearing environment and
0% of their genes. If the same parents also have one or more biological children,
who share 100% of their rearing environment as well as the approximate 50% of
their segregating genes, the known degrees of genetic and environmental
relatedness between nonbiologically related siblings can be compared and used
to calculate estimates of genetic and environmental influence on personality.
Adoption studies are particularly powerful in examining the effect of the shared
environment; however, they tend to be somewhat more difficult than twin
studies as it is necessary to collect data on biological and/or adoptive parents, or
find adoptive families that contain both biological and adopted siblings.
Like twin designs, adoption studies find substantial influence from the
nonshared environment on personality traits and little to no influence from the
shared environment. However, adoption studies tend to find smaller heritability
estimates than twin studies (Loehlin, Willerman, & Horn, 1987; Plomin, Corley,
Caspi, Fulker, & DeFries, 1998). There are two possibilities that might account
for the difference in heritability estimates between twin and adoption studies.
First, the EEA may be violated in twin samples. Identical twins reared in the
same environment may be more alike because influences from the environment
serve to make them more similar; however, studies of twins reared apart have
reported heritability estimates equivalent to those found in samples of twins
raised together (Bouchard, 1994; Pedersen, Plomin, McClearn, & Friberg, 1988),
which supports the EEA. A second possibility to account for the difference in
heritability estimates is that adoption studies may be better able to detect and
account for the influence of nonadditive genetic effects. Twin studies that have
attempted to test for dominant (nonadditive) genetic influence rarely find
evidence of significant effects and may be limited in their ability to do so. In a
twin-plus-sibling design, Keller and colleagues (2005) found evidence of
nonadditive genetic effects for the Eysenck Personality Questionnaire (EPQ;
Eysenck & Eysenck, 1975) and TCI personality domains. Thus, extensions of
twin-only or adopted-only family designs may be better at uncovering
nonadditive genetic influences.
Extended adoption designs may also be better at identifying family influences
on personality. In an adoption study that included a sample of families with
parents and two adolescent siblings (biologically related siblings, two adopted
siblings, or one adopted and one biologically related sibling), the authors
reported that variation in negative emotionality was attributed mainly to genetic
and nonshared environmental influences and the variation in disinhibition had a
substantial shared environmental component (20%) (Buchanan, McGue, Keyes,
& Iacono, 2009).
Molecular Genetics of Personality

Thus far in this chapter, quantitative methods of exploring the relative
influence of nature and nurture have been discussed. In this section, more
qualitative methods, that is, molecular genetics, will be considered. As noted
above, molecular genetic methods allow researchers to identify specific genetic
markers that may contribute to a phenotype, such as personality. Uncovering
these markers may be important in identifying biological processes that
contribute to personality development. Here, several methods used in the rapidly
developing field of molecular genetics will be reviewed, but interested readers
are also referred to more in-depth explanations of these techniques (Neale,
Ferreira, Medland, & Posthuma, 2008; Yang, Lee, Goddard, & Visscher, 2011).
Candidate Gene Analysis

Candidate gene analysis is the earliest developed method in the molecular
genetics field. This method identifies a polymorphism or allele (form of a gene)
and examines its statistical relationship to a phenotype of interest. In personality
research, the candidate gene is often hypothesized to be related to the phenotype
and is chosen based on animal research or what is known about its mechanism of
action. The first molecular genetic studies of personality appeared in the 1990s,
reporting evidence that NEO Extraversion and Tridimensional Personality
Questionnaire (TPQ; Cloninger et al., 1993) Novelty Seeking were associated
with the dopaminergic gene DRD4 (see Ebstein, 2006). Other studies have
focused on a promoter region of serotonin transporter gene 5-HTTLPR. One
investigation suggested that 5-HTTLPR may be related to TPQ Harm Avoidance.
However, replications by independent research groups produced little additional
support for this association (Ebstein, 2006). Discrepancies between studies may
have been due, in part, to the measurement of personality traits.
More recently, meta-analyses have been used to address some of the
limitations of candidate gene analyses conducted by independent researchers.
For example, two of these meta-analyses suggest that 5-HTTLPR is associated
with avoidance-related traits (Munafò et al., 2003; Sen, Burmeister, & Ghosh,
2004). It has also been suggested that 5-HTTLPR and aggressive traits are
significantly associated, and dopaminergic genes (DRD3 and DRD4) are related
to approach and avoidance traits; however, these associations were reduced to
nonsignificance when age, ethnicity, and sex were added as covariates (Munafò
et al., 2003). Of note, some of the studies in this meta-analysis measured traits
using NEO assessments, whereas others measured traits with different
personality inventories such as the TPQ and TCI.
Gene × environment interactions and gene × gene interactions. It has been
difficult for researchers to find strong main effects of specific genetic
polymorphisms and personality domains. As a result, some molecular genetics
studies have also incorporated the idea of G×E. Just as biometric moderation
models were developed to investigate whether genetic influences on a trait
depend on certain environmental conditions, molecular genetic investigations of
G×E have examined interactions of specific alleles and environmental variables.
One study found support for G×E examining the monoamine oxidase A
(MAOA) candidate, childhood maltreatment, and antisocial behavior (Caspi et
al., 2002). It was reported that one MAOA genotype was protective against the
effects of maltreatment, as individuals with this genotype presented fewer
antisocial behaviors. However, this finding has been difficult to replicate. A
meta-analysis found that MAOA genotype status moderated the effect of
maltreatment on mental health outcomes but moderation was not significant for
antisocial behavior (Kim-Cohen et al., 2006).
Not only do genes and environment interact with one another, but there is also
the possibility for gene–gene interactions and epistasis. Epistasis is the effect of
one allele moderated by another, or multiple other, alleles. For example, the
serotonin transporter gene, SLC6A4, is modulated by dopaminergic (e.g., DRD4)
and gamma-aminobutyric acid (e.g., GABA[A]) genes (Ebstein, 2006). More
complex interactions (e.g., DRD4 × 5-HTTPLR × COMT) have been identified in
the personality literature, as well. In a study examining genetic associations
between serotonin-related genes and antisocial personality disorder (ASPD) in
prison inmates, an interaction between tryptophan hydroxylase, COMT, and 5-
HTR2A was found (Arias et al., 2011). This work elucidates the biological
pathways involved in ASPD, suggesting that the epistatic interaction of these
genes affects the dopamine pathway and moderates serotonin levels. Thus far,
studies of gene–gene interactions have not incorporated FFM measures of
personality. Work within this area should incorporate these measures in the
future.
Linkage Analysis
Linkage analysis identifies regions of interest on the chromosome from data
collected from family pedigrees. Because DNA is not copied exactly from the
mother and father, but is rearranged to create a new pattern, it is assumed that
genes located closer to one another on the chromosome are linked and therefore
have a higher probability of being transmitted together. One limitation of linkage
analysis is that it is most appropriately used to identify genes of large effect size,
and most complex human traits, including personality, are presumed to be
influenced by many genes of small effect sizes. Despite this limitation, several
linkage studies have been conducted linking chromosome regions to specific
personality traits. One study found five loci (locations on chromosomes)
associated with Neuroticism, as measured by the EPQ (Fullerton et al., 2003).
Another found linkage to Neuroticism on several chromosomes, replicating
some of the findings from the earlier study (Kuo et al., 2007). Only one study to
date has conducted a linkage analysis for all of the FFM domains (Amin et al.,
2012). The authors used a sample of 2,657 individuals from a genetically
isolated region of the Netherlands and found that each of the NEO-FFI domains
was associated with several loci for individuals in the top 10% of these traits.
This suggested that there may be genes with moderate to large effect sizes that
contribute to personality traits; however, replication of these findings is needed.
Genome-Wide Association Studies

Association studies attempt to determine if individuals who differ on the
amount of a phenotype, such as personality, also differ on the allele of a gene.
This involves measuring the polymorphism and determining whether it is
significantly associated with the phenotype of interest. Of the molecular genetic
methods covered thus far, these studies are the least driven by a priori
hypotheses. Until recently, association analyses could examine only a few
locations on the genome, but advances in technology and genome sequencing
have allowed researchers to examine hundreds of thousands of locations at one
time. This method is called genome-wide association study (GWAS). One
strength of this method, as compared to linkage analysis, is that it is much better
powered to detect genes of small effect. GWAS also has the ability to determine
the exact location of the gene. GWAS centers on single base pair changes in
DNA, referred to as single nucleotide polymorphisms (SNPs). The human
genome contains approximately 3 billion base pairs, many of which are the same
across all individuals; however, sometimes there is a change in a single base pair
and this change may lead to some individuals presenting a less common minor
allele. At present, over a million SNPs can be examined at one time (e.g., Li, Li,
& Guan, 2008). Although this enables researchers to examine a large portion of
the genome, it also allows for the possibility of false positives. Typically,
GWAS studies are corrected for the number of statistical tests, and a SNP has to
be significant at the level of p < 5 × 10–8. This means that GWAS must require
very large sample sizes.
To date, there has been one meta-analytic GWAS study of the FFM. The
authors examined approximately 2.4 million SNPs from over 20,000 individuals
across the world (de Moor et al., 2012). The analysis included 10 discovery
samples and 5 replication samples, all of European ancestry. All participants
completed the NEO-FFI. They did not find significant associations between any
of the SNPs and the traits of Neuroticism, Extraversion, or Agreeableness.
Significant associations were found between two SNPs on chromosome 5 and
Openness and one SNP on chromosome 18 and Conscientiousness; however,
these results did not replicate entirely across samples. Furthermore, the effect
sizes were small and the authors could only theorize about the mechanism of
effect between these SNPs and the related personality traits. Although this study
showed some promise for identifying specific genetic mechanisms involved in
personality, it appears that replication studies with much larger samples are
needed. At present, a research group is currently combining personality data
across an even larger number of samples in order to conduct additional analyses;
as part of this effort, researchers are conducting intensive item response theory
(IRT) analysis to determine the equivalency of the FFM measures across
samples (de Moor, van den Berg, & Boomsma, 2013).
Other approaches. In recent years GWAS approaches have been combined
with other methods in an attempt to expand the understanding of complex traits,
such as personality. For example, imaging genomics combines functional
neuroimaging with molecular genetics to determine how genetic variation is
related to mechanisms within the brain (e.g., Hariri & Holmes, 2006; Munafò,
Brown, & Hariri, 2008). There has also been a movement to shift the focus from
individual genes to gene systems. One study drew from a subset of participants
from the Study of Addiction: Genetics and Environment (SAGE; Bierut et al.,
2010) and pooled SNP information from eight dopaminergic genes to examine
associations between these genes and sensation seeking as measured by
Zuckerman’s Sensation Seeking Scale Form V (SSS-V; Derringer et al., 2010;
Zuckerman, Eysenck, & Eysenck, 1978). This investigation found that a number
of SNPs within four dopaminergic genes were associated with sensation-seeking
behavior. Another study used a summary score from SNPs that were presumed
to be related to NEO PI-R personality domains in order to form “molecular
personality scales” (MPSs; McCrae, Scally, Terracciano, Abecasis, & Costa,
2010). Researchers collected personality data and genomic scans of nearly 4,000
individuals from a genetically isolated province of Sardinia, and MPSs were
assessed in a GWAS. This investigation found that MPSs were related to four of
the five FFM domains (all but Extraversion). Of course, replication is needed,
but these novel approaches show some promise for identifying genetic and
biological mechanisms associated with personality.
Genome-Wide Complex Trait Analysis

The most recently developed molecular genetics method is genome-wide
complex trait analysis (GCTA; see Yang, Lee, Goddard, & Visscher, 2011).
GCTA uses very large samples to scan hundreds of thousands of SNPs across all
chromosomes or the whole genome to determine whether similarity in a
phenotype is associated with a large number of polymorphisms. Unlike GWAS,
which attempts to identify specific genes associated with a phenotype, GCTA
estimates the genetic variance that can be accounted for by shared SNPs. GCTA
requires that individuals in a sample are unrelated. Critics of twin studies view
this as an advantage of GCTA because it removes the possibility of inflated
heritability estimates due to increased similarity of MZ twins (see Equal
Environments Assumption, above). To date, there are no studies using GCTA to
examine FFM personality traits. However, GCTA studies looking at other
outcomes, such as callous-unemotional behavior (Viding et al., 2013) and
cognitive ability in children (Plomin, Haworth, Meaburn, Price, & Davis, 2013),
have in fact found that the proportion of variance explained by shared SNPs is
lower than the heritability estimates found in twin studies. Using a combined
sample of twins and their parents and adoptive families, one study gathered
heritability estimates from a subsample of twins, used GCTA, and created
genetic risk scores from SNPs to examine the associations between behavioral
disinhibition and substance use (Vrieze, McGue, Miller, Hicks, & Iacono, 2013).
Again, it was found that the estimates derived from GCTA were lower than the
heritability estimates from the biometric models, but both the twin analyses and
the risk scores suggested genetic overlap between behavioral disinhibition and
substance use.
GCTA is still a budding method. Although this method assumes that genetic
variance is additive, it is anticipated that analysis will be extended in the future
to include dominance and gene–gene interactions effects, as well as genetic
correlations and G×E. This opens up the possibility of asking complex questions
about gene–environment interplay that may not be captured using the twin
method.
Summary and Future Directions

In this chapter, an overview of how behavior and molecular genetics methods
can be used to uncover the nature of personality was discussed. Overall, these
methods have played a pivotal role in demonstrating that genes and environment
work in conjunction to manifest in individual differences and have helped us
focus on specific genetic mechanisms that may be involved. Much more work on
personality has been done within these fields, and much of it has focused on the
FFM personality trait model. However, much more still needs to be done,
specifically with regard to the FFM domains and facets.
Although it has been suggested that twin studies have outlived their
usefulness, it is important to consider that this method can provide us with more
than simple heritability estimates. As noted earlier, behavior genetics may help
us understand the etiologic structure of personality. This may prove useful in
understanding the phenotypic structure of personality as it emerges across time
and culture, and how personality is related to the etiology of psychopathology, a
concern of particular interest to the validity of the FFM. There is much that can
still be done using the twin and adoption methods as applied to the FFM. For
instance, work can be done to understand any potential sex differences in the
FFM domains and traits, as this will have implications for molecular genetic
investigations of these traits. Additionally, many researchers have conducted
complex multivariate modeling of personality (e.g., Markon, Krueger, &
Watson, 2005) and psychopathology (e.g., Hettema et al., 2006; Krueger,
Markon, Patrick, & Iacono, 2005), andthis work needs to be extended to
biometric modeling, specifically using measures of the FFM. For instance, a
study could employ a biometric multivariate modeling extension of the work
done by Markon and colleagues to investigate whether the genetic structure of
various personality measures, including the FFM, maps onto the phenotypic
structure.
There are also behavior genetic methods that have never been used with the
FFM domains and facets. For instance, the monozygotic twin differences design
identifies MZ twin pairs, matched for their genetic material and shared
environment, who are discordant on a particular environmental factor. This
provides an opportunity to examine differences between twins that may be
accounted for by their different standings on this environmental variable. In
essence, the discordant MZ design tests the counterfactual, allowing researchers
to see how one twin might have looked in the absence of the unique
environmental experience (see McGue, Osler, & Christensen, 2010 for a more
thorough discussion of this topic).
The field of molecular genetics has also encountered a number of criticisms,
primarily regarding the lack of replication across studies. Indeed, it has been
very difficult to identify specific polymorphisms that are associated with any
behavioral or health-related outcome, let alone personality. In addition,
challenges in molecular genetics are magnified when we consider that most
complex traits are believed to involve small effects of many genes and genes
may interact with one another or their environment to manifest in specific traits.
However, molecular genetics may be able to reliably identify genetic
polymorphisms related to personality through large collaborative efforts. For
example, in the largest GWAS meta-analysis to date, three SNPs were identified
as being involved in educational attainment, and this finding did replicate
(Rietveld et al., 2013). The large consortia that have been formed to address
some of the limitations of GWAS studies provide hope that future endeavors will
uncover specific genetic mechanisms involved in personality.
References
Agrawal, A., Jacobson, K. C., Prescott, C. A., & Kendler, K. S. (2004). A twin study of
personality and illicit drug use and abuse/dependence. Twin Research, 7, 72–81.
Amin, N., Schuur, M., Gusareva, E. S., Isaacs, A., Aulchenko, Y. S., Kirichenko, A. V.,
… van Duijn, C. M. (2012). A genome-wide linkage study of individuals with high
scores on NEO personality traits. Molecular Psychiatry, 17(10), 1031–1041.
Ando, J., Suzuki, A., Yamagata, S., Kijima, N., Maekawa, H., Ono, Y., & Jang, K. L.
(2004). Genetic and environmental structure of Cloninger’s temperament and character
dimensions. Journal of Personality Disorders, 18, 379–393.
Arias, J. M. C., Acosta, C. A. P., Valencia, J. G., Montoya, G. J., Viana, J. C. A., Nieto,
O. C., … Ruiz-Linares, A. (2011). Exploring epistasis in candidate genes for antisocial
personality disorder. Psychiatric Genetics, 21, 115–124.
Bierut, L. J., Agrawal, A., Bucholz, K. K., Doheny, K. F., Laurie, C., Pugh, E., … Rice, J.
P. (2010). A genome-wide association study of alcohol dependence. Proceedings of the
National Academy of Sciences of the United States of America, 107, 5082–5087.
Bleidorn, W., Kandler, C., Riemann, R., Spinath, F. M., and Angleitner, A. (2009).
Patterns and sources of adult personality development: Growth curve analyses of the
NEO PI-R scales in a longitudinal twin study. Journal of Personality and Social
Psychology, 97(1), 142–155.
Borkenau, P., Riemann, R., Angleitner, A., & Spinath, F. M. (2001). Genetic and
environmental influences on observed personality: Evidence from the German
Observational Study of Adult Twins. Journal of Personality and Social Psychology,
80, 655–668.
Bouchard, T. J. (1994). Genes, environment, and personality. Science (New York, N.Y.),
264, 1700–1701.
Bouchard, T. J., & Loehlin, J. C. (2001). Genes, evolution, and personality. Behavior
Genetics, 31, 243–273.
Buchanan, J. P., McGue, M., Keyes, M., & Iacono, W. G. (2009). Are there shared
environmental influences on adolescent behavior? Evidence from a study of adoptive
siblings. Behavior Genetics, 39, 532–540.
Caspi, A., McClay, J., Moffitt, T. E., Mill, J., Martin, J., Craig, I. W., … Poulton, R.
(2002). Role of genotype in the cycle of violence in maltreated children. Science (New
York, N.Y.), 297, 851–854.
Cloninger, C. R., Svrakic, D. M., & Przybeck, T. R. (1993). A psychobiological model of
temperament and character. Archives of General Psychiatry, 50, 975–990.
Collins, W. A., Maccoby, E. E., Steinberg, L., Hetherington, E. M., & Bornstein, M. H.
(2000). Contemporary research on parenting: The case for nature and nurture.
Costa, P. T. J., & McCrae, R. R. (1992). NEO-PI-R professional manual: Revised NEO
personality and NEO Five-Factor Inventory (NEO-FFI). Odessa, FL: Psychological
de Moor, M. H. M., Costa, P. T., Terracciano, A., Krueger, R. F., de Geus, E. J. C.,
Toshiko, T., … Boomsma, D. I. (2012). Meta-analysis of genome-wide association
studies for personality. Molecular Psychiatry, 17(3), 337–349.
de Moor M., van den Berg, S., & Boomsma, D. (2013, July). Finding genetic loci for
neuroticism and extraversion: Results from the Genetics of Personality Consortium
(GPC). Paper presented at the meeting of the Behavior Genetics Association, Marseille,
France.
Derringer, J., Krueger, R. F., Dick, D. M., Saccone, S., Grucza, R. A., Agrawal, A., …
Bierut, L. J. (2010). Predicting sensation seeking from dopamine genes. A candidate-
system approach. Psychological Science, 21(9), 1282–1290.
Eaves, L. J., Heath, A. C., Neale, M. C., Hewitt, J. K., & Martin, N. G. (1998). Sex
differences and non-additivity in the effects of genes on personality. Twin Research:
The Official Journal of the International Society for Twin Studies, 1, 131–137.
Ebstein, R. P. (2006). The molecular genetic architecture of human personality: Beyond
self-report questionnaires. Molecular Psychiatry, 11(5), 427–445.
Questionnaire. San Diego, CA: Educational and Industrial Testing Service.
Finkel, D., & McGue, M. (1997). Sex differences and nonadditivity in heritability of the
Multidimensional Personality Questionnaire Scales. Journal of Personality and Social
Psychology, 72(4), 929–938.
Fullerton, J., Cubin, M., Tiwari, H., Wang, C., Bomhra, A., Davidson, S., … Flint, J.
(2003). Linkage analysis of extremely discordant and concordant sibling pairs
identifies quantitative-trait loci that influence variation in the human personality trait
neuroticism. American Journal of Human Genetics, 72(4), 879–890.
Hariri, A. R., & Holmes, A. (2006). Genetics of emotional regulation: The role of the
serotonin transporter in neural function. Trends in Cognitive Sciences, 10(4), 182–191.
Harris, J. R. (1998). The nurture assumption: Why children turn out the way they do. New
York, NY: Free Press.
Hoekstra, R. A., Bartels, M., Hudziak, J. J., Van Beijsterveldt, T. C. E. M., & Boomsma,
D. I. (2008). Genetic and environmental influences on the stability of withdrawn
behavior in children: A longitudinal, multi-informant twin study. Behavior Genetics,
38, 447–461.
Hopwood, C. J., Donnellan, M. B., Blonigen, D. M., Krueger, R. F., McGue, M., Iacono,
W. G., & Burt, S. A. (2011). Genetic and environmental influences on personality trait
stability and growth during the transition to adulthood: A three-wave longitudinal
study. Journal of Personality and Social Psychology, 100(3), 545–556.
Isen, J. D., Baker, L. A., Raine, A., & Bezdjian, S. (2009). Genetic and environmental
influences on the junior temperament and character inventory in a preadolescent twin
sample. Behavior Genetics, 39, 36–47.
Jang, K. L., Dick, D. M., Wolf, H., Livesley, W. J., & Paris, J. (2005). Psychosocial
adversity and emotional instability: An application of gene-environment interaction
models. European Journal of Personality, 19, 359–372.
Jang, K. L., Livesley, W. J., Angleitner, A., Riemann, R., & Vernon, P. A. (2002).
Genetic and environmental influences on the covariance of facets defining the domains
of the five-factor model of personality. Personality and Individual Differences, 33, 83–
101.
Jang, K. L., Livesley, W. J., & Vernon, P. A. (1996). Heritability of the big five
personality dimensions and their facets: A twin study. Journal of Personality, 64(3),
577–591.
1556–1565.
Johnson, W., & Krueger, R. F. (2004). Genetic and environmental structure of adjectives
describing the domains of the Big Five Model of personality: A nationwide US twin
study. Journal of Research in Personality, 38(5), 448–472.
Johnson, W., Krueger, R. F., Bouchard, T. R., & McGue, M. (2002). The personalities of
twins: Just ordinary folks. Twin Research, 5(2), 125–131.
Johnson, W., McGue, M., & Krueger, R. F. (2005). Personality stability in late adulthood:
A behavioral genetic analysis. Journal of Personality, 73(2), 523–552.
Kandler, C., Riemann, R., & Kämpfe, N. (2009). Genetic and environmental mediation
between measures of personality and family environment in twins reared together.
Keller, M. C., Coventry, W. L., Heath, A. C., & Martin, N. G. (2005). Widespread
evidence for non-additive genetic variation in Cloninger’s and Eysenck’s personality
dimensions using a twin plus sibling design. Behavior Genetics, 35(6), 707–721.
Kendler, K. S., Aggen, S. H., Jacobson, K. C., & Neale, M. C. (2003). Does the level of
family dysfunction moderate the impact of genetic factors on the personality trait of
neuroticism? Psychological Medicine, 33, 817–825.
Kendler, K. S., & Baker, J. H. (2007). Genetic influences on measures of the
environment: A systematic review. Psychological Medicine, 37, 615–626.
depression. Archives of General Psychiatry, 63, 1113–1120.
Kendler, K. S., Thornton, L. M., Gilman, S. E., & Kessler, R. C. (2000). Sexual
orientation in a U.S. national sample of twin and nontwin sibling pairs. The American
Kim-Cohen, J., Caspi, A., Taylor, A., Williams, B., Newcombe, R., Craig, I. W., &
Moffitt, T. E. (2006). MAOA, maltreatment, and gene-environment interaction
predicting children’s mental health: New evidence and a meta-analysis. Molecular
Krueger, R. F., Hicks, B. M., & McGue, M. (2001). Altruism and antisocial behavior:
Independent tendencies, unique personality correlates, distinct etiologies.
Psychological Science: A Journal of the American Psychological Society/APS, 12,
397–402.
Krueger, R. F., Markon, K. E., Patrick, C. J., & Iacono, W. G. (2005). Externalizing
psychopathology in adulthood: A dimensional-spectrum conceptualization and its
implications for DSM-V. Journal of Abnormal Psychology, 114(4), 537–550.
Krueger, R. F., South, S. C., Johnson, W., & Iacono, W. (2008). The heritability of
personality is not always 50%: Gene-environment interactions and correlations
between personality and parenting. Journal of Personality, 76, 1485–1522.
Kuo, P.-H., Neale, M. C., Riley, B. P., Patterson, D. G., Walsh, D., Prescott, C. A., &
Kendler, K. S. (2007). A genome-wide linkage analysis for the personality trait
neuroticism in the Irish affected sib-pair study of alcohol dependence. American
Journal of Medical Genetics. Part B, Neuropsychiatric Genetics: The Official
Publication of the International Society of Psychiatric Genetics, 144B(4), 463–468.
Li, M., Li, C., & Guan, W. (2008). Evaluation of coverage variation of SNP chips for
genome-wide association studies. European Journal of Human Genetics: EJHG, 16(5),
635–643.
Loehlin, J. C., & Nichols, R. C. (1976). Heredity, environment and personality. Austin,
TX: University of Texas Press.
Loehlin, J. C., Willerman, L., & Horn, J. M. (1987). Personality resemblance in adoptive
families: A 10-year follow-up. Journal of Personality and Social Psychology, 53, 961–
969.
McCrae, R. R., Scally, M., Terracciano, A., Abecasis, G. R., & Costa, P. T. (2010). An
alternative to the search for single polymorphisms: Toward molecular personality
scales for the five-factor model. Journal of Personality and Social Psychology, 99(6),
1014–1024.
McGue, M., & Bouchard, T. J. (1984). Adjustment of twin data for the effects of age and
sex. Behavior, 14(4), 325–343.
McGue, M., Osler, M., & Christensen, K. (2010). Causal inference and observational
research: The utility of twins. Perspectives on Psychological Science, 5(5), 546–556.
Munafò, M. R., Brown, S. M., & Hariri, A. R. (2008). Serotonin transporter (5-HTTLPR)
genotype and amygdala activation: A meta-analysis. Biological Psychiatry, 63, 852–
857.
Munafò, M. R., Clark, T. G., Moore, L. R., Payne, E., Walton, R., & Flint, J. (2003).
Genetic polymorphisms and personality in healthy adults: A systematic review and
meta-analysis. Molecular Psychiatry, 8, 471–484.
Neale, B. M., Ferreira, M. A. R., Medland, S. E., & Posthuma, D. (Eds.). (2008).
Statistical genetics: Gene mapping through linkage and association. London, England:
Taylor & Francis.
Pedersen, N. L., Plomin, R., McClearn, G. E., & Friberg, L. (1988). Neuroticism,
extraversion, and related traits in adult twins reared apart and reared together. Journal
Plomin, R., Corley, R., Caspi, A., Fulker, D. W., & DeFries, J. (1998). Adoption results
for self-reported personality: Evidence for nonadditive genetic effects? Journal of
Plomin, R., DeFries, J. C., McClearn, G. E., & McGuffin, P. (2008). Behavioral genetics
(5th ed.). New York, NY: Worth Publishers.
Plomin, R., Haworth, C. M., Meaburn, E. L., Price, T. S., & Davis, O. S. P. (2013).
Common DNA markers can account for more than half of the genetic influence on
cognitive abilities. Psychological Science, 24(4), 562–568.
FFI scales. Journal of Personality, 65, 449–475.
Riemann, R., Kandler, C., & Bleidorn, W. (2012). Behavioral genetic analyses of parent
twin relationship quality. Personality and Individual Differences, 53(4), 398–404.
Rietveld, C. A., Medland, S. E., Derringer, J., Yang, J., Esko, T., Martin, N. W., …
Koellinger, P. D. (2013). GWAS of 126,559 individuals identifies genetic variants
associated with educational attainment. Science (New York, N.Y.), 340(6139), 1467–
1471.
Scarr, S., & Carter-Saltzman, L. (1979). Twin method: Defense of a critical assumption.
Behavior Genetics, 9(6), 527–542.
Sen, S., Burmeister, M., & Ghosh, D. (2004). Meta-analysis of the association between a
serotonin transporter promoter polymorphism (5-HTTLPR) and anxiety-related
personality traits. American Journal of Medical Genetics. Part B, Neuropsychiatric
Genetics: The Official Publication of the International Society of Psychiatric Genetics,
127B, 85–89.
South, S. C., & Krueger, R. F. (2008). Marital quality moderates genetic and
environmental influences on the internalizing spectrum. Journal of Abnormal
Psychology, 117(4), 826–837.
and its relationship to aging. Developmental Psychology, 50(5), 1362–1376.
Spotts, E. L., Lichtenstein, P., Pedersen, N., Neiderhiser, J. M., Hansson, K., Cederblad,
M., & Reiss, D. (2005). Personality and marital satisfaction: A behavioural genetic
analysis, European Journal of Personality, 227(October 2004), 205–227.
Tellegen, A. (1982). Brief manual for the Multidimensional Personality Questionnaire.
Tellegen, A., Lykken, D. T., Bouchard, T. J., Wilcox, K. J., Segal, N. L., & Rich, S.
(1988). Personality similarity in twins reared apart and together. Journal of Personality
Turkheimer, E. (2000). Three laws of behavior genetics and what they mean. Current
Directions in Psychological Science, 9(5), 160–164.
Turkheimer, E., & Waldron, M. (2000). Nonshared environment: A theoretical,
methodological, and quantitative review. Psychological Bulletin, 126, 78–108.
Viding, E., Price, T. S., Jaffee, S. R., Trzaskowski, M., Davis, O. S. P., Meaburn, E. L.,
… Plomin, R. (2013). Genetics of callous-unemotional behavior in children. PloS One,
8(7), e65789.
Vrieze, S. I., McGue, M., Miller, M. B., Hicks, B. M., & Iacono, W. G. (2013). Three
mutually informative ways to understand the genetic relationships among behavioral
disinhibition, alcohol use, drug use, nicotine use/dependence, and their co-occurrence:
Twin biometry, GCTA, and genome-wide scoring. Behavior Genetics, 43(2), 97–107.
Watson, D., Klohnen, E. C., Casillas, A., Simms, E. N., Haig, J., & Berry, D. S. (2004).
Weiss, A., Bates, T. C., & Luciano, M. (2008). Happiness is a personal(ity) thing.
Wolf, H., Angleitner, A., Spinath, F. M., Riemann, R., & Strelau, J. (2004). Genetic and
environmental influences on the EPQ-RS scales: A twin study using self- and peer
reports. Personality and Individual Differences, 37, 579–590.
Yang, J., Lee, S. H., Goddard, M. E., & Visscher, P. M. (2011). GCTA: A tool for
genome-wide complex trait analysis. American Journal of Human Genetics, 88(1), 76–
82.
Young-Wolff, K. C., Enoch, M.-A., & Prescott, C. A. (2011). The influence of gene-
environment interactions on alcohol consumption and alcohol use disorders: A
comprehensive review. Clinical Psychology Review, 31(5), 800–816.
Zuckerman, M., Eysenck, S., & Eysenck, H. J. (1978). Sensation seeking in England and
America: Cross-cultural, age, and sex comparisons. Journal of Consulting and Clinical
Personality Neuroscience and the Five Factor Model
Timothy A. Allen and Colin G. DeYoung
Abstract
Personality psychology seeks both to understand how individuals differ from one
another in behavior, motivation, emotion, and cognition and to explain the causes
of those differences. The goal of personality neuroscience is to identify the
underlying sources of personality traits in neurobiological systems. This chapter
reviews neuroscience research on the traits of the Five Factor Model (the Big
Five: Extraversion, Neuroticism, Openness/Intellect, Conscientiousness, and
Agreeableness). The review emphasizes the importance of theoretically informed
neuroscience by framing results in light of a theory of the psychological functions
underlying each of the Big Five. The chapter additionally reviews the various
neuroscientific methods available for personality research and highlights pitfalls
and best practices in personality neuroscience.
Key Words: personality, Five Factor Model, neuroscience, neurobiology,
Cybernetic Big Five Theory, individual differences, traits
Personality psychologists pursue at least three fundamental questions

regarding human nature: First, how do individuals meaningfully differ from one
another? Second, what are the causes of these individual differences? And third,
what are their consequences? In relation to the first question, a major problem
historically was identification of the most important dimensions of variation in
personality. The emergence of the Five Factor Model (FFM) or “Big Five” has
gone a long way toward solving this problem (Costa & McCrae, 1992; Goldberg,
1990; John, Naumann, & Soto, 2008; Markon, Krueger, & Watson, 2005). The
discovery of five consistent broad dimensions of covariation among specific
traits, in both lexical and questionnaire assessments of personality, has allowed
the field to begin moving beyond questions of taxonomy toward the systematic
accumulation of evidence regarding the causes and consequences of trait
differences. At this point, the consequences of variation in the Big Five have
been studied extensively; the five factors—Extraversion, Neuroticism,
Openness/Intellect, Conscientiousness, and Agreeableness—matter for many life
outcomes, in academic and industrial success, in relationships, in physical and
mental health, etc. (Ozer & Benet-Martinez, 2006). Their causes are not as
thoroughly researched, however, and this chapter reviews the progress that has
been made in identifying the neurobiological basis of the Big Five.
Personality neuroscience rests on the premise that all reasonably persistent
individual differences in thought, cognition, motivation, and emotion (that is,
personality) must entail patterns of consistency in the functioning of the brain
(DeYoung, 2010b; DeYoung & Gray, 2009). From this perspective, the brain is
the proximal source of all personality characteristics, and it is only by affecting
the brain that more distal influences in the genome and environment are able to
influence personality. As a result, two major goals of personality neuroscience
are to identify the neural substrates of personality and to better understand how
genetic and environmental forces, over the course of development, create the
relatively stable patterns of brain function that produce personality. So far, more
progress has been made on the first of these goals than on the second.
The rise of neuroscience technologies for brain imaging and molecular
genetics has led to a rapid proliferation of empirical reports over the past decade.
Research in personality neuroscience has employed many different personality
measures, behavioral tasks, and neurobiological techniques to shed light on the
workings of the human system, and it can be difficult to integrate all of these
into a coherent understanding. Here, we take advantage of the fact that the FFM
can categorize most personality trait measures in order to synthesize findings
from personality neuroscience over the past several decades. We begin by
describing the various tools available for personality neuroscience. Previous
reviews have highlighted a number of methodological limitations in personality
neuroscience research to date (DeYoung, 2010b; Yarkoni, 2015). We echo many
of these cautions and make a concerted effort, throughout the chapter, to
highlight methodologically rigorous research and to provide caveats regarding
findings that are suggestive but flawed.
After reviewing methods, we discuss theories of the psychological functions
underlying each of the Big Five. Beyond brain scanners and gene-identification
chips, theory is one of the most important tools in personality neuroscience.
Atheoretical research is sometimes published in this field, examining
associations of personality traits with brain structure or function or genetic
variation in a purely exploratory manner, but such an approach often makes it
difficult to achieve sufficient statistical power, given the need to correct for
multiple statistical tests when examining associations throughout large portions
of the brain. It also increases the temptation to develop post hoc explanations of
findings, even when they may be merely false positives. Theoretical approaches
to the FFM can provide hypotheses to guide research in personality
neuroscience.
Methodological Issues in Personality Neuroscience

Personality neuroscience, at the intersection of two fields, must contend with
the limitations of measurement in both. Most measurement of personality relies
on self-reports using questionnaires. Better questionnaire assessment can be
achieved by collecting informant reports from knowledgeable peers, in addition
to self-reports (Connelly & Ones, 2010; Vazire, 2010). Still, questionnaires do
not exhaust the possible methods of personality assessment. Various behavioral
and cognitive tasks may also be used to assess stable personality traits. Because
the FFM was discovered and established in questionnaire data, we focus
primarily on such data in this chapter. Nonetheless, we believe nonquestionnaire
methodologies are likely to grow in importance in personality neuroscience (and
personality psychology more generally), as researchers attempt to capture
consistencies in thought, behavior, emotion, and motivation in more diverse
ways.
Whereas personality psychology is largely dominated by a single type of
measure, neuroscience is a field burgeoning with technologies that allow
researchers to explore previously inaccessible details of the structure and
function of the human brain. Neurobiological methods in personality
neuroscience mostly fall into five general categories:
1. Neuroimaging techniques. The most prominent and frequently used method in

personality neuroscience is magnetic resonance imaging (MRI), which creates images of
the brain based on the magnetic properties of different tissue types. MRI is popular not
only because it is noninvasive but also because, in addition to measuring brain structure,
it can also be used to measure brain function, by taking advantage of the fact that blood
flow and oxygen use increase with neural activity. The blood-oxygen-level-dependent
(BOLD) signal from functional MRI (fMRI), therefore, can be used to indicate when
different regions of the brain are more or less active.
Researchers most often use fMRI while participants are engaged in some
computerized task in the scanner. One limitation of task-based fMRI is that
relative rather than absolute levels of neural activation must be studied;
activation during the task of interest (or during a particular type of event within a
task) must be contrasted with activation during other parts of the scan (which
could be a control task, a resting period, or other events within the same task).
Increasingly, however, fMRI researchers are also investigating patterns of
functional connectivity, rather than relative activation, which do not require a
contrast between tasks. Functional connectivity refers to the patterns of temporal
synchrony between different parts of the brain. If brain regions show a similar
temporal pattern of activation and deactivation during some portion of a scan,
they are said to be functionally connected. Analysis of functional connectivity
during periods of rest in the scanner has demonstrated that brain networks that
are spontaneously active closely resemble networks that are activated by specific
tasks (Laird et al., 2011; Smith et al., 2009). This discovery has led to an effort
to map the major networks of the brain using functional connectivity, and the
resulting maps provide useful clues about the brain’s large-scale functional
organization (Choi, Yeo, & Buckner, 2012; Yeo et al., 2011).
One of these networks in particular is worth introducing briefly here because
of its rather opaque label, the “default network” (also called the “default mode
network”), and because of its importance for several personality traits. The
default network received its label because it was discovered more or less by
accident as a function of the fact that neural activation must be studied through
contrasts (Buckner, Andrews-Hanna, & Schacter, 2008). In contrasts of task
versus rest, it was noted that a particular set of brain regions was frequently
more active during rest than during task. Hence, this pattern of activation was
considered the brain’s default mode, what the brain is likely to do when
participants are asked simply to rest and not to attend to external demands.
Subsequent research has determined that the default network is responsible for
simulating experience in a variety of contexts, including times when we
remember events in the past, imagine the future (or any other hypothetical state),
take on another person’s perspective, or evaluate ourselves (Andrews-Hanna,
Smallwood, & Spreng, 2014). These are the kinds of things that people tend to
do when they are not engaged by their immediate surroundings and their minds
are free to wander, but these processes can also be engaged by specific tasks
(e.g., memory or perspective-taking tasks). Here is a case in which the limitation
that task-based analysis of fMRI requires a contrast between two conditions led
to an important discovery.
Another neuroimaging technique, positron emission tomography (PET), has
also been used in personality neuroscience. It has the great advantage of
allowing measurement of receptors for particular neurotransmitters but the
disadvantage of being invasive, as it requires injection of radioactive tracers into
the bloodstream. Both MRI and PET are valuable for their spatial resolution.
2. Electrophysiological techniques. Electroencephelography (EEG) measures neural

activity by recording electrical activity along the scalp. It has a much higher temporal
resolution than fMRI, capable of tracking differences in brain activity on the order of
milliseconds (as opposed to seconds for fMRI), but has greatly reduced spatial resolution.
Other electrophysiological techniques, such as electrocardiography and assessment of
electrodermal activity, use peripheral nervous system activity to draw inferences about
brain processes related to emotion and motivation.
3. Molecular genetics. Variation in the genes that build the brain can be measured
through analysis of DNA. Commonly used molecular genetic techniques in personality
neuroscience include candidate gene studies, in which particular genes are investigated
because of their hypothesized relevance to personality, and genome-wide association
studies (GWAS), in which the entire genome is scanned for variation associated with
some trait or traits (see also the chapter by Jarnecke and South).
4. Psychopharmacological manipulation. Specific chemicals can be administered as
drugs in an attempt to implicate a given neurotransmitter, receptor, or other brain
molecule in the expression of a trait. Effects of the manipulation are examined either on
behavior or on some neurobiological assay. If the effects of the manipulation are
moderated by the trait, or vice versa, this implicates the targeted molecule in the trait.
5. Assays of endogenous psychoactive substances. Measurements of substances such as
hormones or neurotransmitter metabolites, in blood, saliva, urine, or spinal fluid, can be
used to implicate specific neurobiological systems in personality.
The expense of neuroimaging contributes to the largest methodological problem

in the field: low statistical power. Many studies are published with samples that
are far too small for good research on individual differences. A study of 461
structural MRI studies published between 2006 and 2009 found the median
power to be only 8% (Button et al., 2013; Ioannidis, 2011). Another study
reported, in a random sample of 241 neuroimaging papers published after 2007,
the median sample size was just 15 for one-group studies and 14.75 for each
group in two-group studies (Carp, 2012). This trend undoubtedly accounts for
some of the inconsistencies that exist in findings in personality neuroscience
(DeYoung, 2010b; Yarkoni, 2009, 2015). Fifteen is a small sample even for
studying many of the within-person effects that are most commonly researched
in neuroimaging, in which brain activity in one condition is compared to that in
another. Fifteen (or even 30) is ridiculously small for the study of individual or
group differences, and yet many MRI papers have reported correlations of
personality traits with neural variables in samples smaller than 20. Correlations
in small samples are highly susceptible to outliers and to sampling variability
more generally. Further, small sample sizes increase the likelihood that a given
sample will fail to represent variation across the full distribution of the trait of
interest, especially in the tails of the distribution (Mar, Spreng, & DeYoung,
2013). The likelihood of accurately assessing a correlation in a small sample is
very low (Schonbrodt & Perugini, 2013). Whenever possible, therefore, we
focus our review in this chapter on studies with larger sample sizes.
One method for increasing power in smaller samples is the use of extreme
groups, in which participants very high and very low on the trait of interest are
recruited based on a previous assessment of that trait. This is likely to yield a
larger effect size (the difference between high and low groups on the biological
variable of interest) than the correlation across the full range of the trait. This
tactic has pitfalls, however. First, the degree to which the expected effect size
increases is unpredictable, making power calculations difficult. Second, it
prevents any meaningful analysis of variables other than the trait used for
selection and may alter the effects of covariates in unpredictable ways. We
recommend an extreme-groups design only in cases in which a single, clear
hypothesis is being tested, funds are limited, and any covariates are handled at
the time of recruitment rather than in analysis. Important covariates, such as
gender and age, should be balanced when recruiting the extreme groups.
Crucially, something that should never be done is to analyze a subset of a larger
existing sample by identifying extreme groups within it and excluding the rest of
the participants from the analysis even though they have all relevant variables
assessed. Nor should a continuous variable ever be dichotomized (or
trichotomized) and analyzed as if it were a categorical variable. These strategies
entail an unacceptable loss of power compared to analyzing continuous variables
in the whole sample (MacCallum, Zhang, Preacher, & Rucker, 2002).
Chronically low power in personality neuroscience has a number of important
implications. Most obvious of these is increased Type II error rates—that is,
failures to detect real effects as significant. Two-thirds of the significant effects
reported in psychology are smaller than r = .3 (Hemphill, 2003), and there is no
reason to assume that effects in personality neuroscience should be larger. An
observed correlation of .3 will not be significant at p < .05 with a sample size
less than 40, and, with a sample of 40, the power to detect a true correlation of .3
is only about 50%, meaning that a Type II error would result about half the time,
as the observed correlation fluctuates due to sampling variability. Given that the
middle third of effect sizes in psychology is between r = .2 and .3 (Hemphill,
2003) and that the average effect size in personality research has been estimated
at .21 (Richard, Bond, & Stokes-Zoota, 2003), researchers should attempt to
ensure that they have the power to detect effects of at least r = .2. To have 80%
power to detect a correlation of .2 at p < .05 requires a sample of 194.
One promising strategy for acquiring sufficiently large samples in MRI is to
aggregate across many smaller studies of different tasks by including standard
structural scans or brief resting-state scans in each study. If a database of
subjects’ contact information is maintained, this method can be used to carry out
new MRI studies of individual differences without collecting additional MRI
data (Mar et al., 2013). Even with a large sample, however, the need to carry out
large numbers of statistical tests to examine the whole brain can lead to
problems with power. MRI studies typically divide the brain into a three-
dimensional grid of small “voxels” and often involve testing whether an effect is
present in thousands of individual voxels. In whole-brain analyses, researchers
sometimes choose a stringent threshold for significance at the voxel level (e.g., p
< .001) and then correct to p < .05 for the analysis as a whole based on the size
of clusters (adjacent significant voxels). This can lead to Type II errors because
the effect of interest may not be large enough to achieve significance at p < .001
in any voxel, even in a sample large enough to detect the same effect at a higher
p-value. We recommend setting a voxel-level threshold that will be capable of
detecting effects equivalent to r = .2 or larger, given one’s sample size (then
subsequently correcting to p < .05 for the whole analysis).
A less well-known but perhaps even more troubling result of low power is that
it increases the proportion of significant results that are Type I errors, false
positives (Green et al., 2008; Yarkoni, 2009, 2015). As the sample size
decreases, sampling variability increases and precision decreases. Even if the
true effect is zero, in small samples it is more likely to be sufficiently
misestimated as to appear significant. Testing effects in many small samples and
publishing only those that are large enough to achieve significance is a recipe for
the publication of many false positives, which then distort the literature and are
likely to mislead other researchers (Button et al., 2013). When the true effect is
not zero, low power still has the pernicious effect of artificially inflating
significant effect sizes, a problem that is exacerbated in MRI and other methods
that involve making many statistical tests in the same study. Estimates of the
effect will vary across voxels, and in small samples it is likely that only voxels
that greatly overestimate the effect will be significant (Yarkoni, 2009). This
leads not only to overestimated effect sizes, but also to the false impression that
effects are localized to very narrow regions of the brain, when the true effects
are likely to be much weaker but to be present in much broader swathes of brain
tissue (Yarkoni, 2015). The situation is made even worse when researchers
identify voxels of interest using a significance test (a threshold) with some
neural variable and then aggregate across those voxels before inappropriately
carrying out another, nonindependent significance test involving that variable
(Vul, Harris, Winkielman, & Pashler, 2009).
Beyond small samples, another potential cause of inconsistencies within the
neuroimaging literature is the wide variability in the methods that researchers
employ. Carp’s (2012) review of recent neuroimaging studies indicated that
nearly all (223 of 241) of the reviewed studies reported using different analytical
techniques. Even using different versions of the same software package for MRI
analysis or using the same version on different computers can lead to different
results (Gronenschild et al., 2012). The wide range of methods available may
contribute to the presence of excess significance bias within the neuroimaging
literature—and the psychological literature more generally (Ioannidis, 2011;
Jennings & Van Horn, 2012). One reason for the disproportionate number of
significant findings may be selective reporting bias, in which researchers try
multiple analytical methods and choose one that yields the most statistically
significant results, or those best matching their hypotheses, even when other
analytical methods may not support such a conclusion (Ioannidis, 2011). These
practices increase Type I error. Of course, the great variety of methods available
can lead to Type II errors as well, if methods are chosen that obscure effects of
interest (Henley et al., 2010).
A related issue is simply that some methods are better than others, but their
relative quality is not always clear or widely known. In the area of structural
MRI, for example, the most common method for assessing the relative volume
of different brain structures is voxel-based morphometry (VBM). In VBM,
structural brain images are spatially normalized (deformed) to match a template
brain, partitioned into gray and white matter, and smoothed so that each voxel
reflects the average percentage of gray matter within itself and the voxels
surrounding it (Ashburner & Friston, 2000). VBM has been criticized on several
grounds. First, it has been noted that if registration to the template were perfect,
there would be no individual differences for VBM to detect; thus, the method
relies problematically on imperfections in processing the data (Bookstein, 2001).
Further, because VBM relies on the density of gray matter in each voxel, it may
accurately detect differences in structure only near the gray–white matter
boundary and, even there, only when the differences are not expressed on an axis
parallel to the boundary (Bookstein, 2001; Davatzikos, 2004). Finally, VBM is
poor at detecting nonlinear differences in brain morphology, which are likely to
be common (Davatzikos, 2004). A better approach to structural MRI may be
deformation- or tensor-based morphometry (TBM), using the nonlinear portion
of the transformation that aligns each brain image to the template brain as the
index of relative local volume (e.g., DeYoung et al., 2010). Newer versions of
the VBM toolbox in the software program SPM integrate this TBM method as
an option under the label “modulation” (see
http://dbm.neuro.unijena.de/vbm/segmentation/modulation/), and we
recommend selecting modulation for nonlinear effects in any VBM study of
personality. The fact that many structural MRI studies of the FFM have used
VBM without modulation may account for some of their inconsistency.
Neuroimaging is not the only area of personality neuroscience in which low
power and inconsistent findings are problems. In molecular genetics, well-
replicated findings are rare. The first candidate gene studies of personality were
published 20 years ago (Benjamin et al., 1996; Ebstein et al., 1996), linking a
particular polymorphism of the dopamine D4 receptor gene (DRD4) to both
Extraversion and Novelty Seeking (a complex trait reflecting primarily low
Conscientiousness but also high Extraversion and potentially also low
Agreeableness and high Openness/Intellect; DeYoung & Gray, 2009). A later
meta-analysis of 36 studies found both effects to be nonsignificant, although a
different polymorphism in the same gene appeared to be associated with Novelty
Seeking but not Extraversion (Munafo, Yalcin, Willis-Owen, & Flint, 2008).
Such failures to replicate are typical of candidate gene studies, which is perhaps
not surprising given that well-powered GWAS studies in much larger samples
have also largely failed to identify genetic variants associated with the Big Five
(de Moor et al., 2012; Terracciano et al., 2008). These failures do not indicate a
lack of genetic influences on personality (the Big Five are substantially
heritable; Johnson & Krueger, 2004; Loehlin, McCrae, Costa, & John, 1998;
Riemann, Angleitner, & Strelau, 1997); rather, they are indicative of the fact that
complex traits are massively polygenic—that is, influenced by many thousands
of variations in the genome—with most having only a miniscule effect on the
trait in question (Munafo & Flint, 2011). Superficially, candidate gene studies of
personality may seem to have reasonably large sample sizes, often in the
hundreds, but these are probably often nowhere near large enough given the tiny
effects of interest. It seems likely that the situation with the FFM will resemble
that with schizophrenia: once sample sizes for GWAS exceeded 30,000, many
genes began to be robustly implicated (Need & Goldstein, 2014). Because
GWAS studies of the FFM are still not that large, the current review will largely
ignore molecular genetic findings and will usually provide caveats when they are
cited.
Theories of Psychological Function in the FFM

The FFM has long been criticized for being descriptive rather than
explanatory (e.g., Block, 1995). We would argue that the establishment of an
accurate descriptive model is not a flaw but rather a prerequisite for good
science. Nonetheless, having identified the major dimensions of personality, the
field must next strive to explain them. Personality neuroscience is aimed at
neurobiological explanations, but in order to develop neurobiological hypotheses
it is very helpful to begin with theories of the psychological functions underlying
each of the Big Five. Based on what is known about how different psychological
functions are carried out by the brain, it is possible to derive corresponding
neurobiological hypotheses.
Decades of behavioral and biological research on personality have led to the
development of a number of theories specifying the psychological functions
associated with each of the Big Five (Denissen & Penke, 2008; DeYoung,
2015a; MacDonald, 1995; Nettle, 2006, 2007; Van Egeren, 2009). These
theories come to very similar conclusions about each of the five dimensions, and
this level of agreement suggests that the available data point fairly clearly toward
some broad conclusions. For the purposes of this chapter, we will adopt the
perspective of the most thoroughly elaborated of these theories, Cybernetic Big
Five Theory (CB5T; DeYoung, 2015a).
Cybernetics is the study of goal-directed, self-regulating systems (Carver &
Scheier, 1998; Wiener, 1965). It is a useful and perhaps even necessary approach
for understanding living systems (Gray, 2004). CB5T defines personality traits
as “probabilistic descriptions of relatively stable patterns of emotion, motivation,
cognition, and behavior, in response to classes of stimuli that have been present
in human cultures over evolutionary time” and attributes the existence of traits to
variations in the parameters of evolved cybernetic mechanisms (DeYoung,
2015a). (Importantly, CB5T recognizes that these parameters are influenced by
both genetic and environmental forces; the substantial heritability of the Big
Five does not render them impervious to life experience.) The cybernetic
mechanisms that underlie traits allow people to identify goals, to be motivated to
attain goals, to select and carry out appropriate actions to move toward their
goals, to interpret feedback about the current state of the world (including the
organism itself), and to detect whether the current state matches their goal state.
CB5T adopts a MIMIC (multiple indicators, multiple causes) approach (cf.
Kievit et al., 2012), which posits that a shared psychological function causes
covariance among the specific traits (the multiple indicators) that are
encompassed by each of the Big Five, but that this psychological function is
instantiated by complex brain systems with many parameters (the multiple
causes) that vary to create individual differences in that function. In other words,
CB5T does not attempt to identify just a single biological parameter responsible
for a given trait because it recognizes that various biological mechanisms with
many parameters contribute to any given psychological function.
One advantage of CB5T over the other, similar theories cited above is that it
specifies mechanisms for traits at three levels of the personality hierarchy, not
just the Big Five (Figure 15.1 and Table 15.1). The fact that personality is
structured hierarchically means that the Big Five are not the only traits of
interest in personality psychology or neuroscience. They are merely the most
prominent major dimensions of covariation among more specific traits. The
variance of those more specific traits, below the Big Five in the hierarchy, is not
fully explained by the Big Five, in either phenotypic or genotypic analysis (Jang
et al., 1998, 2002). This means that, in addition to investigating mechanisms for
the Big Five, personality neuroscience should also investigate mechanisms that
differentiate specific traits within each of the Big Five domains.
Figure 15.1. A personality trait hierarchy based on the Five Factor Model. First (top)
level: metatraits. Second level: Big Five domains. Third level: aspects. Fourth level:
facets. The minus sign indicates that Neuroticism is negatively related to Stability.
Additionally, the Big Five themselves are not entirely independent; they show
relatively weak but consistent correlations with each other. Based on these
correlations, a considerable body of research demonstrates the existence of two
higher-order factors above the Big Five in the trait hierarchy, called metatraits
(DeYoung, 2006; Digman, 1997; McCrae et al., 2008). When modeled using
ratings from multiple informants, the correlation between the metatraits is near
zero, suggesting that there is no nonartifactual “general factor of personality”
above them (Chang, Connelly, & Geeza, 2012; DeYoung, 2006; Revelle & Wilt,
2013). CB5T includes hypotheses regarding the mechanisms associated with the
metatraits, as well as a level of traits below the Big Five, in addition to the Big
Five themselves.
The metatraits, Stability and Plasticity, are not given a separate section in this
chapter because most of the evidence for their biological basis comes from
studies of the Big Five considered individually, rather than in terms of their
shared variance, and these studies will be reviewed in the sections on each of the
Big Five. This evidence suggests that serotonin influences Stability and
dopamine influences Plasticity (DeYoung, 2006, 2010b, 2013). Serotonin
stabilizes information processing in many brain systems, helping to maintain
ongoing cybernetic function by facilitating both resistance to disruption by
impulses and focus on ongoing goals (Carver et al., 2008; Gray & McNaughton,
2000; Spoont, 1992). Stability represents the shared variance of
Conscientiousness, Agreeableness, and low Neuroticism. Each of these traits
reflects a different kind of stability: low Neuroticism reflects emotional stability,
Conscientiousness reflects motivational stability, and Agreeableness reflects
social stability (maintaining social harmony). Serotonergic neurons project from
the raphe nuclei in the brainstem to innervate most cortical and subcortical brain
structures, making serotonin well poised to influence the broad range of
personality traits implicated in Stability.
Dopamine facilitates exploration, approach, learning, and cognitive flexibility
in response to unexpected rewards and cues indicative of the possibility of
reward (Bromberg-Martin, Matsumoto, & Hikosaka, 2010; DeYoung, 2013).
Though not as widespread in the brain as serotonin, it nonetheless influences
most subcortical and frontal cortical structures. Plasticity represents the shared
variance of Extraversion and Openness/Intellect, and CB5T posits that it reflects
a general tendency toward exploration (DeYoung, 2013, 2015a). Whereas
Extraversion reflects behavioral exploration and sensitivity to specific rewards,
Openness/Intellect reflects cognitive exploration and sensitivity to the reward
value of information. The metatraits are important from a cybernetic perspective
because they represent variation in the prioritization of two of the broadest needs
of any cybernetic system that must survive in a complex and changing
environment: (1) to move toward goals consistently (Stability) and (2) to
generate new interpretations, strategies, and goals in order to adapt to the
environment (Plasticity) (DeYoung, 2006, 2015a).
Table 15.1. Psychological Functions Hypothesized to Be Associated with Each of the

Traits Labeled in Figure 15.1
Trait Cybernetic Function
Metatraits
Stability Protection of goals, interpretations, and strategies from disruption by impulses.
Plasticity Exploration: creation of new goals, interpretations, and strategies.
Big Five
Extraversion Behavioral exploration and engagement with specific rewards (i.e., goals to
approach).
Neuroticism Defensive responses to uncertainty, threat, and punishment.
Openness/Intellect Cognitive exploration and engagement with information.
Conscientiousness Protection of nonimmediate or abstract goals and strategies from disruption.
Agreeableness Altruism and cooperation; coordination of goals, interpretations, and strategies
with those of others.
Aspects
Assertiveness Incentive reward sensitivity: drive toward goals.
Enthusiasm Consummatory reward sensitivity: enjoyment of actual or imagined goal
attainment.
Volatility Active defense to avoid or eliminate threats.
Withdrawal (anxiety, Passive avoidance: inhibition of goals, interpretations, and strategies, in response
depression) to uncertainty or error.
Intellect Detection of logical or causal patterns in abstract and semantic information.
Openness to Experience Detection of spatial and temporal correlational patterns in sensory and perceptual
information.
Industriousness Prioritization of nonimmediate goals.
Orderliness Avoidance of entropy by following rules set by self or others.
Compassion Emotional attachment to and concern for others.
Politeness Suppression and avoidance of aggressive or norm-violating impulses and
strategies.
Adapted with permission from DeYoung (2015a).
The third level of traits labeled in Figure 15.1 is described as aspects of the
Big Five, whereas the unlabeled traits at the lowest level of the hierarchy are
known as facets (DeYoung, Quilty, & Peterson, 2007). No consensus exists
regarding the number and identity of facets within each Big Five dimension.
Although the NEO Personality Inventory-Revised (PI-R), a popular measure of
the FFM, identifies six facets for each, its 30 facets were derived rationally
through a review of the personality literature, rather than empirically (Costa &
McCrae, 1992), and other instruments assess different FFM facets (e.g.,
Goldberg, 1999). CB5T focuses on the aspect level of the trait hierarchy,
between the Big Five and their facets, because this level was empirically derived
and, thus, is likely to capture the most important distinctions within each of the
Big Five (DeYoung et al., 2007). This level of the trait hierarchy was first
detected in a behavioral genetic analysis of twins, in which two genetic factors
were needed to model the covariance of the six NEO PI-R facets in each domain
(Jang et al., 2002). If the Big Five were the next level of the hierarchy above the
facets, only a single genetic factor should have been necessary for each domain.
In a different sample, similar factors were subsequently found in nongenetic
factor analysis, using 15 facet scales for each domain, rather than six (DeYoung
et al., 2007). The resulting 10 factors were characterized empirically, based on
their correlations with over 2000 items from the International Personality Item
Pool (Goldberg, 1999), and a public-domain instrument, the Big Five Aspect
Scales (BFAS), was created to measure them (DeYoung et al., 2007). Whenever
possible in the following review, we distinguish between the two aspects in
terms of their neurobiological correlates.
Table 15.1 lists the cybernetic functions hypothesized by CB5T to be
associated with each of the labeled traits in Figure 15.1. An important caveat is
that even the functions associated with the aspects may themselves be broken
down into various interacting psychological mechanisms, each of which is likely
to be instantiated within the brain in different ways (Yarkoni, 2015). Some of
these mechanisms will be associated with specific facets, but even these are
likely to be further decomposable into multiple mechanisms. For example, the
passive avoidance mechanisms associated with the Anxiety facet of the
Withdrawal aspect of Neuroticism involve increased vigilance (attention to both
the external environment and information in memory), involuntary inhibition of
behavior, and increased arousal of the sympathetic nervous system, all of which
have distinct, identifiable neural circuits (Gray & McNaughton, 2000). Further,
specific mechanisms may be involved in multiple traits, so that the mapping of
traits to brain systems will be many-to-many (Yarkoni, 2015; Zuckerman, 2005).
Another caveat is that the hierarchy depicted in Figure 15.1 is oversimplified
in one important way: it depicts personality as having a simple hierarchical
structure, with no cross-loadings. If the diagram were entirely accurate as is,
traits beneath Stability could not be related to traits beneath Plasticity, but this is
not the case at the levels below the Big Five (Costa & McCrae, 1992; DeYoung,
2010b; Hofstee, de Raad, & Goldberg, 1992). For example, Politeness is
negatively related to Assertiveness, and Compassion is positively related to
Enthusiasm (DeYoung, Weisberg, Quilty, & Peterson, 2013). These cross-
connections are potentially important for biological models of personality. In
relation to the example just mentioned, testosterone may be at least partly
responsible for the covariation of Assertiveness and Politeness, given that it is
related to both of these dimensions (DeYoung et al., 2013; Turan, Guo,
Boggiano, & Bedgood, 2014).
The cybernetic perspective on the FFM has a number of advantages for
personality neuroscience. First, the hypothesized functions for each trait provide
a ready jumping-off point for hypotheses about brain function. Second, it
describes traits as the product of variation in a set of integrated mechanisms,
which is consistent with the fact that the brain is a single complex adaptive
system with many interacting subsystems. Considering the interactions among
these mechanisms may help to explain the relations among traits as well as their
manifestation in behavior. Third, by focusing on the psychological functions
underlying the Big Five, rather than just their superficial manifestation in
behavior and experience, we can more easily connect research on personality in
childhood and adulthood. All five factors appear to be present relatively early in
childhood, even though their exact manifestations in behavior shift with age
(Shiner & DeYoung, 2013). For example, a 4-year-old child high in
Openness/Intellect is unlikely to be interested in poetry or philosophy but is
nonetheless likely to express the tendency toward cognitive exploration through
curiosity and imaginative play. By using the FFM in developmental research,
personality neuroscience can shed light on the ontogeny of personality. Finally,
this perspective helps to link human research with the wealth of knowledge from
neuroscience research in other species, in which the brain can be observed and
manipulated more directly. The Big Five can be used to describe individual
differences in other species (Gosling & John, 1999), and, despite important
evolutionary change, much of the anatomy and cybernetic function of the brain
has been conserved by evolution, especially across mammalian species.
Extraversion
CB5T posits that sensitivity to reward is the core function underlying
Extraversion, enabling the individual to be energized by goals (DeYoung, 2013,
2015a). Here CB5T builds on the work of Depue and Collins (1999), who
argued that sensitivity to incentive reward mediated by the dopaminergic system
is the primary driver of Extraversion. Depue and Collins were themselves
influenced by Gray’s Reinforcement Sensitivity Theory, which posited a
Behavioral Approach System (BAS) that mediates the relation between
sensitivity to incentive reward and ensuing approach behavior (Gray, 1982; Gray
& McNaughton, 2000; Pickering & Gray, 1999). Although Gray initially
hypothesized that impulsivity was the personality trait most closely reflecting
BAS sensitivity, evidence has accumulated that Extraversion is a better
candidate, and the questionnaire most commonly used to measure BAS
sensitivity shows reasonable convergent validity with Extraversion (Carver &
White, 1994; Pickering, 2004; Quilty, DeYoung, Oakman, & Bagby, 2014;
Smillie, Pickering, & Jackson, 2006; Wacker, Mueller, Hennig, & Stemmler,
2012). All of these theories highlight the central role of the neurotransmitter
dopamine in the brain’s reward system. (Depue & Collins, 1999; DeYoung,
2013; Pickering & Gray, 1999; Smillie, 2008).
The association of variation in dopaminergic function with Extraversion is one
of the best established findings in personality neuroscience (see also the chapter
by Wilt and Revelle). A number of empirical studies have demonstrated that
Extraversion moderates the effects of pharmacological manipulation of the
dopaminergic system (Chavanon, Wacker, & Stemmler, 2013; Depue, Luciana,
Arbisi, Collins, & Leon, 1994; Mueller et al., 2014; Rammsayer, 1998;
Rammsayer, Netter, & Vogel, 1993; Wacker, Chavanon, & Stemmler, 2006;
Wacker, Mueller, Pizzagalli, Hennig, & Stemmler, 2013; Wacker & Stemmler,
2006). In a particularly impressive demonstration, a recent study by Depue and
Fu (2013) used Pavlovian conditioning in human participants to show that high
Extraversion was associated with greater sensitivity to the rewarding effects of
dopamine. To understand the meaning of this association, we must understand
the difference between incentive and consummatory reward (DeYoung, 2013).
An incentive reward is a cue that one is moving toward a goal, whereas a
consummatory reward is the actual attainment of a goal. Dopamine is
responsible for the drive to attain rewards in response to incentive cues but not
for the hedonic enjoyment of reward; this distinction has been described in terms
of “wanting” versus “liking” (Berridge, Robinson, & Aldridge, 2009). Whereas
the dopaminergic system is responsible for wanting, the opiate system is
responsible for liking (Peciña, Smith, & Berridge, 2006), and the association of
Extraversion with dopamine reflects only that Extraversion is linked to desire for
reward, not enjoyment of reward.
Nonetheless, questionnaire and behavioral research indicates that Extraversion
involves not only increased wanting, but also increased liking of rewards.
Positive emotionality is a facet of Extraversion describing energized positive
emotions such as excitement, enthusiasm, and elation that have a clear hedonic
component, and research indicates that Extraversion predicts the amount of these
positive emotions that people experience in response to incentively rewarding
stimuli (Smillie, Cooper, Wilt, & Revelle, 2012). This suggests that Extraversion
might be related to opiate function as well as to dopamine. CB5T posits that the
two aspects of Extraversion, Assertiveness and Enthusiasm, reflect the
difference between wanting and liking, with Assertiveness reflecting wanting
rather than liking and Enthusiasm reflecting primarily liking and only
secondarily wanting (DeYoung, 2015a). Enthusiasm appears to reflect liking in
an incentive context, with opiate release providing the positive hedonic feelings
that accompany dopaminergic activity (DeYoung, 2013). Research on dopamine
is consistent with this hypothesis, as measures of Assertiveness (usually called
“agentic Extraversion” in this literature) appear to be more strongly related to
dopaminergic variables than do measures of Enthusiasm (often called “affiliative
Extraversion”) (Mueller et al., 2014; Wacker et al., 2012). Further, one study
found that Social Closeness, a good marker of Enthusiasm, moderated the effects
of an opiate manipulation (Depue & Morrone-Strupinsky, 2005; DeYoung et al.,
2013). Whereas Assertiveness encompasses traits such as drive, leadership,
initiative, and activity, Enthusiasm encompasses both sociability or
gregariousness and positive emotionality (DeYoung et al., 2007).
In sum, existing research strongly supports the hypothesis that dopamine is an
important substrate of Extraversion, especially Assertiveness, and shows some
preliminary support for the hypothesis that the opiate system is also important
for Extraversion, particularly Enthusiasm. Note that the strong support for the
dopamine hypothesis leaves much unknown about the specific parameters of the
dopaminergic system that contribute to Extraversion (e.g., parameters related to
the density of different dopamine receptors, mechanisms of neurotransmitter
synthesis, or clearance from the synapse). This is indicative of the state of
personality neuroscience in general, in which even the best established findings
are merely preliminary to a thorough mechanistic understanding.
Electroencephalographic (EEG) research on a phenomenon known as the
“feedback-related negativity” (FRN) also supports the hypothesis that
Extraversion reflects dopaminergically driven sensitivity to incentive reward.
The FRN is an EEG waveform that appears 200–350 milliseconds after
receiving feedback about an outcome and appears to be generated by the dorsal
anterior cingulate cortex (ACC) in response to dopaminergic signaling of
deviations from the expected value of the outcome (Sambrook & Goslin, 2015).
Animal research has shown that one type of dopaminergic neuron encodes a
prediction error learning signal by spiking in response to better-than-expected
outcomes and dropping below baseline levels of activity in response to worse-
than-expected outcomes (Bromberg-Martin, Matsumoto, & Hikosaka, 2010).
The FRN shows the same pattern (becoming most negative for worse-than-
expected outcomes and least negative for better-than-expected outcomes),
indicating that it is a prediction error signal driven by dopamine (Proudfit, 2015;
Sambrook & Goslin, 2015). Several studies have shown that Extraversion
(sometimes measured with the BAS sensitivity scale) is correlated with FRN
amplitude following reward (Bress & Hajcak, 2013; Cooper, Duke, Pickering, &
Smillie, 2014; Lange, Leue, & Beauducel, 2012; Smillie, Cooper, & Pickering,
2011). Implicating dopamine more directly, Mueller et al. (2014) showed that
agentic Extraversion was associated with FRN magnitude following failure (i.e.,
a worse-than-expected outcome), but only when the task was incentivized, and
the association was eliminated by the administration of a dopamine D2 receptor
antagonist (a drug that blocks one type of dopamine receptor).
Turning to neuroimaging research, and considering the brain as a whole, the
most obvious hypothesis about Extraversion is that it should be associated with
function and structure in regions of the brain that are part of the reward system,
including the ventromedial prefrontal cortex (VMPFC; often called the
orbitofrontal cortex, OFC), the nucleus accumbens (often described as the
ventral striatum), the caudate nucleus (part of the dorsal striatum), the ACC, and
the midbrain regions from which dopaminergic neurons project (substantia nigra
and ventral tegmental area [SN/VTA]). That Extraversion should be associated
with amygdala function is another important hypothesis for neuroimaging,
stemming from the observation that the amygdala is crucial for processing
emotional salience related to rewarding as well as threatening stimuli (Stillman,
Van Bavel, & Cunningham, 2015).
Several fMRI studies have supported these hypotheses, showing that
Extraversion predicts neural activation in some or all of these structures in
response to emotionally positive or rewarding stimuli (Canli, Sivers, Whitfield,
Gotlib, & Gabrieli, 2002; Canli et al., 2001; Cohen, Young, Baek, Kessler, &
Ranganath, 2005; Mobbs, Hagan, Azim, Menon, & Reiss, 2005; Schaefer,
Knuth, & Rumpel, 2011). All of these studies, however, had samples smaller
than 20, rendering their evidentiary value questionable at best. Well-powered,
task-based, fMRI studies of the link between Extraversion and reward are
needed. A recent study with a sample of 52 is a step in the right direction,
showing that Extraversion predicted neural activity in the nucleus accumbens
during anticipation of gaining five dollars (Wu, Samanez-Larkin, Katovich, &
Knutson, 2014).
In contrast to the functional studies just mentioned, structural MRI studies
with larger sample sizes are beginning to appear, and the most replicated finding
for Extraversion is that it is associated positively with regional volume in
VMPFC, a brain area that appears to be crucial for maintaining representations
of the value of stimuli (Cremers et al., 2011; DeYoung et al., 2010; Omura,
Constable, & Canli, 2005). One of the largest such studies, which used the BAS
sensitivity scale rather than a more standard measure of Extraversion, found a
positive association with VMPFC in women but found a significant negative
association in men (Li et al., 2014). Other studies have not replicated the
association at all (Bjørnebekk et al., 2013; Hu et al., 2011; Kapogiannis et al.,
2013; Liu et al., 2013). Variation in the populations studied and methods
employed could be at least partly responsible for differing results. Further, all of
these studies reported whole brain analyses, rather than focusing on the VMPFC
as a region of interest. Whole brain analyses require corrections for multiple
tests that could have rendered even the larger studies underpowered to detect a
true association. Additional large primary studies, targeted hypothesis testing,
and meta-analyses will be needed to provide accurate estimates of this effect.
Associations of Extraversion with volume in other brain regions have been even
more inconsistent.
A recent PET study also provided some evidence of an association between
Extraversion and VMPFC, showing that Positive Emotionality (PEM), as
measured by the Multidimensional Personality Questionnaire (MPQ), was
positively associated with resting-state glucose metabolism in this region
(Volkow et al., 2011). MPQ-PEM is a broader construct than its label would
suggest, consisting of subscales measuring Social Potency and Social Closeness,
which are good measures of Extraversion, but also subscales measuring Well-
Being (Extraversion and Neuroticism) and Achievement (Assertiveness,
Conscientiousness, and Openness/Intellect) (DeYoung, 2013; DeYoung et al.,
2013; Markon, Krueger, & Watson, 2005). Although it is primarily a measure of
Extraversion, some caution is warranted about whether findings will generalize
to more traditional Extraversion measures. Another recent study that used this
measure and found a positive association between PEM and left amygdala
volume is worth mentioning here because of its sample size: N = 486 (Lewis et
al., 2014).
Resting EEG hemispheric asymmetry, in which one frontal lobe of the brain is
more active than the other, is another phenomenon that has been linked to
Extraversion and to the motivation to approach that is characteristic of response
to incentive reward. Considerable evidence suggests that the left hemisphere is
biased toward information processing associated with approach motivation and
behavior (Davidson, 1998; Harmon-Jones, Gable, & Peterson, 2010). For the left
hemisphere to be chronically more active than the right, therefore, might reflect
a general tendency toward approach that could be manifested in increased
Extraversion. Indeed, a number of studies have found that Extraversion, or more
specifically its Assertiveness aspect, is related to greater left-dominant
asymmetry (Amodio, Master, Yee, & Taylor, 2008; Coan & Allen, 2003; De
Pascalis, Cozzuto, Caprara, & Alessandri, 2013; Harmon-Jones & Allen, 1997;
Schmidt, 1999; Sutton & Davidson, 1997). However, failures to replicate have
been reported as well, and a meta-analysis found no evidence for the effect
(Wacker, Chavanon, & Stemmler, 2010).
Should the idea of linking Extraversion to hemispheric asymmetry be
abandoned, therefore? Perhaps not; a recent study of an all-male sample found
that the BAS sensitivity scale predicted resting-state asymmetry only for
participants interacting with a female experimenter whom they rated as attractive
(Wacker et al., 2013). Another much smaller EEG study found an analogous
effect; a trait measure of positive affect that is strongly linked to Extraversion
was associated with asymmetry only in a condition of positive mood as opposed
to negative or neutral mood (Coan, Allen, & McKnight, 2006). These studies
suggest that the association of Extraversion with hemispheric asymmetry may be
detectable only when positive emotional states related to incentive motivation
are activated. This possibility is consistent with many trait theories, including
CB5T, which posit that traits represent the tendency to respond in particular
ways to particular classes of stimuli. Without the presence of a relevant stimulus,
the trait may not be manifest, and individual differences in behavior or neural
activity may not be apparent.
Interestingly, one EEG effect measured during rest appears to be more
robustly associated with Extraversion than hemispheric asymmetry. Meta-
analysis has shown that agentic Extraversion is associated with increased
posterior versus anterior theta activity at centerline electrode sites (Koehler et al.,
2011; Wacker et al., 2010). (Frequency bands in EEG are labeled with the names
of Greek letters.) This finding has been extended to the delta frequency band as
well, and this theta/delta anterior–posterior difference appears to reflect activity
in the rostral ACC and to be associated with processing of reward and salience
information (Chavanon, Wacker, & Stemmler, 2011; Knyazev, 2010; Wacker &
Gatt, 2010; Wacker et al., 2010). The association of the anterior–posterior EEG
index with Extraversion has been linked empirically to dopaminergic function.
Several studies have shown that the association of Extraversion with increased
posterior–anterior difference is either negated or reversed when subjects are
administered a dopamine antagonist prior to the EEG recording (Chavanon et al.,
2013; Wacker et al., 2006), and a study combining EEG with molecular genetics
found that variation in the catechol-O-methyltransferase (COMT) gene (which
produces an enzyme that metabolizes dopamine in the synapse and varies in
efficiency depending on genotype) was associated with both agentic
Extraversion and posterior versus frontal resting delta/theta activity (Wacker &
Gatt, 2010).
Several fMRI studies with samples around N = 40–50 have reported
associations of Extraversion with resting-state functional connectivity. Their
results have not been very similar, but, then, neither have their methods: one
examined connectivity only between the amygdala and other brain regions
(Aghajani et al., 2014), one examined connectivity with nine seed regions on the
medial surface of the cortex (Adelstein et al., 2011), one examined connectivity
only within the default network (Sampaio, Soares, Coutinho, Sousa, &
Goncalves, 2014), and one examined connectivity of the midbrain dopaminergic
SN/VTA with other brain regions (Passamonti et al., 2015). With such
heterogeneous methods and small samples, it is hard to draw conclusions. The
most compelling Extraversion findings, from these studies, were that it was
positively associated with (1) connectivity between the amygdala and several
other regions involved in basic emotional and motivational processes (Aghajani
et al., 2014) and (2) connectivity between SN/VTA and the striatum, both key
components of the dopaminergic reward system (Passamonti et al., 2015).
Neuroticism
CB5T posits that Neuroticism reflects individual differences in the sensitivity
of defensive distress systems that become active in the face of threat,
punishment, and uncertainty (DeYoung, 2015a). Uncertainty is innately
threatening because the inability to predict the outcome of an action or
perception may indicate that one does not understand the current situation
sufficiently to be confident in the progress toward one’s goals—sometimes
including goals as fundamental as survival (Gray & McNaughton, 2000; Hirsh,
Mar, & Peterson, 2012; Peterson & Flanders, 2002). Indeed, one EEG study
found that for people high in Neuroticism, ambiguous feedback about task
performance produced a more negative FRN even than negative feedback
(whereas the opposite was true for people low in Neuroticism), consistent with
the theory that Neuroticism is associated with aversion to uncertainty (Hirsh &
Inzlicht, 2008).
Individuals high in Neuroticism are prone to emotional responses to stress that
foster avoidant or defensive behavior, including anxiety, depression, anger,
irritability, and panic (see also the chapter by Tackett and Lahey). Largely
because Neuroticism is the major personality risk factor for psychopathology
(Lahey, 2009), more neuroscientific research is being conducted on Neuroticism
than on any other trait in the FFM. To parse this research, CB5T draws on Gray
and McNaughton’s (2000) theory that Neuroticism reflects the joint sensitivity
of a behavioral inhibition system (BIS), which responds to threats in the form of
conflicts between goals (e.g., approach–avoidance conflict or any other conflict
that generates uncertainty), and a fight–flight–freeze system (FFFS), which
responds to threats without conflict—that is, when the only motivation is to
escape or eliminate the threat. Much is known about the neurobiology of the BIS
and FFFS in the brainstem, hypothalamus, and limbic system, which can aid in
the interpretation of existing research on Neuroticism and inform hypotheses in
future research.
CB5T posits that variations in the BIS and FFFS are likely to be reflected
differentially in the two aspects of Neuroticism. Withdrawal (related to BIS)
reflects the shared variance of traits related to anxiety and depression, which
involve passive avoidance, the tendency to slow or inhibit behavior to avoid
potential punishment or error. Volatility (related to FFFS) encompasses traits
related to irritability, anger, emotional lability, and the tendency to get upset
easily, which involve active defensive responses. In research on children, similar
factors have been described as anxious distress and irritable distress (Rothbart &
Bates, 1998; Shiner & Caspi, 2003). Neuroticism is often studied using scales
such as the BIS sensitivity scale (Carver & White, 1994), Cloninger’s Harm
Avoidance, and various measures of trait anxiety (most of which appear to
measure something broader than just the anxiety facet). Most such scales
measure either a combination of Withdrawal and Volatility or just Withdrawal.
To identify existing neuroscience research specifically relevant to Volatility
requires focusing on measures of anger or hostility as emotional traits (though
not actual aggression, which is more strongly related to Agreeableness than
Neuroticism).
The neurotransmitters serotonin and noradrenaline modulate both the BIS and
the FFFS and, therefore, are likely candidates as contributors to Neuroticism
(Gray & McNaughton, 2000). Several lines of evidence implicate serotonin in
Neuroticism. Serotonergic drugs are used to treat many disorders with symptoms
reflecting severe Neuroticism, including depression, anxiety and panic disorders,
and intermittent explosive disorder. In clinical depression, selective serotonin
reuptake inhibitors (SSRIs) have been shown to reduce Neuroticism, and this
reduction appears to mediate the improvements in depressive symptoms caused
by SSRIs (Du, Bakish, Ravindran, & Hrdina, 2002; Quilty, Meusel, & Bagby,
2008; Tang et al., 2009). A clinical trial has also shown that an SSRI can reduce
irritability and anger (Kamarck et al., 2009). Three PET studies have found that
Neuroticism predicts variation in serotonin receptor or transporter binding
(Frokjaer et al., 2008; Takano et al., 2007; Tauscher et al., 2001), although only
the most recent of these used a sample large enough to be of much interest. Two
studies have shown that response to a fenfluramine pharmacological challenge
(which assesses central serotonergic function) is associated with Neuroticism;
however, gender differences in the effect were apparent in both studies, and the
direction of effect was not consistent for men (Brummett, Boyle, Kuhn, Siegler,
& Williams, 2008; Manuck et al., 1998). Both studies were too small to assess
effects separately by gender with much confidence. Molecular genetic studies
implicating serotonergic genes in Neuroticism are inconclusive (Munafo et al.,
2009). A small body of research exists to suggest an association of noradrenaline
and Neuroticism, which may be more specific to fear and anxiety, and this
hypothesis could use more research (Hennig, 2004; White & Depue, 1999;
Zuckerman, 2005). Other understudied neurotransmitters involved in stress
responses may influence Neuroticism as well. One extensive study using a
variety of methods in neuroscience linked trait anxiety with a variation in levels
of neuropeptide Y, which is released under stress and modulates anxiety and
pain (Zhou et al., 2008).
Substantial evidence documents a link between Neuroticism and increased
activation of the hypothalamic–pituitary–adrenal (HPA) axis, which regulates
the body’s stress response under the control of both BIS and FFFS (Zobel et al.,
2004). Corticotropin-releasing hormone (CRH) is the proximal activator of the
HPA axis, and several studies of variation in the CRH receptor 1 gene have
linked it to depression or Neuroticism in individuals maltreated as children,
though results are complex and may differ by race and type of maltreatment
(Bradley et al., 2008; DeYoung, Cicchetti, & Rogosch, 2011; Grabe et al., 2010;
Kranzler et al., 2011; Polanczyk et al., 2009). A better established link is
between Neuroticism and levels of cortisol, the stress hormone released from the
adrenal cortex at the culmination of the stress response initiated by CRH.
Neuroticism is positively associated with baseline levels of cortisol (Garcia-
Banda et al., 2014; Gerritsen et al., 2009; Miller, Cohen, Rabin, Skoner, &
Doyle, 1999; Nater, Hoppmann, & Klumb, 2010; Polk et al., 2005) as well as
with blunted cortisol responses to specific stressors (Netter, 2004; Oswald et al.,
2006; Phillips, Carroll, Burns, & Drayson, 2005; but see Kirschbaum, Bartussek,
& Strasburger, 1992; Schommer, Kudielka, Hellhammer, & Kirschbaum, 1999,
for failures to replicate). This pattern suggests that people high in Neuroticism
tend to be not only chronically stressed but also less able to engage the resources
necessary to cope with specific stressful situations.
Interestingly, an overabundance of cortisol is known to potentiate excitotoxic
cell death in neurons (Sapolsky, 1994), a fact that Knutson, Momenan,
Rawlings, Fong, and Hommer (2001) suggested as a possible explanation for
their findings and those of others that Neuroticism is negatively related to global
measures of brain volume, such as the volume of cerebral gray matter, the ratio
of brain volume to intracranial volume, and total brain volume (Bjørnebekk et
al., 2013; Jackson, Balota, & Head, 2011; Liu et al., 2013). The chronic stress
associated with high Neuroticism may damage the brain as a whole.
The threat and punishment systems that control HPA activation are the
obvious neural candidates to underlie Neuroticism, and evidence from both
functional and structural MRI supports this broad hypothesis. Until recently,
most fMRI studies reporting that Neuroticism predicts neural responses to
aversive stimuli used samples so small as to preclude confidence in their results.
Of 21 samples in a recent meta-analysis of these effects (Servaas et al., 2013b),
only seven of them were larger than 25, and only one was larger than 60. Meta-
analysis cannot solve the problems created by underpowered samples because
meta-analytic conclusions are likely to be biased by their inclusion. One study
not included in this meta-analysis, with a sample of 52, found that Neuroticism
predicted right insula activation in anticipation of a loss of five dollars, and that
this insula activation showed trait-like stability over a period of 2.5 years (Wu et
al., 2014).
Many theoretical accounts of the neurobiology of Neuroticism highlight a role
for the amygdala, given its central role in BIS, FFFS, and mobilization of
negative affect and stress responses. Although the meta-analysis by Servaas et
al. (2013b) did not implicate the amygdala, some larger fMRI studies have found
associations between Neuroticism and amygdala response to aversive stimuli,
although methods have differed and the findings cannot be easily integrated. One
study reported that Neuroticism predicted a slower decrease in amygdala activity
after viewing aversive images (N = 120; Schuyler et al., 2014), and another
reported that Neuroticism was positively correlated with amygdala activity in
response to aversive images, but only in participants generally lacking in social
support (N = 103; Hyde, Gorka, Manuck, & Hariri, 2011). A region considered
part of the “extended amygdala,” known as the bed nucleus of the stria
terminalis (BNST), has been specifically linked to anxious vigilance, and its
activation to a persistent threat cue was predicted by Neuroticism (Somerville,
Whalen, & Kelley, 2010; N = 50).
Structural neuroimaging studies linking Neuroticism to amygdala volume
have been inconsistent, much like studies of Extraversion and VMPFC volume.
Several studies have found a positive correlation (Barros-Loscertales et al.,
2006; Iidaka et al., 2006; Koelsch, Skouras, & Jentschke, 2013), but several
others have not (Cherbuin et al., 2008; DeYoung et al., 2010; Fuentes et al.,
2012; Liu et al., 2013). Luckily, in this case, a nearly definitive study has been
carried out in a sample of over 1000 people that found that Neuroticism scores
based on the average of several commonly used questionnaire measures were
indeed correlated with amygdala volume (controlling for total brain volume),
albeit weakly (r = .1; Holmes et al., 2012). Only one other subcortical structure,
the hippocampus, was also significantly correlated with Neuroticism (r = .1),
which is salient both because the hippocampus is a core component of the BIS
and because resting-state hippocampal activity has previously been linked to
Neuroticism using PET (Gray & McNaughton, 2000; Sutin, Beason-Held,
Dotson, Resnick, & Costa, 2010).
Given the small effects detected by Holmes et al. (2012), previous
inconsistencies are likely to reflect a lack of statistical power. Another
possibility is that the amygdala effect is suppressed because it differs for
different subfactors of Neuroticism. One study found that a measure of trait
anger was associated negatively with left amygdala volume (Reuter, Weber,
Fiebach, Elger, & Montag, 2009). Although this study was small (N = 47) and,
therefore, may have misestimated the correlation of anger with amygdala
volume, it does raise the possibility that facets encompassed by Volatility might
show a different association with amygdala volume than those encompassed by
Withdrawal.
In addition to the volume of subcortical structures, Holmes et al. (2012) also
examined cortical thickness and found that Neuroticism was negatively
associated with the thickness of a region of left rostral ACC and adjacent medial
PFC (r = –.1). Interestingly, in a subset of 206 members of their sample who
completed additional questionnaire measures, Holmes et al. (2012) found that
the thickness of this region was correlated (r = –.2) with measures of social
dysfunction that appear to assess low Extraversion (perhaps blended with
Neuroticism). This finding represents a notable parallel to the findings described
above of positive correlations between Extraversion and nearby regions of the
VMPFC. Another study that examined cortical area as well as thickness found
that Neuroticism was associated negatively with cortical area in a very similar
region of ACC and medial PFC in the right hemisphere (Bjørnebekk et al.,
2013).
Given the size of the sample of Holmes et al. (2012), this is likely to be the
only region of the cortex in which thickness is associated with Neuroticism;
however, other types of structural measures may nonetheless implicate
additional cortical regions. Two studies of volume instead of thickness, with
samples over 100, have found that Neuroticism was negatively associated with
other regions of the PFC (DeYoung et al., 2010; Fuentes et al., 2012). Reduced
volume and thickness in the medial PFC may be linked to the low self-esteem
and poor regulation of emotion that are characteristic of Neuroticism, as this
region is part of the default network crucially involved in self-evaluation and
regulation of emotion (Andrews-Hanna et al., 2014). Three fMRI studies are
consistent with this hypothesis: Lemogne et al. (2011) found that Neuroticism
was associated with increased activation of both the medial PFC and the
posterior cingulate cortex and adjacent precuneus (another core hub of the
default network) when participants judged whether negative pictures were
related to themselves. Williams et al. (2006) found that Neuroticism predicted
age-related decreases in medial PFC responses to happy faces and increases in
responses in that region to fear faces. And Haas, Constable, and Canli (2008)
found that Neuroticism was associated with activity in medial PFC when
viewing blocks of sad facial expressions, but not fearful or happy facial
expressions (though in a small sample; N = 29).
The emotion regulation hypothesis is also consistent with a number of studies
of both functional and structural connectivity, which have found that
Neuroticism predicts reduced connectivity between frontal cortical regions and
the amygdala (sometimes in conjunction with increased connectivity of the
amygdala with other limbic regions). In functional studies, methods vary and
results are hard to integrate; larger samples would be helpful. Mujica-Parodi et
al. (2009) reported reduced synchrony between the amygdala and PFC regions
while viewing neutral, fearful, and happy faces. Servaas et al. (2013a) found that
Neuroticism was negatively correlated with the synchrony of amygdala and
hippocampus with dorsomedial and dorsolateral PFC during a scan preceded by
criticism from the experimenter (prerecorded to ensure standardization) relative
to a standard resting-state scan. In a more typical resting-state study by the same
group, Neuroticism was associated with weaker functional connections
throughout the brain, including connections in frontoparietal, sensory, and
default mode networks, but with stronger connectivity between affective regions,
including the amygdala, hippocampus, and insula (Servaas et al., 2015). This is
not entirely consistent with smaller resting-state studies that found that
Neuroticism was negatively associated with connectivity of the amygdala with
temporal lobe regions and the insula (Aghajani et al., 2014) and positively
associated with connectivity in the default network (between dorsomedial PFC
and the precuneus; Adelstein et al., 2011). Finally, a larger resting-state study (N
= 178) found that Neuroticism was positively associated with connectivity
between the amygdala and fusiform gyrus (a region crucial for visual processing
of faces), which may be related to the fact that Neuroticism is associated with
greater neural reactivity to negative facial expressions (Cremers et al., 2010).
Structural studies have found a more consistent pattern of reduced
connectivity associated with Neuroticism. Structural connectivity is measured in
MRI using diffusion tensor imaging (DTI) to assess the integrity of the white
matter (axon) tracts that connect different parts of the brain. Neuroticism is
associated with reductions in white matter integrity in tracts connecting cortical
and subcortical regions (Bjørnebekk et al., 2013; Taddei, Tettamanti, Zanoni,
Cappa, & Battaglia, 2012; Westlye, Bjørnebekk, Grydeland, Fjell, & Walhovd,
2011; Xu & Potenza, 2012).
Interestingly, although Holmes et al. (2012) did not examine structural or
functional connectivity, they did find that in individuals scoring highest in
Neuroticism (more than one standard deviation above the mean), cortical
thickness in the ACC and medial PFC region was negatively correlated with
amygdala volume (whereas they were unrelated in the rest of the sample). In
sum, the evidence suggests that Neuroticism is associated with an imbalance
between control of behavior and experience by subcortical negative emotional
systems versus frontal cortical systems.
Another consistent finding regarding Neuroticism comes from EEG research
demonstrating a pattern of greater activation in the right frontal lobe relative to
the left when viewing stimuli and while at rest (Gale, Edwards, Morris, Moore,
& Forrester, 2001; Shackman, McMenamin, Maxwell, Greischar, & Davidson,
2009; Sutton & Davidson, 1997), and this has been confirmed by meta-analysis
(Wacker et al., 2010). Similarly, near-infrared reflection spectroscopy (a
technique that uses light to measure regional cerebral oxygenated hemoglobin)
has shown that cerebral blood flow in the right frontal lobe is positively
correlated with Neuroticism during anticipation of a shock (Morinaga et al.,
2007). A lesion study, comparing 199 brain-damaged patients to 50 healthy
controls using MRI, found that focal damage to the left dorsolateral prefrontal
cortex was associated with higher scores on Neuroticism, especially the anxiety
facet (Forbes et al., 2014). Lesions of the left hemisphere lead to dominance of
right hemisphere function. Whereas most evidence suggests that the association
of Neuroticism with lateralization is driven by differences in frontal activation,
one large EEG study found a similar effect in posterior portions of the right
hemisphere (Schmidtke & Heller, 2004).
Importantly, not all components of Neuroticism show the same association
with hemispheric asymmetry. The right-dominant asymmetry appears to apply
only to traits in the Withdrawal subfactor, such as anxiety and depression, which
are linked to passive avoidance. In contrast, traits in the Volatility subfactor,
such as anger-proneness and hostility, which involve active defense, are
associated with greater left-dominant frontal asymmetry (Everhart, Demaree, &
Harrison, 2008; Harmon-Jones, 2004; Harmon-Jones & Allen, 1998).
Bearing in mind the caveat that different aspects of Neuroticism may show
different relations to hemispheric asymmetry, it is worth considering two non-
EEG studies that found that Neuroticism predicted hemispheric asymmetry in
connectivity. (Importantly, most global measures of Neuroticism—including
those used in these two studies—emphasize Withdrawal more than Volatility.)
Madsen et al. (2012) found that Neuroticism was associated with higher right,
relative to left, white matter integrity in the major white matter tract (the
cingulum) connecting limbic regions. Cremers et al. (2010) found that
Neuroticism predicted reduced synchrony between the left amygdala and medial
PFC when viewing negative versus neutral emotion faces, but increased
synchrony between these structures in the right hemisphere.
We conclude this section with a call for more studies that explicitly
distinguish between Withdrawal and Volatility. One otherwise exemplary study
unfortunately used a sample of only 18 (Cunningham et al., 2010), but its
innovative methodology is worth describing, in the hope of encouraging
replication attempts in larger samples. Participants in fMRI viewed positive,
negative, and neutral images and were required either to approach them (by
pressing a button that enlarged the image, creating the illusion of approach) or to
avoid them (by pressing a button that shrank the image). Withdrawal was found
to predict amygdala reactivity to approach relative to avoidance (independently
of stimulus valence), whereas Volatility was found to predict amygdala
reactivity to negative stimuli relative to neutral and positive stimuli
(independently of behavioral direction). These findings, if replicated, would
support the hypothesis that Withdrawal reflects sensitivity to conflict (especially
approach–avoidance conflict), thus leading to increased vigilance and behavioral
inhibition when approaching any stimulus, whereas Volatility reflects sensitivity
to all negatively valenced proximal stimuli.
Openness/Intellect
CB5T posits that Openness/Intellect reflects individual differences in the
cognitive exploration that generates new interpretations of experience in terms of
causal and correlational patterns and connections. Cognition here is conceived
broadly to include both reasoning and perceptual processes (DeYoung, 2015b).
People high in Openness/Intellect are imaginative, curious, innovative,
perceptive, thoughtful, and creative. The trait’s compound label stems from the
debate about whether to label it “Openness to Experience” or “Intellect” (Costa
& McCrae, 1992; Goldberg, 1990). This debate has been resolved by the
recognition that these two labels capture two major distinct subfactors of the
trait, with Intellect reflecting cognitive engagement with abstract information
and ideas (intellectual interests) and Openness reflecting cognitive engagement
with perceptual and sensory information (artistic and aesthetic interests)
(DeYoung et al., 2007; DeYoung, Grazioplene, & Peterson, 2012; Johnson,
1994; Saucier, 1992). When we refer to “Openness/Intellect,” we are referring to
the broad FFM dimension; when we refer to either “Intellect” or “Openness”
alone, we are referring to just one aspect of Openness/Intellect (see also the
chapter by Sutin).
The curiosity and innovation that are common to both Openness and Intellect
are likely to be driven by dopamine—specifically, a type of dopaminergic
neuron that codes for salience instead of value, is activated by both positive and
negative information, and innervates different brain regions than do the value-
coding neurons implicated in Extraversion (Bromberg-Martin et al., 2010;
DeYoung, 2013). The evidence for dopaminergic involvement in
Openness/Intellect is more circumstantial than the evidence for Extraversion,
although there have been two molecular genetic studies showing associations
with the DRD4 and COMT genes in three samples (DeYoung, Cicchetti,
Rogosch, Gray, & Grigorenko, 2011; Harris et al., 2005). The adult sample
investigated by DeYoung et al. (2011) exhibited an interaction effect between
DRD4 and COMT, which, if replicated, could explain the failure of these genes
to be identified in larger GWAS studies of the FFM.
The original hypothesis that dopamine is involved in the biological substrate
of Openness/Intellect was based on several lines of indirect evidence (DeYoung,
Peterson, & Higgins, 2002, 2005): (1) the involvement of dopamine in curiosity
and exploratory behavior is well-established in animal research (Panksepp,
1998); (2) dopamine is involved in the working-memory attentional mechanisms
that allow maintenance and manipulation of information in short-term memory,
and Openness/Intellect (specifically its Intellect aspect) is the only FFM trait
positively associated with working memory ability (DeYoung et al., 2005,
2009); and (3) Openness/Intellect is associated with reduced latent inhibition, an
automatic preconscious process that blocks stimuli previously categorized as
irrelevant from entering awareness (Peterson & Carson, 2000; Peterson, Smith,
& Carson, 2002). Dopamine is the primary neuromodulator of latent inhibition,
with increased dopaminergic activity producing reduced latent inhibition
(Kumari et al., 1999), and Openness/Intellect may reflect individual differences
in the automatic tendency to perceive salient information in everyday
experience.
One fMRI study tested hypotheses derived explicitly from the dopamine
theory of Openness/Intellect. Although dopaminergic activity cannot be studied
directly in fMRI, neural activity can be assessed in regions that are core to the
dopaminergic system, with the inference that activation there is probably
reflective of dopaminergic function (much like the FRN in EEG). Passamonti et
al. (2015) examined functional connectivity between the midbrain SN/VTA,
where the dopaminergic system originates, and other brain regions, not only
during resting state but also in two tasks involving sensory experience. In the
first, participants were presented with pleasant food odors through a special
apparatus, contrasted with smelling pure air. In the second, participants viewed
appealing pictures of food, contrasted with viewing a fixation cross. In all three
tasks, Openness/Intellect positively predicted connectivity of SN/VTA with
dorsolateral PFC, a region crucial for voluntary control of attention and working
memory. This circuit may help to explain why people high in Openness/Intellect
find sensory experiences interesting and rewarding.
The association of Intellect with working memory has been demonstrated
neurally as well as behaviorally. An fMRI study using the Ideas facet of the
NEO PI-R as a measure of Intellect found that it was the only facet associated
with brain activity predicting accurate working memory performance in the
scanner (DeYoung et al., 2009). Associations were found in two regions of the
PFC, the left frontal pole of the lateral PFC and a posterior region of the medial
PFC. The frontal pole is crucial for integrating the outputs of various simpler
cognitive operations and for making abstract analogies (Gilbert et al., 2006;
Green, Fugelsang, Kraemer, Shamosh, & Dunbar, 2006; Ramnani & Owen,
2004). The medial PFC region in question is known to be involved in monitoring
goal-directed performance, which might be particularly important for those high
in Intellect, who are motivated to do well in cognitive tasks (Brown & Braver,
2005; Ridderinkhof, Ullsperger, Crone, & Nieuwenhuis, 2004). A PET study,
which did not separate Intellect from Openness, found that Openness/Intellect
was associated with neural activity while participants were at rest, in brain areas
not identical to but near the two areas just described, in regions of lateral PFC
and anterior cingulate cortex associated with working memory and error
detection (Sutin, Beason-Held, Resnick, & Costa, 2009).
Given the centrality of imagination for Openness/Intellect (“Imagination” was
even suggested as an alternative label for the whole dimension; Saucier, 1992),
we might expect that the default network would be an important substrate of the
trait, especially the Openness aspect, which encompasses fantasy-proneness as
one of its facets (DeYoung, 2015b). Two relatively small functional connectivity
studies offer some tentative preliminary support for this hypothesis. One found
that Openness/Intellect was associated with increased connectivity between the
main midline hubs of the default network, in medial PFC and precuneus
(Adelstein et al., 2011), whereas the other found that Openness/Intellect was
associated with connectivity in more parietal components of the default network
instead (Sampaio et al., 2014).
Studies of the association of Openness/Intellect with the volume of regions
throughout the brain have been inconsistent, often finding no significant effects
despite samples larger than 100 (Bjørnebekk et al., 2013; DeYoung et al., 2010;
Hu et al., 2011; Kapogiannis et al., 2013; Li et al., 2014; Liu et al., 2013). An
MRI study of change in brain structure in 274 adults (M = 51, SD = 12 years)
over a period of 6–9 years found that Openness/Intellect was negatively
correlated with an age-related decline in gray matter volume in the right inferior
parietal lobule, a region linked to intelligence and creativity (Taki et al., 2013).
The volume of this area was previously found to be associated positively with
Openness/Intellect, though in a region too small to be significant after correction
for multiple tests (DeYoung et al., 2010). Clearly, this area would be a sensible
region of interest for future research.
Two DTI studies have found apparently contradictory findings for
Openness/Intellect, which may be reconcilable through consideration of the
differences between Openness and Intellect in their associations with IQ and
positive schizotypy or psychoticism (comprising magical ideation and perceptual
aberrations). The first study found a negative association between
Openness/Intellect and white matter integrity in the frontal lobes (Jung,
Grazioplene, Caprihan, Chavez, & Haier, 2010), whereas the second study found
a positive association (Xu & Potenza, 2012). The major difference between the
two studies appears to be that the first controlled for IQ whereas the second did
not. Importantly, frontal white matter integrity is positively associated with IQ
but negatively related to psychoticism (Chiang et al., 2009; Nelson et al., 2011).
Intellect is independently associated with IQ, whereas Openness is not
(DeYoung, Quilty, Peterson, & Gray, 2014), so controlling for IQ should render
the residual Openness/Intellect scores closer to Openness. Further, Openness is
positively related to psychoticism, whereas Intellect is negatively related to it
(Chmielewski et al., 2014; DeYoung et al., 2012). In combination, these pieces
of evidence suggest that Openness and Intellect might be differentially related to
frontal white matter integrity, and future research should measure them
separately.
We close this section by noting the possibility that serotonin may play some
role in Openness/Intellect. A PET study of 50 people (Kalbitzer et al., 2009)
found that Openness/Intellect predicted serotonin transporter binding in the
midbrain (whereas Neuroticism did not). In a sample that small, this finding
might simply be a false positive. However, the involvement of serotonin in
Openness/Intellect is rendered more plausible by the fact that most
hallucinogenic drugs act directly on the serotonergic system. A longitudinal
study of 52 hallucinogen-naive adults who received doses of psilocybin (the
active serotonergic agent in hallucinogenic mushrooms) or an active placebo
(methylphenidate) found that participants showed increases in
Openness/Intellect following psilocybin but not placebo (MacLean, Johnson, &
Griffiths, 2011). Even more dramatically, Openness/Intellect remained elevated
over a year later for the 30 participants who had had mystical experiences while
on psilocybin. No other FFM traits were affected. Of course, it is possible that
dramatic disruptions of the serotonergic system by hallucinogens might
influence Openness/Intellect even if normal variation in that system does not.
Nonetheless, people high in Openness (especially when also low in Intellect)
appear to be susceptible to cognitive and perceptual distortions of the kind that
are greatly exaggerated in hallucination (i.e., to psychoticism), and these might
be associated with reduced serotonergic function (Chmielewski et al., 2014;
DeYoung et al., 2012).
Conscientiousness
CB5T posits that the function of Conscientiousness is to facilitate the pursuit
of nonimmediate goals and rule-based behavior (DeYoung, 2015a). This
function is critical to the successful navigation of human culture, and, indeed,
Conscientiousness is typically the best psychological predictor, after
intelligence, of academic and occupational success, as well as health-promoting
behaviors and longevity (Ozer & Benet-Martinez, 2006; Roberts, Lejuez,
Krueger, Richards, & Hill, 2014). The two aspects of Conscientiousness are
Industriousness, reflecting the ability and tendency to suppress disruptive
impulses and persist in working toward nonimmediate goals, and Orderliness,
which involves a tendency to adopt and follow rules, whether these rules are
self-generated or imposed by others (DeYoung et al., 2007; see also the chapter
by Jackson and Roberts).
The low pole of the Conscientiousness dimension is often described as
“impulsivity,” but impulsivity is a complex construct, and multiple types of
impulsivity can be identified, not all of which are equivalent to low
Conscientiousness (DeYoung, 2010a). The UPPS model (Whiteside & Lynam,
2001) identifies four types of impulsivity, of which lack of Perseverance is the
most clearly related to Conscientiousness, being essentially equivalent to low
Industriousness. Lack of Premeditation, the tendency to act quickly without
deliberation, is also clearly linked to Conscientiousness, but it appears to be a
blend of low Conscientiousness and high Extraversion and may therefore have
somewhat different biological substrates than other traits in the
Conscientiousness domain. For example, one fMRI study found that reward-
related activity in the ventral striatum was positively associated with scores on
the Barratt Impulsivity Scale, a commonly used measure that corresponds most
closely to lack of Premeditation (Forbes et al., 2009; Whiteside & Lynam, 2001).
This finding seems likely to have been driven by reward-related variance linked
to Extraversion. The other two types of impulsivity in the UPPS system are
Urgency, which reflects the broader Stability metatrait, and Sensation Seeking,
most closely linked to Extraversion (DeYoung, 2010a).
Humans are highly unusual in their ability to follow explicit systems of rules
and plan for the distant future, so it is perhaps not surprising that chimpanzees
are the only other species in which a trait analogous to Conscientiousness has
been identified (Freeman & Gosling, 2010; Gosling & John, 1999). Other
species obviously need to inhibit disruptive impulses, but individual differences
in impulse control may simply be reflected in dimensions analogous to
Neuroticism and Agreeableness that are related to more immediate goals and are
influenced by serotonin. As noted above, CB5T hypothesizes that the variance
Conscientiousness shares with Neuroticism and Agreeableness is linked to
serotonin. A fenfluramine challenge study found that Conscientiousness was
positively associated with central serotonergic function in men (Manuck et al.,
1998). Another study failed to replicate this effect, but its sample was only half
as large (Brummett et al., 2008). In a study of 75 men, Manuck, Flory, Ferrell,
Mann, and Muldoon (2000) used a fenfluramine challenge to show that central
serotonergic function was negatively associated with a combined measure of
Hostility, Aggression, and lack of Premeditation (the latter assessed by the
Barratt Impulsivity Scale), a composite that is probably a good indicator of low
Stability. Serotonin remains a plausible component of the substrate of
Conscientiousness, but more research is needed.
Considerable evidence exists to implicate the PFC in Conscientiousness,
which is sensible given the central role of PFC in following rules and
maintaining goal representations (Bunge & Zelazo, 2006; Miller & Cohen,
2001). The PFC is the brain region most expanded in human evolution (Deacon,
1997; Hill et al., 2010), so this association is consistent with the fact that only
humans and their closest evolutionary relatives appear to have a distinct trait of
Conscientiousness. Multiple MRI studies have found that Conscientiousness was
positively associated with the volume of regions in the dorsolateral PFC
(DeYoung et al., 2010; Jackson et al., 2011; Kapogiannis et al., 2013), though
other studies have not replicated these findings (Bjørnebekk et al., 2013; Hu et
al., 2011; Liu et al., 2013). An MRI study comparing 199 brain-damaged
patients to 50 healthy controls found that focal damage to the left dorsolateral
prefrontal cortex was associated with lower scores on Conscientiousness,
especially the self-discipline facet, which is a marker of Industriousness (Forbes
et al., 2014).
The association of Conscientiousness with dorsolateral PFC raises an
interesting question about the differentiation of Conscientiousness from other
traits that have been linked to dorsolateral PFC, particularly Intellect,
intelligence, and working memory capacity. The latter three traits are all related
and can be grouped together in the Intellect dimension (DeYoung, 2015b;
DeYoung et al., 2009, 2012), whereas Conscientiousness is not related to either
intelligence or working memory (except for a possible weak negative correlation
with intelligence; DeYoung, 2011; DeYoung et al., 2014). We propose that
Intellect and Conscientiousness may reflect variations in two different large-
scale neural networks, both of which involve dorsolateral PFC.
Functional connectivity maps have identified two strongly interdigitated
networks in the lateral PFC, anterior insula, putamen, ACC and adjacent medial
PFC, lateral parietal cortex, and posterior temporal cortex (Choi et al., 2012; Yeo
et al., 2011). The first, known as the frontoparietal or cognitive control network,
is the major substrate of working memory and intelligence, and parts of it have
been associated with both Openness/Intellect in general and Intellect in
particular (DeYoung et al., 2009, 2010; Taki et al., 2013). The second, known as
the ventral attention or salience network, is a good candidate as a substrate of
Conscientiousness (DeYoung, 2015a). Its broad function appears to entail
reorienting attention away from distractions and toward stimuli important for
goal pursuit (Fox, Corbetta, Snyder, Vincent, & Raichle, 2006). It is often called
“ventral” due to research focusing on two important nodes of the network, in the
right inferior frontal gyrus and the temperoparietal junction, but it nonetheless
incorporates regions of the dorsal PFC as well, including the region of the
middle frontal gyrus where Conscientiousness has been found to correlate
positively with volume (DeYoung et al., 2010; Kapogiannis et al., 2013; Yeo et
al., 2011). Not only that, but other regions in which Conscientiousness has been
linked to brain structure and function fall within this network, as we will now
review.
Several studies have linked Conscientiousness or the Barratt Impulsivity Scale
to variations in the anterior insula (in what follows, we describe the impulsivity
findings in terms of “Premeditation,” so that they are keyed in the same direction
as Conscientiousness). One structural MRI study found that Conscientiousness
was negatively associated with white matter volume in the insula and adjacent
putamen, caudate, and ACC (Liu et al., 2013), and another found that the cortical
thickness of the anterior insula was negatively correlated with Premeditation
(Churchwell & Yurgelun-Todd, 2013). In an fMRI study of response inhibition,
Premeditation was positively associated with activation of the anterior insula and
lateral frontal cortex on trials when inhibition was required. It was also
associated during those trials with greater functional connectivity of the right
anterior insula with regions of the PFC and visual cortex (Farr et al., 2012).
Several MRI studies have implicated the dorsal ACC and adjacent medial
PFC in Conscientiousness. One structural study found that Premeditation was
negatively related to volume in the left ACC (Matsuo et al., 2009; this study also
found positive associations with VMPFC volumes). Another found that a
measure of Conscientiousness in adolescents (Effortful Control) predicted a
leftward asymmetry in dorsal ACC anatomy (Whittle et al., 2009). In an fMRI
study of response inhibition, Premeditation was negatively associated with
activity in the dorsal ACC and caudate (Brown, Manuck, Flory, & Hariri, 2006).
A resting-state fMRI study found that Conscientiousness was associated with
functional connectivity in the ACC and adjacent medial PFC (Adelstein et al.,
2011).
The overall pattern that emerges suggests that Conscientiousness is associated
with greater volume in the lateral PFC but with reduced volume in other areas of
the ventral attention network. This suggests the hypothesis that
Conscientiousness depends in part on the balance between the portions of this
network that generate signals of motivational salience and those that engage in
attentional and behavioral control in response to those signals. This hypothesis is
also reasonably consistent with the fMRI finding, mentioned above, that
Premeditation predicted greater connectivity of the insula with the lateral PFC
when response inhibition was required than when it was not (Farr et al., 2012).
Some caution is needed moving forward, however, because Premeditation is a
fairly peripheral Conscientiousness facet, not strongly linked to either
Industriousness or Orderliness (DeYoung, 2010a), so findings may not
generalize easily to the broader Conscientiousness dimension.
We close our discussion of Conscientiousness by noting one brain region that
has been associated with Conscientiousness in multiple studies but has not been
identified as part of the ventral attention network—namely, the fusiform gyrus.
In one large structural MRI study, Conscientiousness was negatively correlated
with white matter volume in the left fusiform gyrus (Liu et al., 2013). In another,
which did not separate gray and white matter, Conscientiousness was also
negatively associated with volume in the fusiform gyrus (DeYoung et al., 2010).
Many studies of brain structure consider gray matter volume only, and future
studies may benefit from considering both gray and white matter. Finally, a
study of personality and neurological change in frontotemporal dementia found
that declines in Conscientiousness were associated with relative preservation of
gray matter in the fusiform gyrus (Mahoney, Rohrer, Omar, Rossor, & Warren,
2011); this study was quite small (N = 30), but we mention it because of the
interesting parallel with structural studies of healthy adults.
Agreeableness
CB5T posits that cooperation and altruism—that is, the processes of
coordinating our own goals with those of others—are the core functions
underlying Agreeableness (see also the chapter by Graziano and Tobin). This
entails that Agreeableness should be associated with the ability and tendency to
understand the perspectives of others and to adjust our own behavior to
accommodate them (Nettle & Liddle, 2008). The most obvious candidates as a
neural substrate for Agreeableness are the many parts of the default network that
are involved in decoding the mental states of others (Andrews-Hanna et al.,
2014). Two resting-state fMRI studies have reported that Agreeableness is
positively associated with functional connectivity among major hubs of the
default network (Adelstein et al., 2011; Sampaio et al., 2014).
Two reasonably large structural MRI studies have found no association of
regional brain volumes with Agreeableness (Bjørnebekk et al., 2013; Liu et al.,
2013), and others have found associations that were not consistent (DeYoung et
al., 2010; Hu et al., 2011; Kapogiannis et al., 2013). Two of the latter studies
reported a negative correlation of Agreeableness with a region of the posterior
superior temporal gyrus and sulcus that is part of the default network and is
important for interpreting the actions and intentions of others by decoding
biological motion, but one study found the effect in the left hemisphere and one
in the right (DeYoung et al., 2010; Kapogiannis et al., 2013). Clearly, further
research is necessary on this brain region’s relation to Agreeableness.
The two aspects of Agreeableness are Compassion, reflecting empathy and
sympathy (the tendency to care about others emotionally), and Politeness, the
tendency to conform to social norms and to refrain from belligerence and
exploitation of others. In surveying the relatively sparse neuroscience research
on Agreeableness, it is important to note that measures of empathy reflect
Compassion, whereas measures of aggression reflect low Politeness (DeYoung
et al., 2007, 2013). Compassion scales include the Empathic Concern subscale
(and potentially the Perspective Taking subscale) of the Interpersonal Reactivity
Index (IRI; Davis, 1983), the Balanced Emotional Empathy Scale (Mehrabian &
Epstein, 1972), and the Empathy Quotient (Baron-Cohen & Wheelwright, 2004).
MRI research suggests two general types of neural processes involved in
empathy. The first involves the default network and the ability to simulate the
mental states of others. The second involves what can be called “mirroring”—
neural activation that occurs, while observing someone else, in the same sensory
networks that would be active if the observer were having an experience similar
to that of the observed person. The most studied form of empathy in fMRI is
empathy for pain, and here regions of the anterior insula (involved in integrating
emotional and sensory information with cognitive processes) and the mid-
cingulate cortex appear to constitute the circuit that is active in mirroring (i.e.,
they are active for both one’s own pain and the pain of others), whereas default
network regions are involved in recruiting those pain-related regions by
decoding the experience of others (Lamm, Decety, & Singer, 2011). A number
of fMRI studies of empathy for pain have reported an association between trait
levels of empathy and neural responses, with inconsistent results. As with many
traits, however, most of these studies have been too small to detect individual
differences adequately. In a recent meta-analysis, for example, none of the 15
studies that examined trait effects had a sample larger than 30 (Lamm et al.,
2011, Appendix B).
Social or emotional pain has been found to activate brain systems similar to
physical pain, and one larger fMRI study found that trait empathy predicted
greater functional connectivity of the anterior insula with the PFC and limbic
regions while watching videos of the suffering of others (Bernhardt, Klimecki,
Leiberg, & Singer, 2014). (The default network, like the ventral attention and
frontoparietal networks, includes regions of anterior insula; Yeo et al., 2011.)
Two structural MRI studies found empathy to be positively associated with
regional volume in the anterior insula (Mutschler, Reinbold, Wankerl, Seifritz, &
Ball, 2013; Sassa et al., 2012), but one found no association (Takeuchi et al.,
2014). Another study, with a sample of 118, found a negative correlation of
empathy with anterior insula volume; however, this study used all four subscales
of the IRI as simultaneous predictors, and the process of residualization may
have shifted the meaning of the Empathic Concern subscale (Banissy, Kanai,
Walsh, & Rees, 2012). We would not recommend partialling out shared variance
from the IRI subscales without a clear theoretical justification. One DTI study
found that empathy was widely positively correlated with white matter integrity
in tracts connecting affective, perceptual, and action-oriented brain regions,
which is potentially consistent with the sophisticated integration of different
types of information necessary for both understanding and sharing the emotional
experience of others (Parkinson & Wheatley, 2014).
Agreeableness in general, and Politeness specifically, are likely to be
associated with emotion regulation. Agreeableness predicts suppression of
aggressive impulses and other socially disruptive emotions (Meier, Robinson, &
Wilkowski, 2006), and one fairly small fMRI study found that Agreeableness
predicted greater right lateral PFC activation in response to fearful compared to
neutral faces (Haas, Omura, Constable, & Canli, 2007), which the authors
argued might reflect automatic engagement of emotion regulation when facing
stimuli signaling potential threat or conflict. In a structural MRI of 56 men
drawn from a larger cohort studied since childhood, amygdala volume at age 26
was negatively associated with both current aggression and a history of
aggression (Pardini, Raine, Erickson, & Loeber, 2014).
Inasmuch as Agreeableness involves the ability to suppress aggressive
impulses, it is likely to be facilitated by serotonin (Montoya, Terberg, Bos, &
Van Honk, 2012). An interview-based life history of aggression measure was
negatively associated with serotonin function in men but not women (Manuck et
al., 1998), whereas a 2-month trial on an SSRI significantly reduced aggression
in women but not men (Kamarck et al., 2009). One twin study found that
variation in the serotonin transporter gene accounted for 10% of the genetic
correlation between Neuroticism and Agreeableness (Jang et al., 2001).
Other neurotransmitters likely to be involved in Agreeableness include
testosterone and oxytocin. Testosterone levels appear to be negatively associated
with Agreeableness, particularly Politeness versus Aggression (DeYoung et al.,
2013; Montoya et al., 2012; Turan et al., 2014). Oxytocin is critically involved in
processes of social bonding and attachment. Trait empathy has been found to
moderate the effects of acute oxytocin administration (Perry, Mankuta, &
Shamay-Tsoory, 2015). Difficulties in the assessment of oxytocin levels suggest
the need for caution in research on their association with personality
(Christensen, Shiyanov, Estepp, & Schlager, 2014).
Future Directions
Much new personality neuroscience research has appeared in recent years, as
is evident when comparing this chapter with previous reviews of the field
(DeYoung, 2010b; DeYoung & Gray, 2009; Zuckerman, 2005). Further,
personality neuroscience research is improving in quality, allowing this review
to be reasonably critical and to focus on larger studies. Still, because personality
neuroscience is such a young field, its future is wide open. Very few findings
about the neurobiological sources of the FFM are sufficiently well-supported to
have the status of fact. Every trait needs much additional research before we
begin to have anything like a clear picture of the many biological parameters that
account for its variation.
We have two major recommendations for those interested in pursuing
personality neuroscience. First, work with existing or new theories in order to
develop specific testable hypotheses, rather than pursuing purely exploratory
research. Readers should be able to glean from this chapter many hypotheses
that can be tested by future research. In theory-driven research, it will often be
advantageous to test associations with regions of interest in the brain specified a
priori. Second, collect samples large enough for good research on individual
differences—near 100 at a minimum, preferably over 200. We believe that
existing theories of the psychological functions underlying the FFM, such as
CB5T, are sufficiently well developed to allow rapid advancement of our
understanding of the biological basis of traits, as long as rigorous methods are
employed.
References
Adelstein, J. S., Shehzad, Z., Mennes, M., DeYoung, C. G., Zuo, X.-N., Kelly, C., …
Milham, M. P. (2011). Personality is reflected in the brain’s intrinsic functional
architecture. PloS One, 6(11), e27633.
Aghajani, M., Veer, I. M., van Tol, M.-J., Aleman, A., van Buchem, M. A., Veltman, D.
J., … van der Wee, N. J. (2014). Neuroticism and extraversion are associated with
amygdala resting-state functional connectivity. Cognitive, Affective, & Behavioral
Neuroscience, 14(2), 836–848.
Amodio, D. M., Master, S. L., Yee, C. M., & Taylor, S. E. (2008). Neurocognitive
components of the behavioral inhibition and activation systems: Implications for
theories of self-regulation. Psychophysiology, 45(1), 11–19.
Andrews-Hanna, J. R., Smallwood, J., & Spreng, R. N. (2014). The default network and
self-generated thought: Component processes, dynamic control, and clinical relevance.
Annals of the New York Academy of Sciences, 1316, 29–52.
Ashburner, J., & Friston, K. J. (2000). Voxel-based morphometry—the methods.
NeuroImage, 11, 805–821.
Banissy, M. J., Kanai, R., Walsh, V., & Rees, G. (2012). Inter-individual differences in
empathy are reflected in human brain structure. NeuroImage, 62(3), 2034–2039.
Baron-Cohen, S., & Wheelwright, S. (2004). The empathy quotient: An investigation of
adults with Asperger Syndrome or high functioning autism, and normal sex differences.
Journal of Autism and Developmental Disorders, 34(2), 163–175.
Barrós-Loscertales, A., Meseguer, V., Sanjuán, A., Belloch, V., Parcet, M. A., Torrubia,
R., & Avila, C. (2006). Behavioral inhibition system activity is associated with
increased amygdala and hippocampal gray matter volume: A voxel-based morphometry
study. NeuroImage, 33(3), 1011–1015.
Benjamin, J., Li, L., Patterson, C., Greenberg, B. D., Murphy, D. L., & Hamer, D. H.
(1996). Population and familial association between the D4 dopamine receptor gene
and measures of novelty seeking. Nature Genetics, 12(1), 81–84.
Bernhardt, B. C., Klimecki, O. M., Leiberg, S., & Singer, T. (2014). Structural covariance
networks of the dorsal anterior insula predict females’ individual differences in
empathic responding. Cerebral Cortex, 24(8), 2189–2198.
Berridge, K. C., Robinson, T. E., & Aldridge, J. W. (2009). Dissecting components of
reward: “Liking,” “wanting,” and learning. Current Opinion in Pharmacology, 9(1),
65–73.
Bjørnebekk, A., Fjell, A. M., Walhovd, K. B., Grydeland, H., Torgersen, S., & Westlye,
L. T. (2013). Neuronal correlates of the Five Factor Model (FFM) of human
personality: Multimodal imaging in a large healthy sample. NeuroImage, 65, 194–208.
Block, J. (1995). A contrarian view of the Five-Factor approach to personality
description. Psychological Bulletin, 117(2), 187–215.
Bookstein, F. L. (2001). “Voxel-based morphometry” should not be used with
imperfectly registered images. NeuroImage, 14(6), 1454–1562.
Bradley, R., Binder, E., Epstein, M., Tang, Y., Nair, H., Liu, W., … Ressler, K. (2008).
Influence of child abuse on adult depression: Moderation by the corticotropin-releasing
hormone receptor gene. Archives of General Psychiatry, 65(2), 190–200.
Bress, J. N., & Hajcak, G. (2013). Self-report and behavioral measures of reward
sensitivity predict the feedback negativity. Psychophysiology, 50(7), 610–616.
Bromberg-Martin, E. S., Matsumoto, M., & Hikosaka, O. (2010). Dopamine in
motivational control: Rewarding, aversive, and alerting. Neuron, 68(5), 815–834.
Brown, J. W., & Braver, T. S. (2005). Learned predictions of error likelihood in the
anterior cingulate cortex. Science, 307(5712), 1118–1121.
Brown, S. M., Manuck, S. B., Flory, J. D., & Hariri, A. R. (2006). Neural basis of
individual differences in impulsivity: Contributions of corticolimbic circuits for
behavioral arousal and control. Emotion, 6(2), 239–245.
Brummett, B. H., Boyle, S. H., Kuhn, C. M., Siegler, I. C., & Williams, R. B. (2008).
Associations among central nervous system serotonergic function and neuroticism are
moderated by gender. Biological Psychology, 78(2), 200–203.
Buckner, R. L., Andrews-Hanna, J. R., & Schacter, D. L. (2008). The brain’s default
network: Anatomy, function, and relevance to disease. Annals of the New York
Academy of Sciences, 1124, 1–38.
Bunge, S. A., & Zelazo, P. D. (2006). A brain-based account of the development of rule
use in childhood. Current Directions in Psychological Science, 15(3), 118–121.
Button, K. S., Ioannidis, J. P. A., Mokrysz, C., Nosek, B. A., Flint, J., Robinson, E. S. J.,
& Munafò, M. R. (2013). Power failure: Why small sample size undermines the
reliability of neuroscience. Nature Reviews. Neuroscience, 14(5), 365–376.
Canli, T., Sivers, H., Whitfield, S. L., Gotlib, I. H., & Gabrieli, J. D. E. (2002). Amygdala
response to happy faces as a function of extraversion. Science, 296(5576), 2191.
Canli, T., Zhao, Z., Desmond, J. E., Kang, E., Gross, J., & Gabrieli, J. D. E. (2001). An
fMRI study of personality influences on brain reactivity to emotional stimuli.
Behavioral Neuroscience, 115(1), 33–42.
Carp, J. (2012). The secret lives of experiments: Methods reporting in the fMRI literature.
NeuroImage, 63(1), 289–300.
Carver, C. S., Johnson, S. L., & Joormann, J. (2008). Serotonergic function, two-mode
models of self-regulation, and vulnerability to depression: What depression has in
common with impulsive aggression. Psychological Bulletin, 134(6), 912–943.
Carver, C. S., & Scheier, M. F. (1998). On the self-regulation of behavior. New York,
NY: Cambridge University Press.
Carver, C. S., & White, T. L. (1994). Behavioral inhibition, behavioral activation, and
affective responses to impending reward and punishment: The BIS/BAS Scales.
Chang, L., Connelly, B. S., & Geeza, A. A. (2012). Separating method factors and higher
order traits of the Big Five: A meta-analytic multitrait–multimethod approach. Journal
Chavanon, M.-L., Wacker, J., & Stemmler, G. (2011). Rostral anterior cingulate activity
generates posterior versus anterior theta activity linked to agentic extraversion.
Cognitive, Affective, & Behavioral Neuroscience, 11(2), 172–185.
Chavanon, M.-L., Wacker, J., & Stemmler, G. (2013). Paradoxical dopaminergic drug
effects in extraversion: Dose- and time-dependent effects of sulpiride on EEG theta
activity. Frontiers in Human Neuroscience, 7, 117.
Cherbuin, N., Windsor, T. D., Anstey, K. J., Maller, J. J., Meslin, C., & Sachdev, P. S.
(2008). Hippocampal volume is positively associated with behavioural inhibition (BIS)
in a large community-based sample of mid-life adults: The PATH through life study.
Social Cognitive and Affective Neuroscience, 3(3), 262–269.
Chiang, M.-C., Barysheva, M., Shattuck, D. W., Lee, A. D., Madsen, S. K., Avedissian,
C., … Thompson, P. M. (2009). Genetics of brain fiber architecture and intellectual
performance. The Journal of Neuroscience, 29(7), 2212–2224.
Chmielewski, M., Bagby, R. M., Markon, K., Ring, A. J., & Ryder, A. G. (2014).
Openness to experience, intellect, schizotypal personality disorder, and psychoticism:
Resolving the controversy. Journal of Personality Disorders, 28(4), 483–499.
Choi, E. Y., Yeo, B. T. T., & Buckner, R. L. (2012). The organization of the human
striatum estimated by intrinsic functional connectivity. Journal of Neurophysiology,
108(8), 2242–2263.
Christensen, J. C., Shiyanov, P. A., Estepp, J. R., & Schlager, J. J. (2014). Lack of
association between human plasma oxytocin and interpersonal trust in a prisoner’s
dilemma paradigm. PloS One, 9(12), e116172.
Churchwell, J. C., & Yurgelun-Todd, D. A. (2013). Age-related changes in insula cortical
thickness and impulsivity: Significance for emotional development and decision-
making. Developmental Cognitive Neuroscience, 6, 80–86.
Coan, J. A., & Allen, J. J. B. (2003). Frontal EEG asymmetry and the behavioral
activation and inhibition systems. Psychophysiology, 40(1), 106–114.
Coan, J. A., Allen, J. J. B., & McKnight, P. E. (2006). A capability model of individual
differences in frontal EEG asymmetry. Biological Psychology, 72(2), 198–207.
Cohen, M. X., Young, J., Baek, J.-M., Kessler, C., & Ranganath, C. (2005). Individual
differences in extraversion and dopamine genetics predict neural reward responses.
Cognitive Brain Research, 25(3), 851–861.
Connelly, B. S., & Ones, D. S. (2010). Another perspective on personality: Meta-analytic
integration of observers’ accuracy and predictive validity. Psychological Bulletin,
136(6), 1092–1122.
Cooper, A. J., Duke, E., Pickering, A. D., & Smillie, L. D. (2014). Individual differences
in reward prediction error: Contrasting relations between feedback-related negativity
and trait measures of reward sensitivity, impulsivity and extraversion. Frontiers in
Human Neuroscience, 8, 248.
Costa, P. T., Jr., & McCrae, R. R. (1992). NEO PI-R professional manual. Odessa, FL:
Cremers, H. R., Demenescu, L. R., Aleman, A., Renken, R., van Tol, M. J., van der Wee,
N. J., … Roelofs, K. (2010). Neuroticism modulates amygdala—prefrontal
connectivity in response to negative emotional facial expressions. NeuroImage, 49(1),
963–970.
Cremers, H. R., van Tol, M.-J., Roelofs, K., Aleman, A., Zitman, F. G., van Buchem, M.
A., … van der Wee, N. J. A. (2011). Extraversion is linked to volume of the
orbitofrontal cortex and amygdala. PloS One, 6(12), e28421.
Cunningham, W. A., Arbuckle, N. L., Jahn, A., Mowrer, S. M., & Abduljalil, A. M.
(2010). Aspects of neuroticism and the amygdala: Chronic tuning from motivational
styles. Neuropsychologia, 48(12), 3399–3404.
Davatzikos, C. (2004). Why voxel-based morphometric analysis should be used with
great caution when characterizing group differences. NeuroImage, 23(1), 17–20.
Davidson, R. J. (1998). Anterior electrophysiological asymmetries, emotion, and
depression: Conceptual and methodological conundrums. Psychophysiology, 35(5),
607–614.
Davis, M. H. (1983). Measuring individual differences in empathy: Evidence for a
multidimensional approach. Journal of Personality and Social Psychology, 44(1), 113–
126.
de Moor, M. H., Costa, P. T., Terracciano, A., Krueger, R. F., De Geus, E. J., Toshiko, T.,
… Metspalu, A. (2012). Meta-analysis of genome-wide association studies for
personality. Molecular Psychiatry, 17(3), 337–349.
De Pascalis, V., Cozzuto, G., Caprara, G. V., & Alessandri, G. (2013). Relations among
EEG-alpha asymmetry, BIS/BAS, and dispositional optimism. Biological Psychology,
94(1), 198–209.
Deacon, T. W. (1997). What makes the human brain different? Annual Review of
Anthropology, 26(1), 337–357.
Denissen, J. J. A., & Penke, L. (2008). Motivational individual reaction norms underlying
the Five-Factor Model of personality: First steps towards a theory-based conceptual
framework. Journal of Research in Personality, 42(5), 1285–1302.
Depue, R. A., & Collins, P. F. (1999). Neurobiology of the structure of personality:
Dopamine, facilitation of incentive motivation, and extraversion. Behavioral and Brain
Sciences, 22(3), 491–517.
Depue, R. A., & Fu, Y. (2013). On the nature of extraversion: Variation in conditioned
contextual activation of dopamine-facilitated affective, cognitive, and motor processes.
Frontiers in Human Neuroscience, 7, 288.
Depue, R. A., Luciana, M., Arbisi, P., Collins, P., & Leon, A. (1994). Dopamine and the
structure of personality: Relation of agonist-induced dopamine activity to positive
emotionality. Journal of Personality and Social Psychology, 67(3), 485–498.
Depue, R. A., & Morrone-Strupinsky, J. V. (2005). A neurobehavioral model of
affiliative bonding: Implications for conceptualizing a human trait of affiliation.
Behavioral and Brain Sciences, 28(3), 313–349.
DeYoung, C. G. (2010a). Impulsivity as a personality trait. In K. D. Vohs & R. F.
Baumeister (Eds.), Handbook of self-regulation: Research, theory, and applications
(2nd ed., pp. 485–502). New York, NY: Guilford Press.
DeYoung, C. G. (2010b). Personality neuroscience and the biology of traits. Social and
DeYoung, C. G. (2011). Intelligence and personality. In R. J. Sternberg & S. B. Kaufman
(Eds.), The Cambridge handbook of intelligence (pp. 711–737). New York: Cambridge
University Press.
DeYoung, C. G. (2013). The neuromodulator of exploration: A unifying theory of the role
of dopamine in personality. Frontiers in Human Neuroscience, 7, 762.
DeYoung, C. G. (2015a). Cybernetic Big Five Theory. Journal of Research in
DeYoung, C. G. (2015b). Openness/Intellect: A dimension of personality reflecting
cognitive exploration. In M. L. Cooper& R. J. Larsen (Eds.), The APA handbook of
personality and social psychology: Personality processes and individual differences
(Vol. 4, pp. 369–399). Washington, DC: American Psychological Association.
DeYoung, C. G., Cicchetti, D., & Rogosch, F. A. (2011). Moderation of the association
between childhood maltreatment and neuroticism by the corticotropin-releasing
hormone receptor 1 gene. Journal of Child Psychology and Psychiatry, 52(8), 898–906.
DeYoung, C. G., Cicchetti, D., Rogosch, F. A., Gray, J. R., Eastman, M., & Grigorenko,
dopamine system predicts openness/intellect. Journal of Research in Personality,
45(4), 364–371.
DeYoung, C. G., & Gray, J. R. (2009). Personality neuroscience: Explaining individual
differences in affect, behavior, and cognition. In P. J. Corr & G. Matthews (Eds.), The
Cambridge handbook of personality (pp. 323–346). New York, NY: Cambridge
University Press.
The openness/intellect trait domain as a paradoxical simplex. Journal of Research in
Personality, 46(1), 63–78.
(2010). Testing predictions from personality neuroscience. Brain structure and the Big
Five. Psychological Science, 21(6), 820–828.
Differences, 33(4), 533–552.
DeYoung, C. G., Peterson, J. B., & Higgins, D. M. (2005). Sources of openness/intellect:
Cognitive and neuropsychological correlates of the fifth factor of personality. Journal
of Personality, 73(4), 825–858.
DeYoung, C. G., Quilty, L. C., Peterson, J. B., & Gray, J. R. (2014). Openness to
experience, intellect, and cognitive ability. Journal of Personality Assessment, 96(1),
46–52.
DeYoung, C. G., Shamosh, N. A., Green, A. E., Braver, T. S., & Gray, J. R. (2009).
Intellect as distinct from openness: Differences revealed by fMRI of working memory.
DeYoung, C. G., Weisberg, Y. J., Quilty, L. C., & Peterson, J. B. (2013). Unifying the
aspects of the Big Five, the interpersonal circumplex, and trait affiliation. Journal of
Personality, 81(5), 465–475.
Du, L., Bakish, D., Ravindran, A. V, & Hrdina, P. D. (2002). Does fluoxetine influence
major depression by modifying Five-Factor personality traits? Journal of Affective
Disorders, 71(1–3), 235–241.
Ebstein, R. P., Novick, O., Umansky, R., Priel, B., Osher, Y., Blaine, D., … Belmaker, R.
H. (1996). Dopamine D4 receptor (D4DR) exon III polymorphism associated with the
human personality trait of novelty seeking. Nature Genetics, 12(1), 78–80.
Everhart, D. E., Demaree, H. A., & Harrison, D. W. (2008). The influence of hostility on
electroencephalographic activity and memory functioning during an affective memory
task. Clinical Neurophysiology, 119(1), 134–143.
Farr, O. M., Hu, S., Zhang, S., & Li, C. S. R. (2012). Decreased saliency processing as a
neural measure of Barratt impulsivity in healthy adults. NeuroImage, 63(3), 1070–
1077.
Forbes, C. E., Poore, J. C., Krueger, F., Barbey, A. K., Solomon, J., & Grafman, J.
(2014). The role of executive function and the dorsolateral prefrontal cortex in the
expression of neuroticism and conscientiousness. Social Neuroscience, 9(2), 139–151.
Forbes, E. E., Brown, S. M., Kimak, M., Ferrell, R. E., Manuck, S. B., & Hariri, A. R.
(2009). Genetic variation in components of dopamine neurotransmission impacts
ventral striatal reactivity associated with impulsivity. Molecular Psychiatry, 14(1), 60–
70.
Fox, M. D., Corbetta, M., Snyder, A. Z., Vincent, J. L., & Raichle, M. E. (2006).
Spontaneous neuronal activity distinguishes human dorsal and ventral attention
systems. Proceedings of the National Academy of Sciences of the United States of
America, 103(26), 10046–10051.
Freeman, H. D., & Gosling, S. D. (2010). Personality in nonhuman primates: A review
and evaluation of past research. American Journal of Primatology, 72(8), 653–671.
Frokjaer, V. G., Mortensen, E. L., Nielsen, F. A., Haugbol, S., Pinborg, L. H., Adams, K.
H., … Knudsen, G. M. (2008). Frontolimbic serotonin 2A receptor binding in healthy
subjects is associated with personality risk factors for affective disorder. Biological
Psychiatry, 63(6), 569–576.
Fuentes, P., Barrós-Loscertales, A., Bustamante, J. C., Rosell, P., Costumero, V., &
Ávila, C. (2012). Individual differences in the behavioral inhibition system are
associated with orbitofrontal cortex and precuneus gray matter volume. Cognitive,
Affective, & Behavioral Neuroscience, 12(3), 491–498.
Gale, A., Edwards, J., Morris, P., Moore, R., & Forrester, D. (2001). Extraversion-
introversion, neuroticism-stability, and EEG indicators of positive and negative
empathic mood. Personality and Individual Differences, 30(3), 449–461.
Garcia-Banda, G., Chellew, K., Fornes, J., Perez, G., Servera, M., & Evans, P. (2014).
Neuroticism and cortisol: Pinning down an expected effect. International Journal of
Psychophysiology, 91(2), 132–138.
Gerritsen, L., Geerlings, M., Bremmer, M., Beekman, A., Deeg, D., Penninx, B. W. J. H.,
& Comijs, H. (2009). Personality characteristics and hypothalamic-pituitary-adrenal
axis regulation in older persons. The American Journal of Geriatric Psychiatry, 17(12),
1077–1084.
Gilbert, S. J., Spengler, S., Simons, J. S., Steele, J. D., Lawrie, S. M., Frith, C. D., &
Burgess, P. W. (2006). Functional specialization within rostral prefrontal cortex (area
10): A meta-analysis. Journal of Cognitive Neuroscience, 18(6), 932–948.
Goldberg, L. R. (1990). An alternative “description of personality”: The Big-Five Factor
measuring the lower-level facets of several Five-Factor Models. In I. Mervielde, I.
cross-species review. Current Directions in Psychological Science, 8(3), 69–75.
Grabe, H., Schwahn, C., Appel, K., Mahler, J., Schulz, A., Spitzer, C., … Volzke, H.
(2010). Childhood maltreatment, the corticotropin-releasing hormone receptor gene
and adult depression in the general population. American Journal of Medical Genetics.
Part B: Neuropsychiatric Genetics, 153B(8), 1483–1493.
Gray, J. A. (1982). The neuropsychology of anxiety: An inquiry into the functions of the
septo-hippocampal system. New York, NY: Oxford University Press.
Gray, J. A. (2004). Consciousness: Creeping up on the hard problem. New York, NY:
Gray, J. A., & McNaughton, N. (2000). The neuropsychology of anxiety: An enquiry into
the function of the septo-hippocampal system. New York, NY: Oxford University
Press.
Green, A. E., Fugelsang, J. A., Kraemer, D. J. M., Shamosh, N. A., & Dunbar, K. N.
(2006). Frontopolar cortex mediates abstract integration in analogy. Brain Research,
1096(1), 125–137.
Green, A. E., Munafo, M. R., DeYoung, C. G., Fossella, J. A., Fan, J., & Gray, J. R.
(2008). Using genetic data in cognitive neuroscience: From growing pains to genuine
insights. Nature Reviews Neuroscience, 9, 710–720.
Gronenschild, E. H. B. M., Habets, P., Jacobs, H. I. L., Mengelers, R., Rozendaal, N., van
Os, J., & Marcelis, M. (2012). The effects of FreeSurfer version, workstation type, and
Macintosh operating system version on anatomical volume and cortical thickness
measurements. PloS One, 7(6), e38234.
Haas, B. W., Constable, R. T., & Canli, T. (2008). Stop the sadness: Neuroticism is
associated with sustained medial prefrontal cortex response to emotional facial
expressions. NeuroImage, 42(1), 385–392.
Haas, B. W., Omura, K., Constable, R. T., & Canli, T. (2007). Is automatic emotion
regulation associated with agreeableness? A perspective using a social neuroscience
approach. Psychological Science, 18(2), 130–132.
Harmon-Jones, E. (2004). Contributions from research on anger and cognitive dissonance
to understanding the motivational functions of asymmetrical frontal brain activity.
Biological Psychology, 67(1–2), 51–76.
Harmon-Jones, E., & Allen, J. J. B. (1997). Behavioral activation sensitivity and resting
frontal EEG asymmetry: Covariation of putative indicators related to risk for mood
disorders. Journal of Abnormal Psychology, 106(1), 159–163.
Harmon-Jones, E., & Allen, J. J. B. (1998). Anger and frontal brain activity: EEG
asymmetry consistent with approach motivation despite negative affective valence.
Harmon-Jones, E., Gable, P. A., & Peterson, C. K. (2010). The role of asymmetric frontal
cortical activity in emotion-related phenomena: A review and update. Biological
Psychology, 84(3), 451–462.
Harris, S. E., Wright, A. F., Hayward, C., Starr, J. M., Whalley, L. J., & Deary, I. J.
(2005). The functional COMT polymorphism, Val 158 Met, is associated with logical
memory and the personality trait intellect/imagination in a cohort of healthy 79 year
olds. Neuroscience Letters, 385(1), 1–6.
Hemphill, J. F. (2003). Interpreting the magnitudes of correlation coefficients. The
Henley, S. M. D., Ridgway, G. R., Scahill, R. I., Klöppel, S., Tabrizi, S. J., Fox, N. C., &
Kassubek, J. (2010). Pitfalls in the use of voxel-based morphometry as a biomarker:
Examples from Huntington disease. American Journal of Neuroradiology, 31(4), 711–
719.
Hennig, J. (2004). Personality, serotonin, and noradrenaline. In R. M. Stelmack (Ed.), On
the psychobiology of personality: Essays in honor of Marvin Zuckerman (pp. 379–408).
New York, NY: Elsevier.
Hill, J., Inder, T., Neil, J., Dierker, D., Harwell, J., & Van Essen, D. (2010). Similar
patterns of cortical expansion during human development and evolution. Proceedings
of the National Academy of Sciences of the United States of America, 107(29), 13135–
13140.
Hirsh, J. B., & Inzlicht, M. (2008). The devil you know: Neuroticism predicts neural
response to uncertainty. Psychological Science, 19(10), 962–967.
Hirsh, J. B., Mar, R. A., & Peterson, J. B. (2012). Psychological entropy: A framework
for understanding uncertainty-related anxiety. Psychological Review, 119(2), 304–320.
Hofstee, W., de Raad, B., & Goldberg, L. R. (1992). Integration of the Big Five and
63(1), 146–163.
Holmes, A. J., Lee, P. H., Hollinshead, M. O., Bakst, L., Roffman, J. L., Smoller, J. W.,
& Buckner, R. L. (2012). Individual differences in amygdala-medial prefrontal
anatomy link negative affect, impaired social functioning, and polygenic depression
risk. The Journal of Neuroscience, 32(50), 18087–18100.
Hu, X., Erb, M., Ackermann, H., Martin, J. A., Grodd, W., & Reiterer, S. M. (2011).
Voxel-based morphometry studies of personality: Issue of statistical model
specification-effect of nuisance covariates. NeuroImage, 54(3), 1994–2005.
Hyde, L. W., Gorka, A., Manuck, S. B., & Hariri, A. R. (2011). Perceived social support
moderates the link between threat-related amygdala reactivity and trait anxiety.
Neuropsychologia, 49(4), 651–656.
Iidaka, T., Matsumoto, A., Ozaki, N., Suzuki, T., Iwata, N., Yamamoto, Y., … Sadato, N.
(2006). Volume of left amygdala subregion predicted temperamental trait of harm
avoidance in female young subjects. A voxel-based morphometry study. Brain
Research, 1125(1), 85–93.
Ioannidis, J. P. A. (2011). Excess significance bias in the literature on brain volume
abnormalities. Archives of General Psychiatry, 68(8), 773–780.
Jackson, J., Balota, D. A., & Head, D. (2011). Exploring the relationship between
personality and regional brain volume in healthy aging. Neurobiology of Aging, 32(12),
2162–2171.
of the Five-Factor Model of personality. Personality and Individual Differences, 33(1),
83–101.
Jang, K. L., Livesley, W. J., Riemann, R., Vernon, P. A., Hu, S., Angleitner, A., …
Hamer, D. H. (2001). Covariance structure of neuroticism and agreeableness: A twin
and molecular genetic analysis of the role of the serotonin transporter gene. Journal of
hierarchical model of personality. Journal of Personality and Social Psychology, 74(6),
1556–1565.
Jennings, R. G., & Van Horn, J. D. (2012). Publication bias in neuroimaging research:
Implications for meta-analyses. Neuroinformatics, 10(1), 67–80.
John, O., Naumann, L., & Soto, C. (2008). Paradigm shift to the integrative Big Five trait
taxonomy. In O. P. John, R. W. Robins, & L. A. Pervin (Eds.), Handbook of
personality: Theory and research (pp. 114–158). New York, NY: Guilford Press.
Johnson, J. A. (1994). Clarification of factor five with the help of the AB5C model.
Johnson, W., & Krueger, R. F. (2004). Genetic and environmental structure of adjectives
describing the domains of the Big Five Model of personality: A nationwide US twin
study. Journal of Research in Personality, 38(5), 448–472.
Jung, R. E., Grazioplene, R., Caprihan, A., Chavez, R. S., & Haier, R. J. (2010). White
matter integrity, creativity, and psychopathology: Disentangling constructs with
diffusion tensor imaging. PloS One, 5(3), e9818.
Kalbitzer, J., Frokjaer, V. G., Erritzoe, D., Svarer, C., Cumming, P., Nielsen, F. A., …
Knudsen, G. M. (2009). The personality trait openness is related to cerebral 5-HTT
levels. NeuroImage, 45(2), 280–285.
Kamarck, T. W., Haskett, R. F., Muldoon, M., Flory, J. D., Anderson, B., Bies, R., …
Manuck, S. B. (2009). Citalopram intervention for hostility: Results of a randomized
clinical trial. Journal of Consulting and Clinical Psychology, 77(1), 174–188.
Kapogiannis, D., Sutin, A., Davatzikos, C., Costa, P., & Resnick, S. (2013). The five
factors of personality and regional cortical variability in the Baltimore longitudinal
study of aging. Human Brain Mapping, 34(11), 2829–2840.
Kievit, R. A., van Rooijen, H., Wicherts, J. M., Waldorp, L. J., Kan, K.-J., Scholte, H. S.,
& Borsboom, D. (2012). Intelligence and the brain: A model-based approach.
Cognitive Neuroscience, 3(2), 89–97.
Kirschbaum, C., Bartussek, D., & Strasburger, C. (1992). Cortisol responses to
psychological stress and correlations with personality traits. Personality and Individual
Differences, 13(12), 1353–1357.
Knutson, B., Momenan, R., Rawlings, R. R., Fong, G. W., & Hommer, D. (2001).
Negative association of neuroticism with brain volume ratio in healthy humans.
Biological Psychiatry, 50(9), 685–690.
Knyazev, G. G. (2010). Antero-posterior EEG spectral power gradient as a correlate of
extraversion and behavioral inhibition. The Open Neuroimaging Journal, 4, 114–120.
Koehler, S., Wacker, J., Odorfer, T., Reif, A., Gallinat, J., Fallgatter, A. J., & Herrmann,
M. J. (2011). Resting posterior minus frontal EEG slow oscillations is associated with
extraversion and DRD2 genotype. Biological Psychology, 87(3), 407–413.
Koelsch, S., Skouras, S., & Jentschke, S. (2013). Neural correlates of emotional
personality: A structural and functional magnetic resonance imaging study. PloS One,
8(11), e77196.
Kranzler, H., Feinn, R., Nelson, E. C., Covault, J., Anton, R. F., Farrer, L., & Gelernter, J.
(2011). A CRHR1 haplotype moderates the effect of adverse childhood experiences on
lifetime risk of major depressive episode in African-American women. American
Journal of Medical Genetics. Part B: Neuropsychiatric Genetics, 156(8), 960–968.
Kumari, V., Cotter, P. A., Mulligan, O. F., Checkley, S. A., Gray, N. S., Hemsley, D. R.,
… Gray, J. A. (1999). Effects of d-amphetamine and haloperidol on latent inhibition in
healthy male volunteers. Journal of Psychopharmacology, 13(4), 398–405.
64(4), 241–256.
Laird, A. R., Fox, P. M., Eickhoff, S. B., Turner, J. A., Ray, K. L., McKay, D. R., … Fox,
P. T. (2011). Behavioral interpretations of intrinsic connectivity networks. Journal of
Cognitive Neuroscience, 23(12), 4022–4037.
Lamm, C., Decety, J., & Singer, T. (2011). Meta-analytic evidence for common and
distinct neural networks associated with directly experienced pain and empathy for
pain. NeuroImage, 54(3), 2492–2502.
Lange, S., Leue, A., & Beauducel, A. (2012). Behavioral approach and reward
processing: Results on feedback-related negativity and P3 component. Biological
Psychology, 89(2), 416–425.
Lemogne, C., Gorwood, P., Bergouignan, L., Pélissolo, A., Lehéricy, S., & Fossati, P.
(2011). Negative affectivity, self-referential processing and the cortical midline
structures. Social Cognitive and Affective Neuroscience, 6(4), 426–433.
Lewis, G. J., Panizzon, M. S., Eyler, L., Chen, C., Neale, M. C., Jernigan, T. L., … Franz,
C. E. (2014). Heritable influences on amygdala and orbitofrontal cortex contribute to
genetic variation in core dimensions of personality. NeuroImage, 103, 309–315.
Li, Y., Qiao, L., Sun, J., Wei, D., Li, W., Qiu, J., … Shi, H. (2014). Gender-specific
neuroanatomical basis of behavioral inhibition/approach systems (BIS/BAS) in a large
sample of young adults: A voxel-based morphometric investigation. Behavioral Brain
Research, 274, 400–408.
Liu, W.-Y., Weber, B., Reuter, M., Markett, S., Chu, W.-C., & Montag, C. (2013). The
Big Five of personality and structural imaging revisited: A VBM-DARTEL study.
Neuroreport, 24(7), 375–380.
Loehlin, J. C., McCrae, R. R., Costa, P. T., & John, O. P. (1998). Heritabilities of
common and measure-specific components of the Big Five personality factors. Journal
MacCallum, R. C., Zhang, S., Preacher, K. J., & Rucker, D. D. (2002). On the practice of
dichotomization of quantitative variables. Psychological Methods, 7(1), 19–40.
MacDonald, K. (1995). Evolution, the Five-Factor Model, and levels of personality.
MacLean, K. A., Johnson, M. W., & Griffiths, R. R. (2011). Mystical experiences
occasioned by the hallucinogen psilocybin lead to increases in the personality domain
of openness. Journal of Psychopharmacology, 25(11), 1453–1461.
Madsen, K. S., Jernigan, T. L., Iversen, P., Frokjaer, V. G., Mortensen, E. L., Knudsen,
G. M., & Baaré, W. F. C. (2012). Cortisol awakening response and negative
emotionality linked to asymmetry in major limbic fibre bundle architecture. Psychiatry
Research, 201(1), 63–72.
Mahoney, C. J., Rohrer, J. D., Omar, R., Rossor, M. N., & Warren, J. D. (2011).
Neuroanatomical profiles of personality change in frontotemporal lobar degeneration.
The British Journal of Psychiatry, 198(5), 365–372.
Manuck, S. B., Flory, J. D., Ferrell, R. E., Mann, J. J., & Muldoon, M. F. (2000). A
regulatory polymorphism of the monoamine oxidase-A gene may be associated with
variability in aggression, impulsivity, and central nervous system serotonergic
responsivity. Psychiatry Research, 95(1), 9–23.
Manuck, S. B., Flory, J. D., McCaffery, J. M., Matthews, K. A., Mann, J. J., & Muldoon,
M. F. (1998). Aggression, impulsivity, and central nervous system serotonergic
responsivity in a nonpatient sample. Neuropsychopharmacology, 19(4), 287–299.
Mar, R. A., Spreng, R. N., & DeYoung, C. G. (2013). How to produce personality
neuroscience research with high statistical power and low additional cost. Cognitive,
Affective, and Behavioral Neuroscience, 13, 674–685.
Matsuo, K., Nicoletti, M., Nemoto, K., Hatch, J. P., Peluso, M. A. M., Nery, F. G., &
Soares, J. C. (2009). A voxel-based morphometry study of frontal gray matter
correlates of impulsivity. Human Brain Mapping, 30(4), 1188–1195.
McCrae, R. R., Yamagata, S., Jang, K. L., Riemann, R., Ando, J., Ono, Y., … Spinath, F.
M. (2008). Substance and artifact in the higher-order factors of the Big Five. Journal of
Mehrabian, A., & Epstein, N. (1972). A measure of emotional empathy. Journal of
Personality, 40(4), 525–543.
Meier, B. P., Robinson, M. D., & Wilkowski, B. M. (2006). Turning the other cheek.
Agreeableness and the regulation of aggression-related primes. Psychological Science,
17(2), 136–142.
Miller, E. K., & Cohen, J. D. (2001). An integrative theory of prefrontal cortex function.
Annual review of neuroscience, 24(1), 167–202.
Miller, G., Cohen, S., Rabin, B., Skoner, D. P., & Doyle, W. J. (1999). Personality and
tonic cardiovascular, neuroendocrine, and immune parameters. Brain, Behavior, and
Immunity, 13(2), 109–123.
Mobbs, D., Hagan, C., Azim, E., Menon, V., & Reiss, A. L. (2005). Personality predicts
activity in reward and emotional regions associated with humor. Proceedings of the
National Academy of Sciences of the United States of America, 102(45), 16502–16506.
Montoya, E. R., Terburg, D., Bos, P. A., & Van Honk, J. (2012). Testosterone, cortisol,
and serotonin as key regulators of social aggression: A review and theoretical
perspective. Motivation and emotion, 36(1), 65–73.
Morinaga, K., Akiyoshi, J., Matsushita, H., Ichioka, S., Tanaka, Y., Tsuru, J., & Hanada,
H. (2007). Anticipatory anxiety-induced changes in human lateral prefrontal cortex
activity. Biological Psychology, 74(1), 34–38.
Mueller, E. M., Burgdorf, C., Chavanon, M. L., Schweiger, D., Wacker, J., & Stemmler,
G. (2014). Dopamine modulates frontomedial failure processing of agentic introverts
versus extraverts in incentive contexts. Cognitive, Affective, & Behavioral
Neuroscience, 14, 756–768.
Mujica-Parodi, L. R., Korgaonkar, M., Ravindranath, B., Greenberg, T., Tomasi, D.,
Wagshul, M., … Malaspina, D. (2009). Limbic dysregulation is associated with
lowered heart rate variability and increased trait in healthy adults. Human Brain
Mapping, 30(1), 47–58.
Munafò, M. R., & Flint, J. (2011). Dissecting the genetic architecture of human
personality. Trends in Cognitive Sciences, 15(9), 395–400.
Munafò, M. R., Freimer, N. B., Ng, W., Ophoff, R., Veijola, J., Miettunen, J., … Flint, J.
(2009). 5-HTTLPR genotype and anxiety-related personality traits: A meta-analysis
and new data. American Journal of Medical Genetics Part B: Neuropsychiatric
Genetics, 150B(2), 271–281.
Munafò, M. R., Yalcin, B., Willis-Owen, S. A., & Flint, J. (2008). Association of the
dopamine D4 receptor (DRD4) gene and approach-related personality traits: Meta-
analysis and new data. Biological Psychiatry, 63(2), 197–206.
Mutschler, I., Reinbold, C., Wankerl, J., Seifritz, E., & Ball, T. (2013). Structural basis of
empathy and the domain general region in the anterior insular cortex. Frontiers in
Human Neuroscience, 7, 177.
Nater, U., Hoppmann, C., & Klumb, P. (2010). Neuroticism and conscientiousness are
associated with cortisol diurnal profiles in adults—role of positive and negative affect.
Psychoneuroendocrinology, 35(10), 1573–1577.
Need, A. C., & Goldstein, D. B. (2014). Schizophrenia genetics comes of age. Neuron,
83(4), 760–763.
Nelson, M. T., Seal, M. L., Phillips, L. J., Merritt, A. H., Wilson, R., & Pantelis, C.
(2011). An investigation of the relationship between cortical connectivity and
schizotypy in the general population. The Journal of Nervous and Mental Disease,
199(5), 348–353.
Netter, P. (2004). Personality and hormones. In R. Stelmack (Ed.), On the psychobiology
of personality: Essays in honor of Marvin Zuckerman (pp. 353–377). New York, NY:
Elsevier.
The American Psychologist, 61(6), 622–631.
Nettle, D. (2007). Personality: What makes you the way you are. New York, NY: Oxford
University Press.
Nettle, D., & Liddle, B. (2008). Agreeableness is related to social-cognitive, but not
social-perceptual, theory of mind. European Journal of Personality, 22(4), 323–335.
Omura, K., Constable, R. T., & Canli, T. (2005). Amygdala gray matter concentration is
associated with extraversion and neuroticism. NeuroReport, 16(17), 1905–1908.
Oswald, L., Zandi, P., Nestadt, G., Potash, J. B., Kalaydjian, A. E., & Wand, G. S.
(2006). Relationship between cortisol responses to stress and personality.
Neuropsychopharmacology, 31(7), 1583–1591.
Panksepp, J. (1998). Affective neuroscience: The foundations of human and animal
emotion. New York, NY: Oxford University Press.
Pardini, D. A., Raine, A., Erickson, K., & Loeber, R. (2014). Lower amygdala volume in
men is associated with childhood aggression, early psychopathic traits, and future
violence. Biological Psychiatry, 75(1), 73–80.
Parkinson, C., & Wheatley, T. (2014). Relating anatomical and social connectivity: White
matter microstructure predicts emotional empathy. Cerebral Cortex, 24, 614–625.
Passamonti, L., Terracciano, A., Riccelli, R., Donzuso, G., Cerasa, A., Vaccaro, M., …
Quattrone, A. (2015). Increased functional connectivity within mesocortical networks
in open people. NeuroImage, 104, 301–309.
Peciña, S., Smith, K. S., & Berridge, K. C. (2006). Hedonic hot spots in the brain. The
Neuroscientist, 12(6), 500–511.
Perry, A., Mankuta, D., & Shamay-Tsoory, S. G. (2015). OT promotes closer
interpersonal distance among highly empathic individuals. Social Cognitive and
Affective Neuroscience, 10(1), 3–9.
Peterson, J. B., & Carson, S. (2000). Latent inhibition and openness to experience in a
high-achieving student population. Personality and Individual Differences, 28(2), 323–
332.
Peterson, J. B., & Flanders, J. L. (2002). Complexity management theory: Motivation for
ideological rigidity and social conflict. Cortex, 38(3), 429–458.
Peterson, J. B., Smith, K. W., & Carson, S. (2002). Openness and extraversion are
associated with reduced latent inhibition: Replication and commentary. Personality and
Phillips, A., Carroll, D., Burns, V. E., & Drayson, M. (2005). Neuroticism, cortisol
reactivity, and antibody response to vaccination. Psychophysiology, 42(2), 232–238.
Pickering, A. D. (2004). The neuropsychology of impulsive antisocial sensation seeking
personality traits: From dopamine to hippocampal function? In R. M. Stelmack (Ed.),
On the psychobiology of personality: Essays in honor of Marvin Zuckerman (pp. 453–
477). New York, NY: Elsevier.
Pickering, A. D., & Gray, J. A. (1999). The neuroscience of personality. In L. A. Pervin
& O. P. John (Eds.), Handbook of personality (2nd ed., pp. 277–299). New York, NY:
Guilford Press.
Polanczyk, G., Caspi, A., Williams, B., Price, T. S., Danese, A., Sugden, K., … Moffitt,
T. E. (2009). Protective effect of CRHR1 gene variants on the development of adult
depression following childhood maltreatment: Replication and extension. Archives of
General Psychiatry, 66(9), 978–985.
Polk, D., Cohen, S., Doyle, W., Skoner, D. P., & Kirschbaum, C. (2005). State and trait
affect as predictors of salivary cortisol in healthy adults. Psychoneuroendocrinology,
30(3), 261–272.
Proudfit, G. H. (2015). The reward positivity: From basic research on reward to a
biomarker for depression. Psychophysiology, 52(4), 449–459.
Quilty, L. C., DeYoung, C. G., Oakman, J. M., & Bagby, R. M. (2014). Extraversion and
behavioral activation: Integrating the components of approach. Journal of Personality
Quilty, L. C., Meusel, L.-A. C., & Bagby, R. M. (2008). Neuroticism as a mediator of
treatment response to SSRIs in major depressive disorder. Journal of Affective
Disorders, 111(1), 67–73.
Rammsayer, T. H. (1998). Extraversion and dopamine: Individual differences in response
to changes in dopaminergic activity as a possible biological basis of extraversion.
European Psychologist, 3(1), 37–50.
Rammsayer, T., Netter, P., & Vogel, W. H. (1993). A neurochemical model underlying
differences in reaction times between introverts and extraverts. Personality and
Ramnani, N., & Owen, A. M. (2004). Anterior prefrontal cortex: Insights into function
from anatomy and neuroimaging. Nature Reviews. Neuroscience, 5(3), 184–194.
Reuter, M., Weber, B., Fiebach, C. J., Elger, C., & Montag, C. (2009). The biological
basis of anger: Associations with the gene coding for DARPP-32 (PPP1R1B) and with
amygdala volume. Behavioural Brain Research, 202(2), 179–183.
Revelle, W., & Wilt, J. (2013). The general factor of personality: A general critique.
Richard, F. D., Bond, C. F., Jr., & Stokes-Zoota, J. J. (2003). One hundred years of social
psychology quantitatively described. Review of General Psychology, 7(4), 331–363.
Ridderinkhof, K. R., Ullsperger, M., Crone, E. A., & Nieuwenhuis, S. (2004). The role of
the medial frontal cortex in cognitive control. Science, 306(5695), 443–447.
FFI scales. Journal of Personality, 65(3), 449–475.
conscientiousness and how can it be assessed? Developmental Psychology, 50(5),
1315–1330.
Rothbart, M., & Bates, J. (1998). Temperament. In W. Damon & N. Eisenberg (Eds.),
Handbook of child psychology: Vol. 3. Social, emotional, and personality development
(pp. 105–176). New York, NY: Wiley.
Sambrook, T. D., & Goslin, J. (2015). A neural reward prediction error revealed by a
meta-analysis of ERPs using great grand averages. Psychological Bulletin, 141(1),
213–235.
Sampaio, A., Soares, J. M., Coutinho, J., Sousa, N., & Gonçalves, O. F. (2014). The Big
Five default brain: Functional evidence. Brain Structure & Function, 219(6), 1913–
1922.
Sapolsky, R. M. (1994). Glucocorticoids, stress and exacerbation of excitotoxic neuron
death. Seminars in Neuroscience, 6(5), 323–331.
Sassa, Y., Taki, Y., Takeuchi, H., Hashizume, H., Asano, M., Asano, K., … Kawashima,
R. (2012). The correlation between brain gray matter volume and empathizing and
systemizing quotients in healthy children. NeuroImage, 60(4), 2035–2041.
Saucier, G. (1992). Openness versus intellect: Much ado about nothing? European
Schaefer, M., Knuth, M., & Rumpel, F. (2011). Striatal response to favorite brands as a
function of neuroticism and extraversion. Brain Research, 1425, 83–89.
Schmidt, L. A. (1999). Frontal brain electrical activity in shyness and sociability.
Schmidtke, J. I., & Heller, W. (2004). Personality, affect and EEG: Predicting patterns of
regional brain activity related to extraversion and neuroticism. Personality and
Schommer, N., Kudielka, B., Hellhammer, D. H., & Kirschbaum, C. (1999). No evidence
for a close relationship between personality traits and circadian cortisol rhythm or a
single cortisol stress response. Psychological Reports, 84(3), 840–842.
Schönbrodt, F. D., & Perugini, M. (2013). At what sample size do correlations stabilize?
Schuyler, B. S., Kral, T. R. A., Jacquart, J., Burghy, C. A., Weng, H. Y., Perlman, D. M.,
… Davidson, R. J. (2014). Temporal dynamics of emotional responding: Amygdala
recovery predicts emotional traits. Social Cognitive and Affective Neuroscience, 9(2),
176–181.
Servaas, M. N., Geerligs, L., Renken, R. J., Marsman, J. B. C., Ormel, J., Riese, H., &
Aleman, A. (2015). Connectomics and neuroticism: An altered functional network
organization. Neuropsychopharmacology, 40, 296–304.
Servaas, M. N., Riese, H., Renken, R. J., Marsman, J.-B. C., Lambregs, J., Ormel, J., &
Aleman, A. (2013a). The effect of criticism on functional brain connectivity and
associations with neuroticism. PloS One, 8(7), e69606.
Servaas, M. N., van der Velde, J., Costafreda, S. G., Horton, P., Ormel, J., Riese, H., &
Aleman, A. (2013b). Neuroticism and the brain: A quantitative meta-analysis of
neuroimaging studies investigating emotion processing. Neuroscience and
Biobehavioral Reviews, 37(8), 1518–1529.
Shackman, A. J., McMenamin, B. W., Maxwell, J. S., Greischar, L. L., & Davidson, R. J.
(2009). Right dorsolateral prefrontal cortical activity and behavioral inhibition.
Psychiatry, 44(1), 2–32.
Shiner, R. L., & DeYoung, C. (2013). The structure of temperament and personality
traits: A developmental perspective. In P. D. Zelazo (Ed.), The Oxford handbook of
developmental psychology, Vol. 2: Self and other (pp. 113–141). New York, NY:
Smillie, L. (2008). What is reinforcement sensitivity? Neuroscience paradigms for
approach-avoidance process theories of personality. European Journal of Personality,
22(5), 359–384.
Smillie, L. D., Cooper, A. J., & Pickering, A. D. (2011). Individual differences in reward-
prediction-error: Extraversion and feedback-related negativity. Social Cognitive and
Affective Neuroscience, 6(5), 646–652.
Smillie, L. D., Cooper, A. J., Wilt, J., & Revelle, W. (2012). Do extraverts get more bang
for the buck? Refining the affective-reactivity hypothesis of extraversion. Journal of
Smillie, L. D., Pickering, A. D., & Jackson, C. J. (2006). The new reinforcement
sensitivity theory: Implications for personality measurement. Personality and Social
Psychology Review, 10(4), 320–335.
Smith, S. M., Fox, P. T., Miller, K. L., Glahn, D. C., Fox, P. M., Mackay, C. E., …
Beckmann, C. F. (2009). Correspondence of the brain’s functional architecture during
activation and rest. Proceedings of the National Academy of Sciences of the United
States of America, 106(31), 13040–13045.
Somerville, L. H., Whalen, P. J., & Kelley, W. M. (2010). Human bed nucleus of the stria
terminalis indexes hypervigilant threat monitoring. Biological Psychiatry, 68(5), 416–
424.
Spoont, M. R. (1992). Modulatory role of serotonin in neural information processing:
Implications for human psychopathology. Psychological Bulletin, 112(2), 330–350.
Stillman, P. E., Van Bavel, J. J., & Cunningham, W. A. (2015). Valence asymmetries in
the human amygdala: Task relevance modulates amygdala responses to positive more
than negative affective cues. Journal of Cognitive Neuroscience, 27, 842–851.
Sutin, A. R., Beason-Held, L. L., Dotson, V. M., Resnick, S. M., & Costa, P. T. (2010).
The neural correlates of neuroticism differ by sex prospectively mediate depressive
symptoms among older women. Journal of Affective Disorders, 127(1–3), 241–247.
in resting-state neural correlates of openness to experience among older adults.
Cerebral Cortex, 19(12), 2797–2802.
Sutton, S. K., & Davidson, R. J. (1997). Prefrontal brain asymmetry: A biological
substrate of the behavioral approach and inhibition systems. Psychological Science,
8(3), 204–210.
Taddei, M., Tettamanti, M., Zanoni, A., Cappa, S., & Battaglia, M. (2012). Brain white
matter organisation in adolescence is related to childhood cerebral responses to facial
expressions and harm avoidance. NeuroImage, 61(4), 1394–1401.
Takano, A., Arakawa, R., Hayashi, M., Takahashi, H., Ito, H., & Suhara, T. (2007).
Relationship between neuroticism personality trait and serotonin transporter binding.
Takeuchi, H., Taki, Y., Sassa, Y., Hashizume, H., Sekiguchi, A., Fukushima, A., &
Kawashima, R. (2014). Regional gray matter volume is associated with empathizing
and systemizing in young adults. PloS one, 9(1).
Taki, Y., Thyreau, B., Kinomura, S., Sato, K., Goto, R., Wu, K., … Fukuda, H. (2013). A
longitudinal study of the relationship between personality traits and the annual rate of
volume changes in regional gray matter in healthy adults. Human Brain Mapping,
34(12), 3347–3353.
Tang, T. Z., Derubeis, R. J., Hollon, S. D., Amsterdam, J., Shelton, R., & Schalet, B.
(2009). Personality change during depression treatment. Archives of General
Psychiatry, 66(12), 1322–1330.
Tauscher, J., Bagby, R. M., Javanmard, M., Christensen, B. K., Kasper, S., & Kapur, S.
(2001). Inverse relationship between serotonin 5-HT 1A receptor binding and anxiety:
A [11C]WAY-100635 PET investigation in healthy volunteers. American Journal of
Psychiatry, 158(8), 1326–1328.
Terracciano, A., Sanna, S., Uda, M., Deiana, B., Usala, G., Busonero, F., … Costa, P. T.
(2008). Genome-wide association scan for five major dimensions of personality.
Molecular Psychiatry, 15(6), 647–656.
Turan, B., Guo, J., Boggiano, M. M., & Bedgood, D. (2014). Dominant, cold, avoidant,
and lonely: Basal testosterone as a biological marker for an interpersonal style. Journal
of Research in Personality, 50, 84–89.
Van Egeren, L. F. (2009). A cybernetic model of global personality traits. Personality and
Social Psychology Review, 13(2), 92–108.
Vazire, S. (2010). Who knows what about a person? The self–other knowledge
asymmetry (SOKA) model. Journal of Personality and Social Psychology, 98(2), 281–
300.
Volkow, N. D., Tomasi, D., Wang, G.-J., Fowler, J. S., Telang, F., Goldstein, R. Z., …
Alexoff, D. (2011). Positive emotionality is associated with baseline metabolism in
orbitofrontal cortex and in regions of the default network. Molecular Psychiatry, 16(8),
818–825.
Vul, E., Harris, C., Winkielman, P., & Pashler, H. (2009). Puzzlingly high correlations in
fMRI studies of emotion, personality, and social cognition. Perspectives on
Wacker, J., Chavanon, M.-L., & Stemmler, G. (2006). Investigating the dopaminergic
basis of extraversion in humans: A multilevel approach. Journal of Personality and
Wacker, J., Chavanon, M.-L., & Stemmler, G. (2010). Resting EEG signatures of agentic
extraversion: New results and meta-analytic integration. Journal of Research in
Personality, 44(2), 167–179.
Wacker, J., & Gatt, J. M. (2010). Resting posterior versus frontal delta/theta EEG activity
is associated with extraversion and the COMT VAL(158)MET polymorphism.
Neuroscience Letters, 478(2), 88–92.
Wacker, J., Mueller, E. M., Hennig, J., & Stemmler, G. (2012). How to consistently link
extraversion and intelligence to the catechol-O-methyltransferase (COMT) gene: On
defining and measuring psychological phenotypes in neurogenetic research. Journal of
Wacker, J., Mueller, E., Pizzagalli, D. A., Hennig, J., & Stemmler, G. (2013). Dopamine-
D2-receptor blockade reverses the association between trait approach motivation and
frontal asymmetry in an approach-motivation context. Psychological Science, 24(4),
489–497.
Wacker, J., & Stemmler, G. (2006). Agentic extraversion modulates the cardiovascular
effects of the dopamine D2 agonist bromocriptine. Psychophysiology, 43(4), 372–381.
Westlye, L. T., Bjørnebekk, A., Grydeland, H., Fjell, A. M., & Walhovd, K. B. (2011).
Linking an anxiety-related personality trait to brain white matter microstructure:
Diffusion tensor imaging and harm avoidance. Archives of General Psychiatry, 68(4),
369–377.
White, T. L., & Depue, R. A. (1999). Differential association of traits of fear and anxiety
with norepinephrine- and dark-induced pupil reactivity. Journal of Personality and
Differences, 30(4), 669–689.
Whittle, S., Allen, N. B., Fornito, A., Lubman, D. I., Simmons, J. G., Pantelis, C., &
Yücel, M. (2009). Variations in cortical folding patterns are related to individual
differences in temperament. Psychiatry Research: Neuroimaging, 172(1), 68–74.
Wiener, N. (1965). Cybernetics or control and communication in the animal and the
machine (2nd ed., p. 212). New York, NY: MIT Press.
Williams, L. M., Brown, K. J., Palmer, D., Liddell, B. J., Kemp, A. H., Olivieri, G., …
Gordon, E. (2006). The mellow years?: Neural basis of improving emotional stability
over age. The Journal of Neuroscience, 26(24), 6422–6430.
Wu, C. C., Samanez-Larkin, G. R., Katovich, K., & Knutson, B. (2014). Affective traits
link to reliable neural markers of incentive anticipation. NeuroImage, 84, 279–289.
Xu, J., & Potenza, M. N. (2012). White matter integrity and Five-Factor personality
measures in healthy adults. NeuroImage, 59(1), 800–807.
Yarkoni, T. (2009). Big correlations in little studies: Inflated fMRI correlations reflect
low statistical power–commentary on Vul et al. (2009). Perspectives on Psychological
Science, 4(3), 294–298.
Yarkoni, T. (2015). Neurobiological substrates of personality: A critical overview. In M.
Mikulincer & P. R. Shaver (Eds.), APA handbook of personality and social
psychology: Vol. 4. Personality processes and individual differences (pp. 61–84).
Washington DC: American Psychological Association.
Yeo, B., Krienen, F., Sepulcre, J., Sabuncu, M., Lashkari, D., Hollinshead, M., …
Buckner, R. L. (2011). The organization of the human cerebral cortex estimated by
intrinsic functional connectivity. Journal of Neurophysiology, 106(3), 1125–1165.
Zhou, Z., Zhu, G., Hariri, A. R., Enoch, M. A., Scott, D., Sinha, R., … Goldman, D.
(2008). Genetic variation in human NPY expression affects stress response and
emotion. Nature, 452(7190), 997–1001.
Zobel, A., Barkow, K., Schulze-Rauschenbach, S., Von Widdern, O., Metten, M.,
Pfeiffer, U., … Maier, W. (2004). High neuroticism and depressive temperament are
associated with dysfunctional regulation of the hypothalamic-pituitary-adrenocortical
system in healthy volunteers. Acta Psychiatrica Scandinavica, 109(5), 392–399.
Zuckerman, M. (2005). Psychobiology of personality (2nd ed., revised and updated). New
SECTION 3
Applications
Assessment of the Five Factor Model
Leonard J. Simms, Trevor F. Williams, and Ericka Nus1
Abstract
We review the current state of the science with respect to the assessment of the
Five Factor Model (FFM), a robust structural model of personality that emerged
from two distinct traditions: The lexical and questionnaire traditions. The lexical
tradition is predicated on the hypothesis that important individual differences in
personality are encoded as single words in language. This bottom-up tradition has
suggested that five broad factors account for much of the personality variation
observed among individuals: Extraversion (or Surgency), Agreeableness,
Conscientiousness (or Dependability), Neuroticism (vs. Emotional Stability), and
Openness to Experience (or Intellect/Culture). The questionnaire tradition
emphasizes the measurement of similar constructs, largely through top-down
development of measures. We examine the strengths and limitations associated
with existing measures of the FFM and related models, focusing on measures
rooted in the lexical and questionnaire traditions. We also consider maladaptive
FFM measures and conclude by analyzing important issues in the FFM
assessment literature.
Key Words: assessment, Five Factor Model, personality, personality psychology,
lexical tradition, questionnaire tradition, extraversion, agreeableness,
conscientiousness, neuroticism
The primary purpose of this chapter is to describe the current state of the
science with respect to the assessment of the Five Factor Model (FFM). First,
however, some historical context and definitions of terms are needed to organize
and unify the discussion. The FFM represents a robust structural model of
personality that emerged from two distinct traditions in personality psychology.
The earlier of these traditions is rooted in Allport and Odbert’s (1936) dictionary
study of American English. This lexical tradition, as it is called, is predicated on
the hypothesis that important individual differences in personality will come to
be encoded as single words in language (Goldberg, 1993). Factor analytic
studies of responses to such person descriptors have resulted in a robust
literature showing that five broad factors account for much of the personality
variation observed among individuals (e.g., Goldberg, 1990, 1993; John &
Srivastava, 1999; see also the chapter by De Raad and Mlačić). These “Big
Five” traits include (I) Extraversion (or Surgency), (II) Agreeableness, (III)
Conscientiousness (or Dependability), (IV) Neuroticism (vs. Emotional
Stability), and (V) Openness to Experience (or Intellect/Culture). Thus, the Big
Five is a bottom-up model derived from structural work with dictionary person
descriptors. In contrast, the questionnaire tradition has contributed to our
understanding and measurement of similar constructs, largely through top-down
development of measures and models of these five broad constructs and their
associated lower-order facets (e.g., McCrae & Costa, 1997, 1999; see also the
chapter by Costa and McCrae). Although differences among these approaches
exist, the similarities in the resulting models far outweigh their differences.
Thus, for our purposes we will organize our description of assessment practices
in this domain around the FFM, broadly construed.
Notably, the FFM is well represented in the assessment research literature and
the psychological research literature more generally. We conducted a search of
PsycINFO to estimate the amount of assessment research that has been focused
on the FFM. First focusing on peer-reviewed journals in which the “assessment”
appears in the journal title, we counted the number of published papers identified
with the search terms “five factor,” “five factor,” “big five,” “FFM,” or “NEO.”
This search returned 506 articles published since 1986. Of course, this search left
out many FFM-related assessment papers that were published in other outlets
focused more generally on psychopathology, personality, and personnel
psychology. Broadening the search by crossing the above search terms with
“assessment” across all peer-reviewed journals resulted in 1,932 published
articles. Finally, searching the same terms without any connection to assessment
resulted in 12,258 articles being identified. Taken together, these data
demonstrate that assessment of the FFM (and research on the FFM more
generally) is a robust activity in the modern psychological literature.
However, despite the robust study of the FFM in the psychological literature,
it is curious to note that practicing psychologists and other mental health
clinicians have not widely adopted measures of the FFM for use in clinical
practice (e.g., Peterson, Lomas, Neukrug, & Bonner, 2014; Piotrowski, 1999;
Smith, Gorske, Wiggins, & Little, 2010; Watkins, Campbell, Nieberding, &
Hallmark, 1995). More on this point will be discussed in the final section of this
chapter. However, regardless of the reasons for the gap between research and
clinical practice, FFM-based measures exist that are based on modern
personality models and that exhibit good psychometric characteristics. To that
end, the primary purpose of this chapter is to describe the strengths and
limitations associated with existing measures of the FFM and related models. By
doing so, we hope to contribute to the education of potential users of FFM
measures for applied work and research, and highlight areas for further study
and development in the FFM assessment literature. The chapter is arranged in
several sections focused on (1) measures rooted in the questionnaire tradition
(those in the “NEO” family of instruments; i.e., related to the NEO Personality
Inventory-Revised of Costa & McCrae, 1992), (2) measures rooted in the lexical
tradition, (3) maladaptive FFM measures, and (4) important issues in the FFM
assessment literature (e.g., unipolarity vs. bipolarity, lack of a consensus facet
structure, and building FFM measures with applied contexts in mind).
The content for this chapter was drawn from an extensive literature review.
We surveyed the literature for instruments assessing the FFM and closely related
models, and then summarized the primary measure features in Table 16.1. The
first column of the table provides the name and citation(s) for each measure’s
most recent version. The second column lists the domains (i.e., higher-order
factors) each measure assesses, as well as the number of facets within each
domain. Many of the remaining columns provide basic information that may be
of interest to test users (e.g., internal consistency); however, the facet derivation,
polarity, and range columns provide somewhat more complex information that
merits further description. For relevant measures, the facet derivation column
lists whether the facet-level structure of an instrument was developed using a
primarily “top-down” or “bottom-up” approach. These determinations were
primarily based on whether an a priori facet-level structure was put forth during
construction (i.e., top-down), as opposed to starting with a broad pool of specific
traits without a predetermined structure (i.e., bottom-up). Although this
dichotomy does not capture the full variability in facet development, we believe
it still serves as a useful heuristic. Next, the range column indicates whether a
measure was intended to measure normal-range or maladaptive personality traits.
Here we describe whether a measure was developed (i.e., intended) to assess one
or the other, recognizing that in many cases a respective instrument assesses
traits with relevance to both normal and abnormal personality. Finally, the
polarity column describes whether the instrument assesses traits that are
considered unipolar (i.e., representing one construct vs. its absence) or bipolar
(i.e., traits with meaningful constructs at either end). These determinations
largely were based on statements made by test developers and are further
discussed in the polarity section.
Questionnaire Tradition: The NEO and Its Derivatives

What is clear from the citation counts in Table 16.1 and, indeed, from any
casual or formal review of the personality assessment literature is that measures
in the NEO family of instruments are the most widely known and used measures
of the FFM. The initial goal was to develop a comprehensive measure of
personality traits based on the state of the literature at that time. However,
although the FFM enjoys a reasonable consensus today, at least at the higher-
order domain level, no such consensus existed in the 1970s and 1980s. When
McCrae and Costa began work on their measure, the literature showed the
clearest consensus for neuroticism and extraversion as broad domains of
personality, and initial scale and facet development focused on these domains.
Through a rational/conceptual process, McCrae and Costa also identified
openness to experience as a broad domain worthy of inclusion. Rational methods
initially were used to map out the facet structure of these three domains and to
develop item pools for each. Item factor analyses of these items were used to
hone the resultant scales and finalize the first NEO Inventory (NEO-I; McCrae
& Costa, 1983), which included six facet scales each to tap these three broad
domains (18 scales total). Notably, the name of the measure is an acronym
representing the first initials of the original domains included in the instrument.
Table 16.1. Summary of Five Factor Model Measures, Organized by Tradition

Measure Availability Facets Items Response
Options
Questionnaire Tradition Measures

NEO-Personality Psychological Assessment Resources 30, Top- 240, 5-point scale
Inventory-3 down Statements (strongly
(NEO-PI-3; disagree
McCrae, Costa, & strongly agree
Martin, 2005)
NEO-Five Factor Psychological Assessment Resources None 60, Statements 5-point scale
Inventory-3 (strongly
(NEO-FFI-3; disagree
McCrae & Costa, strongly agree
2007)
Structured Psychological Assessment Resources 30, Top- 120, 0 = absent

Interview for the down Statements present and does
Five Factor not esult in
Model of significant
Personality dysfunction
(SIFFM; Trull & present and may
Widiger, 1997) result in
significant
dysfunction
International See http://ipip.ori.org/new Multipleconstructs.htm 30, Top- 300, 5-point scale
Personality Item down Statements (very inaccurate
Pool NEO (IPIP- to very accurate
NEO; Goldberg,
1999; Goldberg et
al., 2006)
Five Factor See 30, Top- 30, Adjective 5-point scale
Model Rating http://samppl.psych.purdue.edu/~dbsamuel/research.html down rating scales7 (extremely low
Form (FFMRF; to extemely high
Mullins-Sweatt et
al., 200613)
Five Factor See Few et al. (2010) 30, Top- 30, Adjective 7-point scale
Model Score down rating scales (problematic,
Sheet (FFMSS; very low on trait
Few et al., 20108) to problematic,
very high on
trait)
Faceted Inventory Contact [email protected] 26, 247, 5-point scale
of the Five Factor Bottom-up Statements (strongly
Model (FI-FFM; disagree
Simms, 2009) strongly agree
Five Factor Contact [email protected] 30, 30, 5-point scale
Form (FFF; Top- Adjective (maladaptively
Rojas & down rating scales low to
Widiger, maladaptively
201417) high)
Lexical Tradition Measures:
Big Five See http://www.ocf.berkeley.edu/~johnlab/bfi.php None 44, Statements 5-point scale
Inventory (BFI; (disagree
John, Donahue, & strongly
Kentle, 1991; strongly
John &
Srivastava, 1999)
Goldberg’s See article 100 339, 9-point scale

Synonym Clusters clusters, Adjectives (extremely
(Goldberg, 1990) Bottom-up inaccurate
extremely
accurate)
Goldberg’s Big See article None 100/509, 9-point scale
Five Markers Adjectives (extremely
(Goldberg, 1992) inaccurate
extremely
accurate)
Ten-Item See article None 10 Adjectives 7-point scale
Personality (disagree
Inventory (TIPI; strongly
Gosling et al., strongly
2003)
Saucier’s Big See article None 40, Adjectives 9-point scale

Five Mini (extremely
Markers (Saucier, inaccurate
1994) extremely
accurate)
Interpersonal See http://paultrapnell.com/measures/ 8, Top- 124, 8-point scale

Adjective Scales– down4 Adjectives (extremely
Revised: Big Five inaccurate
Version (IASR- extremely
B5; Trapnell & accurate
Wiggins, 1990)
Big Five Aspect See http://ipip.ori.org/BFASKeys.htm 10, 100, 5-point scale
Scales (BFAS; Bottom-up Adjectives (very inaccurate
DeYoung, Quilty, to very accurate
& Peterson, 2007)
Big Five Giunti O.S. Organizzazioni Speciali 10, Top- 132, 5-point scale
Questionnaire down Statements (very false for
(BFQ; Caprara, me to
Barbaranelli, for me
Borgogni, &
Perugini, 1993)1
Five Factor Swets Test Publishers. See article for details. Calculable 100, 5-point scale
Personality facets12 Statements (not at all
Inventory (FFPI; applicable
Hendriks et al., entirely
1999) applicable
Big Five Modular See article 36, 90, Adjectives 6-point scale
Markers (Saucier, Bottom-up (very inaccurate
2002) to very accurate
International See http://ipip.ori.org/newAB5CKey.htm 45, 484, 5-point scale

Personality Bottom- Statements (very
Item Pool up inaccurate
Abridged Big very accurate
Five-
Dimensional
Circumplex
Model (IPIP-
AB5C;
Goldberg,
199911;
Goldberg et al.,
2006)
Maladaptive/PSY-5 Measures
MMPI-2-RF University of Minnesota Press None 104, True/False
Personality Statements
Psychopathology5
Scales (PSY-5-
RF; Harkness et
al., 2014)
Personality See 25, 220, 5-point scale
Inventory for http://www.psychiatry.org/practice/dsm/dsm5/online- Bottom-up Statements (very false or
DSM-5 (PID-5; assessment-measures#Personality often false
Krueger et al., very true or
2012) often true)
Computerized Contact [email protected] 33, 188 5-point scale
Adaptive Test Bottom- adaptive/216 (very untrue of
of Personality up static, me to
Disorder (CAT- Statements true of me
PD; Simms et
al., 2011;
Wright &
Simms, 2014)
Note: Measures are sorted within descending order of citation counts within the section.
1 There is an updated version of the BFQ (i.e., BFQ-2; Caprara, Barbaranelli, Borgogni & Vecchione 2007)
in Italian.
2 Alternatively can be used as 90 unipolar facets.
3 Actual response anchors are tailored to each item (i.e., are item dependent).
4 IAS-BR has eight subscales related to Dominance and Love.
5 Single-item facets.
6 Mean alpha, rather than median; Trull et al. (1998).
7 Items consist of a rating scale with descriptive adjectives at the extremities.
8 Few et al. (2010) is the first article to focus on measure; Spitzer et al. (2008) represents the first usage of
the measure and has been cited 48 times.
9 The measure is available as 100 adjectives or 50 bipolar rating scale items.
10 These measures use single items to measure facets.
11 Represents the first instance of an AB5C measure; the model was proposed previously (Hofstee et al.,
1992) and has been cited 374 times.
12 Facets can be calculated as combinations of domain scores (e.g., De Fruyt, McCrae, Szirmák, & Nagy,
2004).
13 Based on Mullins-Sweatt et al. (2006), Lynam and Widger (2001) is the first use of FFMRF and has
been cited 254 times.
14 Based on a combination of John et al. (1991) and John and Srivastava (1999).
15 Goldberg (1999) and Goldberg et al. (2006) focus on many measures; therefore articles citing both
sources were searched for the term “NEO” and the resulting counts were combined.
16 Goldberg (1999) and Goldberg et al. (2006) focus on many measures; therefore articles citing both
sources were searched for the term “AB5C” and the resulting counts were combined.
17 Rojas and Widiger (2014) is the first article focused on this measure; Mullins-Sweatt et al. (2010) is first
use of measure and was cited 52 times.
18 Web of Science, for English versions.
Further scouring of the literature identified two additional domains—

agreeableness and conscientiousness—which were added to the instrument in the
next revision, but only at the domain level (NEO-PI; Costa & McCrae, 1985).
Following rational and factor analytic procedures similar to those employed for
the original NEO facets, the facet structure of agreeableness and
conscientiousness later was elaborated and operationalized in the revised NEO
Personality Inventory (NEO-PI-R; Costa & McCrae, 1992). The NEO-PI-R thus
included a complete FFM with exactly six facets for each of the five broad
domains of the FFM—Neuroticism, Extraversion, Openness, Agreeableness, and
Conscientiousness—and it is this version of the NEO (and indeed the FFM more
generally) that has predominated in the FFM literature over the past two
decades. Finally, another minor revision was introduced in 2005 (NEO-PI-3;
McCrae & Costa, 2010; McCrae, Costa, & Martin, 2005) in which 37 items were
modified to reflect simpler, more readable language and to be more appropriate
for use with adolescents in addition to adults. However, in all other respects, the
NEO-PI-3 includes scales and structure identical to the NEO-PI-R. A summary
of the facets associated with each NEO domain is provided in Table 16.2, along
with the facets associated with several other normal-range and maladaptive FFM
measure exemplars.
As noted above, the NEO instruments have been heavily used in research
settings to represent the FFM across a wide range of disciplines, including
clinical settings (e.g., Piedmont, 1998; Singer, 2005), personality disorder
assessment (e.g., Widiger, Trull, Clarkin, Sanderson, & Costa, 2002), industrial–
organizational psychology (Judge & Bono, 2000; LePine, 2003), behavioral
medicine (Trobst, Herbst, Masters, & Costa, 2002; Weiss & Costa, 2005),
educational psychology (e.g., Noftle & Robins, 2007), and behavioral genetics
(e.g., Jang, McCrae, Angleitner, Riemann, & Livesley, 1998). In addition, the
NEO measures have been translated into many other languages, including
Spanish, Chinese, Croatian, Czech, Dutch, Estonian, French, Farsi, Japanese,
Malay, Polish, Portuguese, Russian, Serbian, Slovakian, Korean, Thai, and
Turkish. Moreover, etic cross-cultural studies of the NEO have demonstrated
reasonably strong structural equivalence across cultures and language
translations (McCrae & Costa, 1997).
However, despite the wide use of the NEO instruments, numerous limitations
have been described in the literature. First, the rational approach to the
development of facet constructs and scales likely resulted in a set of scales that
does not cleanly carve personality structure at its joints, at least not at the lower-
order facet level. Other faceted measures to be discussed later in this chapter
show both points of overlap and distinctiveness with the NEO facets. Moreover,
although the symmetry of the measure—i.e., exactly six facets for each of five
domains and exactly eight items to represent each facet—is appealing on a
number of levels, rigid adherence to this symmetry in the scale development
process likely led to limitations in the measure’s structural validity and other
psychometric characteristics (e.g., Church & Burke, 1994). For example,
although all NEO facets correlate most highly with their respective domains,
several facets correlate more highly with facets from another domain than they
do with one or more of the within-domain facets (e.g., Warmth with both
Extraversion and Agreeableness facets and Impulsivity with both Neuroticism
and Conscientiousness facets; McCrae & Costa, 2010; Zuckerman, Kuhlman,
Joireman, Teta, & Kraft, 1993). Similarly, discriminant validity within domains
has been noted as a concern, as some within-factor facet intercorrelations are
higher than might be ideal, especially in the Neuroticism and Conscientiousness
domains, suggesting that some of the facets could be collapsed into a smaller
number of underlying dimensions (McCrae & Costa, 2010). Based on these
considerations, it is quite likely that the FFM facet structure suggested by the
NEO does not reflect personality truth per se, but rather an imperfect rational
sampling of possible facets associated with each domain.
Table 16.2. Examples of Five Factor Model Facets across Normal-Range and
Maladaptive Measures
Normal-Range Measures Maladaptive Meas
Measure NEO-PI-R/3 Goldberg’s Big Five FI-FFM (E. PID-5 (
(McCrae & Costa, AB5C Facets Modular Simms, 2009) et al., 2012
2010) (Goldberg, 1999) Markers
(Saucier, 2002)
Neuroticism vs. Anxiety Stability Fretfulness Anxiety Anxiousness
Emotional
Stability
Angry Hostility Happiness Anxiety Depression Emotional
Lability
Depression Calmness Emotional Anger Proneness Hostility
Excitability
Self-Consciousness Moderation Jealousy/Envy Somatic Perseveration
Complaints
Impulsiveness Toughness Hyperdevotedness Envy Restricted
Affectivity
Vulnerability Impulse Control Separation
Insecurity
Imperturbability Submissiveness
Cool-headedness
Tranquility
Extraversion vs. Warmth Gregariousness Talkativeness Positive Anhedonia

Detachment Temperament
Gregariousness Friendliness Sociability Sociability Depressivity
Assertiveness Assertiveness Assertiveness Ascendance Intimacy
Avoidance
Activity Poise Spontaneity Venturesomeness Suspiciousness
Excitement- Leadership Adventurousness Frankness Withdrawal

Seeking
Positive Emotions Provocativeness Restraint
Self-Disclosure Shyness
Talkativeness
Sociability
Openness/Oddity Fantasy Intellect Analytical Inquiry Intellectance Eccentricity
Aesthetics Ingenuity Reflectiveness Novel Experience Cognitive and
Seeking Perceptual
Feelings Reflection Intellectuality Nontraditionalism Dysregulation
Actions Competence Unconventionality Unusual Beliefs

and
Ideas Quickness Experiences
Values Introspection Emotional
Resonance
Creativity Unusual
Experiences
Imagination Eccentric Beliefs
Depth
Agreeableness vs. Trust Understanding Kindness Empathy Attention
Antagonism Seeking
Straightforwardness Warmth Warmth Trust vs. Callousness
Cynicism
Altruism Morality Sympathy Straightness vs. Deceitfulness
Compliance Pleasantness Agreeableness Manipulativeness Grandiosity
Modesty Empathy Sensitivity Modesty Manipulativeness
Tender-Mindedness Cooperation Toughness
Sympathy Slyness
Tenderness Criticalness
Nurturance Demandingness
Conscientiousness Competence Conscientiousness Efficiency Self-Discipline Distractibility
vs. Disinhibition
Order Efficiency Organization Dutifulness Impulsivity
Dutifulness Dutifulness Perfectionism Deliberation vs. Irresponsibility

Impulsivity
Achievement Purposefulness Decisiveness Achievement Rigid
Striving Striving Perfectionism
Self-Discipline Organization Caution Order Risk Taking
Deliberation Cautiousness Ambition
Rationality Forgetfulness
Perfectionism
Orderliness
Another concern likely rooted in the forced symmetry of the instrument is

inconsistent support for internal consistency across the facet scales. Although
data presented in the NEO manual (McCrae & Costa, 2010) indicate that the
domains demonstrate strong internal consistency (Mdn alpha = .89; range = .87
to .92), the same cannot generally be said of the facet scales (Mdn alpha = .75;
range = .48 to .81). Notably, only two facet scales reach .80, which is a common
threshold suggested by some scale development experts (e.g., Clark & Watson,
1995), and 13 of 30 facets yield alpha coefficients below .70, a threshold deemed
minimally adequate by some experts for exploratory research contexts (e.g.,
Nunnally & Bernstein, 1994). Notably, though, such low alphas will tend to
attenuate validity effects in research settings and increase the size of confidence
intervals in applied contexts. Especially in settings in which point estimates of
traits are used to make important decisions about individuals, data such as these
suggest that NEO domains will yield more trustworthy point estimates than will
NEO facets.
Although the NEO initially was developed as a single measure, since its initial
development, given its utility and popularity, a number of alternative versions
and independent measures have been published that are based on the same
domains and/or facets as the NEO. For example, the NEO instruments include
not only the fully faceted NEO-PI-R and NEO-PI-3 as described above, but a
brief 60-item form focused only on measurement of the FFM domains—the
NEO Five Factor Inventory (NEO-FFI)—also is available. Similarly, the NEO
authors recently published a 120-item short form of the measure that includes
only the first four items from each facet—NEO-PI-3 First Half (NEO-PI-3FH;
McCrae & Costa, 2007)—which balances the desire for efficient measurement
with reasonable coverage of the lower-order facets. Notably, though, McCrae
and Costa (2007) reported expectably lower internal consistencies for the
shortened facet scales (Mdn alpha = .64 in adult participants). However, they
noted strong fidelity with the full-length facet scales (Mdn r = .91), good
evidence of self–other agreement (Mdn r = .44), and structural similarity with
the full-length scales (congruence coefficients ranged from .96 to .98 across the
five domains).
An important limitation of the NEO measures is its proprietary nature, which
makes using it more costly than some users can afford. To that end, Goldberg’s
(1999) International Personality Item Pool (IPIP), composed of a growing
number of public domain personality items and scales, includes numerous scales
that were designed to directly map the NEO domains and facets. The full IPIP-
NEO parallels the NEO-PI-R/III scales and structure with 30 facets scales
measured by 300 items rated on a standardized five-point IPIP response scale (1
= very inaccurate to 5 = very accurate). Goldberg (1999) reported that these
alternative IPIP-NEO scales strongly relate to their NEO counterparts (M
corrected convergent correlation = .94) and are slightly more internally
consistent (M alphas for the IPIP-NEO and NEO scales = .80 and .75,
respectively), likely owing to their longer length. Johnson (2011) also developed
a 120-item short form of the IPIP-NEO facet scales, with alphas that generally
exceed those of the same-sized NEO-PI-3FH measure described above. In
addition to the faceted IPIP-NEO versions, the IPIP offers domain-only NEO
scales in 50- and 100-item variations, with similarly favorable psychometric
properties. Notably, all of these scales are in the public domain and are free to
use.
Observer and Interview Measures of the FFM

In addition to the self-report versions of the NEO measures described above,
measures have been developed to rate the same FFM domains and facets from
the perspective of others. Given the dangers associated with single assessment
methods and shared method variance (Campbell & Fiske, 1959; e.g., inflated
validity coefficients, self-report biases, self-presentation strategies, and making
inferences based on a single source of data), the availability of instruments to
assess NEO domains and facets from multiple perspectives is a clear advantage
and a testament to the large research base existing on the NEO/FFM-based
domains and facets. In fact, the NEO authors themselves acknowledge the
importance of considering non-self-report responses in both applied and research
contexts; to that end, the standard NEO materials come with Form R, which is
worded appropriately to rate others (as opposed to the self-rated Form S that is
most commonly used and reported). Form R is available for both the full NEO-
PI-3 and the nonfaceted NEO-FFI, and psychometric evidence presented in the
manual suggests that it yields roughly equivalent internal consistency compared
to Form S and generates evidence of self–other agreement that is in line with the
broader personality literature (Mdn self–other rs = .44 and .51 in adolescents and
adults, respectively; McCrae & Costa, 2010).
In addition, Trull and Widiger (1997) published the Structured Interview for
the Five Factor Model of Personality (SIFFM), which is a semistructured
interview designed to provide researchers and applied practitioners with a tool to
assess the presence or absence of the domains and facets of the FFM, largely
rooted in the structure provided by the NEO measures. Similar to the rationale
for developing Form R, Trull and Widiger (1997) noted that researchers would
benefit from an interview measure of the FFM to the extent that multiple
methods are useful to distinguish meaningful construct variance from that
attributable to using shared methods across all measures. Similarly, they noted
that applied practitioners often desire interview methods, especially when
measures are used to support psychiatric diagnoses, treatment plans, or other
important recommendations they might be asked to make for a given patient. To
that end, the SIFFM is structured to assess the FFM across a wide range of each
trait and includes items designed specifically to tap maladaptive levels of these
traits.
The SIFFM interview includes 120 items that are proportionally distributed
among the FFM domains and facets such that each facet is represented by four
items and each domain is represented by 24 items. Given these features,
expectably, the domain scores demonstrate stronger internal consistency (Mdn
alpha = .80, range = .72 to .89) than do the facet scales (Mdn alphas = .72, .75,
.46, .49, and .54 across the N, E, O, A, and C domains, respectively; Trull &
Widiger, 1997), a feature that has implications, especially for the applied use of
these scales. Despite these sometimes low internal consistencies, the facet (and
domain) scores generally demonstrate strong test–retest reliability over a 2-week
interval (Mdn rs = .73, .83, .80, .75, and .74 across the N, E, O, A, and C
domains, respectively; Trull & Widiger, 1997). Moreover, the manual reports
uniformly high interrater reliabilities across domains and facets based on a
subset of independent ratings of audiotaped interviews.
Measures from the Lexical Tradition

Adjective-Based Measures
As noted above, the lexical tradition emerged from structural studies of
dictionary person descriptors. Although the lexical hypothesis can be traced all
the way back to Sir Francis Galton, most in the modern trait literature point to
the seminal work of Allport and Odbert (1936) as the starting point for the
studies that ultimately led to the Big Five model of personality.2 Allport and
Odbert scoured the second unabridged edition of Webster’s New International
Dictionary for terms describing personality and behavior, identifying nearly
18,000 such terms and sorting them into four categories: (1) 4,504 personal trait
terms, (2) 4,541 temporary mood or state terms, (3) 5,226 social evaluation
terms, and (4) 3,682 obscure and/or miscellaneous terms that could not be
readily sorted into one of the previous three categories. Notably, Allport and
Odbert (1936) believed that only the first group of terms represented genuine
personality trait descriptors, and it is these terms—as well as an updated set
identified by Norman (1967)—that have served as the foundation for much of
the factor analytic work that followed, leading to the Big Five model and, later,
the FFM.
In the decades that followed publication of these terms, numerous scholars—
including Cattell (1943, 1945), Tupes and Christal (1961), and Norman (1963)—
collected and factor analyzed responses to subsets of these terms, leading to
converging lines of evidence for the Big Five model. Arguably, though, Lew
Goldberg and his colleagues have done the most to solidify the Big Five traits as
a ubiquitous structural model underlying the most basic individual differences in
personality. Goldberg conducted a number of studies developing markers and
measures of the Big Five model. Goldberg (1990) classified 479 trait adjectives
into 133 synonym clusters with the use of dictionaries, synonym finders, and
independent judgments of lexicographers. These then were pared down to 100
clusters measured by 339 trait adjectives. The 339 trait adjectives are rated on a
nine-point scale ranging from 1 (extremely inaccurate as a self-description) to 9
(extremely accurate as a self-description), and scales can be scored for each of
the 100 synonym clusters. Furthermore, as the mean number of adjectives
included in each synonym cluster was 3.4, it is not surprising that the modal
alphas were in the .60 range. Nonetheless, factor analyses of self-rated and peer-
rated responses for these adjectives revealed strong evidence for the Big Five
factors.
Although Goldberg’s (1990) 339 trait adjectives provided good breadth of
content, and the synonym clusters likely provide a reasonably inclusive set of
possible facets underlying each of the Big Five domains, the measure is long and
inefficient, especially if you wish to score only the higher-order domains. As
such, Goldberg (1992) investigated several alternative sets of bipolar and
unipolar trait adjective markers for the Big Five model in multiple samples. The
strongest evidence emerged for a set of unipolar markers. Starting with 566 trait
adjectives, Goldberg iteratively reduced them to a set of 100 adjectives that
displayed a strong pattern of convergent and discriminant validity with respect to
the Big Five factors. The resulting 100 unipolar markers include 20 for each Big
Five domain, largely are balanced in terms of positively and negatively keyed
items, and demonstrate strong levels of internal consistency across multiple
development samples (all alphas were greater than .80 across multiple data sets).
The final measure includes all 100 adjectives, each rated on a nine-point scale
ranging from 1 (extremely inaccurate) to 9 (extremely accurate).
Saucier (1994) extended Goldberg’s work, reducing the 100 unipolar markers
to 40 adjectives—the Big Five Mini-Markers—which exhibit reasonable
psychometric characteristics despite shortening the scales to 40% of their
original length. Moreover, Saucier has been quite active in elaborating the
lexical Big Five personality space (and, indeed, the variable space of all person
descriptors more generally) and developing variations on the measures
previously described, such as the Big Five Modular Markers, Orthogonal-40
Markers, and Mini-Modular Markers, all of which were designed to maximize
orthogonality (see Saucier, 2002, for a summary of these alternative Big Five
marker sets). For comparison purposes, the parcels/facets associated with his
modular markers appear in Table 16.2 alongside the facets of competing Big
Five and FFM measures.
Statement-Based Measures
Notably, the lexical measures reviewed thus far, and indeed the lexical
literature more generally, are based on single adjectives rather than more fully
contextualized and/or behaviorally referent sentences or phrases. However, the
use of single adjectives as items can cause reduced precision and clarity for the
respondent relative to longer statements (e.g., Widiger & Trull, 1997). In
particular, a single word may have variable meanings and be interpreted
differently across individuals, whereas a longer phrase or sentence provides a
greater opportunity for the test developer to clarify the intended meaning of the
item. To that end, some lexically influenced measures have attempted to
translate what we know from the trait adjective literature into measures
composed of longer statements that are more contextually and behaviorally
specific. Notably, Goldberg’s (1999) IPIP has facilitated the translation of trait
adjectives into longer statement-based items.
Among the most prominent of these lexically influenced measures is the IPIP
version of Hofstee, de Raad, and Goldberg’s (1992) Abridged Big Five-
Dimensional Circumplex (AB5C), which is a circular model of the Big Five
personality domain. In the AB5C model, the Big Five domains are crossed in an
elegant series of 10 two-dimensional circumplex models, such that each
circumplex includes two orthogonal axes representing two of the Big Five
domains. Trait facets are then conceptualized as axes at 45-degree increments
around each circle. In the IPIP incarnation of the AB5C model, for example,
extraversion facets are represented by a single axis reflecting “pure” extraversion
(gregariousness in the AB5C model) plus eight facets derived from crossing
extraversion with high or low variants of the remaining four Big Five domains:
Extraversion crossed with (1) high and low agreeableness = friendliness and
provocativeness, respectively, (2) high and low neuroticism = talkativeness and
poise, respectively, (3) high and low conscientiousness = assertiveness and self-
disclosure, respectively, and (4) high and low openness/intellect = leadership and
sociability, respectively. The full IPIP-AB5C measure includes 486 items
tapping 45 facets. A summary of all IPIP-AB5C facets is presented in Table
16.2, organized by FFM domain. Like all IPIP-based scales, the items are rated
on a five-point scale ranging from 1 (very inaccurate) to 5 (very accurate).
Based on responses from the Eugene-Springfield Community Sample, the
IPIP-AB5C scales demonstrate adequate to good internal consistency (Goldberg,
1999; mean alpha = .78 across scales). Structural work on the IPIP-AB5C has
supported its use as a faceted measure of the Big Five (e.g., Bäckström, Larsson,
& Maddux, 2009); however, the facet scales often display significant cross-
loadings, as would be expected given their circumplex origins, but only partially
supporting the particular cross-loadings that would be expected given the
original AB5C model proposed by Hofstee et al. (1992). In addition, the IPIP-
AB5C scales include some item overlap, which will serve to artificially inflate
interscale correlations and influence structural results. Finally, the measure is
limited given its long length. Regardless, it serves as the most elaborated facet
model of the Big Five model, and its circumplex underpinnings provide an
elegant basis for modeling uniform coverage of the Big Five facet space.
Other Short Big Five Scales

In addition to scales mapping the AB5C and NEO-PI-R/III models, the IPIP
also includes 50-item and 100-item public domain scales mapping the Big Five
factors as operationalized by Goldberg’s unipolar markers. Goldberg reported
strong evidence of internal consistency for these scales (M alphas = .84 and .90
for the short and long versions, respectively) and strong convergence with the
unipolar marker scales (M convergent correlations corrected for unreliability =
.81 and .78 for the short and long scales, respectively). Others have developed
lexically informed measures tapping the Big Five domains. The most widely
used of these is the Big Five Inventory (BFI; John, Donahue, & Kentle, 1991;
John & Srivastava, 1999), which was designed to be a brief statement-based
measure of the broad Big Five factors. The BFI includes 44 items (8 to 10 items
per domain) rated on a five-point scale ranging from 1 (strongly disagree) to 5
(strongly agree). Items were developed to reflect elaborated versions of
prototypic trait adjectives for each Big Five domain. For example, as noted by
John and Srivastava (1999), the openness adjective “original” was elaborated for
the BFI to read “is original, comes up with new ideas.” Strengths of the measure
include strong evidence of internal consistency (M alpha = .83 across scales),
convergent validity with respect to several other domain-based measures (M
corrected convergent correlations = .92 and .95 with matched NEO-FFI and
unipolar marker scales, respectively), and strong traction in the research
literature (see Table 16.1).
If the 44-item BFI or the other brief measures described above are too long for
what is needed, other even briefer measures of the Big Five have been
developed. Donnellan, Oswald, Baird, and Lucas (2006) reduced the IPIP Big
Five scales to 20 items total, four for each Big Five factor (IPIP-20). Gosling,
Rentfrow, and Swann (2003) went even further, developing 5-item and 10-item
personality inventories (FIPI and TIPI, respectively) designed to briefly tap each
Big Five domain. Although the psychometrics of these reduced scales clearly are
less favorable than their longer counterparts, they may be minimally adequate
for certain research designs in which there are pressures to reduce total test
battery length. Of course, extremely short measures such as these come with
huge caveats. The reduced measurement precision will attenuate validity
correlations in research studies, which is a salient threat to internal validity. In
applied settings in which psychological measures are used to help make
important decisions about individuals’ lives, the reduced precision of such short
measures likely confers an unacceptable degree of true score uncertainty. Thus,
we do not advocate for the use of such short measures; however, they are
presented here as a service to those who need quick tools for Big Five
assessment and who are willing to tolerate the marked increase in statistical
noise that accompanies them.
Related Lexical Models and Measures

As noted above, the lexical personality literature is rooted in dictionary studies
such as that conducted by Allport and Odbert (1936). However, most structural
studies of such descriptors have focused solely on responses to their first
category terms—personal trait terms—to the exclusion of descriptors reflecting
states, evaluative terms, or other types of person description. Such exclusions
have led some lexical scholars to speculate that additional personality-relevant
factors might exist beyond the Big Five traits. To that end, a literature has
developed examining the impact of using more inclusive sets of person
descriptors as the basis of lexical personality structural work. Doing this has
resulted in a range of models and measures that goes beyond the five broad traits
of the Big Five model. In one strand of this literature, using an inclusive set of
person descriptors led to a seven-factor structure—termed the Big Seven model
by Tellegen and Waller (1987)—that includes factors corresponding closely to
four of the Big Five traits, an unconventionality dimension that is related to but
not isomorphic with traditional conceptualizations of openness/intellect, and
positive and negative valence dimensions that reflect extremely positive (e.g.,
describing yourself as exceptional, important, smart) and negative (e.g.,
describing yourself as evil, immoral, disgusting) self-evaluations, respectively.
The Inventory of Personal Characteristics-7 (IPC-7; Benet & Waller, 1995;
Tellegen, Grove, & Waller, 1991) was developed to measure the Big Seven
model, which has facilitated studies of this alternative lexical structure model.
Notably, others have presented similar findings (e.g., Saucier, 1997), but such
inclusive models are not without their critics (e.g., Ashton & Lee, 2001; McCrae
& Costa, 1995; Widiger, 1993; Widiger & Trull, 1992), who have argued that
the valence dimensions may represent (1) response style dimensions, (2)
artifactual factors due to highly skewed descriptive terms, (3) indices of positive
and negative self-esteem processes, or (4) maladaptive or extreme variants of the
existing Big Five factors, as opposed to personality traits independent of the Big
Five. Research related to these critiques is relatively sparse, but some work has
demonstrated that positive and negative valence can be measured reliably, can be
consensually rated (i.e., moderate levels of self–peer agreement, especially for
negative valence), and significantly improves the prediction of personality
disorder after accounting for traditional Big Five markers (Simms, 2007; Simms,
Yufik, & Gros, 2010; Simms, Zelazny, Yam, & Gros, 2010).
Similar to the Big Seven literature—but coming to different conclusions—is
the HEXACO literature, which has supported a six-factor model and measure
emerging from lexical studies in multiple languages (e.g., Ashton & Lee, 2009;
Ashton et al., 2004; Lee & Ashton, 2004). Specifically, Ashton and Lee have
conducted an impressive number of studies showing that structural analyses of
an inclusive, representative set of personality descriptors result in six broad
factors: Honesty-Humility (H), Emotionality (E), Extraversion (X),
Agreeableness (A), Conscientiousness (C), and Openness to Experience (O). The
differences between the HEXACO and Big Five models largely are a matter of
elaboration and factor analytic rotation. Three of the Big Five factors are similar
to the extraversion, conscientiousness, and openness to experience factors of the
HEXACO model. HEXACO agreeableness and emotional stability, however,
represent rotated variants of their Big Five counterparts. For example, whereas
anger and quick temper are associated with low emotional stability in the Big
Five, these characteristics reflect low Agreeableness in the HEXACO model. In
addition, the HEXACO factor of honesty-humility—which includes facets of
sincerity, fairness, greed avoidance, and modesty—is not directly included
among the Big Five domains (although some aspects are related to Big Five
agreeableness). The HEXACO model can be measured using 60-, 100-, or 200-
item questionnaires that are in the public domain; the 100- and 200-item versions
include four facet scales for each of the six domains. Moreover, a growing body
of research has supported the psychometric characteristic of the measures and
their construct validity vis-à-vis the Big Five and other models of personality
(e.g., Ashton & Lee, 2009; Ashton, Lee, & de Vries, 2014; Lee & Ashton,
2004).
Maladaptive FFM Measures

Although the NEO family of measures initially was developed with the
intention to map and measure individual differences in normal-range personality
traits, a robust literature has emerged showing the value of the NEO and its
component domains and facets as a possible trait-dimensional framework for
organizing and understanding personality disorder (PD). Much of this literature
has focused on studying the NEO domain-level and facet-level profiles
associated with each PD listed in various editions of the Diagnostic and
Statistical Manual of Mental Disorders [DSM; American Psychiatric Association
(APA), 1994, 2000, 2013]. Expert consensus studies (e.g., Lynam & Widiger,
2001) and meta-analytic work (Samuel & Widiger, 2008; Saulsman & Page,
2004) have summarized this literature and revealed clear patterns of relation
among NEO-based FFM traits and DSM-based personality disorders, which
resulted in numerous scholars calling for the FFM to serve as the basis for
personality disorder classification in the recently revised DSM-5 (American
Psychiatric Association, 2013). Although a pathologically flavored version of
the FFM [i.e., the Personality Psychopathology Five (PSY-5); Harkness &
McNulty, 1994] ultimately emerged as the organizing scheme underlying the
experimental PD trait system in DSM-5, there initially was resistance to using
the FFM as a basis for the PD classification. This resistance was complex and
multifaceted; for our purposes here, it is notable that some of this resistance
likely was due to the conflation of the FFM with the NEO family of measures,
which are both (1) proprietary in nature and (2) designed to measure normal-
range personality variation. The latter point is particularly important, since the
NEO scales largely lack enough maladaptive personality variance to serve as a
clinically useful measure (e.g., Haigler & Widiger, 2001).
Maladaptive NEO-Based Measures

In response, scholars interested in advancing the links between the FFM and
personality disorder have developed a series of measures designed to tap both
normal-range and maladaptive personality variation within an FFM context. As
noted above, the SIFFM (Trull & Widiger, 1997) is a semistructured interview
designed to capture the full range of FFM variance. However, this measure has
not been widely used since its initial publication, perhaps in part due to the time
and clinician resources needed to complete it. Thus, Widiger and colleagues
have also developed a series of brief self- and/or clinician-rated measures to
assess the FFM domains and facets as structured in the NEO family of measures.
The Five Factor Model Rating Form (FFMRF; Mullins-Sweatt, Jamerson,
Samuel, Olson, & Widiger, 2006) is the most widely cited of these measures.
The FFMRF is a one-page measure composed of 30 items, one for each facet of
the NEO model. For each facet, the facet name is provided, along with several
descriptors indicative of the high and low poles of each facet. For example, for
the FFM facet of Angry Hostility, “angry, bitter” and “even-tempered” are
provided to represent the high and low poles of the facet, respectively.
Responses are provided on a five-point scale ranging from 1 (extremely low) to 5
(extremely high). Although the brevity of the measure is a huge advantage,
especially for those desiring facet-level information regarding respondents, the
measure has shown limitations in the assessment of maladaptive FFM variation
given that the response format focuses on the degree of traitedness rather than
maladaptivity per se.
To that end, two similarly brief alternatives have been developed. First, the
Five Factor Model Score Sheet (FFMSS; Few et al., 2010; Widiger & Spitzer,
2002) also is a one-page rating sheet consisting of 30 items representing each of
the 30 facets of the NEO model. Identical to the FFMRF, each item includes the
FFM trait name and several descriptors of those high and low on that trait. The
difference between the measures largely is in the response format, which
attempts to disentangle trait extremity from maladaptivity. Each item is rated on
a seven-point scale, with the extreme points focusing on problematically high or
low manifestations of the trait and the middle options reflecting normal-range
variation. However, this measure is limited to the extent that it leaves to the
user’s imagination what normal-range and problematic variants of the facets
might look like. In response, the Five Factor Form (FFF; Rojas & Widiger,
2014) was developed to provide more clarity regarding the normal-range and
maladaptive variants of each FFM facet. The FFF also consists of one item for
each FFM facet, each rated on a five-point scale including the following options:
1 (maladaptive low), 2 (normal low), 3 (neutral), 4 (normal high), and 5
(maladaptive high). In addition, and in contrast to the FFMSS, each item also
includes exemplar descriptors of both the maladaptive and normal-range options.
For example, for the facet of Warmth, 1 = “cold, distant” and 2 = “formal-
reserved” on the low end and 4 = “affectionate, warm” and 5 = “intense
attachments” on the high end. Thus, options 1 and 5 reflect maladaptively low
and high manifestations of warmth, respectively, whereas options 2 and 4 reflect
normal-range variations in warmth.
Although promising as extensions of the FFM to reflect maladaptive variation,
research on the FFMRF, FFMSS, and FFF measures remains relatively limited
relative to research focused on the full NEO measures. Additional work is
needed to evaluate the reliability and construct validity of these measures. In
particular, research is needed to study the presumed bipolarity of the items
within these measures. For example, to what extent do the two FFMRF and four
FFF descriptors for each facet item statistically cohere to form a single
dimension? We return to the issue of unipolarity vs. bipolarity of FFM traits and
scales in a later section.
PSY-5 Measures
In contrast to the maladaptive FFM variants described above, all of which
have remained wedded to the particular domain and facet conceptualizations of
the NEO model, several measures have been developed from the ground up to
reflect the related but maladaptively tinged PSY-5 model (Harkness & McNulty
1994). Until recently, direct measurement of this model was focused on higher-
order scales developed using the item pool of the Minnesota Multiphasic
Personality Inventory-2 (MMPI-2; Butcher et al., 2001; Harkness, McNulty, &
Ben-Porath, 1995) and, more recently, its restructured form (MMPI-2-RF; Ben-
Porath & Tellegen, 2008; Harkness et al., 2014). In short, for the past two
decades, Harkness and colleagues have argued that the PSY-5 model—
composed of traits reflecting negative emotionality/neuroticism and
introversion/low positive emotionality, aggressiveness, disconstraint, and
psychoticism—represents an important review of basic psychological and
neurobehavioral systems useful for organizing both clinical data as well as the
science underlying these traits (Harkness, Reynolds, & Lilienfeld, 2014). Interest
in these dimensions has grown steadily in recent years. However, reliance on the
MMPI tools and the lack of official facet scales3 likely have limited the potential
value of these scales, especially in research settings in which the MMPI is not
routinely part of test batteries.
To that end, two measures have been developed from the ground up to
integrate across multiple models to better represent the maladaptive traits
underlying PD. The first of these—the Personality Inventory for DSM-5 (PID-5;
Krueger, Derringer, Markon, Watson, & Skodol, 2012)—was developed as part
of the official revision for the DSM-5. Deliberations regarding the constructs
thought to be central to PD, by members of the Personality and Personality
Disorders workgroup and their consultants, led to an initial list of 37 lower-order
traits hypothesized to be nested within six higher-order domains. Following data
collection, these then were reduced to 25 traits that load on five higher-order
domains—Negative Affectivity, Detachment, Antagonism, Disinhibition, and
Psychoticism—that strongly resemble the PSY-5 model described above.
Notably, the American Psychiatric Association opted against adopting these 25
traits as the basis for official PD classification in DSM-5; however, to spur
additional research into trait-based alternatives, they included these traits in
Section III of the manual, which is focused on “Emerging Measures and
Models.” Thus, the PID-5 is a useful measure of maladaptive PD traits to the
extent that it directly maps the alternative PD model in DSM-5.
Although a relatively new measure, the PID-5 has been the subject of a
growing literature examining its construct validity and the Section III PD
approach more generally. The measure includes 220 self-report items tapping the
25 lower-order traits (see Table 16.2 for a listing of PID-5 facets, organized by
domain). PID-5 items are rated on a four-point scale ranging from 0 (very false
or often false) to 3 (very true or often true). Krueger et al. (2012) reported
adequate to good internal consistencies based on a U.S. representative sample,
Mdn alpha = .86; range = .72 to .96 across scales. Moreover, accumulating
evidence supports the construct validity of the PID-5 as a broad measure of PD-
relevant traits (Anderson et al., 2013; Hopwood et al., 2012, 2013; Wright,
Pincus, et al., 2012; Wright, Thomas, et al., 2012). However, some concern has
been expressed regarding the adequacy of the measure’s coverage of the full
universe of PD-relevant content (e.g., Simms et al., 2011; Evans & Simms,
under review). Moreover, although this goes beyond the psychometric features
of the PID-5 per se, others have questioned the adequacy of the trait-to-PD
mappings elaborated in Section III of DSM-5 (e.g., Yam & Simms, 2014).
Nonetheless, the PID-5 represents a promising new measure of maladaptive
traits relevant to FFM.
A second measure of PD traits—the Computerized Adaptive Test of
Personality Disorder (CAT-PD; see Simms et al., 2011)—was developed
concurrently to and independent of the PID-5. The CAT-PD project was funded
by the National Institute of Mental Health, with the primary goals of building (1)
an integrative measure of maladaptive traits and (2) an efficient measure to
assess those traits. Similar to the PID-5, early efforts were focused on
explicating the universe of PD-relevant constructs to be considered for the model
and measure. These construct development details are described elsewhere
(Simms et al., 2011). In brief, literature reviews and consultations with PD
experts yielded 59 candidate traits, organized conceptually within a PSY-5
framework. These traits initially were measured by a total of 2,589 items, 1,570
items from the IPIP (Goldberg, 1999; Goldberg et al., 2006) as well as 1,019
new items written to fill construct and severity gaps in the IPIP. Responses to
these items were collected in community (N = 1,268) and patient samples (N =
628). Iterative factor- and content-analytic procedures of these responses
resulted in a final set of 33 traits and scales to tap each (see Table 16.2 for a
listing of CAT-PD facets, organized by domain), including several potentially
important PD-relevant traits not directly taped by the PID-5 (e.g., self-harm,
norm violation, and health anxiety).
In addition, the CAT-PD project resulted in an efficient form of measurement
not utilized to date in the FFM literature: computerized adaptive testing (CAT).
Rooted in item response theory, CAT permits tests to be individually tailored to
individuals such that items selected for administration are based on responses to
previous items in the session. That is, the computer algorithm, aided by known
item characteristics, attempts to identify maximally informative items for a given
person given their estimated trait level. For example, consider a hypothetical
CAT to assess anger. On such a test, if a given individual endorses the item “I
get into lots of fistfights,” then it likely is safe to assume that he or she also
would endorse a less severe anger item such as “I get angry sometimes.” IRT
provides a statistical foundation for grading items in terms of their severity and
information value. CATs use this information, thereby permitting fewer items to
be administered by not presenting items that essentially are irrelevant given what
we iteratively learn about an individual’s trait level.
To facilitate CAT, the full CAT-PD scales are long by design—1,366 total
items; M scale length = 44 items (SD = 12)—of which 188 are administered to
any given examinee. In addition, a 216-item static form (CAT-PD-SF) was
developed using a combination of statistical and content validity considerations
to facilitate quick and standardized assessment across studies. Responses to
items are made on a five-point scale ranging from 1 (very untrue of me) to 5
(very true of me). The static scales demonstrate good internal consistency, Mdn
alphas = .83 and .85 in the community and patient samples, respectively. Only
recently finalized, studies of the CAT-PD measure’s construct validity now are
starting to emerge (e.g., Simms et al., 2011; Wright & Simms, 2014), with others
at various stages of the publication pipeline. Nonetheless, the CAT-PD model
offers an alternative representation of the PD trait space that should be useful as
the literature evaluates the adequacy of the Section III trait model in preparation
for future revisions to the official PD nosology.
Unresolved Issues in FFM Assessment

The measures reviewed thus far reflect the diversity found across the FFM and
Big Five literatures. However, despite some of the conceptual and structural
differences we highlighted across measures, an inescapable conclusion is that
there is a reasonable degree of consensus regarding the broad structure of the
basic individual differences in personality. That said, several important questions
remain in the FFM assessment literature. Here we have elected to discuss several
of these. First, what do we know about the unipolarity vs. bipolarity of FFM
traits? And how well do existing FFM measures reflect these assumptions
regarding polarity? Second, although there is reasonable consensus at the higher-
order domain level, how close are we to a consensual lower-order facet
structure? Finally, although FFM measures are relatively ubiquitous in the
research literature, why have they failed to gain much traction in applied
settings?
Trait and Scale Polarity

An interesting argument often made in the FFM assessment literature is that
FFM domains and facets are inherently bipolar, meaning that both poles of each
trait reflect meaningful but oppositely valenced variation. However, little has
been written to describe exactly how scale bipolarity should be operationalized
or how to study whether such bipolarity is present in a given scale. In our review
of the FFM assessment literature, we attempted to classify extant measures with
respect to their purported unipolarity or bipolarity. To that end, Table 16.1
includes a “Conceptual Polarity” column that refers to whether the scales
generally are considered (e.g., stated by the instrument authors) to measure
bipolar or unipolar dimensions. However, in many cases what is meant by scale
polarity is poorly defined and not thoroughly discussed. This is unfortunate, as
polarity may have important implications for understanding constructs (e.g.,
Russell & Carroll, 1999; Wright & Simms, 2014), their measurement (e.g.,
Polak, de Rooj, & Heiser, 2012), and their use in applied settings (e.g., Krueger
et al., 2011; Samuel, 2011). In this sense, the polarity of a latent dimension, and
the extent to which a scale correctly represents this, represents an aspect of
construct validity (Cronbach & Meehl, 1955).
At the most basic level, a dimension’s polarity depends upon whether both
ends of the continuum can be meaningfully defined and labeled. As such, a
unipolar dimension would have one end representing the presence, or extremity,
of a construct and another end representing the absence (i.e., low-level, or
“zero”) of the same construct. This is most apparent in measures that focus on
pathological personality (Harkness et al., 2014; Krueger et al., 2012; Simms et
al., 2011). In contrast, a bipolar dimension would involve two separate, but
quantitatively linked constructs, with the poles representing the extreme
presence of one construct and the middle of the scale representing the absence of
either construct. For instance, as described above, the FFMRF uses response
options that imply the extreme presence of a construct on each end (i.e.,
extremely high vs. low) and “Neither high nor low” in the middle (Lynam &
Widiger, 2001). This basic definition provides a useful starting point for
discussing polarity.
Going beyond a general definition of polarity, more explicit definitions can be
examined that have implications for what evidence may be brought to bear upon
the polarity of a dimension. One possible definition of polarity is that latent
dimensions assessed by scales with positively and negatively keyed items (or
lower-order scales that together represent a higher-order domain) are bipolar,
and those with items keyed in one direction are unipolar. Thus, evidence for a
bipolar dimension would be that it is measured by a scale with items correlated
both positively and negatively with the scale total score. Many of the measures
in Table 16.1 would be bipolar by this standard. However, such a view of
bipolarity does not specify the constructs at either end of the dimension, requires
inappropriate assumptions regarding item keying (e.g., disagreeing with an
extreme low item implies agreement with an extreme high item; Spector, Van
Katwyk, Brannick, & Chen, 1997), and does not necessarily lead to adequate
measurement of the middle range of a bipolar dimension (i.e., absence of
construct; Chernyshenko, Stark, Drasgow, & Roberts, 2007).
A more rigorous definition of bipolarity would be that both poles of an
underlying trait dimension are theoretically definable, such that a nomological
net can be established around each. In this definition, evidence involves
examining the external correlates of a scale. For a pole to represent a meaningful
construct, it should relate to the distinct presence of conceptually meaningful
thoughts, feelings, and behaviors and not just the absence of such features
related to the high end of the underlying dimension. For example, one possible
manifestation of this definition is that the extremity of a given pole should relate
to impairment and the absence of this extreme is adaptive; thus, bipolar
dimensions would include extreme poles and also would include adaptively high
and low variation in the middle range. Such a definition is relevant to a number
of measures in Table 16.1 (Lynam & Widiger, 2001; Mullins-Sweatt et al., 2010;
Trull & Widiger, 1997; Widiger & Spitzer, 2002) that emphasize maladaptivity
(i.e., impairment) existing at both high and low levels of a trait, although not at
moderate levels.
Despite the implied and/or explicit claims regarding the polarity of many FFM
scales, such polarity claims generally go untested. To that end, research aiming
to understand the polarity of FFM traits and measures is sorely needed. Creative
research designs may be needed for this task. One approach would be to study
whether the opposite poles of purportedly bipolar measures line up empirically
as single dimensions. For example, consider the bipolar items/facets of the
FFMRF described above. Recall that FFMRF items are composed of several
adjectives reflecting opposite poles of a given FFM facet. As such, each item
could be split into two separate items, each one reflecting the high and low ends
of the dimension (e.g., for the warmth facet, the high and low items would be
“cordial, affectionate, attached” and “cold, aloof, indifferent,” respectively). This
approach would lead to a 60-item measure composed of three-adjective clusters
that measure the high and low poles of each facet. Correlational and structural
analyses then could be conducted to determine the extent to which the poles line
up as single scales (or organized accurately within domains) as would be
predicted from the structure of the original measure. In addition, the external
correlates of each pole could be examined independently to support construct
validity (e.g., are opposite poles associated with psychosocial impairment to the
same or different extent). Finally, more sophisticated psychometric models (e.g.,
dominance or ideal-point item response theory models) may serve as useful tools
in the determination of a given scale’s polarity or, perhaps more effectively, in
the development of scales built from the ground up to be bipolar (e.g.,
Chernyshenko et al., 2007; Pettersson et al., 2014; Polak et al., 2012; Spector et
al., 1997; van Schuur & Kiers, 1994).
Overall, the issues raised in this section suggest that researchers can, and
should, consider the polarity of FFM personality traits as an aspect of construct
validation. In particular, as multiple definitions of polarity were noted,
researchers should indicate how they are defining polarity when they refer to this
property of their scales. Doing so will assist other researchers in evaluating their
work. Future research in this area has the potential to enrich our understanding
of FFM domains and facets, increasing theoretical understanding of relations
with external constructs (e.g., impairment), improving measurement scales, and
potentially resolving debates in applied contexts (e.g., Krueger et al., 2011;
Samuel, 2011).
Alternative FFM Facet Models and Measures

As noted above, Table 16.2 lists the facets associated with several normal-
range and maladaptive FFM measures. Although numerous points of
convergence can be seen throughout the table (e.g., facets tapping anxiety or fear
appear across all listed models in the neuroticism domain), other facets are
relatively limited to fewer measures [e.g., the hyperdevotedness facet in
Saucier’s (2002) Big Five Modular Markers] or are inconsistently located and
operationalized across measures (e.g., warmth appears in the extraversion
domain on the NEO, in the agreeableness domain in the lexical measures, and
likely is represented only indirectly in the maladaptive measures). Some scholars
have attempted conceptually and empirically to map the facets associated with
numerous FFM domains, such as conscientiousness (e.g., Roberts, Bogg,
Walton, Chernyshenko, & Stark, 2004) and openness (e.g., Connelly, Ones, &
Chernyshenko, 2014; DeYoung, Quilty, & Peterson, 2007). Others have
developed alternative faceted models and measures of the FFM across all
domains simultaneously, at various levels of abstraction. However, consensus
remains elusive at the lower-order level of all FFM domains.
Intermediate facet measures. Notably, whereas measures such as the NEO
PI-R, AB5C, PID-5, and CAT-PD typically include trait variables at two levels
of abstraction—higher-order domains and lower-order facets assumed to reflect
the same level of narrowness or generality—it is quite possible, perhaps even
likely, that facets exist that vary with respect to their breadth or narrowness. To
that end, several measures have been developed following conceptual and
empirical work revealing facets at an intermediate level between higher-order
domains and lower-order facets. The first of these is the Big Five Questionnaire
(BFQ; Barbaranelli & Caprara, 2002; Caprara et al., 1993), which was created as
a rational bifurcation of each FFM domain. The BFQ includes 132 items
assessing exactly two facets for each FFM domain: dynamism and dominance
for the energy (i.e., extraversion) domain, cooperativeness and politeness for the
friendliness (i.e., agreeableness) domain, scrupulousness and perseverance for
the conscientiousness domain, emotion control and impulse control for the
emotional stability (i.e., neuroticism) domain, and openness to culture and
openness to experiences for the openness domain. These subscales were
developed conceptually based on a literature review, and their appropriateness
for assessing the proper FFM domain was verified through structural analyses.
Notably, however, as each factor is composed of only two facets, multiple
content subdomains appear to be compacted into a single facet in this
instrument. For example, the scrupulousness facet assesses both order and
dependability, and emotion control combines items reflecting both anxiety and
depression. Moreover, the facets scales’ internal consistencies often are lower
than would be ideal (all but two of the 10 facet scales have alphas below .80 and
four of the scales fall below .70), and the measure has not been widely studied
(see Table 16.1).
DeYoung et al. (2007) also identified two intermediate facets within each of
the FFM domains—labeled aspects by the authors—that are operationalized
using the Big Five Aspect Scales (BFAS). The BFAS model was the product of
structural studies of NEO and AB5C facets in the Eugene-Springfield data set,
and scale development followed the results of those structural analyses. The final
BFAS includes 100 items assessing two facets for each FFM domain:
withdrawal and volatility for the neuroticism domain, compassion and politeness
for the agreeableness domain, industriousness and orderliness for the
conscientiousness domain, enthusiasm and assertiveness for the extraversion
domain, and intellect and openness for the openness domain. The psychometric
characteristics of the BFAS were extensively studied by DeYoung et al. (2007),
who reported adequate to strong internal consistencies across aspect scales, good
test–retest reliability, strong evidence of convergent and discriminant validity
vis-à-vis other FFM measures, and promising connections with genetic FFM
factors identified by Jang et al. (2002). Although a promising new measure of
FFM facets, work is needed to validate the measure in independent samples and
in connection with other faceted measures of the FFM. To that end, a recent
structural study using a similar set of measures in the same Eugene-Springfield
sample reported evidence for a more differentiated facet structure within each
domain (Naragon-Gainey & Watson, 2014). Moreover, the utility of an
intermediate level of facet extraction embodied by both the BFAS and BFQ
remains unclear.
A fully faceted FFM alternative. As noted above, among the concerns raised
regarding the NEO version of the FFM is that the particular facets included
reflect a more rational/conceptual than empirical sampling of the facets
underlying each broad domain. Moreover, inconsistencies abound among the
questionnaire-based and lexically-based measures of the FFM. To that end, the
Faceted Inventory of the Five Factor Model (FI-FFM; E. Simms, 2009) was
developed following an extensive literature review and data collection. The
resulting measure includes 247 items assessing 26 lower-order traits organized
within an FFM framework.4 Items are rated on a five-point scale ranging from
strongly disagree to strongly agree. Final scale and item selection decisions
were driven by factor analysis.
The FI-FFM includes five neuroticism facets (anxiety, depression, anger
proneness, somatic complaints, and envy), five extraversion facets (positive
temperament, sociability, ascendance, venturesomeness, and frankness), four
agreeableness facets (empathy, trust vs. cynicism, straightforwardness vs.
manipulativeness, and modesty), five conscientiousness facets (self-discipline,
dutifulness, deliberation vs. impulsivity, achievement striving, and order), and
three openness facets (intellectance, novel experience seeking, and
nontraditionalism). Comparisons between these facets and those of other
prominent FFM measures appear in Table 16.2. In addition, the FI-FFM includes
four interstitial scales that do not cleanly fit within any single FFM domain:
dependency, emotional resonance, unusual experiences, and eccentric beliefs. E.
Simms (2009) reported extensive evidence supporting the FI-FFM scales,
including internal consistency, test–retest reliability, self–other agreement,
convergent and discriminant validity vis-à-vis other measures of the FFM, and
prediction of conceptually matched behaviors.
Taken together with the other faceted measures described in Table 16.2 and
elsewhere in this chapter, consensus regarding the nature and number of facets
underlying the FFM remains elusive. Although some developments point to the
emergence of some consensus in the field (e.g., the marked similarities across
the maladaptive PID-5 and CAT-PD facets) much work remains in this regard.
Notably, much of the facet work has been conducted at the level of the full FFM,
often in the service of building measures. Given the large number of possible
facets underlying the FFM (e.g., Goldberg, 1990; Widiger & Simonsen, 2005),
tackling this issue simultaneously across all FFM domains likely is too
ambitious a task. Rather, the more profitable way forward likely is to study the
lower-order structure of each domain individually and to reach consensus
conceptually or meta-analytically across such domain-specific studies. As noted
above, work such as this has begun in certain domains (e.g., Connelly et al.,
2014; Roberts et al., 2004), but much more is needed.
Clinical Utility
As noted above, an interesting duality exists for FFM assessment. Whereas
the FFM and its measurement are hot topics in the research literature (see the
citation counts listed in Table 16.1), the model’s research traction has not
translated into routine use of the model in applied contexts (e.g., clinical or
industrial–organizational). For example, across surveys of assessment practices
of practicing psychologists, even the NEO (i.e., the most widely known and used
measure of the FFM) is not listed among the measures commonly used in
clinical practice (e.g., Peterson et al., 2014; Piotrowski, 1999; Smith et al., 2010;
Watkins et al., 1995). The disconnect between measures used in clinical practice
and those suggested by the modern empirical literature is notable, but the
reasons for this gap are unclear. One possibility is inertia among psychologists
who traditionally have been trained in the use of assessment methods rooted in
the pre-FFM days, such as the MMPI, Rorschach Inkblot methods, Thematic
Apperception Test, and various sentence completion tasks. Despite the roots of
such measures in antiquated models of personality and psychopathology (and in
some cases questionable construct validity and psychometric characteristics), it
is likely that they enjoy favor among clinicians because these were the measures
covered in their graduate school assessment training. To that end, exposing
budding clinical psychologists to the potential applied benefits of the FFM is one
possible way to close the usage gap.
Another possibility worth exploring is that existing FFM measures may not be
amenable to the needs of practicing clinicians, who often desire measures that
include methods for detecting unreliable or invalid response patterns (i.e.,
validity scales). However, across the measures described in this chapter,
relatively few include methods to gauge test-takers’ engagement in or approach
toward the test. Notably, the NEO measures, which are the most widely studied
and marketed of the FFM measures, do not include any official, dedicated scales
to detect potentially problematic aspects of response distortion.5 And for the
most part, the few FFM measures that do include such methods do so in a very
limited way. For example, the BFQ includes a single “Lie” scale designed to
detect positive impression management. In contrast, no established scales of the
FFM include validity scales designed to detect a full range of possible response
distortions. The MMPI-2 and MMPI-2-RF, in contrast, include a broad range of
validity scales designed to detect inconsistent responding, acquiescence vs.
denial, malingering, and impression management. The only FFM measure listed
in Table 16.1 to include a reasonably broad set of response distortions is the
nascent CAT-PD measure—which includes optional scales to detect inconsistent
responding, positive impression management, and possible malingering—but
these are new and relatively untested.
A final possible reason for the disconnect between research and applied use of
the FFM is that clinical users may not readily see the connection between FFM
constructs and the more idiosyncratic (and perhaps idiographic) constructs that
their clients bring into the consulting room. Although studies show that
clinicians favor the FFM approach when its features and advantages compared to
other methods are described directly to them (e.g., Lowe & Widiger, 2009;
Mullins-Sweatt & Widiger, 2011), the usage gap has not been narrowed
appreciably. The MMPI measures, again, provide a useful point of comparison
here. Although it can be argued that these measures are based, at least partially,
on antiquated notions of personality and psychopathology—and also include
many scales with questionable psychometric characteristics—their continued
popularity among clinicians cannot be ignored. Aside from the inertia
explanation described above, it also is likely that clinicians favor the MMPI (and
similar measures) because substantial didactic materials (e.g., test interpretation
texts, computerized interpretations, workshops geared toward clinicians rather
than researchers, etc.) have been developed to aide clinicians in translating test
results into clinically useful case conceptualizations, treatment plans, and
estimates of prognosis. Although the NEO PI-R’s publisher, Psychological
Assessment Resources, offers some such materials, they pale in comparison to
the rich clinical materials offered in support of the MMPI and other popular
clinical measures. Thus, if the FFM is to become the basis for applied practice,
much more work will be needed from FFM measure authors to translate the
mounds of supporting evidence into measures and materials that are maximally
useful for applied users.
Concluding Remarks
As should be clear from this review, there is no shortage of measures designed
to assess the FFM and related models. These measures share much in common,
but also vary considerably in their origins, structure, length, literature support,
and psychometric features. There is strong support for the FFM as a consensus
structure to represent the broad individual differences in personality. Across the
questionnaire, lexical, and maladaptive trait traditions, four traits in particular—
neuroticism/negative emotionality vs. emotional stability, extraversion/positive
emotionality vs. detachment, agreeableness vs. antagonism, disinhibition vs.
constraint/conscientiousness—appear to reflect a core of broad traits upon which
there is significant agreement. The fifth trait enjoys less consensus. Lexical work
has focused on some combination of intellect, openness, creativity, and
unconventionality as reflecting the fifth broad factor. The NEO-FFM literature
has focused more squarely on open to experience as the core of the fifth factor.
Finally, oddity/psychoticism is the fifth broad domain in maladaptive models
and measures. Work evaluating the extent to which these different operational
definitions of the fifth factor reflect the same or different constructs has revealed
mixed evidence, especially for the relation between oddity and openness (e.g.,
Chmielewski, Bagby, Markon, Ring, & Ryder, 2014; Piedmont et al., 2009;
Samuel & Widiger, 2008; Watson, Clark, & Chmielewski, 2008). Moreover,
studies of more inclusive sets of person descriptors have revealed the possibility
that additional meaningful factors exist outside of the FFM (e.g., Ashton et al.,
2004; Simms, 2007).
Regardless of these debates around the edges of the FFM literature, this
review demonstrates that the FFM is a vibrant, important organizing model of
personality and that many options exist for assessing the FFM in both research
and applied contexts. Increasing the clinical utility of and elaborating a
consensus facet structure of the FFM are the most important targets for future
work.
Notes
1. Ericka Nus formerly was Ericka Simms.
2. Although we elsewhere subsume models and measures of different origin under the
“FFM” umbrella term, in this section we use “Big Five” as appropriate since that is
the term most commonly used by lexical personality theorists.
3. Notably, attempts at post-hoc PSY-5 facet scales have appeared in the literature (e.g.,
Bolinskey, Arnau, Archer, & Handel, 2004), but these have been limited
psychometrically, likely due to limitations with the MMPI-2 item pool (Quilty &
Bagby, 2007).
4. Note that only 22 of 26 scales were deemed to be clean indicators of single FFM
domains (i.e., some scales appear interstitial in nature).
5. Notably, experimental validity scales have been built from the NEO item pool
(Schinka, Kinder, & Kremer, 1997), but these have not gathered much research
support and are not included in the official scoring materials for the measure.
References
Psychological Monographs, 47(1), 1–171.
disorders (4th ed., text revision). Washington, DC: Author.
Anderson, J. L., Sellbom, M., Bagby, R. M., Quilty, L. C., Veltri, C. O. C., Markon, K.
E., & Krueger, R. F. (2013). On the convergence between PSY-5 domains and PID-5
domains and facets: Implications for assessment of DSM-5 personality traits.
Assessment, 20(3), 286–294.
Ashton, M. C., & Lee, K. (2009). The HEXACO-60: A short measure of the major
dimensions of personality. Journal of Personality Assessment, 91(4), 340–345.
Ashton, M. C., Lee, K., & de Vries, R. E. (2014). The HEXACO Honesty-Humility,
Agreeableness, and Emotionality Factors: A review of research and theory. Personality
and Social Psychology Review, 18(2), 139–152.
Ashton, M. C., Lee, K., Perugini, M., Szarota, P., de Vries, R. E., Di Blas, L., … De
Raad, B. (2004). A six-factor structure of personality-descriptive adjectives: Solutions
Psychology, 86(2), 356–366.
Bäckström, M., Larsson, M. R., & Maddux, R. E. (2009). A structural validation of an
inventory based on the abridged five factor circumplex model (AB5C). Journal of
Barbaranelli, C., & Caprara, G. V. (2002). Studies of the Big Five Questionnaire. In B. de
Raad & M. Perugini (Eds.), Big five assessment (pp. 109–124). Ashland, OH: Hogrefe
& Huber.
Ben-Porath, Y. S., & Tellegen, A. (2008). MMPI–2–RF: Manual for administration,
scoring, and interpretation. Minneapolis, MN: University of Minnesota Press.
Benet, V., & Waller, N. G. (1995). The big seven factor model of personality description:
Evidence for its cross-cultural generality in a Spanish sample. Journal of Personality
Bolinskey, P. K., Arnau, R. C., Archer, R. P., & Handel, R. W. (2004). A replication of
the MMPI-A PSY-5 scales and development of facet subscales. Assessment 11(1), 40–
48.
Butcher, J. N., Graham, J. R., Ben-Porath, Y. S., Tellegen, A., Dahlstrom, W. G., &
Kaemmer, B. (2001). Minnesota Multiphasic Personality Inventory–2 (MMPI-2):
Manual for administration and scoring (Rev. ed.). Minneapolis, MN: University of
Minnesota Press.
Campbell, D. T., & Fiske, D. W. (1959). Convergent and discriminant validation by the
multitrait-multimethod matrix. Psychological Bulletin, 56(2), 81–105.
Caprara, G. V., Barbaranelli, C., Borgogni, L., & Perugini, M. (1993). The “Big Five
Questionnaire”: A new questionnaire to assess the five factor model. Personality and
Caprara, G. V., Barbaranelli, C., Borgogni, L., & Vecchione, M. (2007). Big Five
Questionnaire-2: Manuale. Firenze, Italy: Giunti O.S. Organizzazioni Speciali.
Cattell, R. B. (1943). The description of personality: Basic traits resolved into clusters.
Journal of Abnormal and Social Psychology, 38(4), 476–506.
Chernyshenko, O. S., Stark, S., Drasgow, F., & Roberts, B. W. (2007). Constructing
personality scales under the assumptions of an ideal point response process: Toward
increasing the flexibility of personality measures. Psychological Assessment, 19(1),
88–106.
Chmielewski, M., Bagby, R. M., Markon, K. E., Ring, A. J., & Ryder, A. G. (2014).
Church, T. A., & Burke, P. J. (1994). Exploratory and confirmatory tests of the big five
and Tellegen’s three- and four-dimensional models. Journal of Personality and Social
Psychology, 66(1), 93–114.
Clark, L. A., & Watson, D. (1995). Constructing validity: Basic issues in objective scale
development. Psychological Assessment, 7(3), 309–319.
nomological net. Journal of Personality Assessment, 96(1), 1–16.
R) and NEO Five Factor Inventory (NEO-FFI) professional manual. Odessa, FL:
Cronbach, L. J., & Meehl, P. E. (1955). Construct validity in psychological tests.
Psychological Bulletin, 52(4), 281.
De Fruyt, F., McCrae, R. R., Szirmák, Z., & Nagy, J. (2004). The Five Factor Personality
Inventory as a measure of the five factor model: Belgian, American, and Hungarian
comparisons with the NEO-PI-R. Assessment, 11(3), 207–215.
Donnellan, M. B., Oswald, F. L., Baird, B. M., & Lucas, R. E. (2006). The mini-IPIP
scales: Tiny-yet-effective measures of the Big Five factors of personality.
Evans, C., & Simms, L. J. (under review). Comparing between the Section II and Section
III Conceptualizations of Borderline Personality Disorder in DSM-5.
Few, L. R., Miller, J. D., Morse, J. Q., Yaggi, K. E., Reynolds, S. K., & Pilkonis, P. A.
(2010). Examining the reliability and validity of clinician ratings on the Five Factor
Model Score Sheet. Assessment, 17(4), 440–453.
Goldberg, L. R. (1990). An alternative “description of personality”: The Big Five factor
measuring the lower-order level facets of several Five factor models. In I. Mervielde, I.
J. Deary, F. De Fruyt, & F. Ostendorf (Eds.), Personality psychology in Europe (Vol. 7,
pp. 7–28). Tilburg, Netherlands: Tilburg University Press.
public-domain personality measures. Journal of Research in Personality, 40(1), 84–96.
Gosling, S. D., Rentfrow, P. J., & Swann, W. B. (2003). A very brief measure of the Big-
Five personality domains. Journal of Research in Personality, 37(6), 504–528.
Haigler, E. D., & Widiger, T. A. (2001). Experimental manipulation of NEO PI-R items.
Harkness, A. R., & McNulty, J. L. (1994). The Personality Psychopathology Five (PSY–
York, NY: Springer.
Psychopathology Five (PSY–5): Constructs and MMPI–2 scales. Psychological
Assessment, 7(1), 104–114.
Harkness, A. R., McNulty, J. L., Finn, J. A., Reynolds, S. M., Shields, S. M., & Arbisi, P.
(2014). The MMPI–2–RF Personality Psychopathology Five (PSY–5–RF) Scales:
Development and validity research. Journal of Personality Assessment, 96(2), 140–
150.
Harkness, A. R., Reynolds, S. M., & Lilienfeld, S. O. (2014). A review of systems for
psychology and psychiatry: Adaptive systems, Personality Psychopathology Five
(PSY–5), and the DSM–5. Journal of Personality Assessment, 96(2), 121–139.
Hendriks, A. A. J., Hofstee, W. K. B., & De Raad, B. (1999). The Five Factor Personality
Inventory (FFPI). Personality and Individual Differences, 27, 307–325.
Hofstee, W. K. B., de Raad, B., & Goldberg, L. R. (1992). Integration of the big five and
63(1), 146–163.
Hopwood, C. J., Thomas, K. M., Markon, K. E., Wright, A. G. C., & Krueger, R. F.
(2012). DSM-5 personality traits and DSM-IV personality disorders. Journal of
Hopwood, C. J., Wright, A. G. C., Krueger, R. F., Schade, N., Markon, K. E., & Morey,
L. C. (2013). DSM-5 pathological personality traits and the Personality Assessment
Inventory. Assessment, 20(3), 269–285.
Jang, K. L., Livesley, W. J., Angleitner, A., Reimann, R., & Vernon, P. A. (2002).
of the five factor model of personality. Personality and Individual Differences, 33(1),
83–101.
hierarchical model of personality. Journal of Personality and Social Psychology, 74(6),
1556.
John, O. P., & Srivastava, S. (1999). The Big Five Trait taxonomy: History,
measurement, and theoretical perspectives. In L. A. Pervin & O. P. John (Eds.),
Handbook of personality: Theory and research (2nd ed., pp. 102–138). New York, NY:
Guilford Press.
Johnson, J. A. (2011, June). Development of a short form of the IPIP-NEO Personality
Inventory. Poster presented at the 2nd Biennial Meeting of the Association for
Research in Personality, Riverside, CA.
Judge, T. A., & Bono, J. E. (2000). Five Factor model of personality and transformational
leadership. Journal of Applied Psychology, 85(5), 751.
Krueger, R. F., Derringer, R., Markon, K. E., Watson, D., & Skodol, A. E. (2012). Initial
Psychological Medicine, 42(9), 1879–1890.
Krueger, R. F., Eaton, N. R., Clark, L. A., Watson, D., Markon, K. E., Derringer, J. D., …
& Livesley, W. J. (2011). Deriving an empirical structure of personality pathology for
DSM-5. Journal of Personality Disorders, 25(2), 170–191.
Inventory. Multivariate Behavioral Research, 39(2), 329–358.
LePine, J. A. (2003). Team adaptation and postchange performance: Effects of team
composition in terms of members’ cognitive ability and personality. Journal of Applied
Psychology, 88(1), 27.
Lowe, J. R., & Widiger, T. A. (2009). Clinicians’ judgments of clinical utility: A
comparison of the DSM-IV with dimensional models of general personality. Journal of
Lynam, D. R., & Widiger, T. A. (2001). Using the five factor model to represent the
DSM-IV personality disorders: An expert consensus approach. Journal of Abnormal
Psychology, 110(3), 401–412.
McCrae, R. R., & Costa, P. T., Jr. (1983). Joint factors in self-reports and ratings:
Neuroticism, extraversion and openness to experience. Personality and Individual
Differences, 4(3), 245–255.
McCrae, R. R., & Costa, P. T., Jr. (1995). Positive and negative valence within the five
factor model. Journal of Research in Personality, 29(4), 443–460.
McCrae, R. R., & Costa, P. T., Jr. (1999). A five factor theory of personality. In L. A.
Pervin & O. P. John (Eds.), Handbook of personality (2nd ed., pp. 139–153). New
McCrae, R. R., & Costa, P. T., Jr. (2007). Brief versions of the NEO-PI-3. Journal of
McCrae, R. R., & Costa P. T., Jr. (2010). NEO Inventories professional manual. Odessa,
McCrae, R. R., Costa, P. T., Jr., & Martin, T. A. (2005). The NEO-PI-3: A more readable
Revised NEO Personality Inventory. Journal of Personality Assessment, 84(3), 261–
270.
Mullins-Sweatt, S. N., Glover, N. G., Derefinko, K. J., Miller, J. D., & Widiger, T. A.
(2010). The search for the successful psychopath. Journal of Research in Personality,
44, 554–558.
(2006). Psychometric properties of an abbreviated instrument of the five factor model.
Assessment, 13(2), 119–137.
Mullins-Sweatt, S. N., & Widiger, T A. (2011). Clinician’s judgments of the utility of the
DSM-IV and five factor models for personality disordered patients. Journal of
Naragon-Gainey, K., & Watson, D. (2014). Consensually defined facets of personality as
prospective predictors of change in depression symptoms. Assessment, 21(4), 387–403.
93(1), 116–130.
Norman, W. T. (1967). 2800 personality trait descriptors: Normative operating
characteristics for a university population. Ann Arbor, MI: Department of Psychology,
University of Michigan.
Nunnally, J. C., & Bernstein, I. H., (1994). Psychometric theory (3rd ed.). New York,
NY: McGraw-Hill.
Peterson, C. H., Lomas, G. I., Neukrug, E. S., & Bonner, M. W. (2014). Assessment use
by counselors in the United States: Implications for policy and practice. Journal of
Counseling & Development, 92(1), 90–98.
traits? Psychological Assessment, 26(2), 433–446.
applications. New York, NY: Plenum.
Piedmont, R. L., Sherman, M. F., Sherman, N. C., Dy-Liacco, G. S., & Williams, J. E. G.
(2009). Using the Five Factor Model to identify a new personality disorder domain:
The case for experiential permeability. Journal of Personality and Social Psychology,
96(6), 1245–1258.
Piotrowski, C. (1999). Assessment practices in the era of managed care: Current status
and future directions. Journal of Clinical Psychology, 55(7), 787–796.
Polak, M., de Rooj, M., & Heiser, W. J. (2012). A model-free diagnostic for single-
peakedness of item responses using ordered conditional means. Multivariate
Behavioral Research, 47(5), 743–770.
Quilty, L. C., & Bagby, R. M. (2007). Psychometric and structural analysis of the MMPI-
2 Personality Psychopathology-5 (PSY-5) facet subscales. Assessment, 14(4), 375–384.
Roberts, B. W., Bogg, T., Walton, K., Chernyshenko, O., & Stark, S. (2004). A lexical
approach to identifying the lower-order structure of conscientiousness. Journal of
Rojas, S. L., & Widiger, T. A. (2014). Convergent and discriminant validity of the Five
Factor Form. Assessment, 21(2), 143–157.
Russell, J. A., & Carroll, J. M. (1999). On the bipolarity of positive and negative affect.
Samuel, D. B. (2011). Assessing personality in DSM-5: The utility of bipolar constructs.
Samuel, D. B., & Widiger, T. A. (2008). A meta-analytic review of the relationships
between the five factor model and DSM-IV-TR personality disorders: A facet level
analysis. Clinical Psychology Review, 28(8), 1326–1342.
markers. Journal of Personality Assessment, 63(3), 506–516.
Saucier, G. (1997). Effects of variable selection on the factor structure of person
descriptors. Journal of Personality and Social Psychology, 73(6), 1296–1312.
Saucier, G. (2002). Orthogonal markers for orthogonal factors: The case of the big five.
Saulsman, L. M., & Page, A. C. (2004). The five factor model and personality disorder
empirical literature: A meta-analytic review. Clinical Psychology Review, 23(8), 1055–
1085.
Schinka, J. A., Kinder, B. N., & Kremer, T. (1997). Research validity scales for the NEO-
PI-R: Development and initial validation. Journal of Personality Assessment, 68(1),
127–138.
Simms, E. E. (2009). Assessment of the facets of the five factor model: Further
development and validation of a new personality measure. Unpublished doctoral
dissertation, University of Iowa, Iowa City, IA.
Simms, L. J. (2007). The Big Seven model of personality and its relevance to personality
pathology. Journal of Personality, 75(1), 65–94.
CAT-PD project. Journal of Personality Assessment, 93(4), 380–389.
Simms, L. J., Yufik, T., & Gros, D. F. (2010). Incremental validity of positive and
negative valence in predicting personality disorder. Personality Disorders: Theory,
Research, and Treatment, 1(2), 77–86.
Simms, L. J., Zelazny, K., Yam, W. H., & Gros, D. F. (2010). Self-informant agreement
for personality and evaluative person descriptors: Comparing methods for creating
informant measures. European Journal of Personality, 24(3), 207–221.
Singer, J. A. (2005). Personality and psychotherapy: Treating the whole person. New
Smith, S. R., Gorske, T. T., Wiggins, C., & Little, J. A. (2010). Personality assessment
use by clinical neuropsychologists. International Journal of Testing, 10(1), 6–20.
Spector, P. E., Van Katwyk, P. T., Brannick, M. T., & Chen, P. Y. (1997). When two
factors don’t reflect two constructs: How item characteristics can produce artifactual
factors. Journal of Management, 23(5), 659–677.
Spitzer, R. L., First, M. B., Shedler, J., Westen, D., & Skodol, A. (2008). Clinical utility
of five dimensional systems for personality diagnosis: A “consumer preference” study.
Journal of Nervous and Mental Disease, 196(5), 356–374.
Tellegen, A., Grove, W., & Waller, N. G. (1991). Inventory of Personal Characteristics
#7. Unpublished measure, University of Minnesota.
Tellegen, A., & Waller, N. G. (1987, August). Re-examining basic dimensions of natural
language trait descriptors. Paper presented at the 95th annual meeting of the American
Psychological Association, New York.
to include the big five of personality. Journal of Personality and Social Psychology,
59(4), 781–790.
Trobst, K. K., Herbst, J. H., Masters, H. L., III, & Costa, P. T., Jr. (2002). Personality
pathways to unsafe sex: Personality, condom use, and HIV risk behaviors. Journal of
Trull, T. J., & Widiger, T. A. (1997). Structured interview for the Five Factor Model of
Personality. Odessa, FL: Psychological Assessment Resources.
Trull, T. J., Widiger, T. A., Useda, J. D., Holcomb, J., Doan, B., Axelrod, S. R., …
Gershuny, B. S. (1998). A structured interview for the assessment of the five factor
model of personality. Psychological Assessment, 10(3), 229–240.
Tupes, E. C., & Christal, R. E. (1961). Recurrent personality factors based on trait
ratings (USAF ASD Tech. Rep. No. 61-97). Lackland Airforce Base, TX: U. S. Air
Force.
van Schuur, W. H., & Kiers, H. A. L. (1994). Why factor analysis often is the incorrect
model for analyzing bipolar concepts, and what model to use instead. Applied
Psychological Measurement, 18(2), 97–110.
Watkins, C. E., Campbell, V. L., Nieberding, R., & Hallmark, R. (1995). Contemporary
practice of psychological assessment by clinical psychologists. Professional
Psychology: Research and Practice, 26(1), 54–60.
structure. Journal of Personality, 76(6), 1545–1586.
mortality among Medicare patients aged 65 to 100. Psychosomatic Medicine, 67(5),
724–733.
Widiger, T. A. (1993). The DSM-III-R categorical personality disorder diagnoses: A
critique and an alternative. Psychological Inquiry, 4(2), 75–90.
Widiger, T. A., & Spitzer, R. L. (2002). Five Factor Model Score Sheet. Unpublished
measure.
Widiger, T. A., & Trull, T. J. (1992). Personality and psychopathology: An application of
the Five Factor Model. Journal of Personality, 60(2), 363–393.
Widiger, T. A., & Trull, T. J. (1997). Assessment of the five factor model of personality.
disorder: Finding a common ground. Journal of Personality Disorders, 19(2), 110–130.
Widiger, T. A., Trull, T. J., Clarkin, J. F., Sanderson, C., & Costa, P. T., Jr. (2002). A
description of the DSM-IV personality disorders with the five factor model of
personality. In P. T. Costa Jr. & T. A. Widiger (Eds.), Personality disorders and the
five factor model of personality (2nd ed., pp. 89–99). Washington, DC: American
Wright, A. G. C., Pincus, A. L., Hopwood, C. J., Thomas, K. M., Markon, K. E., &
Krueger, R. F. (2012). An interpersonal analysis of pathological personality traits in
DSM-5. Assessment, 19(3), 263–275.
Conjoint analyses of the CAT-PD, PID-5, NEO-PI-3 trait models. Personality
traits. Journal of Abnormal Psychology, 121(4), 951–957.
Yam, W. H., & Simms, L. J. (2014). Comparing the criterion- and trait-based personality
disorder diagnostic systems in DSM-5. Journal of Abnormal Psychology, 123(4), 802–
808.
Zuckerman, M., Kuhlman, D. M., Joireman, J., Teta, P., & Kraft, M. (1993). A
comparison of three structural models for personality: The Big Three, the Big Five, and
the Alternative Five. Journal of Personality and Social Psychology, 65(4), 757–768.
The Five Factor Model of Personality in Business and
Industry
Scott E. Seibert and David S. DeGeest
Abstract
Personality traits have played a central role in industrial/organizational
psychology, human resource management, and organizational behavior, the key
fields in the application of psychology to business and industry. In the early years,
excessive optimism led scholars to unrealistic expectations about the value of
personality traits at work. This was followed by a period of profound pessimism
regarding the value of personality as an explanatory variable when the unrealistic
expectations were inevitably disappointed. More recently, advances in theory and
methodology have led scholars to re-examine the role of personality with more
realistic expectations. The Five Factor Model (FFM) has predominated as an
integrative personality structure for conceptualizing and researching the
relationship of personality to workplace outcomes. Five specific domains of
research are considered herein: personnel selection; employee motivation,
attitudes, and behavior; leadership; teams; and entrepreneurship. The chapter ends
with open questions for future research in this domain.
Key Words: personality traits, Five Factor Model, industry, organizational
behavior, human resources, personnel selection, leadership, teams,
entrepreneurship
Individual differences have played a central role in industrial/organizational

(I/O) psychology, human resource management, and organizational behavior, the
key fields interested in the application of psychology to business and industry.
Interest in personality variables has followed a recognizable pattern in each area,
even if the initial dates and time frames over which the pattern unfolds vary. In
the early years, scholars expected their research to provide an unambiguous
personality profile that distinguishes those who will excel at the phenomenon of
interest and those who will fail, whether it is job performance, leadership,
entrepreneurship, or any other outcome. Pessimism regarding the value of
personality as explanatory variables sets in when these unrealistic expectations
are inevitably not fulfilled, leading to the neglect of personality variables as
researchers move on to seemingly more promising theories and approaches.
Finally, some scholars return to reexamine the role of personality in the
phenomenon of interest with realistic expectations, tempered perhaps by modest
results from other research avenues, indicating that personality traits are a factor
that contribute to effective functioning without providing a complete account of
the phenomenon.
Several critical developments have encouraged the resurgence of interest in
the study of individual differences, and more specifically personality, in
workplace behavior and phenomena. First, personality psychologists verified the
development of the Five Factor Model (FFM) of personality (Costa & McCrae,
1992). These developments will by now be familiar to readers of this text, so we
will spend little time recounting them here. Second, scholars developed
sophisticated methods of statistical analysis, such as meta-analysis (Hunter &
Schmidt, 2004) and structural equation modeling (Anderson & Gerbing, 1988),
that allow us to draw more accurate conclusions regarding the effects of
variables, including personality variables. Often it is not the magnitude of the
observed effect that has changed over time, but the consistency or reliability
with which we can speak about the results with these new techniques. Third,
empirical findings regarding within-person stability over long periods of time
and multiple settings renewed theoretical interest in individual differences, rather
than contextual variables as explanatory constructs (Arvey, Bouchard, Segal, &
Abraham, 1989; Barrick & Mount, 1991; Scarr, Webber, Weinberg, & Wittig,
1981; Staw & Ross, 1985). This paved the way for more rigorous investigations
of how personality might influence workplace outcomes that continue today.
As this brief review suggests, developments in social and personality
psychology have been a critical driver of developments in the role of personality
in business and industry. The fields of industrial/organizational (I/O)
psychology, organizational behavior (OB), and human resource management
(HR) are rooted in the source disciplines of psychology (among others), and
therefore it is not surprising that they have closely followed these developments.
Five specific domains of research within the IO/OB/HR field have been most
concerned with personality effects: personnel selection; employee motivation,
attitudes, and behavior; leadership; teams; and entrepreneurship and careers. In
the sections below, we provide a narrative review of trends and findings related
to personality research in each domain, then evaluate and develop critical
questions that remain for future personality research in the IO/OB/HR domains.
Personnel Selection/Job Performance

Personnel selection is the process used to recruit, select, and hire individuals
who have the knowledge, skills, and abilities and other characteristics to perform
the jobs in which they are placed successfully (Hough & Oswald, 2000). The
identification of factors that predict the likely job performance of the individual
is the core of a personnel selection system. Selection tools typically include
ability tests (e.g., cognitive, physical, and psychomotor), knowledge tests,
structured interviews, biographical data, work samples, and personality tests
(Rynes, Brown, & Colbert, 2002). There is a robust literature on the relationship
between personality and job performance, and the utility of personality as a
selection tool in industry has often been contested. Three significant phases can
be identified in the evaluation of personality in selection contexts. The first
phase ranges from the beginnings of modern psychology in the early 1900s,
when a great deal of optimism for personality as a selection tool was first
manifest, to the mid-1980s, when pessimism prevailed. The second phase,
ranging from the mid-1980s to the late 2000s, saw a resurgence of interest in
personality in no small part due to the development of the Big Five Model of
personality and its application to the selection context. The third phase,
beginning roughly in the new century and continuing to this day, involves a
return to the critiques of the earlier era and the modern responses these critiques
have engendered. Although there is little doubt now in the literature that
personality, and in particular, conscientiousness, has a demonstrable,
generalized, positive relationship with job performance (see also the chapter by
Jackson and Roberts, this volume), many questions still remain, and the
influence of personality on work behaviors and job performance remains a
robust area of research and debate.
One of the earliest and most persistent applications of differential psychology
in industry was to problems of personnel selection and placement. These issues
arose in relation to the needs of the Army to select and place soldiers during
World War I (Landy, 1997), but the scale of the selection and placement
problem only grew as veterans returned from World War II. Although the
original focus was on mental measurements or intelligence, by the 1950s
personality testing for selection was a widespread practice in industry (Spriegel
& Dale, 1953), reflecting a growing interest not only in individual ability (what
the individual “can do”), but in individual motivation as well (what the
individual “will do”) (Guion & Gottier, 1965). The first phase of the research
literature on personality therefore focused on the need to identify valid
personality instruments that could be used to predict employee performance.
Typically, studies were conducted that estimated the relationship among
individual scales from a number of different personality inventories and various
dimensions and operationalizations of job performance (Barrick, Mount, &
Judge, 2001). However, after years of hopeful searching, researchers drew the
general conclusion that personality had little relationship to job performance. A
narrative review of the literature by Guion and Gottier (1965) had a major
impact on the direction of the field. Finding that only 12% of the studies they
reviewed yielded a prediction of performance significantly different from zero,
they concluded that no generalizable evidence existed to suggest that personality
inventories could be used as good or practical tools for selection purposes. The
reasons for this pessimistic conclusion arose from a number of factors, including
poor research design in the primary studies, issues with construct deficiency and
contamination in the measurement of personality traits, the lack of clear
theoretical logic linking traits to performance or a general theory of work
performance, little emphasis on replication, the lack of an organizing paradigm
that might reveal consistent relationships of personality traits to performance,
and the lack of a psychometrically sound way to quantitatively summarize the
results of primary studies. These problems made it difficult to identify
relationships and advance our understanding of the relationship between
personality and work behavior. This conclusion led to a more widespread
assumption in the I/O scholarly community that personality had little to do with
workplace behavior and future research should focus on other types of selection
tools.
Two major developments in the mid-1980s rose to challenge this assumption
and opened new possibilities for research on personality in the work domain.
The first development was the scholarly consensus that formed around the Five
Factor Model (FFM) of personality as a useful way to organize the plethora of
personality scales and inventories within a relatively parsimonious set of
constructs (Costa & McCrae, 1980; Digman, 1990; McCrae & Costa, 1987).
While debate continued over the lower-order factors that should be included in
such a higher-order model (Goldberg, 1993), research has consistently identified
extraversion, agreeableness, conscientiousness, neuroticism, and oppenness as
five key personality constructs under which we can organize the wide variety of
lower-level facet measures. Most research in the OB/HR domain today uses
measures of personality that either explicitly measure one of these higher-order
constructs directly or uses the FFM framework to categorize scales (Barrick et
al., 2001). The second major development was the widespread acceptance and
use of psychometric meta-analytic techniques, which allowed researchers to
address the relationship between personality and job performance in a way that
no primary study could: by estimating the personality–performance relationship
and account for both sampling error and statistical and measurement artifacts
that biased correlations downward (DeGeest & Schmidt, 2010; Hunter &
Schmidt, 2004). By combining the advances in statistical analysis with the
development of the FFM, both Barrick and Mount (1991) and Tett, Jackson, and
Rothstein (1991) were able to make significant strides in challenging the long-
held assumption that personality had little to do with workplace behavior,
reinvigorating research on personality in the field of organizational psychology.
Perhaps the two most significant studies of this time period were Barrick and
Mount’s (1991) meta-analysis, which showed that conscientiousness had a
generalized, positive relationship to job performance, and the meta-analysis of
Tett et al. (1991), which produced similar results. During the period from 1990
to 2006, 16 meta-analytic studies of the relationship between personality and job
performance were conducted (Rothstein & Goffin, 2006), all of which used the
FFM to organize studies and the personality traits. Barrick et al. (2001)
conducted a second-order analysis of 11 unique meta-analyses and confirmed
that personality, and in particular the conscientiousness trait, is significantly
related to job performance with a corrected multiple correlation as high as .30. In
addition, other meta-analyses showed that personality had relationships to other
important job-related workplace behavior such as citizenship behaviors (LePine,
Erez, & Johnson, 2002), counterproductive work behaviors (Berry, Ones, &
Sackett, 2007; Salgado, 2002), and job performance in other non-U.S. contexts
(Salgado, 1997). Finally, other studies showed that the validity of personality
traits in different occupational contexts may change; for example, Mount,
Barrick, and Stewart (1998) demonstrated that the meta-analytic relationships of
extraversion and agreeableness with job performance were relatively strong in
service-oriented occupations but weaker in others. Finally, research on integrity,
a compound personality trait consisting of conscientiousness, agreeableness, and
emotional stability, showed considerable predictive validity for both job
performance and counterproductive behaviors (Ones, Viswesvaran, & Schmidt,
1993, 2003). In general, meta-analyses during this time generally demonstrated a
consistent if somewhat modest relationship between personality and job
performance.
Although these meta-analytic reports generally show support for personality
as a predictor of job performance, the results of both Barrick and Mount (1991)
and Tett et al. (1991) showed substantial variability according to type of
personality measure, occupational category, and validation strategies used.
Therefore, during this period a number of primary studies were conducted that
focused on identifying occupational and environmental characteristics that
increase the validity of personality traits as a predictor of performance. For
example, Hogan and Holland (2003) noted that extraversion, a trait associated
with assertiveness, sociality, and gregariousness, was an important predictor of
performance in jobs in which “getting ahead” of others was an important
outcome of the job. Barrick et al. (2001) found that extraversion predicted job
performance in sales (ρ = .15) and management jobs (ρ = .18). Similarly, in jobs
in which working on a team is a significant component of performance,
agreeableness has a positive relationship with performance (Bell, 2007; Mount et
al., 1998; Peeters, Van Tuijl, Rutte, & Reymen, 2006). Limited evidence also
supports the idea that when job performance requires you to learn new
information or skills, openness can be a positive predictor of performance
(LePine, Colquitt, & Erez, 2000; Pulakos et al., 2002).
Other primary studies in this era focused on addressing limitations in
personality testing in high-stakes testing situations. These include laboratory and
field studies that have investigated the role of faking in personality tests (Barrick
& Mount, 1996; Ellingson, Sackett, & Connelly, 2007; Hogan, Barrett, &
Hogan, 2007; Ones, Dilchert, Viswesvaran, & Judge, 2007; Schmitt & Oswald,
2006). These studies have generally shown that although impression
management does not appear to have an effect on the predictive validity of
personality tests, individuals in laboratory settings can clearly alter their scores
to “fake good” on conventional self-report personality inventories (Griffith &
Peterson, 2006). Other studies investigated the degree to which social
desirability may influence the predictive validity of the FFM traits (Tett et al.,
2006; Viswesvaran, Ones, & Hough, 2001). A handful of studies also
investigated the use of observer ratings of personality (e.g., Mount, Barrick, &
Strauss, 1994). Overall, the studies during this period suggested that the
relationship between personality and performance was contingent on a number
of factors, but generally played a role in job performance and other workplace
behaviors affecting the functioning of the organization.
A third phase in the research on the utility of personality in selection contexts
began as a result of a series of critical papers published in Personnel Psychology
over the past decade. Although the critical perspective questions the value of
personality as a selection tool, these scholars rely on the FFM framework in
making their critique, recognizing the FFM as the dominant paradigm for
personality research in organizational psychology. For example, Morgeson et al.
(2007a, 2007b) reviewed the meta-analytic and primary literature from the 1990s
through 2006 using the FFM framework to draw conclusions regarding the
relationship of personality to job performance. However, unlike previous
scholars, they came to several relatively pessimistic conclusions regarding the
use of self-report personality testing in selection contexts, which can be
summarized around three points. First, the critique notes that the uncorrected
validities of personality traits tended to be quite low and have not improved over
time, with the uncorrected correlations ranging from .03 to .15 (Hurtz &
Donovan, 2000). Rather, corrections for measurement error and other statistical
artifacts have grown larger, causing corrected validities to increase to as high as
.31 (Barrick et al., 2001). Morgeson et al. (2007b) note that when organizations
use personality tests, they must use observed scores, not corrected scores, so
corrections for lack of reliability in the personality variables are not appropriate.
A fair consensus has emerged in the literature that corrections for unreliability in
the dependent variable and corrections for range restriction when data are
available indicating that such restriction has occurred are appropriate, rendering
what is referred to as operational validity—the validity of the instrument as used
in the practical context of selection. Following Barrick and Mount (1991), many
meta-analyses have made corrections for construct invalidity—imperfect
correlation between the lower order traits used in the primary studies and the
global FFM traits. A cleaner design, without the necessity of estimates, would
examine the validity for direct measures of the FFM traits alone. Hurtz and
Donovan (2000) conducted a meta-analysis using only direct measures of the
FFM constructs and the corrections for sampling error, range restriction, and
criterion reliability. They reported operational validities ranging from .06 to .20,
with conscientiousness (ρ = .20) the strongest predictor and emotional stability
(ρ = .13) also producing a small but reliable result. Note that there is little
disagreement in the literature regarding the size of the personality effect (Barrick
et al., 2001; Hurtz & Donovan, 2000). Rather, the disagreement centers on
whether effects of this size have practical utility in selection contexts.
A second critique raised by Morgeson et al. (2007a, 2007b) is that individuals
clearly have the ability to fake (i.e., artificially inflate) their scores on self-report
personality inventories. Although the effects of social desirability have long
been a concern in the personality domain, intentional faking of responses may be
much more likely and more problematic in high-stakes situations, such as
applying for a job. However, the authors of the critique, chosen because of their
roles as journal editors, did not come to a consensus on the degree to which
faking is problematic in such high-stakes situations. The review also noted that
although faking may not change the operational validity of the measures, it may
change other characteristics of personality measures, such as the
intercorrelations among the traits.
The third problem with the use of personality measures in selection contexts,
including direct measures of the FFM, is that applicants tend to have negative
reactions to personality tests (Jayne & Rauschenberger, 2000). This effect may
occur due to low face validity and questionable content validity of some
commercially available self-report personality measures. Because negative
applicant reactions can drive away the very applicants an organization wishes to
attract, the utility of personality scales in practice may be less than implied even
by operational validity coefficients. Based on these considerations, these authors
recommended using selection tools with at least as much validity but associated
with less negative applicant reactions, such as cognitive tests, work samples, and
assessments of past performance (Rynes, Colbert, & Brown, 2002). Fourth and
finally, future research on personality in selection contexts should focus either
on criteria that are more strongly related to personality (e.g., job attitudes,
leadership, counterproductive work behaviors, contextual performance) or
develop alternative ways to operationalize personality constructs aside from self-
report personality measures. The result of this somewhat pessimistic review of
the value of personality testing in selection settings has changed the subsequent
focus of research on personality to address the concerns raised in this review and
critique.
Since the publications of Morgeson et al. (2007a, 2007b), a number of
researchers have crafted responses to move the literature on personality in the
workplace forward. For example, one of the recommendations to come out of the
review and the discussion in Morgeson et al. (2007a, 2007b) concerned different
operationalizations of personality, including observer ratings of the FFM traits.
Oh, Wang, and Mount (2011) completed work on a meta-analysis of observer
ratings of personality, finding that the predictive validities for observer ratings
were significantly higher than those of self-reports and, in addition, showing that
observer ratings provide incremental validity over self-reports of personality. In
a similar vein, Morgeson et al. (2007a, 2007b) also discussed the use of
situation-specific personality inventories. In these types of formats, question
stems typically include a phrase such as “at work” to indicate a situation-specific
manifestation of personality. A recent meta-analysis (Shaffer & Postlethwaite,
2012) showed that the use of contextualized, situation-specific FFM measures
changed the mean validity across all traits from a corrected value of .11 to a
corrected value of .24. Conscientiousness still had the largest effect, showing a
corrected correlation of .30 with performance as opposed to the .22 of
noncontextualized measures. The contextualized measures of emotional stability
(ρ = .27, contextualized; ρ = .11, noncontextualized) and extraversion (ρ = .25,
contextualized; ρ = .08, noncontextualized) also had much higher correlations.
Finally, the recommendations of Morgeson et al. (2007a, 2007b) included
focusing on domains other than overall measures of job performance, where
personality traits may have more influence. Recent meta-analyses have done so,
focusing on behaviors such as citizenship behaviors (Chiaburu, Oh, Berry, Li, &
Gardner, 2011), adaptive performance (Huang, Ryan, Zabel, & Palmer, 2014), or
on situational moderators that influence the relationship between
conscientiousness and task performance (Meyer, Dalal, & Bonaccio, 2009).
Most recently, Neal, Yeo, Koy, and Xiao (2012) demonstrated that although
conscientiousness has relatively consistent effects on task, adaptive and
proactive job performance, characteristics such as openness and agreeableness
have differential and sometimes countervailing effects on these different
dimensions of performance. For example, in this study, openness had positive
effects on proactivity (β = .14) whereas agreeableness had negative effects (β =
–.13).
Although situation-specific and observer ratings show potential, they also
raise questions about practicality. For example, it remains unclear how an
organization would acquire a large number of observer ratings of applicant
personality or workplace-based personality since job applicants would not be
working in the organization at the time they apply. Such ratings might be used
internally within an organization for the purpose of promotion decisions, but in
this case it is unclear why an organization would rely on observer ratings of
personality when supervisor and peer ratings of the employees’ actual
performance are available, are likely to have higher validities (Rynes et al.,
2000), and are likely to generate less negative applicant reactions.
Other research responding to the critique has focused on the way the effects of
personality on performance may change over time. For example, Lievens, Ones,
and Dilchert (2009) showed that the relationship between conscientiousness and
job performance for medical students changed over time, generally increasing as
the students’ responsibilities shifted from learning new knowledge to managing
interpersonal relationships with co-workers and managing the care of patients.
Similarly, a study conducted by Thoresen, Bradley, Bliese, and Thoresen (2004)
showed in two samples of sales managers that although only conscientiousness
and extraversion were associated with between-person differences in overall
sales, both conscientiousness and other personality traits were associated with
growth in sales over time. Future research can investigate the degree to which
personality may be associated not simply with overall levels in performance, but
with changes in performance trajectories over time. Organizations may care
about these issues because time to proficiency in a position or increases in
proficiency over time can matter to the overall productivity of an individual
(Levin & Stephan, 1991; Zyphur, Chaturvedi, & Arvey, 2008).
In conclusion, the FFM has made major contributions to the field of personnel
selection and job performance. It has become the dominant personality
framework for personnel selection (Hurtz & Donovan, 2000). As an organizing
framework, the FFM has made it possible to cumulate and communicate
consistent conclusions from the literature regarding the validity of the
personality traits as predictors of job performance and has changed previous
conclusions in the field regarding the utility of traits for personnel selection. The
most important substantive conclusion is that conscientiousness is the best
predictor of what an employee “will do” at work in terms of job performance. At
the same time, controversies remain regarding the utility of personality as a
personnel selection tool for organizations. But even here, the FFM has framed
the debate. Further investigation into when, how, and why personality influences
work behavior remains a continued area of research.
Employee Job Attitudes, Motivation, and Behavior

Although the literature on personality and job performance is focused on the
use of personality constructs for personnel selection purposes, there is also a
significant literature on the relationship between personality and job attitudes
and other work-related motivational states and behaviors. This literature is
focused on a theoretical understanding of the behavior of employees at work and
on explaining the way personality influences employee behavior, with the
ultimate goal of managing employees to achieve organizational and personal
goals. Again, the FFM plays an important role in providing a framework for
theorizing as well as organizing and communicating accumulating results. The
literature on the FFM and job attitudes has primarily focused on three principal
areas: job satisfaction, organizational commitment, and work-related motivation.
Typically, theoretical models link personality to a job attitude, which is in turn
linked to behavioral intentions, which are then in turn linked to workplace
behaviors such as performance or turnover. In these models, theoretical
perspectives such as the theory of the planned behavior (Ajzen, 1991) often
provide a theoretical framework, although new theories have begun to emerge.
The earliest research on the relationship between personality and job attitudes
suggested there was at least a modest dispositional component to job attitudes
(e.g., Hoppock, 1935). However, following the trends in the literature on
personality and performance, research on the relationship between personality
and job attitudes waned in the 1970s and 1980s. The sources of job attitudes
such as work satisfaction, organizational commitment, and work motivation
were sought in aspects of an individuals’ environment, such as the design of the
job, the supervisor, the organizational climate, or the compensation, training, and
other policies of the organization (Judge & Kammeyer-Mueller, 2012b).
But, again, following trends in the research on job performance, research on
the relationship between personality and job attitudes experienced a resurgence
following seminal studies by Staw and colleagues as well as Arvey and
colleagues. Using archival data sources, Staw and Ross (1985) were able to
follow employees’ attitudes over the span of their working life. Their results
showed that measures of job satisfactions were relatively stable over time, even
after individuals changed employers and/or occupations. Staw, Bell, and Clausen
(1986) went further in this respect and demonstrated that affective disposition
measured in adolescence correlated relatively strongly (r = .34; p < .05) with job
satisfaction assessed at ages 54–62 years. Arvey et al. (1989) provided further
evidence for a nonenvironmental explanation of job attitudes by showing that
there was a significant correlation between job satisfaction among monozygotic
twins reared apart from early childhood. These studies helped to create an
impetus to determine specific dispositional factors that influence job satisfaction
and other work-related attitudes at a time when the research literature eschewed
or otherwise discounted dispositional explanations of work-related behaviors and
attitudes (Judge & Kammeyer-Mueller, 2012a).
Since then, a number of studies have used the FFM framework to investigate
the relationship between personality and job satisfaction (Judge, Heller, &
Mount, 2002; Ng, Eby, Sorensen, & Feldman, 2005). Meta-analytic results show
that conscientiousness (ρ = .26), extraversion (ρ = .25), and emotional stability
(ρ = .29) tend to have the strongest relationships with job satisfaction (Judge,
Heller, et al., 2002). One study even showed that these traits assessed in
adolescence (age 12–14 years) predicted adult job satisfaction (Judge, Higgins,
Thoresen, & Barrick, 1999). Given that job satisfaction also has a relatively
strong relationship with individual task performance (Judge, Thoresen, Bono, &
Patton, 2001), some researchers argue that part of the relationship between
personality and performance is driven through the effect of personality on job
satisfaction (Barrick & Mount, 2012). Mount, Ilies, and Johnson (2006) showed
that satisfaction mediated the relationship between different FFM dimensions
and counterproductive work behaviors. Similarly, Zimmerman (2008) showed
that job satisfaction mediated the relationship between FFM dimensions and
turnover. Overall, results suggest that the FFM traits, particularly
conscientiousness, extraversion, and emotional stability, have positive,
significant relationships with job satisfaction, and that satisfaction may be an
attitudinal mechanism through which FFM traits influence workplace behavior.
Research has also investigated the relationship between FFM traits and
organizational commitment. Organizational commitment generally refers to an
individual’s personal involvement with his or her employing organization, a
feeling of obligation toward the organization, and the perceived costs associated
with leaving that organization (Mathieu & Zajac, 1990). Current meta-analytic
evidence suggests that conscientiousness (ρ = .32) and emotional stability (ρ =
.30) have significant relationships with organizational commitment (Choi, Oh, &
Colbert, 2015).
In addition to the literature on attitudes, meta-analytic research has also
investigated the degree to which the FFM is related to work-related motivation.
For example, Judge and Ilies (2002) conducted a meta-analysis to investigate the
relationships of the FFM traits to constructs tied to three central theories of
motivation—expectancy theory, goal-setting theory, and self-efficacy
motivation. Results from this study showed that conscientiousness and emotional
stability had consistent, generalizable relationships with all three motivational
constructs. Extraversion also had consistent relationships, but the magnitude of
the effects were weaker. These results indicate that being conscientious,
emotionally stable, and extraverted is linked to the motivational tendencies to
create rewards for yourself, set goals at work, and believe that you can complete
your work effectively. These results are important because they show that
conscientiousness and emotional stability and, to a lesser extent, extraversion,
may be linked to job performance via their effects on work-related motivation.
Additionally, this research addresses one of the key limitations often discussed
in the literature on the personality–performance relationship: the lack of a
theoretical framework linking personality to work-related behaviors (Kanfer &
Heggestad, 1997). Other meta-analytic studies have shown that the FFM traits
are related to work-related motivational states such as engagement and burnout
(Christian, Garza, & Slaughter, 2011; Swider & Zimmerman, 2010). In addition
to meta-analytic evidence, primary studies have also suggested that the FFM
traits are related to proximal motivational characteristics that could link traits,
especially conscientiousness and emotional stability, to performance. For
example, Mount and Barrick (1995) show that conscientiousness is positively
correlated with effort exerted at work. Similarly, Barrick, Mount, and Strauss
(1993) show that conscientiousness is also related to the amount of time spent on
a task.
Current research on the relationship between the FFM and work-related
attitudes has attempted to bridge gaps across different literature in the OB/HR
domain and, more specifically, to address common complaints about the FFM
and its apparent lack of a precise theoretical specification. Two recent studies are
notable in their attempts to provide a clearer theoretical link between personality
traits and work-related motivation. First, the theory of purposeful work behavior
(TPWB; Barrick, Mount, & Li, 2013) is a recent theoretical model developed to
integrate the FFM traits with both the extensive literature on the job
characteristics model and to research on motivational strivings. This article
provides a systematic theoretical framework for when and how the FFM
personality traits should be expected to relate to proximal work motivation.
According to this theory, the FFM personality traits lead individuals to engage in
motivational strivings linked to personality traits, which in turn lead individuals
to develop proximal motivational states (i.e., self-efficacy, specific goals,
expectancy motivation), which in turn lead to work-related behaviors and
attitudes (e.g., job performance, satisfaction, citizenship behaviors). A key point
of this model stipulates that situational job characteristics can modify the
relationship between personality and proximal work motivation via the
activation and stimulation of motivational strivings. For example, in the TWPB,
extraversion is associated with status strivings, which in turn links to proximal
work motivations. However, job characteristics such as the level of power in the
position and task significance may alter the relationship between extraversion
and subsequent proximal motivations. Although this model currently remains
untested, use of its conceptual components is likely to determine future research
on the relationship between personality and both work motivation and job
attitudes.
In sum, substantial research evidence suggests that the FFM personality traits
are linked to both job attitudes and proximal work motivation characteristics.
Future research on the relationship between personality and job attitudes and
workplace motivation will likely focus on examining how different situational
characteristics magnify or constrain the effects of FFM traits on attitudes and
motivation, and how attitudes and motivation serve a theoretical link between
FFM traits and work-related behaviors and outcomes.
Leadership
The FFM has also been useful for organizing the large and varied literature on
personality and leadership effectiveness. More recently, it has led to theoretical
and empirical work linking personality traits with effectiveness through their
relationship with leadership behavior. As in other areas, a number of important
debates continue to resonate, including issues of predictive validity, the
theoretical mechanisms by which personality works, and the appropriate facet
level at which to conduct future research. As business scholars have observed,
leadership is a folk term rather than a technical term and so carries considerable
ambiguity as a scientific construct (Alvesson & Sveningsson, 2003; Janda,
1960). Nevertheless, most definitions of leadership suggest that it is a process of
influencing or motivating others to agree about what needs to be done and how
to do it, including actions that facilitate individual and collective efforts to
accomplish shared objectives and satisfy the needs of important stakeholders
(e.g., Yukl, 2012).
Personal traits and characteristics have been regarded as central to successful
leadership at least since the “Great Man” view of history espoused by Thomas
Carlyle in the nineteenth century. According to this view, heroic individuals
shape history through the exercise of their exceptional personal attributes such as
intelligence, wisdom, charisma, or political skill. Based on almost typological
notions of greatness heavily influenced by evolutionary theory, the search for the
set of enduring traits that distinguishes leaders from those who would not
become leaders seemed justified. This perspective inspired hundreds of studies
in the 1930s and 1940s designed to identify the individual leader traits associated
with the emergence (who becomes the leader of a leaderless group?) or
effectiveness (who performs well in the leader role?) of leaders across a wide
range of contexts, from school children to military and industry settings.
Researchers examined a broad range of characteristics, including height,
intelligence, and verbal fluency, as well as aspects of personality, such as
adaptability, extroversion, dominance, initiative, persistence, responsibility, self-
confidence, and emotional control (Stogdill, 1948). Enthusiasm for this approach
was severely dampened when several influential reviews of the literature (Mann,
1959; Stogdill, 1948, 1974) found only weak and inconsistent relationships
between traits and leadership emergence or effectiveness. For example, Mann
(1959) reviewed the literature on leadership emergence in small groups and
concluded that intelligence, adjustment, extroversion, dominance, masculinity,
and interpersonal sensitivity are positively related, whereas conservatism is
negatively related, to leadership emergence in small groups. However, he noted
that situational conditions altered the relationships and that in no case were
correlations higher than .25, with median correlations close to .15 (p. 266).
Based on these reviews, prominent scholars over the ensuing years (e.g., House
& Aditya, 1997; Landy, 1985; Muchinsky, 1983) regularly reported that traits
had little to do with leader emergence or effectiveness, redirecting scholarly
attention toward behavioral or situational theories of leadership.
Not until 2002 did the techniques of meta-analysis and the framework of the
FFM come together on the topic of personality and leadership in the work of
Judge, Bono, Ilies, and Gerhardt (2002). As these authors observed, “one reason
for the inconsistent and disappointing results from previous reviews is that, until
recently, we have lacked a taxonomic structure for classifying and organizing
traits” (p. 766). They used the FFM to organize the multitude of traits studied in
the leadership literature into the set of five consistent and coherent dimensions of
personality. Judge, Bono, et al. (2002) found four of the Big Five personality
dimensions related to leadership, defined as a broad construct including both
emergence and effectiveness; extraversion (ρ = .31), conscientiousness (ρ = .28),
neuroticism (ρ = –.24), and openness to experience (ρ = .24). Only agreeableness
was unrelated to overall leadership. Judge, Bono, et al. (2002) also calculated a
credibility interval to estimate the variability of individual correlations across a
population of studies. Although the four personality dimensions that were related
to leadership demonstrated credibility intervals indicating that 90% of the
individual correlations would exclude zero, these intervals were nevertheless
quite large, suggesting the presence of moderators. Judge, Bono, et al. (2002) did
find evidence suggesting both situational and methodological moderation. For
example, the FFM traits predicted student perceptions of leadership better than
leadership perceptions in business or government settings. They also predicted
leadership emergence more strongly than leadership effectiveness. The multiple
regression coefficient for all five FFM traits was .53 when predicting leadership
emergence, but only .39 when predicting leadership effectiveness.
As valuable as the study of Judge, Bono, et al. (2002) was in documenting the
role of the FFM traits in leadership, a number of important theoretical and
empirical challenges remain unanswered. For example, Judge, Bono, et al.
(2002) attempted to address the bandwidth–fidelity dilemma (Cronbach &
Gleser, 1965; Ones & Viswesvaran, 1996) by examining the predictive validity
of narrow traits relative to the five broad constructs identified by the FFM. Some
scholars have argued that narrow traits might better account for variance that is
situation specific or better predict narrowly defined outcomes than more broadly
defined trait constructs (Hogan & Roberts, 1996; Paunonen, 1998). Judge, Bono,
et al. (2002) sought to address this issue, but found what they described as mixed
results. In fact, the data available to Judge, Bono, et al. (2002) may not have
allowed an accurate comparison of broad versus narrow traits since few primary
studies included direct measures of the broad FFM traits. A true examination of
the validity of global FFM traits relative to their narrow domain counterparts
remains to be conducted.
A second issue that remains unresolved concerns the theoretical process
explaining the effect of personality on leadership. Judge, Bono, et al. (2002)
provided behavioral, rather than attributional explanations linking personality to
leader emergence or effectiveness. For example, they argued that extraverted
individuals would be more active, energetic, and assertive and that conscientious
individuals would show more initiative, persistence, and task competence,
leading them to emerge as leaders or be more effective in a leadership role.
Although results were consistent with these explanations, they were in some
ways better explained by implicit leadership theory (Lord, DeVader, & Alliger,
1986; Lord, Foti, & Phillips, 1982). This social cognitive perspective argues that
lay people hold shared beliefs about the traits and behaviors of leaders that they
use as cues to categorize individuals. People who possess characteristics that
correspond to these implicit beliefs are more likely to emerge as a leader in
groups and teams, regardless of the extent to which these characteristics actually
make someone a better leader. A common example would be height: naive
observers might hold an implicit belief that leaders are tall, leading them to
respond positively to the ideas and influence attempts of the tallest person in the
group. The results of Judge, Bono, et al. (2002) tended to show the FFM traits
are more strongly related to leader emergence than leader effectiveness,
suggesting these traits are more strongly related to followers’ perceptions of
appropriate leader behavior than to actual performance as a leader. Likewise, the
FFM traits were more strongly associated with leadership in student samples
than in samples drawn from business or government/military contexts. Students
may rely more on the automatic use of their implicit theories of leadership than
do samples of business or government employees because classroom settings
provide less motivation to form accurate perceptions of leader effectiveness and
because students have less experience with effective leader behavior on which to
draw, causing them to rely on preconceived cues to a greater extent.
Of course these two explanations for the relationship between traits and
leadership effectiveness are not incompatible. As Judge, Bono, et al. (2002) note,
peoples’ implicit beliefs about the traits and behaviors of effective leaders may
be correct in the sense that they truly reflect the traits that contribute to
effectiveness in a leadership role. Thus, the difference in magnitude between the
leader effectiveness (multiple R = .39) and leader emergence (multiple R = .53)
ratings may provide us with an estimate of the upper limit of the size of the
perceptual biases introduced by the implicit leadership effects.
Conscientiousness and agreeableness have the most interesting pattern of results
when the difference between leader emergence and leader effectiveness is
examined. Although conscientiousness might be expected to be a stronger
predictor of performance in a leader role, in fact it is much more strongly
associated with emergence (ρ = .33) than it is with effectiveness (ρ = .16). On
the other hand, agreeableness is not significantly related to leader emergence (ρ
= .05), but is almost as strongly related to effectiveness (ρ = .21) as are any of
the other traits. If the difference between leader emergence and effectiveness is
regarded as an indicator of implicit theories, then these findings suggest that
assumptions about desirable leadership traits are not completely consistent with
the traits that predict effectiveness. The contribution of the FFM to the
leadership field could be made stronger if a clear theoretical link could be made
between the FFM traits and behaviors clearly related to leadership effectiveness.
Such connections would demonstrate that the FFM can provide theoretical
insights into the leadership process, rather than serving merely as a structure for
organizing previously examined traits. Furthermore, it would mitigate concerns
that the contribution of personality to leadership is largely due to the implicit
assumptions of the followers rather than to the concrete skills and behaviors of
the individual in the leadership role.
The earliest behavioral leadership work was conducted by researchers at the
Ohio State University (e.g., Fleishman, 1953; Stogdill & Coons, 1957), who
focused on two dimensions of leader behavior that they identified through factor
analysis. Initiation of structure refers to leader behaviors directed toward the
organization of effort toward completion of the group’s task. Examples include
assigning tasks to subordinates, emphasizing deadlines, and criticizing poor
performance (Yukl, 2012). Consideration refers to leader behaviors that show
concern for the needs and feelings of subordinates. Examples include listening to
subordinates’ problems, accepting suggestions from subordinates, and treating
your subordinate as an equal (Yukl, 2012). More recently, leadership scholars
have focused on inspirational and change-oriented leadership behaviors
identified within transformational leadership theory (Bass, 1985, 1996).
Transformational leader behaviors include articulation of a compelling vision,
acting as a role model, increasing followers’ awareness of problems, stimulating
followers to view problems from a new perspective, and supporting the career
and personal development of followers. This theory also identified a distinct but
not mutually exclusive set of transactional leadership behaviors that are assumed
to bring about follower compliance to the direction of the leader if not
enthusiastic commitment. Transactional leader behaviors include the use of
incentives and rewards for achieving work requirements (contingent reward
behavior) and punishments to sanction deviations from acceptable performance
standards (management by exception).
Based on previous meta-analyses (Bono & Judge, 2004; Judge, Bono, et al.,
2002) and new analyses of their own, DeRue, Nahrgang, Wellman, and
Humphrey (2011) tested a meta-analytic path model to examine the extent to
which leader behaviors mediate the relationship between leader traits and leader
effectiveness. This study found that several FFM traits had reliable but small
relationships to leadership behaviors. For example, the task-accomplishment
oriented trait of conscientiousness is related to initiation of structure (ρ = .26)
whereas the interpersonally oriented trait of extraversion is related to
consideration leadership behaviors (ρ = .29) and transformational leadership (ρ =
.20). Agreeableness was also related to consideration behavior (ρ = .21), as the
authors expected. As found in previous research (Judge & Piccolo, 2004; Judge,
Piccolo, & Ilies, 2004), all of these leader behaviors are related to leadership
effectiveness. However, subsequent analysis indicates that the relationship
between the FFM traits and leadership effectiveness is mediated only partially
by these leader behaviors. In fact, in all cases the direct effect of the traits on
leadership effectiveness is stronger than the indirect effect through leader
behavior (see DeRue et al., 2011, Table 7). This suggests that we do not fully
understand the theoretical mechanisms linking the FFM traits to leadership. The
effect of openness to experience is, in particular, poorly explained by the current
leader behavior models. DeRue et al. (2011) expected openness to be related to
transformational leadership behavior due to the association between both of
these variables and a change-orientation, but this relationship failed to emerge.
Thus, although current leadership behavior theory can be used to link FFM
traits to leadership effectiveness, there is more of this relationship still to be
explained. Future leadership research might formulate leadership models that
include behaviors more closely related to the personality traits themselves. For
example, Judge, Bono, et al. (2002) speculated that open individuals might be
better leaders because they are more creative, because they are more likely to
take risks, or because they are more visionary as leaders. However, this returns
us to the bandwidth-fidelity dilemma, because such narrow behaviors might be
more closely related to narrow traits than they are to broad personality
constructs. Other avenues for the further exploration of trait effects include the
consideration of joint combinations of traits, whether in statistical interactions or
profiles, and the search for curvilinear effects between traits and outcomes
(Zaccaro, 2007). In sum, although future work is called for, it is clear that the
FFM has made an important and lasting contribution to understanding
leadership.
Teams
Teams have become an increasingly important area of research in the OB/HR
domain (Wageman, Gardner, & Mortensen, 2012), and as such, research on the
role of personality in teams has also greatly expanded and remains a relatively
untapped area with significant potential for new insights. Much of the key focus
has been on two areas: (1) how the composition of the team in terms of the
personality traits of its members influences team-level processes and outcomes,
and (2) how team-level processes and characteristics influence the relationship
between individuals’ personality traits and their subsequent motivation and
behavior as team or organizational members. A typical question addressed in the
first type of research might be, “How does the mean level of conscientiousness
of team members influence the performance of the team?” A typical question
addressed in the second type of research might be, “How does the relationship
between conscientiousness and individual task performance change based on a
person’s given role in a team?” Both types of questions have merit and are the
subject of both early and contemporary research.
Research on the role of personality in teams began as early as the 1950s and
1960s, when social psychologists examined the role of personality in small-
group research (LePine, Buckman, Crawford, & Methot, 2011). Mann’s (1959)
review of the literature, predating the FFM taxonomy, focused on the role of
personality traits and individual behaviors in the context of teams, rather than the
role of personality composition on team processes and outcomes. Heslin’s
(1964) review focused more explicitly on the role of personality composition in
team-level outcomes. His review noted substantial variability in terms of how
personality was assessed at the team level, how team performance was assessed,
and variations in the social and task structures for different types of teams. In
line with the fallow period of research on personality’s relationship to job
performance and work attitudes, research on the role of personality in teams was
largely dormant in the period from the early 1960s to the mid-1980s. Two trends
contributed to a resurgence of interest in the role of personality in teams. First
was the increasing interest in the late 1980s and early 1990s on the role of teams
in organizational research (Devine, Clayton, Philips, Dunford, & Melner, 1999;
Manz & Sims, 1993). Second was the development of the FFM. The FFM
framework provided a coherent set of constructs that helped researchers focus
their efforts. In sum, research on the FFM personality traits in teams since the
1990s has become increasingly important to understanding the contemporary
literature on teams.
The majority of research on the role of the FFM traits in teams since the 1990s
has focused on how team-level compositions of personality traits relate to team-
level processes and outcomes. One of the key questions in these studies has been
on ways to operationalize FFM traits at the team level. Three models of
operationalization are most common: the mean of members’ individual trait
scores, the variance on team members’ trait scores, or the score of one focal
member (LePine et al., 2011). Two seminal studies systematically investigated
issues of composition type (Barrick, Stewart, Neubert, & Mount, 1998; LePine,
Hollenbeck, Ilgen, & Hedlund, 1997). LePine et al. (1997) studied different
operationalizations of conscientiousness in laboratory teams with different task
settings, finding that the lowest score among team members had the strongest
relationship to overall team performance. The study concluded that, in general,
teams were as strong as their weakest link. Barrick et al. (1998) tested all five
FFM traits with all three operationalizations of personality traits across several
different assessments of team performance in a field setting (e.g., decision-
making accuracy, productivity, viability). Results from this study showed that
teams with higher levels of conscientiousness (i.e., higher means and higher
minimum scores) tended to perform better across all measures of team
performance. In addition, the mean team agreeableness and emotional stability
also predict job performance, whereas the mean team extraversion and emotional
stability predicted team viability. These studies spurred a great deal of research,
though many studies found conflicting, contradictory, or inconsistent results,
even with similar types of teams or operationalizations of personality (e.g.,
Beersma et al., 2003; Ellis et al., 2003; English, Griffith, & Steelman, 2004;
Neuman, Wagner, & Christiansen, 1999).
In an effort to clarify the literature, Bell (2007) conducted a meta-analysis of
the existing research on the role of personality compositional variables in teams.
Results from this meta-analysis found that overall relationships between
personality traits and team performance were relatively modest (i.e., between
values of ρ = .02 and ρ = .10). However, moderator analyses revealed a starkly
different picture: the type of team and operationalization of personality mattered
significantly. Studies in field settings showed that personality composition had
moderate effects on team performance (conscientiousness, ρ = .30;
agreeableness, ρ = .31; extraversion, ρ = .15; openness to experience, r = .20),
with the effects of personality on teams in laboratory settings being negligible.
The results also showed that in general, personality traits had the strongest
relationships to team performance when operationalized as the mean of the team
members’ individual scores; however, the effect of the minimum level of
agreeableness on team performance was also quite large (ρ = .37), suggesting
that operationalizations of personality beyond the mean also matter to team
performance. Prewett, Walvoord, Stilson, Rossi, and Brannick (2009) conducted
a similar meta-analysis with different categorical compositions and found similar
but weaker results
Two conclusions can be drawn from these meta-analyses. First, the
operationalization of personality in teams as the mean level of individuals’ trait
scores appears to provide the strongest relationship between team member
personality and outcomes, though other operationalizations can also influence
outcomes. Second, both meta-analyses suggested that team member personality
matters more to team outcomes and processes when teams are highly
interdependent. In many ways, this result is intuitive: the composition of the
group in terms of personality and the resulting interpersonal interactions matter
more to group outcomes when team members must continually interact with
each other to achieve higher levels of performance.
In addition to research on the effects of team personality on team outcomes,
models of the role of personality in teams has also addressed the effects of team
personality on team processes (Bell, 2007; Hackman, 1987; LePine et al., 2011;
Marks, Mathieu, & Zaccaro, 2001; Marks, Zaccaro, & Mathieu, 2000). Much
like the individual-level model in which the effects of individual personality on
work-related behavior are mediated through personality’s effects on attitudes
and work-related motivation, there is a growing consensus in teams research that
team personality influences proximal team-level processes (e.g., cohesion,
collective efficacy, helping behaviors), which in turn influence team level
outcomes. However, most research on the role of personality in teams has
focused on the personality–team outcome relationship at the expense of
identifying theoretically derived process mechanisms (LePine et al., 2011). Of
the studies that do address these concerns, most focus on the relationship
between team personality and team cohesion. This focus is logical given the
strong relationship between cohesion and effective team functioning (Beal,
Cohen, Burke, & McLendon, 2003; Gully, Devine, & Whitney, 1995; LePine,
Piccolo, Jackson, Mathieu, & Saul, 2008). For example, Van Vianen and De
Dreu (2001) found that team extraversion and emotional stability were related to
different types of team cohesion, which in turn had a positive relationship on
task performance. Most recently, Colbert, Barrick, and Bradley (2014)
demonstrated that mean conscientiousness and extraversion in top management
teams had significant impacts on organizational effectiveness outcomes,
suggesting that the effects of personality on teams can have far-reaching,
organizational-level implications.
Although cohesion has been a consistent focus, a number of other individual
studies have investigated other mediating processes. LePine (2003) found that
characteristics such as the mean level of team member openness had indirect
effects on task performance via a team’s capacity to change the structure of
individual roles among team members. Alternatively, Homan et al. (2008) found
that mean team openness had indirect effects on team performance through team
information elaboration processes. Other studies such as Barry and Stewart
(1997) found that conscientiousness and extraversion were related to team
processes such as focus on tasks, social cohesion, and open communication.
Some studies also examined the mediating effects of specific team behaviors.
For example, Porter et al. (2003) found that the effects of team conscientiousness
and emotional stability were related to team performance via backing up
behaviors, or the degree to which team members provided help and support to
fellow teammates. Finally, recent research such as Bradley, Klotz, Postlethwaite,
and Brown (2013) investigated how team member personality may moderate the
relationship between team processes, such as conflict, and team outcomes.
Overall, research on the role of personality in team-level outcomes continues to
grow, with a need for future studies examining processes linking team
personality to team-level outcomes.
Although the dominant perspective in the literature on team personality has
focused on how team-level operationalizations of personality influence team-
level processes and outcomes, other research has focused on the role of team
personality on individual behaviors, or how team processes influence and impact
the relationship between individual-level personality and work-related behavior.
One common example of this is the relationship between personality and team
roles. Stewart, Fulmer, and Barrick (2005) found that individual personality
traits were related to the type of roles individuals would take on in teams. In this
study, Stewart et al. (2005) found that although agreeableness (β = .27) was
related to an individual engaging in social role behaviors (i.e., facilitating
interpersonal relationships among team members), conscientiousness (β = .21)
and extraversion (β = –.22) were related to task role behaviors (i.e., setting
deadlines for work, allocating tasks to team members). Research in this vein has
also investigated the degree to which personality traits are related to the display
of effective teamwork skills and behaviors, with meta-analytic results showing
that agreeableness, conscientiousness, and emotional stability were all
generalizable predictors of teamwork behaviors (Hough, 1992). Future research
on this topic is most likely to focus on how these types of individual-level
personality–behavior relationships translate to team-level effectiveness (LePine
et al., 2011).
An additional issue in the relationship between team personality and team
effectiveness is the role of cross-level relationships, or the way in which team-
level processes may directly or indirectly influence individual-level behaviors
and relationships. These types of studies typically seek to identify how team-
level personality composition may moderate or alter the relationship between
individual-level personality traits and behavior. An example of this type of
research is Peeters, Rutte, van Tuijl, and Reyman (2006). This study showed that
an individual member’s conscientiousness and extraversion were not directly
related to satisfaction with the team; rather, an individual’s dissimilarity to their
teammates was negatively linked to his or her satisfaction with the team. Team
processes may also moderate the personality–behavior relationship. For
example, Tasa, Sears, and Schat (2011) found that team collective efficacy
moderated the relationship between agreeableness and teamwork behavior such
that the relationship between these variables was strongest when collective
efficacy was higher. In a similar vein, Schmidt, Ogunfowora, and Bourdage
(2012) found that team-level processes, such as cohesion as well as the team’s
extraversion and conscientiousness composition, moderated the relationship
between individual-level personality traits and behaviors in a sample of varsity
athletic teams. However, only a handful of studies on cross-level effects exist,
suggesting that these are topics ripe for future research.
In sum, the current research on the role of FFM personality traits in teams has
a remarkable amount of variability, due in part to the substantial variability in
the many different types of teams and different types of tasks those teams must
complete. Nevertheless, we can draw several conclusions, based on the current
literature. First, the relationship between mean team composition of the FFM
traits and team performance appears to be significantly stronger than the
relationship between individual personality traits and individual job performance
(conscientiousness, ρ team versus individual = .30 versus .24; agreeableness, ρ
team versus individual = .31 versus .11; openness, ρ team versus individual = .20
versus .07; extraversion, ρ team versus individual = .15 versus .15; Bell, 2007).
Given these results, traits other than conscientiousness, including agreeableness
and openness, clearly matter in understanding the effective functioning of teams.
Second, although there are a number of studies that link personality to emergent
team processes or behaviors that support team effectiveness, there is little
research that provides evidence that these processes and behaviors are the link
through which team personality is related to team outcome. Future research
needs to develop models in which team personality is linked to performance via
its effects on team processes and behaviors known to be associated with team
effectiveness. Third, although there is evidence that personality traits are linked
to effective team performance, almost no empirical research on staffing teams
exists (e.g., Humphrey, Hollenbeck, Meyer, & Ilgen, 2007). Future research
needs to investigate the degree to which the conclusions developed from the
literature can be applied in such a way as to have a practical impact on how
organizations choose to staff teams, and what sort of tradeoffs in terms of
personality trait composition may positively impact a team. In conclusion,
although there have been significant advances in research on personality in
teams since Heslin’s (1964) review, there is still a significant amount of
untapped research potential in this area.
Entrepreneurship
The roots of scholarship on entrepreneurship lie in the theorizing of early
twentieth-century economists, in particular Schumpeter (1911) and Knight
(1921). These scholars emphasized the important role that individuals play in
transforming new technologies into viable business products and services, the
individual acting as the instrument of “creative destruction,” the memorable
phrase popularized by Schumpeter (1942). These scholars emphasized the
personal characteristics of the archetypical entrepreneur, much in keeping with
the “Great Man” theories of leadership espoused in the late nineteenth century.
The entrepreneur was thought to have special personal qualities, such as
judgment, perseverance and will, knowledge of the world, and the power to
identify new technologies and bring them into commercial existence. Although
these scholars did not attempt to empirically test their assertions, they left a
lasting impression on the field of entrepreneurship.
Modern academic research on entrepreneurship began with a focus on
personality, in the form of McClelland’s work on achievement motivation
(Baum, Frese, Baron, & Katz, 2007). In The Achieving Society (1961),
McClelland attempted to show that the level of achievement motivation
embodied by the people of a society was associated with entrepreneurial
activities, which in turn was associated with economic growth in that society. In
Motivating Economic Achievement (1969), McClelland and his colleagues
reported on field-based quasiexperiments they had designed to stimulate higher
levels of achievement behavior in practicing entrepreneurs, most notably among
a group of businessmen in India. This work became the theoretical basis for
government and academic programs devised to train and support nascent
entrepreneurs and inspired a deluge of research using the trait approach to
identify potential entrepreneurs in the 1970s (Baum et al., 2007). Among the
most frequently studies traits were need for achievement, need for autonomy or
independence, creativity or openness, locus of control, and risk-taking
preference.
In what should be a familiar theme by now, by the mid-1980s researchers in
the entrepreneurship area became disillusioned with the value of trait-based
research. To some extent this disillusionment was motivated by institutional
issues, as entrepreneurship education and scholarship were increasingly housed
in business rather than psychology programs and the pool of entrepreneurship
researchers was increasingly composed of academics trained in economic and
strategic management paradigms rather than in psychology (Katz, 2003). Still,
many of the same issues plaguing trait research in other areas of business
scholarship were the focus of criticism in the entrepreneurship domain: there are
too many traits with unknown relationships to each other; traits were used in a
“shotgun” manner with no clear theoretical linkage between the trait and
entrepreneurship articulated; and trait correlations with entrepreneurial status or
performance were zero or inconsistent (Brockhaus & Horwitz, 1985). A 1988
article entitled “ ‘Who is an entrepreneur’ is the wrong question” by Gartner
had a major impact in redirecting the field away from trait-based research.
Although much of Gartner’s (1988) criticism focused on ambiguities in the
definition used to identify samples of entrepreneurs, a criticism more of the field
of entrepreneurship than of traits per se, the crux of his frustration was
summarized as follows:
A startling number of traits and characteristics have been attributed to the
entrepreneur, and a “psychological profile” of the entrepreneur assembled from these
studies would portray someone larger than life, full of contradictions, and
conversely, someone so full of traits that (s)he would have to be a sort of generic
“Everyman.”
(Gartner, 1988, p. 21)
Although studies of the traits of entrepreneurs continued, they were often

received with skepticism as the field focused more on firm strategies and
contextual economic factors promoting entrepreneurship. This picture began to
change with the introduction of meta-analytic methods to more accurately
summarize the results of previous work, and more developed theories to link
personality to entrepreneurial behavior and performance emerged (Baron &
Markman, 2000; Rauch & Frese, 2007; Shane, Locke, & Collins, 2003; Zhao,
Seibert, & Hills, 2005). Although early work continued to focus on the factors
that distinguish entrepreneurs from others, similar to the leadership emergence
paradigm reviewed above, later studies began to focus on the outcomes of
entrepreneurial behavior defined in terms of the performance of the venture
founded by the entrepreneur. For example, using meta-analytic techniques,
Stewart and Roth (2001, 2004) found that entrepreneurs were higher in risk
propensity and Collins, Hanges, and Locke (2004) and Stewart and Roth (2007)
found that entrepreneurs were higher on achievement motivation than their
manager counterparts.
The FFM was used to provide a more comprehensive summary of the effects
of traits on entrepreneurial behavior in two studies conducted by Zhao, Seibert,
and their colleagues. Zhao and Seibert (2006) organized dozens of personality
variables used in the field of entrepreneurship within the framework of the FFM
to examine their relationship to entrepreneurial status. The general theoretical
perspective of person–environment fit, and Schneider’s (1987) attraction-
selection-attrition (ASA) model in particular, was used to explain the proposed
relationships. Adapting the ASA model to the context of entrepreneurship, Zhao
and Seibert (2006) suggested that certain FFM traits would lead individuals to be
more attracted to entrepreneurial work, more likely to be provided with financial
and other support and thus “selected” by outside agents for entrepreneurial work,
and more satisfied in the entrepreneurial role. They examined only studies that
compared the traits of entrepreneurs with the traits of practicing managers, a
rigorous comparison considering the similarity of managerial work to the work
of founding and running a new business. Their results showed that entrepreneurs
differed from their managerial counterparts on four of the five FFM traits.
Entrepreneurs were more conscientious (ρ = .22), emotionally stable (ρ = .18),
open to experience (ρ = .18), and slightly less agreeable (ρ = .08). Although
these effects are small, the set of FFM traits together explained 14% of the
variance in entrepreneurial status. The heterogeneity of effect sizes was
considerable, suggesting the possibility of moderators of the relationship.
Subsequent analysis showed that the achievement facet of conscientiousness was
somewhat more strongly related to entrepreneurial status (ρ = .28) than was the
effect for the full set of variables and significantly more strongly related than the
dependability facet (.00, n.s.), which showed almost no relationship. Again, too
few studies had been conducted that used direct measures of the FFM constructs,
so no comparison could be made regarding the predictive validity of global
constructs versus narrow subfactors.
Classically, scholars in the economic tradition had been concerned with the
characteristics of those who undertake the formation of new ventures, almost
equating firm formation with success. More recent scholarship in this field has
taken a more nuanced approach to entrepreneurship and has focused on different
specific aspects or phases of the entrepreneurial process (Baron & Shane, 2007;
Venkataraman, 1997). Perhaps the earliest phase is the formation of the intention
to someday found your own business (Bird, 1988; Krueger, Reilly, & Carsrud,
2000). Zhao, Seibert, and Lumpkin (2010) used the meta-analytic techniques and
the FFM framework to examine the relationship of personality to entrepreneurial
intentions among individuals, usually students or managers, who were not
currently entrepreneurs. The logic relating personality to entrepreneurial
intentions was again built on Schneider’s (1987) ASA theory. The results of
Zhao et al. (2010) showed that openness to experience (ρ = .24) and emotional
stability (ρ = .22) were the strongest trait predictors of entrepreneurial intentions,
followed by conscientiousness (ρ = .19) and extraversion (ρ = .16).
Agreeableness had no reliable relationship to entrepreneurial intentions.
Zhao et al. (2010) also examined the relationship between the FFM traits and
the performance of the new firm founded by the entrepreneur. Their theorizing
extended the logic linking personality to individual performance to the firm
level, based on the view that the activities of the founder and principal manager
of a new venture would have a major impact on the venture’s success. For
example, conscientiousness should be associated with the motivation to work
long hours and set high goals for the firm; openness with the creativity needed
for problem solving; emotional stability with the ability to deal with the stress
and anxiety associated with running your own business; and extraversion with
the interpersonal requirements associated with leading a business team. Their
results again supported a role for all of the FFM traits except agreeableness in
entrepreneurial firm performance. The largest effects were for openness to
experience (ρ = .21), conscientiousness (ρ = .19), and emotional stability (ρ =
.18). Extraversion again demonstrated a small but reliable relationship with firm
performance (ρ = .09).
A comparison of the role of the FFM traits in entrepreneurship to the role they
play in managerial job performance and leadership is instructive.
Conscientiousness and emotional stability are consistently associated with
managerial job performance (Barrick, et al., 2001; Hurtz & Donovan, 2000),
leadership effectiveness (Judge, Bono, et al., 2002), and entrepreneurial firm
performance (Zhao et al., 2010). However, openness is associated with
performance only in the latter two domains. Apparently, leading an
entrepreneurial firm is similar to leadership in other contexts with respect to the
value of openness, a trait not related to performance in a managerial role within
a traditional employment context. On the other hand, extraversion appears to
make a larger contribution to leadership effectiveness than it does to either
managerial job performance or entrepreneurial firm performance. In this way the
FFM helps us to not only make sense of the role of traits within specific fields of
study within the business and industry, but to make comparisons across several
contexts and job roles.
Conclusions
Several major themes run through our review of the use of the Five Factor
Model of personality in business and industry. The first issue is the validity of
personality variables as predictors of important attitudes, behaviors, or outcomes
in work settings. The second issue is the extent to which the validity of
personality variables generalizes across situations or is conditioned by
situational variables. Finally, a third area of controversy is the bandwidth–
fidelity issue. The FFM is central to all of these debates in each of the areas of
application in the following review.
Debates regarding the level of predictive validity for the FFM traits, and
methods to assess that validity, remain. Although this issue has its most
important practical implications in the literature on selection, where hiring
decisions are made based on personality traits, it also has theoretical importance
for domains such as leadership and entrepreneurship. Have the operational
validities of personality measures changed over time, or have scholars in the
field simply become more accepting of small effect sizes? Have the techniques
for correcting measurement error or source effects become more sophisticated
and nuanced or more unrealistic? Future work might explore the literature on
statistical versus practical significance to settle the debate regarding the
importance of personality in work behavior (Aguinis et al., 2010).
Another question that remains for personality is the degree to which
contextual and situational factors likely influence the effect size of the FFM
traits across contexts. Although general effect sizes for personality tend to be
small to modest, most meta-analytic studies included in this review suggest that
moderators exist in the relationship between FFM traits and workplace behaviors
and outcomes. Recent empirical research and theoretical models have suggested
that situational cues and contextual factors may in fact play significant roles in
the effects of personality on workplace behaviors and outcomes (Barrick et al.,
2013; Judge & Zapata, 2015). Future research can and should continue to
investigate this possibility, particularly in light of the limitations of current
theory on person–situation interactions (Barrick et al., 2013).
A third debate in the literature has questioned whether personality should be
assessed via narrow facets, broad traits, or larger, compound traits (Ones &
Viswesvaran, 1996; Judge & Kammeyer-Mueller, 2012b). Although the most
recent meta-analytic research has suggested that facets do in fact contribute to
the explanation of work behaviors and outcomes beyond the effects of broader
traits (Judge, Rodell, Klinger, Simon, & Crawford, 2013), there remains
significant future research possibilities regarding when and how these narrower
facets and compound traits are likely to have an important incremental influence
on outcomes beyond the broad FFM constructs.
In several research domains in our review, the FFM served as a useful
framework around which to organize previous research, estimate construct-level
relationships, and move forward. Several domains within the OB/HR area may
yet benefit from this organizing framework for personality, such as performance
management, recruitment, and top management teams, to mention a few.
However, questions remain regarding the utility of the FFM as a theoretical
framework capable of generating clear propositions linking traits to behaviors
and outcomes. Future theoretical work is required to better integrate personality
and personality processes into established theoretical models of work behavior.
Taking another perspective, some efforts are being made to develop and test
theoretically coherent models of workplace motivation and behavior built around
the FFM (e.g., Barrick et al., 2013), but the utility of this model remains to be
established.
Overall, the FFM has allowed scholars to organize findings regarding the role
of personality in several important domains of research in the OB/HR field.
Often the ability to organize the literature into a small, comprehensive and
coherent framework has rendered conclusions that contradict previously held
views regarding the role of personality traits. Furthermore, research on the FFM
traits and their role in workplace behaviors and outcomes has proven to be one
of the most productive areas in modern OB/HR studies (Judge, Klinger, Simon,
& Yang, 2008). Although this organization has sometimes raised questions and
presented limitations, it has also sparked significant research and suggested
useful directions for research going forward. It is our hope that the FFM of
personality can continue to inspire thoughtful work on the role of individual
differences in workplace behaviors and outcomes.
References
Aguinis, H., Werner, S., Abbott, J. L., Angert, C., Park, J. H., & Kohlhausen, D. (2010).
Customer-centric science: Reporting significant research results with rigor, relevance,
and practical impact in mind. Organizational Research Methods, 13(3), 515–539.
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human
Decision Processes, 50(2), 179–211.
Alvesson, M., & Sveningsson, S. (2003). The great disappearing act: Difficulties in doing
“leadership.” The Leadership Quarterly, 14(3), 359–381.
Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A
review and recommended two-step approach. Psychological Bulletin, 103(3), 411–433.
Arvey, R. D., Bouchard, T. J., Segal, N. L., & Abraham, L. M. (1989). Job satisfaction:
Environmental and genetic components. Journal of Applied Psychology, 74(2), 187–
192.
Barrick, M. R., & Mount, M. K. (2012). Nature and use of personality. In N. Schmitt
(Ed.), The oxford handbook of personnel assessment and selection. New York, NY:
Oxford University Press, 225-251.
Baron, R. A., & Markman, G. D. (2000). Toward a process view of entrepreneurship: The
changing impact of individual level variables across phases of new venture
development. In M. A. Rahim, R. T. Golembiewski, & K. D. McKenzie (Eds.), Current
topics in management (Vol. 9, pp. 45–64). New Brunswick, NJ: Transaction.
Baron, R., & Shane, S. (2007). Entrepreneurship: A process perspective. Stamford, CT:
Cengage Learning.
Barrick, M. R., & Mount, M. K. (1991). The big five personality dimensions and job
performance: A meta-analysis. Personnel Psychology, 44(1), 1–26.
Barrick, M. R., & Mount, M. K. (1996). Effects of impression management and self-
deception on the predictive validity of personality constructs. Journal of Applied
Psychology, 81(3), 261–272.
Barrick, M. R., Mount, M. K., & Judge, T. A. (2001). Personality and performance at the
beginning of the new millennium: What do we know and where do we go next?
International Journal of Selection and Assessment, 9(1–2), 9–30.
Barrick, M. R., Mount, M. K., & Li, N. (2013). The theory of purposeful work behavior:
The role of personality, higher-order goals, and job characteristics. Academy of
Management Review, 38(1), 132–153.
Barrick, M. R., Mount, M. K., & Strauss, J. P. (1993). Conscientiousness and
performance of sales representatives: Test of the mediating effects of goal setting.
Journal of Applied Psychology, 78(5), 715–722.
Barrick, M. R., Stewart, G. L., Neubert, M. J., & Mount, M. K. (1998). Relating member
ability and personality to work-team processes and team effectiveness. Journal of
Barry, B., & Stewart, G. L. (1997). Composition, process, and performance in self-
managed groups: The role of personality. Journal of Applied Psychology, 82(1), 62–78.
Bass, B. M. (1985). Leadership and performance beyond expectations. New York, NY:
Free Press.
Bass, B. M. (1996). A new paradigm for leadership: An inquiry into transformational
leadership. Arlington, VA: U.S. Army Research Institute for the Behavioral and Social
Sciences.
Baum, J. R., Frese, M., Baron, R. A., & Katz, J. A. (2007). Entrepreneurship as an area of
study: An introduction. In J. R. Baum, M. Frese, & R. A. Baron (Eds.), The psychology
of entrepreneurship. London, England: Lawrence Erlbaum Associates.
Beal, D. J., Cohen, R. R., Burke, M. J., & McLendon, C. L. (2003). Cohesion and
performance in groups: A meta-analytic clarification of construct relations. Journal of
Beersma, B., Hollenbeck, J. R., Humphrey, S. E., Moon, H., Conlon, D. E., & Ilgen, D.
R. (2003). Cooperation, competition, and team performance: Toward a contingency
approach. Academy of Management Journal, 46(5), 572–590.
Bell, S. T. (2007). Deep-level composition variables as predictors of team performance: A
meta-analysis. Journal of Applied Psychology, 92(3), 595–615.
Berry, C. M., Ones, D. S., & Sackett, P. R. (2007). Interpersonal deviance, organizational
deviance, and their common correlates: A review and meta-analysis. Journal of Applied
Psychology, 92(2), 410–424.
Bird, B. (1988). Implementing entrepreneurial ideas: The case for intention. Academy of
Management Review, 13(3), 442–453.
Bono, J. E., & Judge, T. A. (2004). Personality and transformational and transactional
leadership: A meta-analysis. Journal of Applied Psychology, 89(5), 901–915.
Bradley, B. H., Klotz, A. C., Postlethwaite, B. E., & Brown, K. G. (2013). Ready to
rumble: How team personality composition and task conflict interact to improve
performance. Journal of Applied Psychology, 98(2), 385–392.
Brockhaus, R. H., & Horwitz. P. S. (1985). The psychology of the entrepreneur. In D. L.
Sexton & R. W. Smilor (Eds.), The art and science of entrepreneurship. Cambridge,
MA: Ballinger.
Chiaburu, D. S., Oh, I. S., Berry, C. M., Li, N., & Gardner, R. G. (2011). The five factor
model of personality traits and organizational citizenship behaviors: A meta-analysis.
Journal of Applied Psychology, 96(6), 1140–1166.
Choi, D., Oh, I.-S., & Colbert, A. E. (2015, March 30). Understanding organizational
commitment: A meta-analytic examination of the roles of the five factor model of
personality and culture. Journal of Applied Psychology. Advance online publication.
http://dx.doi.org/10.1037/apl0000014
Christian, M. S., Garza, A. S., & Slaughter, J. E. (2011). Work engagement: A
quantitative review and test of its relations with task and contextual performance.
Personnel Psychology, 64(1), 89-136.
Colbert, A. E., Barrick, M. R., & Bradley, B. H. (2014). Personality and leadership
composition in top management teams: Implications for organizational effectiveness.
Personnel Psychology, 67, 351–387.
Collins, C. J., Hanges, P. J., & Locke, E. A. (2004). The relationship of achievement
motivation to entrepreneurial behavior: A meta-analysis. Human Performance, 17(1),
95–117.
Costa, P. T., Jr., & McCrae, R. R. (1980). Still stable after all these years: Personality as a
key to some issues in adulthood and old age. In P. B. Baltes & O. G. Brim, Jr. (Eds.),
Life span development and behavior (Vol. 3, pp. 65–102). New York, NY: Academic
Press.
Costa, P. T., & McCrae, R. R. (1992). Revised Neo Personality Inventory (NEO-PI-R)
and NEO Five Factor Inventory (NEO-FFI). Psychological Assessment Resources.
Cronbach, L. J., & Gleser, G. C. (1965). Psychological tests and personnel decisions.
Champaign, IL: University of Illinois Press.
DeGeest, D. S., & Schmidt, F. L. (2010). The impact of research synthesis methods on
industrial–organizational psychology: The road from pessimism to optimism about
cumulative knowledge. Research Synthesis Methods, 1(3–4), 185–197.
DeRue, D. S., Nahrgang, J. D., Wellman, N. E. D., & Humphrey, S. E. (2011). Trait and
behavioral theories of leadership: An integration and meta-analytic test of their relative
validity. Personnel Psychology, 64(1), 7–52.
Devine, D. J., Clayton, L. D., Philips, J. L., Dunford, B. B., & Melner, S. B. (1999).
Teams in organizations prevalence, characteristics, and effectiveness. Small Group
Research, 30(6), 678–711.
Digman, J. M. (1990). Personality structure: Emergence of the five factor model. Annual
Review of Psychology, 41(1), 417–440.
Ellingson, J. E., Sackett, P. R., & Connelly, B. S. (2007). Personality assessment across
selection and development contexts: Insights into response distortion. Journal of
Ellis, A. P., Hollenbeck, J. R., Ilgen, D. R., Porter, C. O., West, B. J., & Moon, H. (2003).
Team learning: Collectively connecting the dots. Journal of Applied Psychology, 88(5),
821–835.
English, A., Griffith, R. L., & Steelman, L. A. (2004). Team performance The effect of
team conscientiousness and task type. Small Group Research, 35(6), 643–665.
Fleishman, E. A. (1953). The description of supervisory behavior. Journal of Applied
Psychology, 37(1), 1–25.
Gartner, W. B. (1988). Who is an entrepreneur? is the wrong question. American Journal
of Small Business, 12(4), 11–32.
Griffith, R. L., & Peterson, M. H. (2006). A closer examination of applicant faking.
Greenwich, CT: Information Age Publishing
Guion, R. M., & Gottier, R. F. (1965). Validity of personality measures in personnel
selection. Personnel Psychology, 18(2), 135–164.
Gully, S. M., Devine, D. J., & Whitney, D. J. (1995). A meta-analysis of cohesion and
performance effects of level of analysis and task interdependence. Small Group
Research, 26(4), 497–520.
Hackman, J. R. (1987). The design of work teams. Ariel, 129, 32–197.
Heslin, R. (1964). Predicting group task effectiveness from member characteristics.
Hogan, J., Barrett, P., & Hogan, R. (2007). Personality measurement, faking, and
employment selection. Journal of Applied Psychology, 92(5), 1270–1285.
Hogan, J., & Holland, B. (2003). Using theory to evaluate personality and job-
performance relations: A socioanalytic perspective. Journal of Applied Psychology,
88(1), 100–112.
Hogan, J., & Roberts, B. W. (1996). Issues and non-issues in the fidelity-bandwidth trade-
off. Journal of Organizational Behavior, 17(6), 627–637.
Homan, A. C., Hollenbeck, J. R., Humphrey, S. E., Van Knippenberg, D., Ilgen, D. R., &
Van Kleef, G. A. (2008). Facing differences with an open mind: Openness to
experience, salience of intragroup differences, and performance of diverse work
groups. Academy of Management Journal, 51(6), 1204–1222.
Hoppock, R. (1935). Job satisfaction. New York, NY: Harper.
Hough, L. (1992). The “Big Five” personality variables—construct confusion:
Description versus prediction. Human Performance, 5, 139–155.
Hough, L. M., & Oswald, F. L. (2000). Personnel selection: Looking toward the future––
Remembering the past. Annual Review of Psychology, 51(1), 631–664.
House, R. J., & Aditya, R. N. (1997). The social scientific study of leadership: Quo
vadis? Journal of Management, 23(3), 409–473.
Huang, J. L., Ryan, A. M., Zabel, K. L., & Palmer, A. (2014). Personality and adaptive
performance at work: A meta-analytic investigation. Journal of Applied Psychology,
99(1), 162–179.
Humphrey, S. E., Hollenbeck, J. R., Meyer, C. J., & Ilgen, D. R. (2007). Trait
configurations in self-managed teams: A conceptual examination of the use of seeding
for maximizing and minimizing trait variance in teams. Journal of Applied Psychology,
92(3), 885–892.
Hunter, J. E., & Schmidt, F. L. (Eds.). (2004). Methods of meta-analysis: Correcting
error and bias in research findings. Thousand Oaks, CA: Sage.
Hurtz, G. M., & Donovan, J. J. (2000). Personality and job performance: The Big Five
revisited. Journal of Applied Psychology, 85(6), 869–891.
Janda, K. F. (1960). Towards the explication of the concept of leadership in terms of the
concept of power. Human Relations, 13, 345–363.
Jayne, M. E. A., & Rauschenberger, J. M. (2000). Demonstrating the value of selection in
organizations. In Kehoe, J. (Ed.), Managing selection in changing organizations (pp.
123–157). San Francisco: Jossey-Bass.
Judge, T. A., Bono, J. E., Ilies, R., & Gerhardt, M. W. (2002). Personality and leadership:
A qualitative and quantitative review. Journal of Applied Psychology, 87(4), 765–783.
Judge, T. A., Heller, D., & Mount, M. K. (2002). Five Factor model of personality and
job satisfaction: A meta-analysis. Journal of Applied Psychology, 87(3), 530–541.
Judge, T. A., Higgins, C. A., Thoresen, C. J., & Barrick, M. R. (1999). The big five
personality traits, general mental ability, and career success across the life span.
Judge, T. A., & Ilies, R. (2002). Relationship of personality to performance motivation: A
meta-analytic review. Journal of Applied Psychology, 87(4), 797–807.
Judge, T. A., & Kammeyer-Mueller, J. D. (2012a). Job attitudes. Annual Review of
Judge, T. A., & Kammeyer-Mueller, J. D. (2012b). General and specific measures in
organizational behavior research: Considerations, examples, and recommendations for
researchers. Journal of Organizational Behavior, 33, 161–174.
Judge, T. A., Klinger, R., Simon, L. S., & Yang, I. W. F. (2008). The contributions of
personality to organizational behavior and psychology: Findings, criticisms, and future
research directions. Social and Personality Psychology Compass, 2, 1982–2000.
Judge, T. A., & Piccolo, R. F. (2004). Transformational and transactional leadership: A
meta-analytic test of their relative validity. Journal of Applied Psychology, 89(5), 755–
763.
Judge, T. A., Piccolo, R. F., & Ilies, R. (2004). The forgotten ones? The validity of
consideration and initiating structure in leadership research. Journal of Applied
Judge, T. A., Rodell, J. B., Klinger, R. L., Simon, L. S., & Crawford, E. R. (2013).
Hierarchical representations of the five factor model of personality in predicting job
performance: Integrating three organizing frameworks with two theoretical
perspectives. Journal of Applied Psychology, 98(6), 875–925.
Judge, T. A., Thoresen, C. J., Bono, J. E., & Patton, G. K. (2001). The job satisfaction–
job performance relationship: A qualitative and quantitative review. Psychological
Bulletin, 127(3), 376.
Judge, T. A., & Zapata, C. P. (2015). The person-situation debate revisited: Effect of
situation strength and trait activation on the validity of the big five traits in predicting
job performance. Academy of Management Journal, 58, 1–31.
Kanfer, R., & Heggestad, E. D. (1997). Motivational traits and skills: A person-centered
approach to work motivation. In B. M. Staw & L. L. Cummings (Eds.), Research in
organizational behavior (Vol. 19, pp. 1–56). Greenwich, CT: JAI Press, Inc.
Katz, J. A. (2003). The chronology and intellectual trajectory of American
entrepreneurship education: 1876–1999. Journal of Business Venturing, 18(2), 283–
300.
Knight, F. (1921). Risk, uncertainty, and profit. New York, NY: Augustus Kelly.
Krueger, N. F., Jr., Reilly, M. D., & Carsrud, A. L. (2000). Competing models of
entrepreneurial intentions. Journal of Business Venturing, 15(5), 411–432.
Landy, F. J. (1985). Psychology of work behavior. Homewood, IL: Dorsey Press.
Landy, F. J. (1997). Early influences on the development of industrial and organizational
psychology. Journal of Applied Psychology, 82(4), 467–489.
LePine, J. A. (2003). Team adaptation and postchange performance: Effects of team
composition in terms of members’ cognitive ability and personality. Journal of Applied
LePine, J. A., Buckman, B. R., Crawford, E. R., & Methot, J. R. (2011). A review of
research on personality in teams: Accounting for pathways spanning levels of theory
and analysis. Human Resource Management Review, 21(4), 311–330.
LePine, J. A., Colquitt, J. A., & Erez, A. (2000). Adaptability to changing task contexts:
Effects of general cognitive ability, conscientiousness, and openness to experience.
LePine, J. A., Erez, A., & Johnson, D. E. (2002). The nature and dimensionality of
organizational citizenship behavior: A critical review and meta-analysis. Journal of
LePine, J. A., Hollenbeck, J. R., Ilgen, D. R., & Hedlund, J. (1997). Effects of individual
differences on the performance of hierarchical decision-making teams: Much more than
g. Journal of Applied Psychology, 82(5), 803–811.
LePine, J. A., Piccolo, R. F., Jackson, C. L., Mathieu, J. E., & Saul, J. R. (2008). A meta-
analysis of teamwork processes: Tests of a multidimensional model and relationships
with team effectiveness criteria. Personnel Psychology, 61(2), 273–307.
Levin, S. G., & Stephan, P. E. (1991). Research productivity over the life cycle: Evidence
for academic scientists. The American Economic Review, 81, 114–132.
Lievens, F., Ones, D. S., & Dilchert, S. (2009). Personality scale validities increase
throughout medical school. Journal of Applied Psychology, 94(6), 1514–1535.
Lord, R. G., De Vader, C. L., & Alliger, G. M. (1986). A meta-analysis of the relation
between personality traits and leadership perceptions: An application of validity
generalization procedures. Journal of Applied Psychology, 71(3), 402–421.
Lord, R. G., Foti, R. J., & Phillips, J. S. (1982). A theory of leadership categorization. In
J. G. Hunt, U. Sekaran, & C. Schriesheim (Eds.), Leadership: Beyond establishment
views., Carbondale, IL: Southern Illinois, University Press.
Mann, R. D. (1959). A review of the relationship between personality and performance in
small groups. Psychological Bulletin, 56, 241–270.
Manz, C. C., & Sims, H. P. (1993). Business without bosses: How self-managing teams
are building high-performing companies. New York, NY: Wiley.
Marks, M., Zaccaro, S. J., & Mathieu, J. (2000). Performance implications of leader
briefings and team interaction training for team adaptation to novel environments.
Journal of Applied Psychology, 85, 971–986.
Marks, M. A., Mathieu, J., & Zaccaro, S. J. (2001). A temporally based framework and
taxonomy of team processes. Academy of Management Review, 26, 356–376.
Mathieu, J. E., & Zajac, D. M. (1990). A review and meta-analysis of the antecedents,
correlates, and consequences of organizational commitment. Psychological Bulletin,
108(2), 171–194.
McClelland, D. (1961). The achieving society. New York, NY: Free Press.
McClelland, D. (1969). Motivating economic achievement. New York, NY: Free Press.
McCrae, R. R., & Costa, P. T. (1987). Validation of the five factor model of personality
across instruments and observers. Journal of Personality and Social Psychology, 52(1),
81–90.
Meyer, R. D., Dalal, R. S., & Bonaccio, S. (2009). A meta-analytic investigation into the
moderating effects of situational strength on the conscientiousness–performance
relationship. Journal of Organizational Behavior, 30(8), 1077–1102.
Morgeson, F. P., Campion, M. A., Dipboye, R. L., Hollenbeck, J. R., Murphy, K., &
Schmitt, N. (2007a). Reconsidering the use of personality tests in personnel selection
contexts. Personnel Psychology, 60(3), 683–729.
Morgeson, F. P., Campion, M. A., Dipboye, R. L., Hollenbeck, J. R., Murphy, K., &
Schmitt, N. (2007b). Are we getting fooled again? Coming to terms with limitations in
the use of personality tests for personnel selection. Personnel Psychology, 60(4), 1029–
1049.
Mount, M. K., & Barrick, M. R. (1995). The Big Five personality dimensions:
Implications for theory and practice in human resource management. Research in
Personnel and Human Resource Management, 13, 153–200.
Mount, M. K., Barrick, M. R., & Stewart, G. L. (1998). five factor model of personality
and performance in jobs involving interpersonal interactions. Human Performance,
11(2–3), 145–165.
Mount, M. K., Barrick, M. R., & Strauss, J. P. (1994). Validity of observer ratings of the
big five personality factors. Journal of Applied Psychology, 79(2), 272–280.
Mount, M., Ilies, R., & Johnson, E. (2006). Relationship of personality traits and
counterproductive work behaviors: The mediating effects of job satisfaction. Personnel
Psychology, 59(3), 591–622.
Muchinsky, P. M. (1983). Psychology applied to work: An introduction to industrial and
organizational psychology. Homewood, IL: Dorsey Press.
Neal, A., Yeo, G., Koy, A., & Xiao, T. (2012). Predicting the form and direction of work
role performance from the Big 5 model of personality traits. Journal of Organizational
Behavior, 33(2), 175–192.
Neuman, G. A., Wagner, S. H., & Christiansen, N. D. (1999). The relationship between
work-team personality composition and the job performance of teams. Group &
Organization Management, 24(1), 28–45.
Ng, T. W., Eby, L. T., Sorensen, K. L., & Feldman, D. C. (2005). Predictors of objective
and subjective career success: A meta-analysis. Personnel Psychology, 58(2), 367–408.
Oh, I. S., Wang, G., & Mount, M. K. (2011). Validity of observer ratings of the five
factor model of personality traits: A meta-analysis. Journal of Applied Psychology,
96(4), 762–773.
Ones, D. S., Dilchert, S., Viswesvaran, C., & Judge, T. A. (2007). In support of
personality assessment in organizational settings. Personnel Psychology, 60(4), 995–
1027.
Ones, D. S., & Viswesvaran, C. (1996). Bandwidth–fidelity dilemma in personality
measurement for personnel selection. Journal of Organizational Behavior, 17(6), 609–
626.
Ones, D. S., Viswesvaran, C., & Schmidt, F. L. (1993). Comprehensive meta-analysis of
integrity test validities: Findings and implications for personnel selection and theories
of job performance. Journal of Applied Psychology, 78(4), 679.
Ones, D. S., Viswesvaran, C., & Schmidt, F. L. (2003). Personality and absenteeism: A
meta-analysis of integrity tests. European Journal of Personality, 17(S1), S19–S38.
Peeters, M. A., Rutte, C. G., van Tuijl, H. F., & Reymen, I. M. (2006). The big five
personality traits and individual satisfaction with the team. Small Group Research,
37(2), 187–211.
Peeters, M. A., Van Tuijl, H. F., Rutte, C. G., & Reymen, I. M. (2006). Personality and
team performance: A meta-analysis. European Journal of Personality, 20(5), 377–396.
Porter, C. O., Hollenbeck, J. R., Ilgen, D. R., Ellis, A. P., West, B. J., & Moon, H. (2003).
Backing up behaviors in teams: The role of personality and legitimacy of need. Journal
of Applied Psychology, 88(3), 391–403.
Prewett, M. S., Walvoord, A. A., Stilson, F. R., Rossi, M. E., & Brannick, M. T. (2009).
The team personality–team performance relationship revisited: The impact of criterion
choice, pattern of workflow, and method of aggregation. Human Performance, 22(4),
273–296.
Pulakos, E. D., Schmitt, N., Dorsey, D. W., Arad, S., Hedge, J. W., & Borman, W. C.
(2002). Predicting adaptive performance: Further tests of a model of adaptability.
Human Performance, 15(4), 299–323.
Rauch, A., & Frese, M. (2007). Born to be an entrepreneur? Revisiting the personality
approach to entrepreneurship. In J. R. Baum, M. Frese, & R. A. Baron (Eds.), The
psychology of entrepreneurship (pp. 41–65). Mahwah, NJ: Erlbaum.
Rothstein, M. G., & Goffin, R. D. (2006). The use of personality measures in personnel
selection: What does current research support? Human Resource Management Review,
16(2), 155-180.
Rynes, S. L., Brown, K. G., & Colbert, A. E. (2002). Seven common misconceptions
about human resource practices: Research findings versus practitioner beliefs. The
Academy of Management Executive, 16(3), 92-103.
Rynes, S. L., Colbert, A. E., & Brown, K. G. (2002). HR professionals’ beliefs about
effective human resource practices: Correspondence between research and practice.
Human Resource Management, 41(2), 149-174.
Salgado, J. F. (1997). The Five Factor Model of personality and job performance in the
European Community. Journal of Applied Psychology, 82(1), 30–43.
Salgado, J. F. (2002). The Big Five personality dimensions and counterproductive
behaviors. International Journal of Selection and Assessment, 10(1–2), 117–125.
Scarr, S., Webber, P. L., Weinberg, R. A., & Wittig, M. A. (1981). Personality
resemblance among adolescents and their parents in biologically related and adoptive
families. Journal of Personality and Social Psychology, 40(5), 885.
Schmidt, J. A., Ogunfowora, B., & Bourdage, J. S. (2012). No person is an island: The
effects of group characteristics on individual trait expression. Journal of
Organizational Behavior, 33(7), 925–945.
Schmitt, N., & Oswald, F. L. (2006). The impact of corrections for faking on the validity
of noncognitive measures in selection settings. Journal of Applied Psychology, 91(3),
613–621.
Schneider, B. (1987). The people make the place. Personnel Psychology, 40(3), 437-453.
Schumpeter, J. (1911). The theory of economic development. Cambridge, MA: Harvard
University Press.
Schumpeter, J. (1942). Creative destruction. Capitalism, socialism and democracy. New
York, NY: Harper and Brothers.
Shaffer, J. A., & Postlethwaite, B. E. (2012). A matter of context: A meta-analytic
investigation of the relative validity of contexualized and non-contextualized
personality measures. Personnel Psychology, 65(3), 445–493.
Shane, S., Locke, E. A., & Collins, C. J. (2003). Entrepreneurial motivation. Human
Resource Management Review, 13(2), 257–279.
Spriegel, W. R., & Dale, A. G. (1953). Trends in personnel selection and induction.
Personnel, 30(3), 169–175.
Staw, B. M., Bell, N. E., & Clausen, J. A. (1986). The dispositional approach to job
attitudes: A lifetime longitudinal test. Administrative Science Quarterly, 31(1), 56–77.
Staw, B. M., & Ross, J. (1985). Stability in the midst of change: A dispositional approach
to job attitudes. Journal of Applied Psychology, 70(3), 469–480.
Stewart, G. L., Fulmer, I. S., & Barrick, M. R. (2005). An exploration of member roles as
a multilevel linking mechanism for individual traits and team outcomes. Personnel
Psychology, 58(2), 343–365.
Stewart, W. H., Jr., & Roth, P. L. (2001). Risk propensity differences between
entrepreneurs and managers: A meta-analytic review. Journal of Applied Psychology,
86(1), 145–156.
Stewart, W. H., Jr., & Roth, P. L. (2004). Data quality affects meta-analytic conclusions:
A response to Miner and Raju (2004) concerning entrepreneurial risk propensity.
Journal of Applied Psychology, 89, 14–21.
Stewart, W. H., & Roth, P. L. (2007). A meta-analysis of achievement motivation
differences between entrepreneurs and managers. Journal of Small Business
Management, 45(4), 401–421.
Stogdill, R. M. (1948). Personal factors associated with leadership: A survey of the
literature. The Journal of Psychology, 25(1), 35–71.
Stogdill, R. M. (1974). Leadership traits: 1904-1947. Handbook of leadership. New
York, NY: Free Press.
Stogdill, R. M., & Coons, A. E. (1957). Leader behavior: Its description and
measurement. Columbus, OH: Bureau of Business Research, College of Commerce and
Administration, Ohio State University.
Swider, B. W., & Zimmerman, R. D. (2010). Born to burnout: A meta-analytic path
model of personality, job burnout, and work outcomes. Journal of Vocational
Behavior, 76(3), 487–506.
Tasa, K., Sears, G. J., & Schat, A. C. (2011). Personality and teamwork behavior in
context: The cross-level moderating role of collective efficacy. Journal of
Organizational Behavior, 32(1), 65–85.
Tett, R. P., Anderson, M. G., Ho, C. L., Yang, T. S., Huang, L., & Hanvongse, A. (2006).
Seven nested questions about faking on personality tests: An overview and
interactionist model of item-level response distortion. In R. L. Griffith & M. H.
Peterson (Eds.), A closer examination of applicant faking behavior (pp. 43–83).
Greenwich, CT: Information Age Publishing.
Tett, R. P., Jackson, D. N., & Rothstein, M. (1991). Personality measures as predictors of
job performance: A meta-analytic review. Personnel Psychology, 44(4), 703–742.
Thoresen, C. J., Bradley, J. C., Bliese, P. D., & Thoresen, J. D. (2004). The big five
personality traits and individual job performance growth trajectories in maintenance
and transitional job stages. Journal of Applied Psychology, 89(5), 835–853.
Van Vianen, A. E., & De Dreu, C. K. (2001). Personality in teams: Its relationship to
social cohesion, task cohesion, and team performance. European Journal of Work and
Organizational Psychology, 10(2), 97–120.
Venkataraman, S. (1997). The distinctive domain of entrepreneurship research. Advances
in Entrepreneurship, Firm Emergence and Growth, 3(1), 119–138.
Viswesvaran, C., Ones, D. S., & Hough, L. M. (2001). Do impression management scales
in personality inventories predict managerial job performance ratings? International
Journal of Selection and Assessment, 9(4), 277–289.
Wageman, R., Gardner, H., & Mortensen, M. (2012). The changing ecology of teams:
New directions for teams research. Journal of Organizational Behavior, 33(3), 301–
315.
Yukl, G. (2012). Leadership in organizations (8th ed). Upper Saddle River, NJ: Prentice-
Hall.
Zaccaro, S. J. (2007). Trait-based perspectives of leadership. American Psychologist,
62(1), 6–23.
Zhao, H., & Seibert, S. E. (2006). The big five personality dimensions and entrepreneurial
status: A meta-analytical review. Journal of Applied Psychology, 91(2), 259–271.
Zhao H., Seibert, S. E., & Hills, G. E. (2005). The mediating role of self-efficacy in the
development of entrepreneurial intentions. Journal of Applied Psychology, 90(6),
1265–1280.
Zhao, H., Seibert, S. E., & Lumpkin, G. T. (2010). The relationship of personality to
entrepreneurial intentions and performance: A meta-analytic review. Journal of
Management, 36(2), 381–404.
Zimmerman, R. D. (2008). Understanding the impact of personality traits on individuals’
turnover decisions: A meta-analytic path model. Personnel Psychology, 61(2), 309–
348.
Zyphur, M. J., Chaturvedi, S., & Arvey, R. D. (2008). Job performance over time is a
function of latent trajectories and previous performance. Journal of Applied
Psychology, 93(1), 217–241.
Health Psychology
Margaret L. Kern and Howard S. Friedman
Abstract
As research on personality and health has moved to developing multitrait,
multioutcome models, the five factor approach has shown excellent utility for
understanding health, including physical and mental health, longevity, cognitive
function, social competence, and productivity. Drawing on a growing arsenal of
advanced statistical techniques, studies are testing complex models to explain
how personality influences health. Health behaviors, social situations,
physiological changes, and various indirect and moderating factors are important
pathways connecting personality and health, and reciprocally influence one
another. Future personality research will benefit from interdisciplinary
approaches, including integrative data analyses of archival data, big data analyses,
neuroscientific approaches, and lifespan epidemiology. Bringing together
different types of data, innovative methods, and well-specified theories offers the
potential to understand the personality–health model in ways never before
imagined. Identifying pathways and key factors in turn will inform effective
intervention to help more people live healthier, more productive lives.
Key Words: health outcomes, longitudinal research, lifespan pathways,
conscientiousness, disease-prone personality, self-healing, longevity
People have long dreamed of the fountain of youth. Young people believe
they will live forever, scorning the wrinkles and pains of elders. People may
have surgeries, wear makeup, run marathons, swallow supplements, and color
their hair to make themselves look and feel younger. Older individuals fear the
decline that often surrounds getting older, facing changes in their bodies and
social circles, hoping to somehow be a healthy ager. Health psychology reveals
that there are indeed paths to healthier aging. Some individuals thrive, but there
is no simple route to a happy, healthy old age. Health unfolds over time, and
individual histories, characteristics, and habitual patterns influence the processes
and outcomes that occur. Research over the past three decades has discovered
key reasons why personality is highly relevant to health.
Many factors are associated with health—but these correlations yield different
advice about which patterns are healthy and which are unhealthy. It is
impossible to randomly assign individuals to differing biopsychosocial life
patterns, and so even many of the best studies on health involve disparate
snapshots in time. Longitudinal studies provide the best windows into
personality and health relationships. Dr. George Vaillant, after studying a group
of Harvard men for over 40 years, noted: “No single interview, no single
questionnaire, is ever adequate to reveal the complete man, but the mosaic of
interviews produced by many observers over many years can be most revealing”
(Vaillant, 2012, p. 95). That is, when multiple images are brought together, a
clearer picture appears. In health psychology, a panoramic portrait that has
emerged over the past few decades from many snapshots is the importance of the
Big Five factors for health outcomes.
In this article, we first briefly examine the historical roots of personality and
health psychology. Personality became an important part of the field during a
time when the concept of personality itself was questioned. The Five Factor
Model (FFM) became an organizing model, providing a structure for
understanding personality and health associations. Second, we review a growing
body of literature relating the five factors to health outcomes, ranging from
subjective perspectives to mortality. Of the five factors, considerable evidence
suggests that Conscientiousness is particularly important for health (see also the
chapter by Jackson and Roberts). Third, we examine how and why personality
might relate to health outcomes, beginning with associations and moving to
increasingly sophisticated models. Fourth, growing numbers of tools and
strategies are now available that can be used to address the complex issues that
personality and health raise. Finally, we examine the implications and
applications of the FFM for health psychology.
Historical Perspectives
Links between personality and health have been noted for thousands of years.
In the ancient Greek era, health was seen as the balance of four humors. It was
generally believed that excessive black bile (melancholy) caused depression,
cancer, and degenerative diseases, yellow bile (choler) caused hostility and
fevers, phlegm (apathy) caused rheumatism, and balanced blood (sanguine)
reflected the healthy individual (Friedman, 2007). To Plato, it was not only the
well-balanced body, but also the balanced soul and mind. These dispositions
interacted with behaviors and life circumstances to either maintain a healthy
balance or cause mental and physical problems. Cures aimed at restoring bodily
imbalances.
Such notions of the influence of character on physical health persisted in
various incarnations for the next two millennia. In the mid-twentieth century,
personality characteristics and physical illness and disability were again linked
together. Specific traits were believed to cause specific illnesses. A major
emphasis was on the Type A behavior pattern, a composite of traits such as
tension, hostility, aggression, hurrying, and competitiveness, which was seen as
a primary risk factor for coronary heart disease. But in 1987, Friedman and
Booth-Kewley conducted a meta-analysis and found evidence of a general
“disease-prone personality,” rather than specific traits causing specific diseases.
The findings challenged the field to simultaneously consider multiple traits or
characteristics, as well as multiple health outcomes.
The Five Factors and Health Outcomes

Although early health psychology research focused on Type A behavior and
related domains of hostility, over the past two decades the FFM has become the
dominant organizing framework for integrating studies of personality and health
(Smith & Williams, 1992). A large and growing number of articles have
examined links between the five factors (or related traits) and various health
outcomes. Generally speaking, the five factors have been related to
consequential life outcomes at the intrapersonal, interpersonal, and community
level, including physical, mental, and social well-being; relationships with peers,
family, teachers, employers, and romantic partners; job performance; political
attitudes; crime; and community involvement; with effect sizes equal to or
greater than socioeconomic status and cognitive ability (Heckman, Stixrud, &
Urzua, 2006; Ozer & Benet-Martínez, 2006; Roberts, Kuncel, Shiner, Caspi, &
Goldberg, 2007).
Personality research in health psychology has used a variety of health metrics,
ranging from self-reported health to mortality. In 1948, the World Health
Organization defined health as “physical, mental, and social wellbeing, not
simply the absence of disease.” Yet there are considerable inconsistencies in
how the term is used, including discrepancies between lay and scientific
definitions, disagreements within and across fields, whether health is seen as a
process or an outcome, multidimensionality, and poor measurement.
Here, we define health as an outcome, personality as an independent variable,
and other factors as mediators, moderators, or confounding variables. Studies
(including our own) suggest five to eight meaningful health outcomes: self-rated
health, physical (medical) health, longevity, mental health and subjective well-
being, cognitive function, social competence, and productive function (Aldwin,
Spiro, & Park, 2006; Baltes & Baltes, 1990; Friedman & Kern, 2014). From a
biopsychosocial perspective, the successful life minimizes morbidity until the
final year or so, thus encouraging thriving and accomplishment, while placing
the least social and economic strain on the system (Fries, 1990).
Physical (Biological or Medical) Health

Physical health is perhaps the most common use for the term “health.”
Physical health involves physiological function, organic disease, and energy for
completing daily tasks. On the negative side, it is defined by physician-
diagnosed disease, disability, inability to complete daily activities of living, and
failure to thrive. What health means in the absence of disease is not well
understood. Some evidence suggests that health is marked by balance across
physiological systems, harkening back to notions of humoral balance. For
example, high blood pressure is associated with various diseases and thus is
considered a marker of poor health, but low blood pressure can also indicate
disease or physiological dysfunction. Similarly, a normal body weight is
protective; disease is associated with too much weight (obesity) and too little
weight (frailty, anorexia). However, deviations in blood pressure, weight, and
other classic diagnostic tools of the medical examination may turn out to be
healthy when other variables and circumstances are taken into consideration. In
psychology, this balance idea has been extended to what Friedman (1991)
termed the self-healing personality, but the phenomenon turns out to be a
complex one.
Some studies have focused on biomarkers, which we define as processes, not
final outcomes. Often, suspected biomarkers are not the causal links that they
initially appear to be (Friedman & Kern, 2014). It is more important to focus
directly on morbidity, illness, and physical functioning. Most consistently,
Conscientiousness has been related to better physical health outcomes and
Neuroticism has been related to poorer physical health outcomes. For example,
in a longitudinal study with elderly patients, higher levels of Neuroticism and
lower levels of Conscientiousness were associated with greater physician-rated
illness across a 4-year period (Chapman, Roberts, Lyness, & Duberstein, 2013).
In a twin study with young adults, participants high in Neuroticism at baseline
were more likely to report 13 different physical health conditions 25 years later,
including chronic fatigue syndrome, ulcers, and coronary heart disease (Charles,
Kato, Gatz, & Pedersen, 2008). In a panel study with nearly 7,000 British adults,
child intelligence related to better health status at age 50 years, but this
association was mediated by Emotional Stability and Conscientiousness (Cheng
& Furnham, 2013). In a nationally representative sample of U.S. adults,
Conscientiousness was associated with a significantly reduced likelihood of a
wide range of mental and physical disorders among adults in the general
population, and Neuroticism was associated with increased rates, ranging from
stroke to sciatica, with other five factor traits also sometimes relevant (Goodwin
& Friedman, 2006).
Self-Rated Health
Self-rated health is a component of subjective well-being, although it is often
mistakenly treated as an objective physical health marker. Questions assessing
self-rated health are typically face valid and easy to use (e.g., an item might ask
“In general, how is your health”). Notably, self-rated health is a good predictor
of mortality risk (Idler & Kasl, 1991), but this does not mean that it can be used
as a proxy for objective physical health (Friedman & Kern, 2014); age, sex,
socioeconomic status (SES) and certain personality traits are also good
predictors of mortality risk and are obviously not substitutes for assessing
physical health.
Neuroticism has consistently been linked to lower self-rated health (see also
the chapter by Tackett and Lahey). It is unclear how much of this is a real
association and how much is due to self-selection or measurement biases. Both
stress and self-reported health scales often contain a large negative affect
component, creating noisy measures (Costa & McCrae, 1987; Watson &
Pennebaker, 1989). People high in Neuroticism feel more pain, report more
illness, and seek more care. It is an open question as to whether Neuroticism is
indeed related to worse health, or if links are superficially created through poor
measurement.
The other four personality factors appear to be less susceptible to this
measurement problem, although the subjective nature of the self-report
assessment remains. Conscientiousness has been robustly related to better self-
rated health, predicting better reports both cross-sectionally and longitudinally,
with effects sizes stronger than intelligence and socioeconomic status (SES)
(e.g., Hampson, Goldberg, Vogt, & Dubanoski, 2006; Roberts et al., 2007; Tam
& Wi, 2014). A large-scale Internet study with over 450,000 people worldwide
sponsored by the British Broadcasting Corporation found that low
Conscientiousness related to lower reports of self-rated health, being overweight,
and engaging in substance use; Neuroticism was related to lower self-reported
health, and high Extraversion was related to more frequent substance use
(Atherton, Robins, Rentfrow, & Lamb, 2014). Notably, for self-rated health,
when items referring to stress and emotional problems were removed,
associations with Neuroticism were much weaker, pointing back to the problem
of overlapping predictors and outcomes.
Longevity
Length of life is the clearest measure of health, as it is valid and reliable, and
has consistently been used as a key measure of health worldwide (Friedman &
Kern, 2014). The best longevity studies require longitudinal analyses that track
people over many years or a lifetime, providing a more complete picture of the
trajectories that each person followed. Over the past two decades, there has been
considerable work in personality–health research focused on links between
personality and all-cause mortality. Some studies have also considered other
causes of death such as heart disease and cancer mortality. A problem with
single cause of death analyses is that they can superficially provide “good”
short-term results but “poor” long-term results; if a person is saved from cancer
but dies from heart disease (such that the overall length of life is the same), then
it is not a particularly successful treatment.
Early work focused on Neuroticism/negative affectivity as a primary cause of
early death, but Neuroticism has been inconsistently linked to mortality. Studies
find no association with mortality (e.g., Almada et al., 1991; Iwasa et al., 2008;
Taylor et al., 2009), a negative association (e.g., Korten et al., 1999; Taga,
Friedman, & Martin, 2009; Weiss & Costa, 2005), and a positive association
(e.g., Christensen et al., 2002; Denollet et al., 1996; Shipley et al., 2007; Weiss,
Gale, Batty, & Deary, 2009; Wilson, Mendes de Leon, Bienias, Evans, &
Bennett, 2004). In the U.K. Health and Lifestyle Study, Neuroticism was related
to higher risk, but was reduced to nonsignificance after controlling for SES,
education, smoking, alcohol use, physical activity, and self-rated health (Shipley
et al., 2007). It may be that some aspects of Neuroticism are protective and
others are harmful (Chapman, Roberts, Duberstein, 2011), or it may depend
upon interactions with other personality traits or the social environment (Smith,
Baron, & Grover, 2014). A more complex approach is needed.
The strongest, most consistent findings for personality and longevity have
appeared for Conscientiousness, with higher levels associated with lower
mortality risk (Bogg & Roberts, 2013; Chapman et al., 2011; Friedman et al.,
1993; Kern & Friedman, 2008; Roberts, Lejuez, Krueger, Richards, & Hill,
2014). A meta-analysis of 20 studies with diverse samples found that high levels
of Conscientiousness are robustly predictive of lower risk (Kern & Friedman,
2008). Subsequent studies have consistently confirmed this finding (e.g., Iwasa
et al., 2008; Taylor et al., 2009; Terracciano et al., 2008). Research has now
shifted from establishing that the Conscientiousness-longevity association exists
to explaining possible pathways and mechanisms.
Agreeableness has received less attention than the other five factors, and
associations have been mixed, with many studies finding null results (e.g.,
Christensen et al., 2002; Iwasa et al., 2008; Taylor et al., 2009). Most clearly,
hostility and cynicism (i.e., low Agreeableness) have been linked to heart disease
mortality and all-cause mortality (Almada et al., 1991; Bunde & Suls, 2006;
Chapman et al., 2011). Extraversion generally has not been associated with
mortality risk, most likely because it depends on the pathways involved. That is,
Extraversion is associated with both healthy patterns (e.g., sociability) and
unhealthy patterns (e.g., smoking, drinking, thrill-seeking).
Openness to experience has also demonstrated mixed associations with
longevity, although a meta-analysis with nearly 20,000 people suggested that it
is protective, but its effect is attenuated by other risk factors (e.g., age, social
class) (Ferguson & Bibby, 2012). The extent to which openness matters beyond
SES, education, or intelligence is unclear.
Mental Health and Well-Being

Subjective well-being (SWB) is inherently subjective, and includes emotion,
self-rated health, cognitive evaluation of life, and other related components.
Diener (2012) defines subjective well-being in terms of affect (high positive
affect, low negative affect) and cognition (life satisfaction). Others have
suggested multidimensional approaches to well-being. For example, Ryff and
Keyes (1995) define well-being in terms of six components: autonomy,
environmental mastery, personal growth, positive relationships, purpose in life,
and self-acceptance. Seligman (2011) defined five pillars of well-being: positive
emotion, engagement, relationships, meaning, and accomplishment. Huppert and
So (2013) suggested 10 domains of the flourishing life contrasted against
depression and anxiety: competence, emotional stability, engagement, meaning,
optimism, positive emotion, positive relationships, resilience, self-esteem, and
vitality. Across these different definitions, there appear to be two main
components: hedonic well-being, referring to the emotional side, and
eudaimonic well-being, or the good life (Ryan & Deci, 2001). Life satisfaction
possibly represents a third, cognitive component.
A meta-analysis found that when properly controlled in a multivariate
approach, personality can explain anywhere between 39% and 63% of the
variance in well-being (i.e., happiness, life satisfaction, positive and negative
affect, and quality of life) based upon NEO measures (Steel et al., 2008).
Neuroticism and Extraversion consistently related to lower and higher well-
being, respectively, with moderate to large effect sizes. Conscientiousness was
moderately associated with well-being, with the strongest associations for
quality of life. Agreeableness effects were low to moderate, and Openness was
only weakly related to well-being.
Steel and colleagues (2008) suggested four reasons why personality and SWB
should be correlated (see also Caspi et al., 2014). First, a growing number of
theories and research point to biological elements (such as serotonin, dopamine,
and some genes) that influence both how traits are manifested and proneness
toward mental health or illness. Second, there is conceptual and measurement
confusion. For example, an item on the mental health component of the Short
Form-36 (Ware, Kosinski, & Keller, 1994) is “How much of the time during the
past 4 weeks have you felt downhearted and blue,” and an item assessing
Neuroticism is “Often feel blue” (www.ipip.org). Not surprisingly, the two
measures are correlated. Third, happiness levels tend to have a relative set point
(Lyubomirsky, Sheldon, & Schkade, 2005). Correlations between well-being and
personality measures will increase to the extent that well-being measures capture
stable characteristics of the person rather than momentary states. Finally,
personality influences the situations that people select and are drawn toward, and
these situations have a strong influence on attitudes and emotions (Friedman,
2000).
Cognitive Function
The cognitive aspect of health includes mental processes such as memory,
perception, language, reasoning, decision making, and spatial ability. Numerous
tests have been developed, ranging from brief mental screenings to complex
cognitive batteries that capture both function and timing of declines. At the
negative extreme is dementia and Alzheimer’s disease, in which almost all
cognitive function is gone. Although it is commonly assumed (and perhaps
feared) that cognitive ability decreases over time, studies suggest that although
fluid intelligence (e.g., memory capacity, speed of processing) decreases,
crystallized intelligence (e.g., cultural knowledge) generally remains fairly stable
(Staudinger, Cornelius, & Baltes, 1989). There is also considerable variation;
some individuals can maintain very effective cognitive performance even in old
age (Reynolds et al., 2005; Richardson, 2005).
Studies are increasingly linking personality to cognitive function. Across a 6-
year period, risk for Alzheimer’s disease was associated with high levels of
Neuroticism, low Openness, and low Conscientiousness (Duberstein et al.,
2011). In the Lothian Birth Cohort, low Conscientiousness related to greater loss
of brain tissue and hyperintensities of white matter, with effects partially
mediated by health behaviors (Booth et al., 2014). A meta-analysis of 15 studies
found that high Neuroticism was associated with higher risk for dementia,
Conscientiousness was protective, and Openness, Extraversion, and
Agreeableness were not reliably related (Low, Harrison, & Lackersteen, 2013).
Findings again point to the cumulative physiological effects that
Conscientiousness and Neuroticism seemingly have over the life course.
Social Competence and Productive Function

Social competence and productive function are two final health outcomes that
are valuable both for individuals and for society as a whole. Consistent with the
World Health Organization definition of health (described above), the well-being
of the individual and the well-being of society ought to be mutually reinforcing.
Social competence refers to being able to develop and maintain positive social
relationships with others. It is separate from social support—a contextual
variable that can influence social relationships, but is not the same as how good
those relationships are. It has become clear over the past decade that social
relationships are very important in the thriving life (Tay, Tan, Diener, &
Gonzalez, 2012; Taylor, 2011).
Whether in paid work or social/civic engagement, productivity allows society
to function smoothly. Engaging in work and having a sense of productivity
contributes to better life quality (Bambrick & Bonder, 2005). In his presidential
address to the American Psychological Association, Robert Kaplan (1994)
brought to light the importance of productive engagement. In a Ziggy cartoon,
Ziggy asks the wise sage what the meaning of life is, and is told “life is doin’
stuff … as opposed to death, which is not doing stuff” (p. 452).
Perhaps the clearest markers of social competence and productivity are
successful marriage (versus divorce or isolation) and educational/occupational
attainment (versus unemployment or idleness), respectively. Neuroticism
predicts divorce, and high Agreeableness and Conscientiousness are related to
more stable and better marriages and less divorce (Roberts et al., 2007). In the
lifespan Terman sample, across a 40-year period, Extraversion, Agreeableness,
and Conscientiousness predicted greater social competence, and
Conscientiousness predicted greater productivity (Friedman, Kern, & Reynolds,
2010). Agreeableness is particularly important for maintaining harmonious
relationships with other people (Jensen-Campbell, Knack, & Gomez, 2010; see
also the chapter by Graziano and Tobin). Each of the Big Five traits has been
related to educational and occupational attainment, with Extraversion,
Agreeableness, Conscientiousness, intellect, and low Neuroticism related to
greater achievement and more stable employment (Roberts et al., 2007).
Linking Personality and Health

Now that we know that personality matters to health, the interesting question
quickly becomes: How and why does it matter? Theories and studies are now
breaking open the black box to expose the complex underpinnings of how
personality affects health and how health affects personality.
Mechanisms and Moderators

Various pathways have been proposed to link personality and health outcomes
(Chapman, Hampson, & Clarkin, 2014; Hampson, 2008; Kern & Friedman,
2010; Smith, 2006). Pathways include health behaviors, social situations,
physiological changes, and indirect and moderating factors.
Health behaviors. The clearest pathway linking personality and health is
through the behaviors in which people do and do not engage. Across 194 studies,
Conscientiousness-related traits were positively related to health protective
behaviors and negatively related to risky behaviors (Bogg & Roberts, 2004). In
the Midlife in the United States (MIDUS) study, Conscientiousness and health
links were substantially mediated by excessive alcohol use, smoking, and waist
circumference (Turiano, Chapman, Gruenewald, & Mroczek, 2015).
Conscientious individuals are more likely to adhere to medical advice, with
physical health consequences (Hill & Roberts, 2011; O’Cleirigh, Ironson, Weiss,
& Costa, 2007). Smoking explains part of the Neuroticism–mortality association
(Mroczek, Spiro, & Turiano, 2009; Shipley et al., 2007). Extraversion,
Agreeableness, and Openness have been inconsistently linked to risky and
healthy behaviors.
Although numerous studies find that behaviors mediate the personality–
mortality link, considerable variance consistently remains, suggesting that other
pathways also matter. For example, in the Whitehall II cohort study, associations
between Conscientiousness and mortality were attenuated only 13% when health
behaviors were controlled (Hagger-Johnson et al., 2012). Adding in SES,
physiological risk factors, and minor psychiatric morbidity further attenuated
associations by 16%. Studies using structural equation modeling suggest that
both health behaviors and education mediate Conscientiousness and health
associations (Kern, Hampson, Goldberg, & Friedman, 2014; Lodi-Smith et al.,
2010), but that is only the beginning.
Alternative study designs that unearth how personality plays out in daily life
may be informative. For example, a 28-day study found that Conscientiousness
related to less consumption of high fat snacks, but also higher caffeine intake
and more smoking among those who did smoke (O’Connor, Conner, Jones,
McMillan, & Ferguson, 2009). At the facet level, participants high in order did
more exercise on stressful days, self-efficacy related to eating vegetables, and
smokers high in self-discipline smoked more on stressful days, pointing to
interactions between daily stress and health behaviors. Other studies similarly
suggest that the facet level may be more predictive of behaviors than the five
factors alone (Kern et al., 2013; Paunonen, 1998; Paunonen & Ashton, 2001).
Social pathways. A second set of important pathways connecting personality
and health is through social influences—situation selection, aspects of the social
context, and interactions with other people, all of which influence and affect
personality and health. Personality draws people toward certain situations, which
become self-reinforcing over time (Friedman, 2000). For example, high
Extraversion is related to experiencing more positive life events (both
subjectively and objectively rated), whereas Neuroticism relates to experiencing
more negative life events (Magnus, Diener, Fujita, & Pavot, 1993). Events and
experiences in turn impact self-perceptions.
Self-regulation, social responsibility, Agreeableness, and emotional stability
may help set the stage for better interpersonal relationships. In a group of
adolescents, high Conscientiousness related to better friendships, higher peer
acceptance, and less victimization, with associations mediated by externalizing
behaviors (Jensen-Campbell & Malcolm, 2007). In the Mills Longitudinal Study
of Women, social responsibility predicted family, work, and behavioral
outcomes 20 to 30 years later, and marital quality, work behaviors, and
substance use were associated with subsequent changes in social responsibility
(Roberts & Bogg, 2004). A meta-analysis of 94 studies found that high
Agreeableness, Emotional Stability, and Conscientiousness related to greater
investment in work, family, religion, and volunteering social roles (Lodi-Smith
& Roberts, 2007). Notably, the personality of others can also impact health; a
spouse with high levels of Conscientiousness predicted better physical health
outcomes, over and above the impact of the person’s own personality
(Karademas, & Tsaousis, 2014; Roberts, Smith, Jackson, & Edmonds, 2009).
The company that we keep matters for both who we are as people (our
personality) and for subsequent life outcomes.
Physiological pathways. As tools and methods have improved, considerable
research is now turning toward internal mechanisms, attempting to untangle how
personality affects physiological function, with possible health consequences. In
a sample of healthy adults, participants scoring low on Agreeableness and
Extraversion demonstrated increased sympathetic nervous system activity and,
for Extraversion, higher natural killer cell cytotoxicity, suggesting that
personality is associated with basal physiological levels (Miller et al., 1999). In
adults from Sardinia, high Neuroticism and low Conscientiousness were
associated with higher levels of interleukin-6 (Sutin, Terracciano, Deiana,
Naitza, et al., 2010), and low Conscientiousness was related to lower high-
density lipoprotein (HDL) cholesterol and higher triglycerides (Sutin,
Terracciano, Deiana, Uda, et al., 2010). Serotonin has been linked to impulsivity
(low Conscientiousness) and hostility (low Agreeableness) (Carver & Miller,
2006).
Much of the work on physiological pathways has stemmed from a general
focus on the deleterious effects of stress. According to the basic stress model,
high levels of stressors can trigger a state of homeostatic disruption in which the
heart beats faster, breathing speeds up, cortisol and catecholamine levels
increase, inflammation increases, and immune function is depressed (Kemeny,
2007; McEwen, 2006). Although in the short term, this reaction may be
adaptive, in the long term, a chronic negative response pattern becomes
detrimental, disrupting metabolism, immune function, and physiological
rhythms, and increasing susceptibility to illness, disease, and general breakdown.
Illness in turn further affects psychological functioning, creating a negative cycle
toward disease.
Personality seemingly impacts what stressful events are encountered, the
extent to which stressors are considered stressful, and coping responses (Carver
& Connor-Smith, 2010; Connor-Smith & Flachsbart, 2007; Segerstrom &
O’Connor 2012). Neuroticism is consistently viewed as the trait that increases
the risk for maladaptive responses, chronic stress, and poor health outcomes. For
example, in one study, blood pressure in women high in Neuroticism showed
less recovery after a hostile interaction than after a friendly interaction
(Hutchinson & Ruiz, 2011). In a second study, Neuroticism was related to higher
daily cortisol measures across a 6-day period (Nater, Hoppmann, & Klumb,
2010). Across a 10-year period, hostility was related to less healthy diastolic
blood pressure, although effects varied by gender (Leclerc, Rahn, & Linden,
2006; see also Smith, Glazer, Ruiz, & Gallo, 2004; Williams & Williams, 2012).
Positive traits may buffer the physiological system from stress reactions or
quickly restore balance to the system when stress occurs (Fredrickson, 2001;
Pressman & Cohen, 2005).
Studies addressing this model typically measure personality concurrently with
physiological markers, such as blood pressure, immune and endocrine function,
or cardiovascular reactivity. An assumption of this model is that cross-sectional
and short-term associations between personality and physiological function will
extend to disease outcomes later in life. However, studies are only beginning to
test the process longitudinally, and almost no studies have examined early life
personality, prolonged stress exposure, long-term physiological disruption, and
subsequent disease, at least in the same individuals. In one of the first long-term
tests of the physiological pathway, in the Hawaii Personality and Health study,
low childhood Conscientiousness was related to physiological dysregulation,
greater obesity, and worse lipid profiles 40 years later (Hampson, Edmonds,
Goldberg, Dubanoski, & Hillier, 2013). Personality may affect the immune
system, and the immune system may impact personality, such that causal arrows
are nonexistent (Kemeny, 2007). Longitudinal studies across long life periods,
which include personality, physiological markers and stress measures across
multiple time points, and health outcomes are necessary to truly establish the
physiological pathway.
Third variables. Another way that personality and health can be connected is
through third or confounding variables—factors that relate independently to
personality and health, such that the two appear to be related, when it is really
another factor that drives both relationships. The most notable third factor is
genetic influences, although a host of other factors are potential confounders as
well. Personality has both genetic and environmental influences (Jang, Livesley,
Angleitner, Riemann, & Vernon, 2002; South & Krueger, 2014), and some of
these same influences also impact health outcomes. Genetic studies could reveal
biological systems underlying trait and health differences, and may moderate
how traits are manifested across the lifespan (South & Krueger, 2014). Some
traits may turn on or off at different points in life, with health-related
implications. Much of the work with genetics and personality shows promise,
but the data are messy. As methods are continually refined, genetic studies will
continue to play an important role in the future, but genetics have often been
found to have relatively little predictive value unless environmental influences
are simultaneously considered.
Similarly, the enduring social context, including SES and risky environments,
influences both personality and health. For example, in MIDUS, personality
explained about 20% of the SES gradient in mortality risk, and SES explained
8% of personality risk (Chapman, Fiscella, Kawachi, & Duberstein, 2010), such
that some of the health risk comes from the environment, part is due to
individual dispositions, and part may be the interaction of the person and
environment. Early environments may also promote particular traits and set
various health trajectories. For example, secure attachment has been linked to
optimal self-regulations in childhood and adolescence (Eisenberg, Duckworth,
Spinrad, & Valiente, 2014). More complex models are needed, which consider
how personality and social contexts interact to influence health outcomes across
the lifespan.
Moderators. Finally, personality can also moderate and be moderated by
other variables. In the Terman Life Cycle Study, Conscientiousness moderated
the relationship between career success and mortality, such that
Conscientiousness buffered the otherwise risky effects of career failure (Kern et
al., 2009). In the German Socio-Economic Panel Study, Agreeableness
moderated adaptation to disability, such that those high in Agreeableness
regained general life satisfaction several years later, whereas disagreeable
individuals did not (Boyce & Wood, 2011). In another sample, Openness,
Extraversion, and Neuroticism moderated associations between stress and health
behaviors (Korotkov, 2008). In a French sample, although Neuroticism related to
worse physical functioning, this effect was amplified at low levels of education
and diminished at high levels of education (Jaconelli, Stephan, Canada, &
Chapman, 2013). In a group of children and adolescents, associations between
prosocial behavior, externalizing problems, internalizing problems, and grades
were moderated by the early family environment, with effects being stronger in
more difficult circumstances (Slobodskaya, Ashmetova, & Rippinen, 2014).
Summary. By peering into the black box, we have identified multiple
pathways that connect personality and health, which are often studied separately.
The clearest and perhaps most straightforward pathway is health behaviors.
Interventions need to be sensitive to personality dispositions, and by changing
one’s behaviors, risky characteristics may potentially shift to healthier ones
(Chapman et al., 2014). However, the social context (especially relationships
with other people and socioeconomic status), biological factors, and various
environmental factors selected by and shaping personality are also very
important, and may impact both behaviors and health outcomes. Personality
shapes how a person experiences, interprets, and responds to events throughout
the lifespan (Caspi, Roberts, & Shiner, 2005). It can moderate pathways and
reactions, and effects can be bidirectional, synergetic, and cumulative. To truly
understand the web of influences, models that simultaneously bring together
multiple pathways are needed.
A New Generation of Personality–Health Research

Friedman and colleagues (2014) introduced a new generation of personality
and health models, characterized by complex, dynamic, lifespan models. Rather
than separating out the pathways reviewed above, third generation models
recognize that all of these pathways matter, and most likely reciprocally
influence one another. An analogy for this new generation of models is a ship
traveling on a journey. As the journey begins, its course is set by the captain.
The shape and depth of the water determine how big the ship can be and how
fast it can travel, just as genetic and environmental factors set boundaries for
development. Personality is the rudder, steering the boat through the waters. It
sets a course, which can shift, but takes time and effort. Just as the pathway for
the journey is unknown to the passengers, many of the early influences on
personality may take time to appear. As river conditions and obstacles are
encountered, the captain might adjust the rudder, shifting the course of the
journey, just as health behaviors, social relationships, physiological responses,
and health outcomes affect personality. Thus, it is the dynamic process of a
relatively stable trajectory in the midst of an unsteady and changing context that
results in health-promoting or risky life pathways. By understanding the range of
influences involved, interventions can be done at the right time to avoid danger
and to travel an optimal route.
New theories and models within this framework provide some guidance for
beginning this new generation of studies. For example, Smith and colleagues
(2014) introduced the interpersonal perspective of personality, social, and
clinical psychology in which psychosocial risk accumulates over long periods,
resulting from dynamic interactions between individuals and their social
contexts. Interpersonal experiences such as marriage can trigger physiological
stress responses, which can result in cardiovascular disease or other health
problems as stress becomes chronic. The overt experiences and covert behaviors
reciprocally build either positive or negative patterns, which in turn affect
cardiovascular reactivity, neuroendocrine responses, and other physiological
components. Personality influences who ends up in different relationships, yet it
is the resulting dynamic interaction patterns that have a stronger impact on
chronic health risks. Eaton and colleagues (2012) proposed a framework aimed
at the complex genetic and environmental influences on healthy aging. Rather
than measuring health behaviors at a single period, behaviors should be
measured as patterns over time (see also Friedman & Martin, 2011). They
further suggest that approaches from behavioral genetics, such as co-twin
controls and gene-by-environment analyses, can provide a structure for more
sophisticated models and analyses.
Shanahan and colleagues (2014) proposed the Life Course of Personality
(LCP) model, which suggests that different mechanisms matter to health at
different phases across life, and are moderated by the social context and other
aspects of the person. Associations may depend on the specific facet. For
example, for Conscientiousness, facets such as deliberation, responsibility, self-
discipline, and low impulsivity may decrease chronic stress, whereas facets such
as achievement striving, orderliness, and perseverance may increase daily stress.
Personality is embedded in the social context throughout development,
beginning with an infant’s temperament and early interactions with parents;
behavioral and educational choices in adolescence; decisions about school,
work, and life during the transition to adulthood; and behavioral, work, and
family patterns that occur in adulthood. To study such a model, there is a strong
need to establish large, diverse samples followed over extensive time periods, as
well as specific, strategic samples to study more fine-grained time-limited
processes.
Testing dynamic models will be a challenging endeavor, and personality–
health researchers will benefit from staying current on the ever-developing
statistical tools available, as well as working collaboratively with statisticians
who develop new tools. Dynamic processes can be misconstrued with traditional
statistical models. Data are needed that follow people prospectively over time,
with repeated measures of personality, health behaviors, physiological functions,
social relationships, contextual elements, and health measures. Research needs to
be theory driven, yet still open enough for exploration, with empirical findings
informing the theoretical models. Below, we review some of the innovative
trends that will take personality–health research into the future. To truly have an
impact on health, our models and methods need to move to a greater level of
sophistication.
Drawing on a growing arsenal of advanced statistical techniques, studies are
starting to test more dynamic models. For example, using structural equation
modeling (SEM) to directly model lagged reciprocal effects, physical health
influenced subsequent positive affect, whereas emotion did not affect subsequent
physical health reports (Finch, Baranik, Liu, & West, 2012). Using SEM to
capture multiple processes in the causal chain across a 25-year period, early
adversity was related to more smoking and higher body mass 16 years later,
which in turn were related to inflammatory markers measured 1 to 4 years later
(Raposa et al., 2014). Combining SEM and survival analyses, Neuroticism was
indirectly associated with mortality risk through somatic health, psychological
distress, and smoking pathways, and Extraversion was indirectly associated with
mortality through smoking (Ploubidis & Grundy, 2009).
Beyond single assessments of personality, health, and other variables, several
studies have examined change over time. Across a 3-year period, latent change
models indicated that changes in Conscientiousness were related to changes in
health behaviors and self-perceived physical health (Takahashi, Edmonds,
Jackson, & Roberts, 2013). In MIDUS, becoming less Conscientious and
becoming more Neurotic over a 10-year period were related to worse perceived
health and well-being, a greater presence of metabolic syndrome components,
and a greater likelihood of diagnosis for metabolic syndrome at follow-up
(Human et al., 2013), and changes in Conscientiousness were related to work
limitations due to physical or mental reasons (Turiano et al., 2012). In an
Australian study, increases in Conscientiousness and Extraversion were related
to improved self-reported mental and physical health, and increases in
Neuroticism related to worse self-reported health, but associations were
moderated by birth cohort (Magee, Heaven, & Miller, 2013). In the VA
Normative Aging Study, men with both a high average level of Neuroticism and
increasing levels over time were at a higher mortality risk than those with lower
levels or nonincreasing levels (Mroczek & Spiro, 2007). In the Lothian Birth
Cohort of 1921, decreases in Conscientiousness across the ninth decade of life
were associated with declining physical fitness (Mõttus, Johnson, Starr, &
Deary, 2012).
Tools for the Future of Personality and Health

To overcome the flaws of studies based on self-reported data, future
personality research will benefit from creative, interdisciplinary approaches.
Numerous tools (e.g., innovative statistical approaches), data sources (e.g.,
behavioral observations, big data), and strategies (e.g., personality neuroscience,
computational social science, animal comparisons) are increasingly available
that will propel personality-health research into an exciting future.
The Power of Existing Studies: Long Data

Two decades ago Jack Block (1993) suggested that to truly study personality,
studies should be intentional, open, long enough to observe things, theory-based,
broad and deep, methodologically sound and well-orchestrated, innovative, have
a reasonable sample size, and maintain high quality over time. Fortunately, over
the years, many studies have been conducted that meet many of these qualities.
Much of our existing knowledge of life course personality–health associations
stem from long data—extensive data gathered prospectively across time
documenting people’s lives. For example, much of the personality–health focus
over the past decade has been on the importance of Conscientiousness for health
outcomes. The Conscientiousness–longevity link was first uncovered in an
exploration of data from the lifelong Terman Life Cycle Study (Friedman et al.,
1993). Through over 40 years of working with the Harvard Grant Study, George
Vaillant (2012) noted: “I’ve been studying adult development since I was thirty,
and I know now that many of my past conjectures, apparently accurate at the
time, were contingent or just plain wrong” (p. 1472). Only by looking at detailed
information collected over many years can true relationships be fully revealed.
“Long term longitudinal studies are like mature trees … like a century-old oak,
such studies are rare resources and can add to our knowledge base in ways newer
longitudinal studies cannot” (Mroczek, 2014, p. 1472). Fortunately, studies such
as the Terman Life Cycle Study and the Harvard Grant Study are converging on
similar conclusions (Friedman & Kern, 2014).
Through existing longitudinal studies, it is now possible to model more
complex pathways, especially when multiple measurement occasions were
included. For example, survival analyses estimate mortality risk across different
ages, while accounting for attrition and for long-lived individuals who are still
alive at the point of analysis. Growth curve analyses with two or more time
points can estimate whether the sample as a whole is changing over time and the
extent to which individuals vary from that overall trajectory. Both initial levels
(scores at baseline) and slope (changes over time) can be related to other
predictors and outcomes. Cross-lagged analyses start to untangle causal
pathways (e.g., physical activity at time 1 affecting physical fitness at time 2
versus fitness affecting activity). SEM allows complex associations to be
estimated, while directly including measurement error, direct and indirect
pathways, mechanisms, moderators, control variables, and correlated error in the
model. Growth mixture modeling identifies unobserved groups that change
differently over time. An arsenal of modeling options is increasingly available.
Of course, longitudinal data are far from perfect (Tomlinson-Keasey, 1993).
The original investigators usually were interested in different questions, scales
are often nonexistent, and constructs are often represented by only one or two
items. Data may be missing, although newer statistical techniques provide tools
to deal with missing data problems. Attrition is a notorious challenge, and often
those at highest health risk are those least likely to stay in the study. Studies
were often done on select samples that may not generalize to other population
groups (although this is often less of a problem than it first appears). Work with
archival data needs to be driven by theory and specific research questions. It is a
long process, involving developing an extensive knowledge of the data, carefully
considering validity issues, and the possibility of recasting parts of the archive
(i.e., restructuring existing data to create new measures of a construct) to address
different questions (Elder, Pavalko, & Clipp, 1993). Yet existing studies offer a
wealth of data, representing major investments by prior researchers, funding
agencies, and participants, and it is well worth the effort to work with them and
learn what we can from the data. By building on prior work, we can augment the
field and science as a whole (Friedman et al., 2014).
Integrative Data Analysis

Taking long data to the next level, the existence of many longitudinal datasets
has brought the intriguing possibility of directly combining studies to study
lifespan personality and health processes in greater detail. There are multiple
ways that data can be integrated. A common method, and what many of the
strongest findings in personality and health research are based on, is aggregated
data meta-analysis, in which effect studies from multiple studies are combined,
and moderators are examined. Meta-analyses have identified the clearest
personality and health associations, but gloss over longitudinal trajectories and
data needed to capture more complex trajectories. One of the greatest criticisms
is that meta-analysis can seem to combine apples and oranges; although both are
types of fruit, results are more of a fruit salad than an apple pie.
Pooled integrative data analysis directly combines the data from multiple
studies at an item level, rather than combining the average study effect sizes.
Direct integration ensures that the same items or constructs are being assessed in
each sample, rather than combining divergent constructs that are labeled as the
same. Directly pooling data provides increased statistical power, the ability to
study extended developmental periods, and direct tests of sample heterogeneity.
One of the best examples of integrative studies in health psychology is the
Integrative Analysis of Longitudinal Studies on Aging (IALSA) network, a
collaboration of longitudinal studies on aging, health, and cognition (see
http://www.ialsa.org). The group has pioneered methods to link psychological
studies together at the item level (Bauer & Hussong, 2009; Curran & Hussong,
2009; Hofer & Piccinin, 2009, 2010).
Friedman and colleagues (2014) suggested that the same approach could be
used to examine personality and health associations, integrating studies where
possible to link two or more studies together, and then extending analyses to
unique aspects of each study. For example, Kern and colleagues (2014) directly
integrated data from the lifelong eight-decade Terman Life Cycle Study and the
Hawaii Personality and Health Study. Items in each sample were aligned to the
five factors. Then, using the pooled data, personality was tested as a predictor of
self-rated health 40 years later. Conscientiousness predicted better self-rated
health, directly and indirectly through educational attainment. The aligned
personality factors were then used to test Conscientiousness as a predictor of
longevity in the Terman sample and physiological regulation in the Hawaii
sample, supporting pieces of the lifespan model. Unlike study level meta-
analyses, this process ensures that the same constructs are being used as
predictors. By testing well-defined lifespan theories with integrative data, it may
be possible to empirically piece together the whole life course model.
Ongoing and New Longitudinal Studies
As many of the longitudinal studies are still ongoing, there are opportunities
to add measures within the broader studies to test lifespan development. A
particularly promising approach is the measurement burst design, in which
intensive periods of measurements, such as daily diaries or experience sampling,
are nested within long-term longitudinal studies (Nesselroade, 1991). The design
allows direct consideration of intraindividual variation. With larger, nationally
representative samples, participants from different age groups can be randomly
selected for periods of short-term intensive measures, testing some of the more
fine-grained mechanisms defined by the theoretical models.
New studies will add the most benefit to the field by carefully planning
measures and strategies to test complex models. Cross-sectional studies that
simply show that variables are related offer limited value; prospective studies are
needed. To help manage the time burden for participants, growing numbers of
computer-adapted tests (CATs) are available, such as the National Institute of
Health’s PROMIS (www.nihpromis.org). CATs include a large item bank, in
which items are aligned across an underlying distribution for a given
characteristic. Depending upon how a person answers a question, subsequent
questions are pulled from different parts of the distribution, such that a reliable
estimate of the person’s true score can be determined with four to six questions.
The most discriminating items might be compiled into a short form for off-line
panel studies. Furthermore, planned missingness allows different subsets of
participants to complete measures at each occasion, reducing the burden on
participants, while increasing the number of constructs that can be measured.
Behavioral Data
Beyond self-reported data, a growing body of research has focused on
behavioral health and personality measures. For example, the self-control facet
of Conscientiousness has been measured by delay of gratification tasks and
various executive function tests (e.g., Stroop test, go/no-go task, trail making
task, Balloon Analogue Risk Task). Convergence across self-report, observer
report, delay of gratification measures, and executive function tasks is moderate,
and studies may benefit from including both questionnaire and behavioral tasks
to provide a more complete assessment of the person (Duckworth & Kern,
2011). Behavioral-based measures can also be used with young children who
cannot read and answer questions and with lower SES or cognitively challenged
participants who may have trouble understanding normal personality questions.
Using videotaped interviews, lay people could reliably rate 62 behaviors as
diagnostic of the Big Five traits, suggesting that we often make personality
judgments about others through the things that they do (Funder & Sneed, 1993).
Gosling and colleagues (2002) found that personality is manifested in the
surrounding environment. Observers who briefly saw a person’s bedroom or
office made ratings of the person’s personality; these moderately correlated with
self-ratings and peer-ratings of the same person. Jackson and colleagues (2010)
identified behavioral components of Conscientiousness and developed the
Behavioral Indicators of Conscientiousness, which offers a way to assess what
Conscientious people do in their daily lives. Studying how positive personality
traits are manifested in everyday life may inform interventions.
Taking this a step further, Mehl and Pennebaker (2003) developed the
Electronically Activated Recorder (EAR), which captures 30-second sound bites
throughout a person’s day. Judges rated their impression of the person’s
personality using the recordings; their ratings were generally accurate,
particularly for Extraversion, suggesting that natural observations offer a method
to understand how personality is expressed in everyday behavior (Mehl, Gosling,
& Pennebaker, 2006). Such unobtrusive measures might also point to the
pathways toward health as social interactions, behaviors, and emotional
reactions are captured in the moment throughout the day.
Computational Social Science: Big Data Intrigues

Over the past decade, massive amounts of data, including online social media
(e.g., Facebook, Twitter), electronic health records, online blogs, and search
patterns, have been created, unobtrusively documenting peoples’ lives
(Anderson, Fagan, Woodnutt, & Chamorro-Premuzic, 2012). By combining
tools from computer science, the massive amount of available data, and
personality theory, health psychology researchers can and should take advantage
of these rich data sources.
Using words expressed through online modalities, evidence suggests that
personality is displayed and perceived through the language that people use in
online and mobile environments (Gill, 2004). In nearly 70,000 Facebook users
with 20 million status updates, personality scores were related to language use
(Kern, Eichstaedt, Schwartz, Dziurzynski, et al., 2014). Figure 18.1 shows word
clouds of 50 words and phrases (a series of two to three words, indicated by an
underscore) that were most strongly correlated with Conscientiousness (left) and
Neuroticism (right). The size of the word indicates how strongly the word or
phrase is correlated with the trait (larger equals a stronger correlations). The
method was also used to examine age-related differences (Kern, Eichstaedt,
Schwartz, Park, et al., 2014). Although all analyses were cross-sectional (i.e.,
different people selected from each age group), as social media is used over
time, it will become a tool to follow people’s words, behaviors, and social
interactions over the course of days, months, or even years.
Big data offer opportunities for identifying contextual aspects that may
influence health outcomes. Personality research focuses primarily on the
experience and perspective of individuals, and yet people exist within
ecosystems. Many of the key sociocultural variables that moderate the life
course models are either unmeasured in a particular sample or unidentified, and
big data can identify these sociocultural factors. Key contributors can then be
tested in smaller studies, comparing aspects of different communities, being
cognizant of the individual and social personalities that appear.
Figure 18.1. Word clouds depicting 50 words/phrases that most strongly correlated with
Conscientiousness (left) and Neuroticism (right) in 69,792 Facebook users. Larger words
are more strongly correlated with the factor. Phrases (two- to three-word combinations)
are indicated by an underscore. (Adapted from Kern, Eichstaedt, Schwartz, Dziurzynski,
et al., 2014.)
For example, geo-tagged search queries from consented mobile phones

predicted health care utilization and duration of hospital stays (Yang, White, &
Horvitz, 2013). Monitoring search logs identified adverse drug reactions, and
may serve as a complement to the Food and Drug Administration’s adverse
event reporting system, which relies on spontaneous reports of problematic
reactions (White, Harpaz, Shah, DuMouchel, & Horvitz, 2014). With Twitter
data, U.S. counties scoring high in self-reported life satisfaction expressed
language reflecting productive engagement in life, physical activity, spirituality,
and nature (Schwartz et al., 2013), and counties with high incidence of
atherosclerotic heart disease expressed language reflecting hostility and apathy
(Eichstaedt et al., 2015). The tweets may be reflecting aspects of the social or
cultural context that may be more or less health promoting.
Data from online social media, mobile applications, and other technologies
provide a new experiential sampling method that documents people’s everyday
lives unobtrusively, with relatively low cost and resources. When combined with
long data, there is the possibility for a new type of measurement burst design
(Nesselroade, 1991), with questionnaire and physiological measures collected
over many years complemented by bursts of daily patterns determined through
online data sources. Interventions can be administered and tested remotely and
adjusted to the person’s responses. Altogether, combining long and big data
offers the potential to study and understand the personality–health model in
ways never before imagined.
Personality Neuroscience
Personality neuroscience attempts to find markers of personality in the brain,
and then to map those markers back to gene-by-environment interactions
(DeYoung & Gray, 2009). Personality neuroscience attempts to determine why
people differ from one another, identifying biological sources for differences
between individuals (see also the chapter by Allen and DeYoung). Methods
include neuroimaging, genomics, electrophysiological techniques, assays of
psychoactive substances such as neurotransmitters, and direct changes through
pharmaceuticals. For example, DeYoung and colleagues (2010) mapped neural
images to the Big Five factors. Extraversion was related to the brain area that
processes rewards, Neuroticism was related to the brain regions associated with
negative affect and threat, Conscientiousness was related to the areas involved in
planning and control of behavior, and Agreeableness was associated with areas
involving understanding the intentions and motivations of other people.
Similarly, the five factors correlated with different parts of the default mode
network (a series of stable cortical brain areas thought to serve various cognitive
and emotional functions), regions that have themselves been correlated with
various response patterns for different emotional and cognitive stimuli (Sampaio,
Soares, Countinho, Sousa, & Gonçalves, 2013).
Neuroscience approaches have been relatively neglected in most lifespan
personality–health research, in large part because measures have not been
available within existing studies. If neurological measures are embedded within
larger lifespan studies, neural mechanisms underlying personality and health
processes can potentially be identified. Physiological indicators might be reliably
mapped to the five factors via a self-report measure, and then neurometric
measures of personality could be developed, and the brain circuitry that plays a
part in individual differences could be identified (Patrick, 2014). Combined with
the network of complex models, personality neuroscience may help to further
break open the black box, revealing minute processes that connect personality
processes and subsequent health outcomes.
Other Specie Parallels: Animal Studies

Another approach for studying personality and health involves utilizing
nonhuman species (see also the chapter by Weiss and Gartner). Personality
clearly appears in animals. For example, dog personality has been classified into
dimensions such as reactivity, fearfulness, activity, sociability, responsiveness to
training, submission, and aggression. A meta-analysis of 31 studies suggested
moderate consistency in traits, with personality being more consistent in adult
dogs than in puppies (Fratkin, Sinn, Patall, & Gosling, 2013), similar to the
stability of personality in adulthood versus childhood in humans. Reflecting the
hierarchical structure of the Big Five factors in humans (i.e., two higher order
alpha and beta meta-factors, the five factors, and underlying aspects, facets, and
specific traits), a study of chimpanzees found a similar hierarchical structure of
personality, with two higher order alpha and beta factors hierarchically situated
above a five factor structure (Latzman, Hopkins, Keebaugh, & Young, 2014). In
a study with seven different social groups of rhesus macaques, personality
influenced who successfully intervened in group conflict as well as roles in
grooming and network behavior (McCowan et al., 2011). By using a network
analysis, the study identified interactive effects among personality, social status,
and sex, which together influenced the stability of social groups.
Whereas human studies are almost exclusively correlational in nature, direct
experiments can be done with animals, directly testing cause and effect
associations (Cavigelli, 2005). Furthermore, the entire life course can be
observed over a much shorter period of time. For example, studies suggest that
physical activity extends longevity in animals, and both biological and
nonbiological factors play a role (Sallis, 2000). Stress studies with animals
suggest that persistent activation of the hypothalamic–pituitary–adrenal (HPA)
axis can lead to permanent physiological damage (Repetti, Taylor, & Seeman,
2002). Animals can be bred with particular characteristics to test genetic versus
environmental effects on physiological outcomes. Environments can be created
to mirror different human conditions, testing the effects on physiological
measures, length of life, and changes in brain structure. Altogether, evidence
suggests that animals do indeed have personality characteristics relevant to the
Big Five factors, and can be useful for studies on personality, stress, immunity,
and health (Mehta & Gosling, 2008).
Application: Intervention
The value of identifying pathways and influential factors lies in the potential
to inform interventions that will help more people live happier, healthier,
productive lives, with benefits to society through reduced economic costs and
burden to the system. Currently, there is a large gap between personality
research and the real world context of medicine and behaviors (Mermelstein &
Revenson, 2013). The most successful theories will explicitly specify causal,
direct, and indirect links, will be dynamic in nature and sensitive to context, will
be consistent with real-world evidence, and will be implemented by health care
providers and the general public (Michie, West, & Spring, 2013). By moving
from theoretical models to practical application, interventions may be more
effective, and we can directly observe boundary conditions for the theories.
Thus, the new generation of models will benefit from an iterative process in
which basic research informs application, and application informs basic research
(Rothman, Klein, & Cameron, 2013).
One of the primary places that personality may be informative is in medicine.
There is a growing emphasis in the medical world on the need for patient-
centered approaches to care. Theory-based studies on personality and health can
potentially inform risk models, decision making, and treatment options
(Chapman et al., 2014). Personality assessments may help healthcare
professionals make predictions of future health risks. With high treatment costs,
interventions could be targeted more directly to those most likely to respond,
thus saving time and costs that accrue from nonresponders.
How could this work in practice? At any given appointment, medical teams
often collect background information, including disease history, weight, blood
pressure, and in some cases risky health behaviors (e.g., smoking, drinking,
physical inactivity). Intake assessments could include a brief personality
measure, which would help the physician quickly get to know the patient better.
The information could be included in the electronic health record, sharing the
information with others on the care team, offering some immediate information
about individual differences, risk status, and communication preferences.
Treatments could be adjusted based on different traits. For example, as
Conscientiousness is associated with adherence, those low in Conscientiousness
could be provided with additional monitoring, reminders, and other tools. In
turn, the additional information added to the health records will inform research
on personality influences on patient–provider communication, health service use,
and overall patient outcomes (Israel et al., 2014).
Another reason for specifying and testing specific theories that include causal
clauses is to identify when and how to best intervene. For example, secure
attachment has been related to optimal self-regulation; early interventions that
teach parents parenting skills may help the parents raise self-regulated children
and conscientious adults (Drake, Belsky, & Pasco Fearon, 2014). Motivation,
which has implications for productivity and health, is influenced by parental
behaviors and beliefs, peers, and the school environment. In the Dunedin
Multidisciplinary Health and Development Study, childhood self-control
predicted numerous midlife outcomes (including education, health, wealth,
crime, parenting, and life satisfaction) in a linear fashion, suggesting that even
children with good self-control benefit from improving their self-control skills
further (Israel & Moffitt, 2014). However, the extent to which levels of self-
control and other personality traits change through intervention has yet to be
established.
The idea for personalized medicine and timely interventions certainly has
merit, but the feasibility of such approaches needs to be tested (Israel & Moffitt,
2014), and ethics need to be carefully considered. Again, the theories underlying
personality and health relationships should inform the assessments and
interventions that are conducted.
Conclusions
Once the flaws and limits of single-factor, single-disease approaches such as
the Type A personality were documented, the need for studying multiple
personality factors predicting multiple outcomes emerged (Friedman & Booth-
Kewley, 1987). Since then, the Five Factor Model has provided a useful
heuristic lens for studying the personality side of personality–health relations.
With a small, core set of factors, which at a broad level represents the universe
of personality traits, multidimensional studies can be conducted. The resulting
body of research has produced strikingly replicable results.
Conscientiousness has the clearest and most important influence on health. Its
effects are comparable to or larger than those of many traditional medical risk
factors (Chapman et al., 2011; Friedman & Kern, 2014; Shanahan et al., 2014).
The other four factors are also important, but their effects depend more on
situations and life paths. Personality matters for health, but typically not in
simple or straightforward ways; the two are connected through a complex array
of intertwined circuits and wires. For example, individuals high on Extraversion,
seeking stimulation and interactions with others, may develop unhealthy
drinking and drug abuse if they attend a “party” college and proceed into a
carousing career; but the same level of Extraversion might prove helpful when it
leads individuals to establish good social networks of caring friends. Multiple
pathways occur—behaviors, interactions with other people, situations
encountered, the social context, and inner physiological reactions all are
impacted by personality and in turn shape personality over time.
We have also emphasized the health outcomes side of the equation and the
importance of postulating and testing mechanisms and pathways. Personality is a
naturally valuable concept for understanding health because it captures
biological bases, health behaviors, and situation selection, and it involves
homeostatic forces and changes over time. For example, the neurotransmitter
serotonin is related both to Conscientiousness and to a myriad of bodily
homeostatic functions. As we have seen, conscientious behavior in turn often
involves a variety of health-protective actions, ranging from less likelihood of
smoking and risk-taking activities to safe driving and regular medical check ups.
And Conscientiousness also leads to a host of healthier psychosocial
environments, from more education and career success to better marriages
(Friedman & Martin, 2011; Shanahan et al., 2014). More complete models of
personality and health are important for identifying and clarifying causality,
which in turn influences how and when interventions should occur.
Modern personality research based on a five factor approach is clarifying the
patterns, predictors, and trajectories of health and longevity. Big data,
longitudinal studies with multiple outcomes including longevity, and other
innovative empirical approaches can be combined with a growing array of
advanced statistical methods, making it possible to test complex, dynamic
models of life trajectories. There is no longer a need for short-term studies that
correlate a trait and a disease. Instead, by unraveling the causal pathways, we
can develop interventions that will be better informed and more effective,
providing considerable value to both individuals and society.
References
Aldwin, C. M., Spiro, A., & Park, C. L. (2006). Health, behavior, and optimal aging: A
life span developmental perspective. In J. E Birren & K. W. Schaie (Eds.), Handbook
of the psychology of aging (6th ed., pp. 85–104). New York, NY: Elsevier.
Almada, S. J., Zonderman, A. B., Shekelle, R. B., Dyer, A. R., Daviglus, M. L., Costa, P.
T., Jr., & Stamler, J. (1991). Neuroticism and cynicism and risk of death in middle-
aged men: The Western Electric Study. Psychosomatic Medicine, 53, 165–175.
Anderson, B., Fagan, P., Woodnutt, T., & Chamorro-Premuzic, T. (2012). Facebook
psychology: Popular questions answered by research. Psychology of Popular Media
Culture, 1, 23–37.
Baltes, P. B., & Baltes, M. M. (1990). Psychological perspectives on successful aging:
The model of selective optimization with compensation. In P. B. Baltes & M. M. Baltes
(Eds.), Successful aging: Perspectives from the behavioral sciences (pp. 1–34). New
Bambrick, P., & Bonder, B. (2005). Older adults’ perceptions of work. Work: Journal of
Prevention, Assessment, and Rehabilitation, 24, 77–84.
Bauer, D. J., & Hussong, A. M. (2009). Psychometric approaches for developing
commensurate measure across independent studies: Traditional and new models.
Psychological Methods, 14, 101–125.
Block, J. (1993). Studying personality the long way. In D. C. Funder, R. D. Parke, C.
Tomlinson-Keasey, & K. Widaman (Eds.), Studying lives through time: Personality
and development (pp. 9–41). Washington, DC: American Psychological Association.
Bogg, T., & Roberts, B. W. (2004). Conscientiousness and health-related behaviors: A
meta-analysis of the leading behavioral contributors to mortality. Psychological
Bulletin, 130, 887–919.
Bogg, T., & Roberts, B. W. (2013). The case for conscientiousness: Evidence and
implications for a personality trait marker of health and longevity. Annals of
Behavioral Medicine, 45, 278–288.
Booth, T., Mottus, R., Corley, J., Gow, A. J., Henderson, R. D., Maniega, S. M., …
Deary, I. J. (2014). Personality, health, and brain integrity: The Lothian birth cohort
study 1936. Health Psychology, 33(12), 1477–1486.
Boyce, C. J., & Wood, A. M. (2011). Personality prior to disability determines
adaptation: Agreeable individuals recover lost life satisfaction faster and more
completely. Psychological Science, 22, 1397–1402.
Bunde, J., & Suls, J. (2006). A quantitative analysis of the relationship between the Cook-
Medley Hostility Scale and traditional coronary artery disease risk factors. Health
Carver, C. S., & Connor-Smith, J. (2010). Personality and coping. Annual Review of
Carver, C. S., & Miller, C. J. (2006). Relations of serotonin function to personality:
Current views and a key methodological issue. Psychiatry Research, 144, 1–15.
Caspi, A., Houts, R., Belsky, D. W., Goldman-Mellor, S., Harrington, H. L., Israel, S., …
Moffitt, T. E. (2014). The ‘p factor’: One general psychopathology factor in the
Cavigelli, S. A. (2005). Animal personality and health. Behaviour, 142, 1223–1244.
Chapman, B. P., Fiscella, K., Kawachi, I., & Duberstein, P. R. (2010). Personality,
socioeconomic status, and all-cause mortality in the United States. American Journal of
Epidemiology, 171, 83–92.
Chapman, B. P., Hampson, S., & Clarkin, J. (2014). Personality-informed interventions
for healthy aging: Conclusions from a National Institute on Aging work group.
Chapman, B. P., Roberts, B. W., & Duberstein, P. (2011). Personality and longevity:
Knowns, unknowns, and implications for public health and personalized medicine.
Journal of Aging Research, 759–170.
Chapman, B. P., Roberts, B., Lyness, J., & Duberstein, P. (2013). Personality and
physician-assessed illness burden in older primary care patients over 4 years. American
Journal of Geriatric Psychiatry, 21, 737–746.
Charles, S. T., Kato, K., Gatz, M., & Pedersen, N. L. (2008). Physical health 25 years
later: The predictive ability of neuroticism. Health Psychology, 27, 369–378.
Cheng, H., & Furnham, A. (2013). Factors influencing adult physical health after
controlling for current conditions: Evidence from a British cohort. PLoS One, 8,
e66204.
Christensen, A. J., Ehlers, S. L., Wiebe, J. S., Moran, P. J., Raichle, K., & Lawton, W. J.
(2002). Patient personality and mortality: A 4-year prospective examination of chronic
renal insufficiency. Health Psychology, 21, 315–320.
Connor-Smith, J. K., & Flachsbart, C. (2007). Relations between personality and coping:
A meta-analysis. Journal of Personality and Social Psychology, 93, 1080–1107.
Costa, P. T., & McCrae, R. R. (1987). Neuroticism, somatic complaints, and disease: Is
the bark worse than the bite? Journal of Personality, 55, 299–316.
Curran, P. J., & Hussong, A. M. (2009). Integrative data analysis: The simultaneous
analysis of multiple data sets. Psychological Methods, 14, 81–100.
Denollet, J., Sys, S. U., Stroobant, N., Rombouts, H., Gillebert, T. C., & Brutsaert, D. L.
(1996). Personality as independent predictor of long-term mortality in patients with
coronary heart disease. Lancet, 347, 417–421
Cambridge handbook of personality psychology (pp. 323–346). New York, NY:
(2010). Testing predictions from personality neuroscience: Brain structure and the big
five. Psychological Science, 21, 820–828.
Diener, E. (2012). New findings and future directions for subjective well-being research.
Drake, K., Belsky, J., & Pasco Fearon, R. M. (2014). From early attachment to
engagement with learning in school: The role of self-regulation and persistence.
Duberstein, P. R., Chapman, B. P., Tindle, H. A., Sink, K. M., Bamonti, P., Robbins, J.,
… Franks, P. (2011). Personality and risk for Alzheimer’s disease in adults 72 years of
age and older: A 6-year follow-up. Psychology and Aging, 26, 351–362.
Duckworth, A. L., & Kern, M. L. (2011). A meta-analysis of the convergent validity for
self-control measure. Journal of Research in Personality, 45, 259–268.
Eaton, N. R., Krueger, R. F., South, S. C., Gruenewalk, T. L., Seeman, T. E., & Roberts,
B. W. (2012). Genes, environments, personality, and successful aging: Toward a
comprehensive developmental model in later life. The Journals of Gerontology Series
A, 67A, 480–488.
Eichstaedt, J. C., Schwartz, H. A., Kern, M. L., Park, G., Labarthe, D. R., Merchant, R.
M., … Seligman, M. E. P. (2015). Psychological language on Twitter predicts county-
level heart disease mortality. Psychological Science, 26, 159–169.
Eisenberg, N., Duckworth, A. L., Spinrad, T. L., & Valiente, C. (2014).
Conscientiousness: Origins in childhood? Developmental Psychology, 50, 1331–1349.
Elder, G. H., Jr., Pavalko, E. K., & Clipp, E. C. (1993). Working with archival data:
Studying lives. Newbury Park, CA: Sage.
Ferguson, E., & Bibby, P. A. (2012). Openness to experience and all-cause mortality: A
meta-analysis and requivalent from risk ratios and odds ratios. British Journal of
Health Psychology, 17, 85–102.
Finch, J. F., Baranik, L. E., Liu, Y., & West, S. G. (2012). Physical health, positive and
negative affect, and personality: A longitudinal analysis. Journal of Research in
Fratkin, J. L., Sinn, D. L., Patall, E. A., & Gosling, S. D. (2013). Personality consistency
in dogs: A meta-analysis. PLoS One, 8, e54907.
Fredrickson, B. L. (2001). The role of positive emotions in positive psychology: The
broaden and-build theory of positive emotions. American Psychologist, 56, 218–226.
Friedman, H. S. (1991). Self-healing personality: Why some people achieve health and
others succumb to illness. New York, NY: Henry Holt.
Friedman, H. S. (2000). Long-term relations of personality, health: Dynamisms,
mechanisms, and tropisms. Journal of Personality, 68, 1089–1107.
Friedman, H. S. (2007). Personality, disease, and self-healing. In H. S. Friedman & R. C.
Silver (Eds.), Foundations of health psychology (pp. 172–199). New York, NY: Oxford
University Press.
Friedman, H. S., & Booth-Kewley, S. (1987). The “disease-prone personality”: A meta-
analytic view of the construct. American Psychologist, 42, 539–555.
Friedman, H. S., Kern, M. L., Hampson, S. E., & Duckworth, A. L. (2014). A new
lifespan approach to conscientiousness and health: Combining the pieces of the causal
puzzle. Developmental Psychology, 50, 1377–1389.
Friedman, H. S., Kern, M. L., & Reynolds, C. A. (2010). Personality and health,
subjective well-being, and longevity as adults age. Journal of Personality, 78, 179–
216.
Friedman, H. S. & Martin, L. R. (2011). The longevity project: Surprising discoveries for
health and long life from the landmark eight-decade study. New York, NY: Hudson
Street Press.
Friedman, H. S., Tucker, J., Tomlinson-Keasey, C., Schwartz, J. Wingard, D., & Criqui,
M. H. (1993). Does childhood personality predict longevity? Journal of Personality
Fries, J. F. (1990). Medical perspectives upon successful aging. In P. B. Baltes & M. M.
Baltes (Eds.), Successful aging: Perspectives from the behavioral sciences (pp. 35–49).
New York, NY: Cambridge University Press.
Funder, D. C., & Sneed, C. D. (1993). Behavioral manifestations of personality: An
ecological approach to judgmental accuracy. Journal of Personality and Social
Gill, A. (2004). Personality and language: The projection and perception of personality in
computer-mediated communication. Doctoral dissertation, University of Edinburgh,
United Kingdom. Retrieved from http://citeseerx.ist.psu.edu/viewdoc/download?
doi=10.1.1.65.217&rep=rep1&type=pdf
Goodwin, R. G., & Friedman, H. S. (2006). Health status and the Five Factor personality
traits in a nationally representative sample. Journal of Health Psychology, 11, 643–654.
Gosling, S. D., Ko, S. J., Mannarelli, T., & Morris, M. E. (2002). A room with a cue:
Personality judgments based on offices and bedrooms. Personality Processes and
Hagger-Johnson, G., Sabia, S., Nabi, H., Brunner, E., Kivimaki, M., Shipley, M., &
Singh–Manoux, A. (2012). Low conscientiousness and risk of all-cause, cardiovascular
and cancer mortality over 17 years: Whitehall II cohort study. Journal of
Psychosomatic Research, 73, 98–103.
Hampson, S. E. (2008). Mechanisms by which childhood personality traits influence adult
well-being. Current Directions in Psychological Science, 17, 264–268.
Hampson, S. E., Edmonds, G. W., Goldberg, L. R., Dubanoski, J. P., & Hillier, T. A.
(2013). Childhood conscientiousness relates to objectively measured adult physical
health four decades later. Health Psychology, 32, 925–928.
Hampson, S. E., Goldberg, L. R., Vogt, T. M., & Dubanoski, J. P. (2006). Forty years on:
Teachers’ assessments of children’s personality traits predict self-reported health
behaviors and outcomes at midlife. Health Psychology, 25, 57–64.
Heckman, J. J., Stixrud, J., & Urzua, S. (2006). The effects of cognitive and noncognitive
abilities on labor market outcomes and social behavior. Journal of Labor Economics,
24, 411–482.
Hill, P. L., & Roberts, B. W. (2011). The role of adherence in the relationship between
conscientiousness and perceived health. Health Psychology, 30, 797–804.
Hofer, S. M., & Piccinin, A. M. (2009). Integrative data analysis through coordination of
measurement and analysis protocol across independent longitudinal studies.
Psychological Methods, 14, 150–164.
Hofer, S. M., & Piccinin, A. M. (2010). Toward an integrative science of life-span
development and aging. Journals of Gerontology Series B: Psychological Sciences &
Social Sciences, 65, 269–278.
Human, L. J., Biesanz, J. C., Miller, G. E., Chen, E., Lachman, M. E., & Seeman, T. E.
(2013). Is change bad? Personality change is associated with poorer psychological
health and greater metabolic syndrome in midlife. Journal of Personality, 81, 249–259.
Huppert, F. A., & So, T. T. C. (2013). Flourishing across Europe: Application of a new
conceptual framework for defining well-being. Social Indicators Research, 110, 837–
861.
Hutchinson, J. G., & Ruiz, J. M. (2011). Neuroticism and cardiovascular response in
women: Evidence of effects on blood pressure recovery. Journal of Personality, 79,
277–302.
Idler, E. L., & Kasl, S. V. (1991). Health perceptions and survival: Do global evaluations
of health status really predict mortality? Journal of Gerontology: Social Sciences, 46,
S55–S65.
Israel, S., & Moffitt, T. E. (2014). Conscientiousness in primary care: An opportunity for
disease prevention and health promotion. Developmental Psychology, 50, 1475–1477.
Israel, S., Moffitt, T. E., Belsky, D. W., Hancox, R. J., Poulton, R., Roberts, B., … Caspi,
A. (2014). Translating personality psychology to help personalize preventative
medicine for young adult patients. Journal of Personality and Social Psychology, 106,
484–498.
Iwasa, H., Masui, Y., Gondo, Y., Inagaki, H., Kawaai, C., & Suzuki, T. (2008).
Personality and all-cause mortality among older adults dwelling in a Japanese
community: A five-year population-based prospective cohort study. The American
Journal of Geriatric Psychiatry, 16, 399–405.
What do conscientious people do? Development and validation of the Behavioral
Indicators of Conscientiousness (BIC). Journal of Research in Personality, 44, 501–
511.
Jaconelli, A., Stephan, Y., Canada, B., & Chapman, B. P. (2013). Personality and
physical functioning among older adults: The moderating role of education. The
Journals of Gerontology: Series B: Psychological Sciences & Social Sciences, 68B,
553–557.
of the five factor model of personality. Personality and Individual Differences, 33, 83–
101.
Jensen-Campbell, L. A., Knack, J. M., & Gomez, H. L. (2010). The psychology of nice
people. Social and Personality Psychology Compass, 4, 1042–1056.
Jensen-Campbell, L. A., & Malcolm, K. T. (2007). The importance of conscientiousness
in adolescent interpersonal relationships. Personality and Social Psychology Bulletin,
33, 368–383.
Kaplan, R. M. (1994). The Ziggy theorem: Toward an outcomes-focused health
psychology. Health Psychology, 13, 451–460.
Karademas, E. C., & Tsaousis, I. (2014). The relationship of patient and spouse
personality to cardiac patients’ health: Two observational studies of mediation and
moderation. Annals of Behavioral Medicine, 47, 79–91.
Kemeny, M. E. (2007). Psychoneuroimmunology. In H. S. Friedman & R. C. Silver
(Eds.), Foundations of health psychology (pp. 92–116). New York, NY: Oxford
University Press.
Kern, M. L., Duckworth, A. L., Urzua, S., Loeber, R., Loeber, M., & Lynam, D. (2013).
Do as you’re told! Facets of agreeableness and early adult outcomes for inner city boys.
Kern, M. L., Eichstaedt, J. C., Schwartz, H. A., Dziurzynski, L., Ungar, L. H., Stillwell,
D. J., … Seligman, M. E. P. (2014). The online social self: An open vocabulary
approach to personality. Assessment, 21, 158–169.
Kern, M. L., Eichstaedt, J. C., Schwartz, H. A., Park, G., Ungar, L. H., Stillwell, D. J., …
Seligman, M. E. P. (2014). From “sooo excited!!!” to “so proud”: Using language to
study development. Developmental Psychology, 50, 178–188.
Kern, M. L., & Friedman, H. S. (2008). Do conscientious individuals live longer? A
quantitative review. Health Psychology, 27, 505–512.
Kern, M. L., & Friedman, H. S. (2010). Why do some people thrive while others succumb
to disease and stagnation? Personality, social relations, and resilience. In P. S. Fry & C.
L. M. Keyes (Eds.), Frontiers of resilient aging (pp. 162–184). New York, NY:
Kern, M. L., Friedman, H. S., Martin, L. R., Reynolds, C. A., & Luong, G. (2009).
Personality, career success, and longevity: A lifespan analysis. Annals of Behavioral
Medicine, 37, 154–163.
Kern, M. L., Hampson, S. E., Goldberg, L. R., & Friedman, H. S. (2014). Integrating
prospective longitudinal data: Modeling personality and health in the Terman Life
Cycle and Hawaii Longitudinal studies. Developmental Psychology, 50, 1390–1406.
Korotkov, D. (2008). Does personality moderate the relationship between stress and
health behavior? Expanding the nomological network of the five factor model. Journal
of Research in Personality, 42, 1418–1426.
Korten, A. E., Jorm, A. F., Jiao, Z., Letenneur, L., Jacomb, P. A., Henderson, A. S., …
Rodgers, B. (1999). Health, cognitive, and psychosocial factors as predictors of
mortality in an elderly community sample. Journal of Epidemiology & Community
Health, 53, 83–88.
Latzman, R. D., Hopkins, W. D., Keebaugh, A. C., & Young, L. J. (2014). Personality in
chimpanzees (Pan troglodytes): Exploring the hierarchical structure and associations
with the vasopressin V1A receptor gene. PLoS One, 9, e95741.
Leclerc, J., Rahn, M., & Linden, W. (2006). Does personality predict blood pressure over
a 10-year period? Personality and Individual Differences, 40, 1313–1321.
Lodi-Smith, J., Jackson, J., Bogg, T., Walton, K., Wood, D., Harms, P., & Roberts, B. W.
(2010). Mechanisms of health: Education and health-related behaviours partially
mediate the relationship between conscientiousness and self-reported physical health.
Psychology & Health, 25, 305–319.
Lodi-Smith, J., & Roberts, B. W. (2007). Social investment and personality: A meta-
analysis of the relationship of personality traits to investment in work, family, religion,
and volunteerism. Personality and Social Psychology Review, 11, 1–19.
Low, L.-F., Harrison, F., & Lackersteen, S. M. (2013). Does personality affect risk for
dementia? A systematic review and meta-analysis. American Journal of Geriatric
Lyubomirsky, S., Sheldon, K. M., & Schkade, D. (2005). Pursuing happiness: The
architecture of sustainable change. Review of General Psychology, 9, 111–131.
Magee, C. A., Heaven, P. C. L., & Miller, L M. (2013). Personality change predicts self-
reported mental and physical health. Journal of Personality, 81, 324–334.
McCowan, B., Beisner, B. A., Capitanio, J. P., Jackson, M. E., Cameron, A. N., Seil, S.,
… Fushing, H. (2011). Network stability is a balancing act of personality, power, and
conflict dynamics in rhesus macaque societies. PLoS One, 6, e22350.
McEwen, B. S. (2006). Protective and damaging effects of stress mediators: Central role
of the brain. Dialogues of Clinical Neuroscience, 8, 283–293.
Mehl, M. R., Gosling, S. D., & Pennebaker, J. W. (2006). Personality in its natural
habitat: Manifestations and implicit folk theories of personality in daily life. Journal of
Mehl, M. R., & Pennebaker, J. W. (2003). The sounds of social life: A psychometric
analysis of students’ daily social environments and natural conversations. Journal of
Mehta, P. H., & Gosling, S. D. (2008). Bridging human and animal research: A
comparative approach to personality and health. Brain, Behavior, & Immunity, 22,
651–666.
Mermelstein, R. J., & Revenson, T. A. (2013). Applying theory across settings, behaviors,
and populations: Translational challenges and opportunities. Health Psychology, 32,
592–596.
Michie, S., West, R., & Spring, B. (2013). Moving from theory to practice and back in
social and health psychology. Health Psychology, 32, 581–585.
Miller, G. E., Cohen, S., Rabin, B. S., Skoner, D. P., & Doyle, W. J. (1999). Personality
and tonic cardiovascular, neuroendocrine, and immune parameters. Brain, Behavior,
and Immunity, 13, 109–123.
Mõttus, R., Johnson, W., Starr, J. M., & Deary, I. J. (2012). Correlates of personality trait
levels and their changes in very old age: The Lothian birth cohort 1921. Journal of
Mroczek, D. K. (2014). Personality plasticity, health aging, and interventions.
Mroczek, D. K., & Spiro, A. (2007). Personality change influences mortality in older
men. Psychological Science, 18, 371–376.
Mroczek, D. K., Spiro, A., & Turiano, N. A. (2009). Do health behaviors explain the
effect of neuroticism on mortality? Longitudinal findings from the VA Normative
Aging Study. Journal of Research in Personality, 43, 653–659.
Nater, U. M., Hoppmann, C., & Klumb, P. L. (2010). Neuroticism and conscientiousness
are associated with cortisol diurnal profiles in adults: Role of positive and negative
affect. Psychoneuroendocrinology, 35, 1573–1577.
Nesselroade, J. R. (1991). The warp and woof of the developmental fabric. In R. Downs,
L. Liben, & D. Palermo (Eds.), Visions of development, the environment, and
aesthetics: The legacy of Joachim F. Wohlwill (pp. 213–240). Hillsdale, NJ: Erlbaum.
O’Cleirigh, C., Ironson, G., Weiss, A., & Costa, P. T. (2007). Conscientiousness predicts
disease progression (CD4 number and viral load) in people living with HIV. Health
O’Connor, D. B., Conner, M., Jones, F., McMillan, B., & Ferguson, E. (2009). Exploring
the benefits of conscientiousness: An investigation of the role of daily stressors and
health behaviors. Annals of Behavioral Medicine, 37, 184–196.
Patrick, C. J. (2014). Understanding the role of conscientiousness in healthy aging: Where
does the brain come in? Developmental Psychology, 50, 1465–1469.
behavior. Journal of Personality and Social Psychology, 74, 538–556.
Paunonen, S. V., & Ashton, M. C. (2001). Big Five factors and facets and the prediction
of behavior. Journal of Personality and Social Psychology, 81, 524–539.
Ploubidis, G. B., & Grundy, E. (2009). Personality and all cause mortality: Evidence for
indirect links. Personality and Individual Differences, 47, 203–208.
Pressman, S. D., & Cohen, S. (2005). Does positive affect influence health?
Raposa, E. B., Bower, J. E., Hammen, C. L., Najman, J. M., & Brennan, P. A. (2014). A
developmental pathway from early life stress to inflammation: The role of negative
health behaviors. Psychological Science, 25, 1268–1274.
Repetti, R. L., Taylor, S. E., & Seeman, T. E. (2002). Risky families: Social environments
and the mental and physical health of offspring. Psychological Bulletin, 128, 330–366.
Reynolds, C. A., Finkel, D., McArdle, J. J., Gatz, M., Berg, S., & Pedersen, N. L. (2005).
Quantitative genetic analysis of latent growth curve models of cognitive abilities in
adulthood. Developmental Psychology, 41, 3–16.
Richardson, E. D. (2005). Memory improvement: Are we there yet? PsycCRITIQUES,
50(3), np.
Roberts, B. W., & Bogg, T. (2004). A longitudinal study of the relationships between
conscientiousness and the social-environmental factors and substance-use behaviors
that influence health. Journal of Personality, 72, 325–353.
conscientiousness and how can it be assessed? Developmental Psychology, 50, 1315–
1330.
Roberts, B. W., Smith, J., Jackson, J. J., & Edmonds, G. (2009). Compensatory
conscientiousness and health in older couples. Psychological Science, 20, 553–559.
Rothman, A. J., Klein, W. M. P., & Cameron, L. D. (2013). Advancing innovations in
social/personality psychology and health: Opportunities and challenges. Health
Ryan, R. M., & Deci, E. L. (2001). On happiness and human potential: A review of
research on hedonic and eudaimonic well-being. Annual Review of Psychology, 52,
141–166.
Ryff, C. D., & Keyes, C. L. M. (1995). The structure of psychological well-being
revisited. Journal of Personality and Social Psychology, 69, 719–727.
Sallis, J. F. (2000). Age-related decline in physical activity: A synthesis of human and
animal studies. Medicine and Science in Sports & Exercise, 32, 1598–1600.
Sampaio, A., Soares, J. M., Countinho, J., Sousa, N., & Gonçalves, Ó. F. (2013). The big
five default brain: Functional evidence. Brain Structure and Function, 219, 1913–1922.
Schwartz, H. A., Eichstaedt, J. C., Kern, M. L., Dziurzynski, L., Lucas, R. E., Agrawal,
M., … Ungar, L. H. (2013). Characterizing geographic variation in well-being using
tweets. In Proceedings of the Seventh International AAAI Conference on Weblogs and
Social Media (ICWSM), Boston, MA.
Segerstrom, S. C., & O’Connor, D. B. (2012). Stress, health and illness: Four challenges
for the future. Psychology & Health, 27, 128–140.
Seligman, M. E. P. (2011). Flourish. New York, NY: Simon & Schuster.
Shanahan, M. J., Hill, P. L., Roberts, B. W., Eccles, J., & Friedman, H. S. (2014).
Conscientiousness, health, and aging: The life course of personality model.
Shipley, B. A., Weiss, A., Der, G., Taylor, M. D., & Deary, I. J. (2007). Neuroticism,
extraversion, and mortality in the UK Health and Lifestyle Survey: A 21-year
prospective cohort study. Psychosomatic Medicine, 69, 923–931.
Slobodskaya, H. R., Ashmetova, O. A., & Rippinen, T. O. (2014). Does personality
matter more in difficult circumstances? Journal of Research in Personality, 48, 33–44.
Smith, T. W. (2006). Personality as risk and resilience in physical health. Current
Directions in Psychological Science, 15, 227–231.
Smith, T. W., Baron, C. E., & Grove, J. L. (2014). Personality, emotional adjustment, and
cardiovascular risk: Marriage as a mechanism. Journal of Personality, 82, 502–514.
Smith, T. W., Glazer, K., Ruiz, J. M., & Gallo, L. C. (2004). Hostility, anger,
aggressiveness, and coronary heart disease: An interpersonal perspective on
personality, emotion, and health. Journal of Personality, 72, 1217–1270.
Smith, T. W., & Williams, P. G. (1992). Personality and health: Advantages and
limitations of the five factor model. Journal of Personality, 60, 395–423.
and its relationship to aging. Developmental Psychology, 50, 1362–1376.
Staudinger, U. M., Cornelius, S. W., & Baltes, P. B. (1989). The aging of intelligence:
Potential and limits. Annals of the American Academy of Political and Social Science,
503, 43–59.
Sutin, A. R., Terracciano, A., Deiana, B., Naitza, S., Ferrucci, L., Uda, M., … Costa, P.
T., Jr. (2010). High neuroticism and low conscientiousness are associated with
interleukin-6. Psychological Medicine, 40, 1485–1493.
Sutin, A. R., Terracciano, A., Deiana, B., Uda, M., Schlessinger, D., Lakatta, E. G., &
Costa, P. T. (2010). Cholesterol, triglycerides, and the Five Factor Model of
personality. Biological Psychology, 84, 186–191.
Taga, K. T., Friedman, H. S., & Martin, L. R. (2009). Early personality predictors of
mortality risk following conjugal bereavement. Journal of Personality, 77, 669–690.
Takahashi, Y., Edmonds, G. W., Jackson, J. J., & Roberts, B. W. (2013). Longitudinal
correlated changes in conscientiousness, preventative health-related behaviors, and
self-perceived physical health. Journal of Personality, 81, 417–427.
Tam, T., & Wi, H. F. (2014). Noncognitive traits as fundamental causes of health
inequality: Baseline findings from urban China. Chinese Sociological Review, 46, 32–
62.
Tay, L., Tan, K., Diener, E., & Gonzalez, E. (2012). Social relations, health behaviors,
and health outcomes: A survey and synthesis. Applied Psychology: Health and Well-
Being, 5, 28–78.
Taylor, M. D., Whiteman, M. C., Fowkes, G. R., Lee, A. J., Allerhand, M., & Deary, I. J.
(2009). Five Factor Model personality traits and all cause mortality in the Edinburgh
Artery Study cohort. Psychosomatic Medicine, 71, 631–641.
Taylor, S. E. (2011). Social support: A review. In H. S. Friedman (Ed.), The Oxford
handbook of health psychology (pp. 189–214). New York, NY: Oxford University
Press.
Terracciano, A., Löckenhoff, C. E., Zonderman, A. B., Ferrucci, L., & Costa, P. T.
(2008). Personality predictors of longevity: Activity, emotional stability, and
conscientiousness. Psychosomatic Medicine, 70, 621–627.
Tomlinson-Keasey, C. (1993). Opportunities and challenges posted by archival data sets.
In D. C. Funder, R. D. Parke, C. Tomlinson-Keasey, & K. Widaman (Eds.), Studying
lives through time: Personality and development (pp. 65–92). Washington, DC:
Turiano, N. A., Chapman, B. P., Gruenewald, T. L., & Mroczek, D. K. (2015).
Personality and the leading behavioral contributors of mortality. Health Psychology,
34(1), 51–60.
Turiano, N. A., Pitzer, L., Armour, C., Karlamangla, A., Ryff, C. D., & Mroczek, D. K.
(2012). Personality trait level and change as predictors of health outcomes: Findings
from a National Study of Americans (MIDUS). Journals of Gerontology Series B–
Psychological Sciences and Social Sciences, 67, 4–12.
Vaillant, G. E. (2012). Triumphs of experience: The men of the Harvard Grant Study.
Cambridge, MA: Belknap Press.
Ware, J. E., Kosinski, M., & Keller, S. K. (1994). SF-36 physical and mental health
summary scales: A user’s manual. Boston, MA: The Health Institute.
Watson, D., & Pennebaker, J. W. (1989). Health complaints, stress, and distress:
Exploring the central role of negative affectivity. Psychological Review, 96, 234–254.
mortality among Medicare patients Aged 65 to 100. Psychosomatic Medicine, 67, 715–
723.
Weiss, A., Gale, C. R., Batty, G. D., & Deary, I. J. (2009). Emotionally stable, intelligent
men live longer: The Vietnam Experience Study cohort. Psychosomatic Medicine, 71,
385–394.
White, R. W., Harpaz, R., Shah, N. H., DuMouchel, W., & Horvitz, E. (2014). Toward
enhancing pharmacovigilance using patient-generated data on the internet. Clinical
Pharmacology & Therapeutics, 96, 239–246.
Williams, R., & Williams, V. (2012). Anger kills: Seventeen strategies for controlling
hostility that can harm your health. New York, NY: Harper Collins.
Wilson, R. S., Mendes de Leon, C. F., Bienias, J. L., Evans, D. A., & Bennett, D. A.
(2004). Personality and mortality in old age. Journals of Gerontology B: Psychological
Sciences & Social Sciences, 59B, P110–P116.
Yang, S.-H., White, R. W., & Horvitz, E. (2013, July). Pursuing insights about healthcare
utilization via geocoded search queries. SIGIR’13, Dublin, Ireland.
Cross-Over Analysis Using the Five Factor Model
and NEO Personality Inventory-3 for Assessing
Compatibility and Conflict in Couples
Ralph L. Piedmont and Thomas E. Rodgerson
Abstract
This chapter describes the application of the Five Factor Model (FFM) of
personality description for couple therapy; more specifically, cross over analysis.
Cross over analysis concerns a comparison of each person’s self-description with
the description provided by the spouse. The FFM offers a compelling basis and
means for a couple therapeutic analysis and intervention. It provides a clear,
simple means to understand language for describing motivations and conflict that
couples can easily understand and apply. Second, the availability of a validated
rater form provides an effective and compelling medium for couples to express
their own expectations about each other. Finally, an FFM cross over analysis can
provide for clinicians’ insight into the motivational forces that may be creating
conflict and dissatisfaction for the couple.
Key Words: Five Factor Model, couple therapy, cross over analysis, marital
therapy, marriage and family therapy, NEO PI-3
Although divorce rates of 50% for American couples (National Center for
Health Statistics) and 45% for European couples (Hahlweg, Baucom, Grawe-
Gerber, & Snyder, 2009) could be taken as indicative of the need for couples
therapy, it could also be argued that they are indicative of the failure of much
couple therapy, given the data of outcome studies over the past 30 years.
Jacobson’s (Jacobson & Christensen, 1996) early outcome studies led to the
conclusion that “novice graduate students, with no previous clinical experience,
could produce outcomes with behavioral marital therapy that were as good as or
better than those produced by trained professionals” (p. vii). Applying clinical
significance analysis to a number of published clinical trials, he discovered that a
50% success rate was characteristic of marital therapy outcome research
generally and not unique to a behavioral approach (Jacobson & Addis, 1993;
Jacobson & Christensen, 1996). Other long-term studies of treatment gains in
couple therapy showed that 56–58% of couples were unchanged or deteriorated
from their pretreatment status (Jacobson, Schmaling, & Holtzworth-Munroe,
1987; Snyder, Wills, & Grady-Fletcher, 1991) and only 56.4% of couples in
conjoint forms of therapy and 29.8% of couples treated in nonconjoint formats
were still married 5 years after therapy (Cookerly, 1980). More recent long-term
studies have shown a 51–56% maintenance rate of treatment gains for behavioral
couple therapy and emotionally focused therapy in 6 months to 2 years of
follow-up (Byrne, Carr, & Clark, 2004) and a 45.9% rate of maintained
significant improvement after 5 years for Behavioral Couple Therapy couples
and 50% for Integrated Behavioral Couple Therapy (IBCT; Christensen, Atkins,
Baucom, & Yi, 2010; Jacobson & Christensen, 1996).
The experience of many clinicians may be parallel to such data. On the one
hand, they may experience a high level of demand for couple therapy. On the
other hand, they may still be searching for a form of therapy that consistently
contributes to positive outcomes. It is the intent of this chapter to show that use
of the NEO Personality Inventory-3 (NEO PI-3; McCrae & Costa, 2010) has
significant clinical value in the early assessment phase of couple therapy and has
a unique contribution to make in the construction of a way of understanding that
can contribute to positive outcomes in couple therapy. Information from the
NEO PI-3 can enhance the clinician’s ability to interpret controlling variables, to
provide new meaning for old differences, to set limits, and to foster acceptance.
Whatever the form of couple therapy chosen by the clinician, there usually
follows some form of assessment that leads to an understanding of the
problem(s) and to a subsequent treatment protocol. Depending upon the
assumptive world of the clinician, assessments can range from clinical interview
style questions to pen and paper tests. Although clinical interview questions
grounded in the theoretical approach and experience of the clinician are the sine
qua non of assessment, pen and paper forms of “data” collection provide
additional input into the level and nature of distress, interpersonal versus
intrapersonal issues, the level of commitment, perception of positive and
negative behaviors, direction of desired change, temperaments, and personality
characteristics.
The value of personality assessment to the clinician is 5-fold. First, it provides
an ability to discern if the intrapersonal issues are of such overriding concern
that couple therapy is not warranted at this time. There may be significant mental
health issues in one or both members that need to be addressed before couple
therapy can be effective. Second, assessment enables the clinician to determine
if both parties are invested in the process, and if critical issues such as domestic
violence are a concern. Third, assessment instruments provide a means for the
clinician to gauge the degree of change that might be expected and to guard
against overfunctioning to elicit change. Fourth, measures of personality, such as
the NEO PI-3, provide another, objective way of framing differences that
enables the clinician and the clients to discuss difference in nonreactive ways.
Finally, assessment contributes to a way of understanding that moves from a
focus on derivative variables and techniques to controlling variables that
underlie given behaviors (Jacobson & Christensen, 1996).
By way of illustration, IBCT is a form of therapy that has demonstrated
sustained treatment gains over 5 years and with divorce rates only at 25.7% after
5 years (Christensen et al., 2010). This form of therapy integrates traditional
behavioral approaches with systems theory and Eastern philosophy, and builds
upon the work of Linehan (1993) in her proposed treatment of borderline
personality disorder. The missing link according to this theory has been
acceptance. Jacobson and Christensen (1996) found that although traditional
behavioral therapy might work with couples who were able and willing to
change, it was acceptance that made the difference for those with
incompatibilities and truly irreconcilable differences. Central to the work of
acceptance is the ability to come to a “formulation” in which a couple
understands the core theme, polarization process, and mutual trap in many of
their interactions.
The formulation provides an organizing principle for both therapist and client. Once
therapists have formulations, there is something to refer back to when couples have
conflicts either during or between therapy sessions… . For the partners, the
formulation becomes a context for making sense out of a confusing, desperate,
hopeless, and painful relationship.
(Jacobson & Christensen, 1996, p. 57)
They go on to say, “The primary purpose of assessment is to come up with a

formulation, which will serve as the basis for a treatment plan” (p. 59).
Although this chapter does not intend to advocate for one form of couple
therapy over another, only taking IBCT as an example, it does advocate for the
importance of a clear assessment in any therapeutic approach. Because couple
therapy is a complex, multidimensional experience, assessments can be
conducted on a number of different levels, each with its own insight. The goal of
assessment is to help construct an image and/or formulation of the relationship
that can be discussed and examined by the couple and their therapist. The next
section will outline the three levels on which assessments can be made.
Types of Couples Assessments

The goal of couple’s assessment is to capture important aspects of the
relationship that impact personal satisfaction and dyadic harmony. We see three
types of assessment that approach the relationship from different perspectives
and can be used in addition to the clinical interview. The information each
provides varies in terms of complexity. Each is described in turn.
Level 1 Assessment
Measures at this level of analysis capture individual perspectives on
satisfaction and harmony. These are the most basic and straightforward
measures; they simply query each person separately about how satisfied he or
she is. Measures such as the Locke–Wallace Marital Satisfaction Scale (Locke &
Wallace, 1959) typify scales of this nature. Because the goal of these
assessments is merely to identify an individual’s perspective, no attempt is made
to compare or contrast responses with the partner. It does provide an overall
index of the individual’s levels of satisfaction with the relationship. However,
this level of assessment does not provide any context for evaluating these
perspectives.
Level 2 Assessment
Measures at this level attempt to incorporate information from both members
of the couple about their beliefs, feelings, and attitudes. Essentially, measures at
this level involve the comparison of self-reported scores from each person on a
common set of questions. Responses are then compared and discussion/dialogue
between the members can proceed. The ENRICH/PREPARE (Olson, Fournier,
& Druckman, 1987) scale characterizes measures at this level. Here each
individual expresses his or her expectations about various aspects of the
relationship and these responses are compared between the two.
Level 3 Assessment
At this level an active effort is made to obtain a more complex, dynamic
assessment of the couple. With this mode of assessment, self-reported scores are
obtained and compared with ratings from the partner. This level of analysis
attempts to address the more interactive aspects of the relationships: how
perceptions of your partner may impact the quality of the relationship for the
rater. Such comparisons call attention to how much or how little the individuals
understand themselves and each other. Lack of agreement between the self and
observer rating may indicate specific areas of contention for the rater. At this
level, measurement scales attempt to summarize in static form the on-going
processes that have been characterizing the couples’ patterns of interpersonal
interaction. The Taylor-Johnson Temperament Analysis (Taylor & Morrison,
1984) is an exemplar of this level. Its “criss-cross” paradigm systematically
examines both the extent to which the two self-reports complement one another
and the level of agreement between the self-ratings and the corresponding
partner ratings.
The NEO PI-3 is another instrument well-suited for conducting Level 3
analyses of couples (see Piedmont, 1998). Unlike any other commercial
measure, the NEO PI-3 is based on an empirically robust, comprehensive
taxonomy of personality. This ensures that all relevant personological qualities
have been assessed. The NEO PI-3 also has validated self- and observer rating
forms, which are essential for conducting this type of analysis. It is important
that the self and observer responses are standardized on their own normative
distributions, thus enabling a direct, meaningful comparison of their resulting T-
scores. Cross-over analyses can be easily accomplished using the NEO PI-3
computer scoring program, which can automatically plot the different
combinations of responses on a common graph.
Self and Observer Ratings in the Assessment of Couples

In assessing individuals, there is no “gold standard” for finding the truth about
the person. There is no one method or measure that provides a perfect, or even
near perfect, account of the underlying nature of an individual. As such,
assessors are forced to employ multiple types of measures that assess different
aspects of the person. [There are four data sources: life outcome, self-reports,
observer ratings, and test data. Piedmont (2005) describes the value of each
source.] Although no one method is perfect, using different methods that
complement one another will help reduce error due to responding characteristics
(e.g., halo effects, response styles). Identifying areas of convergence across
methods provides high confidence in the veracity of a finding.
Self-reports are perhaps the most frequently employed method of assessment
because of their ease of administration, scoring, and interpretation. Furthermore,
the underlying assumption of a self-report is that the individual is in the best
position to provide information about himself or herself (e.g., private beliefs and
past behaviors) that may not be readily accessible to other persons. It seems
obvious that if you want to know something about people, ask them. However,
despite such an appealing paradigm, there has always been skepticism and
mistrust surrounding the use of self-reports. Concerns about defensiveness,
faking, and response distortions have always created a pall over such procedures
and have led to the development of many different types of “validity” scales to
correct for these sources of error (e.g., Schinka, Kinder, & Kremer, 1997).
Despite the large literature that validates the use of self-reports, they do not
provide a complete picture of people.
Observer ratings provide a counterpoint to self-reports (e.g., McCrae, 1994).
Individuals providing ratings rarely have the same motivations to distort their
responses as the target himself or herself. Raters are not necessarily committed
to making the target appear “good” or socially desirable. Furthermore, raters
provide another perspective on the individual, such as the type of social
impressions he or she generates or a person’s reputation, which cannot be
obtained from a self-rating. Although ratings have their own sources of error
(e.g., halo effects, stereotypes), these biases do not overlap with the errors that
are inherent to self-reports (McCrae, 1982). Thus, convergence between peer
and self-ratings cannot be attributable to correlated error but to a reliable effect.
A large literature has developed demonstrating the cross-observer validity of
the NEO PI-3, with self–other correlations on the five personality domains
ranging from r = .30 to r = .50. Interestingly, the level of self–other agreement
is, not surprisingly, stronger in married couples than in general samples
[convergence coefficients in couples range as high as r = .74 (Costa & McCrae,
1992)]. That the NEO PI-3 personality domains do have such convergence
makes it an ideal tool for use with couples. As we will show, it will be the
presence of disagreements between these two sources of information that will be
indicative of some type of distortion in the relationship that may be causing
levels of disharmony and conflict. We term the process of examining self–other
agreement in a couple cross over analysis (COA).
Cross-Over Analysis for Assessing Couples
Research has already begun to evaluate the kinds of perceptions individuals
hold of their spouses and the degree to which they correspond to spouses’ self-
perceptions. The underlying hypothesis is that the amount of congruence
between a self-report and the partner’s rating would be revealing of the level of
satisfaction experienced by the rater. This approach goes by different names,
such as insight analysis (Megargee, 1972) or criss-cross testing (Taylor &
Morrison, 1984). Whatever its label, the approach provides a methodology for
couple evaluation that focuses on the images each person holds of the other and
the role these expectations play in facilitating marital adjustment.
The value of a COA is that it focuses on the central dynamic of the
relationship: how each person perceives the motivations underlying the behavior
of his or her partner. When individuals form committed relationships, they have
an inner image of the person they believe their partner to be. These internal
images generate certain expectations about their partner’s behavior, both inside
and outside the relationship. As the couple spends more time together, these
expectations (or hypotheses) are tested. Behaviors are observed and the
correspondence of these behaviors to the “expected” personality is evaluated. A
good “fit” leads to more satisfaction in the relationship because the perceiver is
able to accurately anticipate the behavior of his or her partner (e.g., Gaunt,
2006). Predictability in a relationship is hypothesized to lead to greater
happiness and contentment. However, when the partner’s behaviors are not
consistent with the rater’s implicit personality schema of the partner, the lack of
predictiveness leads to dissatisfaction within the rater. The inability to correctly
anticipate the actions of your partner is hypothesized to lead to an increased
sense of dissatisfaction and interpersonal anomie. Perhaps the failure to predict
is associated with heightened feelings of interpersonal isolation and decreased
feelings of intimacy.
Another reason why the lack of congruence would be related to dissatisfaction
is that the individual is misperceiving the motivations of the partner, which leads
to misattributions of the partner’s behavior. For example, Kevin may believe that
his wife scores low on NEO PI-3 Neuroticism; she is thought by Kevin to be an
emotionally stable person who is not prone to emotional outbursts and prolonged
experiences of negative affect. If, however, his wife is indeed high on
neuroticism, then Kevin will misattribute the reasons for her many complaints,
her nagging, and her yelling at him. Rather than understanding these behaviors
as a sign of her need for succorance and reassurance, he will instead interpret
these behaviors as attacks on him. He may see her as very antagonistic, leading
him to withdraw or to retaliate in kind. This, in turn, only exacerbates her
feelings of loneliness and abandonment. A negative cycle of conflict and
disharmony is created because Kevin does not accurately perceive the needs of
his wife. Or his own personal needs are creating expectations for his wife that
she cannot fulfill.
In either case, the nature and degree of cross-observer congruence between a
rating and a self-report can be a useful barometer of the levels of satisfaction in
the couple. It can outline the motivational dynamics that may be precipitating
and maintaining conflict. It is important to note here that the disagreement
between the rating and the self-report is diagnostic of the lack of satisfaction in
the rater. If the rater has misperceived the partner, then he or she will be unable
to accurately anticipate the partner’s behavior and/or be led to make incorrect
attributions of the partner’s motivations. In either case, the individual has a
tenuous grasp of the interpersonal context of the relationship and this leads to
conflict and unhappiness.
The value of the NEO PI-3 in this context is 3-fold. First, the NEO PI-3
provides a useful language for talking about and describing personality. The Five
Factor Model (FFM) of personality is an empirically robust, comprehensive
model for organizing traditional personality traits. It provides a clear, simple
means to understand language for describing motivations and conflict that
couples can easily understand and apply. Second, the availability of a validated
rater form makes the NEO PI-3 an ideal medium for couples to express their
own expectations about each other. Finally, it provides for clinicians’ insight
into the motivational forces that may be creating conflict and dissatisfaction for
the couple. These patterns may suggest intervention strategies that would benefit
the couple. Taylor and Morrison (1984) provide a good description of the value
of COA:
The test can be effectively used to shift the focus from the immediate complaints to
an examination of the influence of the two personalities; as well as to develop an
understanding of the interpersonal dynamics involved. [COA]…can help the couple
objectify their problems and focus more on the role played by their individual
personalities and behavior in the overall situation. (p. 17)
Aside from identifying aspects of conflict in a relationship, we believe an

evaluation of self-ratings versus observer ratings can also be used to identify
aspects of compatibility for each partner in the relationship. To examine for
compatibility, we would compare the self-report of one person to the ratings that
individual has of his or her partner. To the extent that the partner is viewed as
having qualities that complement the characteristics of the rater, then the rater
would experience higher levels of satisfaction with the relationship. This
approach underscores the assumption that it is the images we carry of our partner
that play a significant role in how we interpret and experience the relationship.
Of course, if the perceptions a person has of his or her partner do not match the
reality of who that person is, then feelings of enduring compatibility may prove
elusive.
Performing a Cross-Over Analysis

To conduct a COA, a couple must complete the NEO PI-3 for both themselves
and their partner. This generates four separate profiles for each couple. The NEO
PI-3 works well in this context because it provides self-report and observer
booklets as well as separate norms. Scores can also be readily translated into T-
scores, and the resulting profiles can be jointly plotted on the same profile sheet.
In addition, the NEO PI-3 computer scoring program has the capacity to
combine a self-report and an observer rating into a single report, and it will
statistically compare the two profile in terms of comparability (an overall profile
congruence coefficient is calculated that quantifies the extent to which the two
sets of ratings are similar) as well as determine whether any facets are
significantly different. The computer program can also generate a simultaneous
plotting of both profiles. The graphs produced for this chapter were generated by
the scoring program.
Step 1: Comparing the Self-Reports

Self-reported scores provide a statement of how people perceive their
strengths and liabilities. These scores work directly from their self-concepts and
presents their perceptions of what motivates them and the goals they are
pursuing. A comparison of these profiles can provide information about how
well the two individuals may complement each other temperamentally. Although
the research literature suggests that happier couples will tend to be more alike
temperamentally than less satisfied couples (Buss, 1984, 1991; Gaunt, 2006;
Keller, Thiessen, & Young, 1996), there are times when differences may not be
problematic. This section will outline some strategies for analyzing these data.
By evaluating these data, a clinician can identify motivational patterns that are
likely to complement each other and those that may not. Furthermore, an
analysis of the self-profiles can also provide insight into the tone and texture of
the relationship itself. Couples may wish to explore the expectations they have
for themselves and the relationship.
We will provide some general hypotheses about how the five personality
domains should relate to one another. Considering facet level agreement would
open up this type of analysis to another level of detail and precision. However,
for the purposes of our brief review here, we will constrain our comments to the
domain level. It should also be kept in mind that the comments we provide are
general ones, and there may be instances in which gender differences in the
combinations may also be important. One final caveat: when analyzing couple
data, it is important that profiles are examined configurally (i.e., a full
consideration is given to how the various facet scores may combine to create a
specific, individual personality style). Specific facet combinations may suggest
potentially good couple compatibility whereas merely considering domains
scores may suggest otherwise. COA is a complex interpretive endeavor that
requires a high level of skill and a broad knowledge of the construct validity of
all 35 scales included in the NEO PI-3 from the practitioner. Reliance on simple
interpretive formulas can quickly become inadequate.
Concerning NEO PI-3 Neuroticism, perhaps the ideal combination would be if
both individuals were low on this dimension, indicating that each is emotionally
stable, hardy, and resilient. Having two people with good coping skills suggests
that together the couple will be well equipped to manage external stressors and
internal conflicts. Each person can be emotionally available to the other in times
of need. Another complementing pattern would be if one member was high and
the other low. Here, one person can provide emotional support and succorance to
the other when needed. Of course, the person low on Neuroticism may find it
difficult to obtain emotional support during his or her times of personal distress,
although his or her own emotional adjustment can be helpful in periods of
personal distress. In this high–low combination, it may be better if the woman
was higher on Neuroticism than if the man was higher. The least preferred
combination would be if both members were high on this domain. High scores
would indicate that this relationship would be characterized by much
emotionally charged energy; interpersonal conflict may be common. Individuals
may find it difficult to have their partner meet their emotional needs, which may
be heavy and continuous. The relationship may carry a lot of tension within it.
With both people in distress, it may be hard for them to find balance and
emotional succor without the help of outside forces (e.g., counselor, family).
Even with support, there may be chronic feelings of dissatisfaction with the
relationship. Couples thinking about entering a committed relationship may want
to explore strategies for finding the emotional support they need.
With regard to NEOI PI-3 Extraversion, there does not seem to be any “toxic”
combination on this domain. When couples are consistent on this domain (e.g.,
both high or both low), they create for themselves an interpersonal world that is
consistent with their own temperaments. Whether being socially active and
ascendant or being quiet and solitary, the couple shares a similar level of
personal arousal and desire for interpersonal contact. Inconsistency on this
domain (e.g., one person high, the other low) may not necessarily be
problematic. Sometimes the introvert may find the energy and social stimulation
that the extraverted partner brings to the relationship engaging and interesting.
The extravert may feel that the introvert helps to keep him or her grounded. It is
only when one partner feels that the other is making unreasonable demands that
problems may occur. The extraverted person may feel that his partner is “a stick
in the mud,” never wanting to socialize or to be involved in any type of social
activities. The introvert may view her partner as a “good time Charlie” or as
running away from issues in the relationship. This may create some resentment,
with the extravert being seen as someone who spends too much time with others
and does not attend sufficiently to the needs of the relationship.
In evaluating NEO PI-3 Openness, it is important to keep in mind that the
dimension does reflect the types of values each person has. Regardless of how
scores may or may not match on this domain, basic values and religious/political
preferences ought to be examined. Does the couple share similar beliefs and
priorities? This is particularly important for those couples who are both low on
this domain. Such individuals may be traditional in their outlook and espouse
conservative values. It would be important to determine that each person be clear
about the values of the other. Lack of agreement on these values may prove to be
the most distressing for the couple. Usually, couples who are low on Openness
share common values and beliefs, having specifically sought individuals who
share their world view. For those high on Openness, the lack of common values
may be less problematic (given each person’s strong capacity for tolerance and
diversity), but it may contribute to the couple drifting apart as they pursue their
own interests. This assessment of values will be of particular importance for
couples that are inconsistent on this dimension. With one person clearly
committed to some set of truths, the other may be more tentative about such
commitments. It would be important to evaluate how the differing value
orientations impact the relationship. Does the high open person find his or her
partner rigid and traditional? Does the high open person feel confined and/or
bored by the habits of the partner? Conversely, does the person low on Openness
feel that he or she is being constantly challenged regarding his or her values?
When the scores of the couple are in opposite directions, it would be important
to discuss how the creative interests of one member are handled within the
conventional orientation of the other.
Agreeableness is an important domain for consideration, given that it includes
attitudes toward others. A couple in which both are high on this domain would
be the ideal. Here, both individuals take a compassionate orientation toward
others; they are willing to be helpful and will sacrifice for the other. They are
trusting and modest and they seek to support and enable others. Such a couple
can find it easy to work together toward common goals. Their trusting nature
makes it easier for them to develop intimacy and build interpersonal bonds.
Couples in which both are low on Agreeableness present an entirely different set
of issues. With both individuals being manipulative, cynical, and distrusting, it
will be hard for this couple to engage cooperatively in their relationship. Intrigue
and dissimulation may be quite apparent as each struggles to find a strategic
advantage in the relationship. Having two individuals low on Agreeableness may
be a toxic combination for a relationship, the key reason being their inability to
reconcile once conflict erupts. Thus, for individuals low on Agreeableness, it
may be important for them to find an agreeable partner. A partner high on
Agreeableness would be less willing to retaliate when injured, would be more
understanding and compassionate, and would be ready to make amends when
emotional upsets occurred. Here again, though, there may be another gender
difference: In discordant pairs having the female higher may be better than the
reverse. We hypothesize that men often seek out partners who are
accommodating, supportive, and “maternal” in orientation. Men sometimes
harbor an expectation that they need to be assertive and dominant and to
sometimes break the rules; after all, “boys will be boys.” In these cases women
are expected to understand men’s churlish behavior. However, the opposite
pattern is believed to be less optimal in this example. A manipulative, selfish,
abusive female would have a more ruinous effect on the relationship, even when
her partner is high on Agreeableness, because such behavior may be perceived
as belittling the man’s sense of masculinity.
With regard to the NEO PI-3 Conscientiousness domain, optimal satisfaction
may be found with couples who are discordant. Couples high on
Conscientiousness are highly organized, motivated toward high standards of
success, and personally organized and reliable. Although these are all laudable
qualities, when both members are high on this domain the couple may be
competing against each other. As with Openness, what are the goals these
individuals are pursuing? Are they in conflict with one another and/or the
relationship? In a dual career relationship, it becomes essential that the couple
find ways to work together at building intimacy and closeness despite a hectic
work schedule. When both individuals are low on Conscientiousness, there is a
more relaxed family atmosphere that focuses on personal needs and gratification.
Although it may be easy for such a couple to find the time to build closeness and
intimacy, there may be issues around managing the logistics of the relationship.
There may be problems with the individuals fulfilling their responsibilities and
obligations, resulting in one person (or both) feeling put upon by the failures of
the other. Discordant pairs may be more optimally configured, although ideally
the “lower” partner on this dimension should be in the average range.
Low Conscientiousness is never a desired quality in a partner by someone
who is high on this dimension. Individuals high on Conscientiousness always
appreciate some level of personal organization and reliability in their partners
who may be expected to provide essential support by managing the logistics of
their lifestyle (e.g., caring for children, running the household). Here again, we
hypothesize another gender difference. We believe in some cases that a woman
scoring low on Conscientiousness presents less of an issue in the relationship
with a man scoring high on Conscientiousness than the reverse. Regardless of
their levels of Conscientiousness, women sometimes prefer men who are
successful and can “pull their weight,” both emotionally and financially. For the
woman low on Conscientiousness, a man average to high on Conscientiousness
provides leadership and financial resources. For a woman high on
Conscientiousness, the high scoring man is seen as a real partner in the
relationship. The man scoring low may be perceived as being a sponge on the
woman’s financial and emotional resources.
The hypotheses presented above simply represent broad interpretive strokes in
an effort to convey the basic sense of how such an analysis should proceed.
Also, some of these comments reflect more stereotypical behaviors that certainly
will not generalize to all couples. Furthermore, we have limited ourselves to
heterosexual couples, and these comments may not be relevant to lesbian and
gay relationships. Given the space limitations of this chapter, we are not able to
discuss what is meant by “high” and “low” scores; certainly different
interpretations would emerge with individuals who are moderate or average on
the various domains. In addition, the magnitude of scores needs to be considered
when giving an interpretation. Most importantly, our interpretations made no
reference to scores on the other personality domains. For example,
interpretations about Conscientiousness scores need to proceed with
consideration of the person’s level of Neuroticism. Extraversion and
Agreeableness are also considered in tandem. Finally, Openness and
Conscientiousness need to be examined together. To understand the
personological implications of any personality score, it must be viewed through
the filter of the other four dimensions. Information from the NEO PI-3 should
also be combined with data obtained from other sources such as clinical
interviews. For a fuller overview of interpretive strategies with the NEO PI-3,
the reader is referred to Piedmont (1998).
Step 2: Cross-Over Analysis: Identifying Areas of Contention

The essential feature of the Cross-Over Analysis is the comparison of a self-
report profile with that of the person’s rated profile generated by an observer. As
we have stated above, such a comparison speaks to the needs and issues of the
person doing the rating. After all, self–other agreement is very strong in couples.
Therefore, when areas of disagreement appear, they should be diagnostically
revealing of the kinds of issues and problems the rater is experiencing with the
partner. Differences are determined by subtracting the self-report T-score from
the observer rating T-score (i.e., rating minus self-report). Differences of 15
points or more ought to be considered statistically significant. Significant
differences on the domains indicate a general class of problems that the rater is
experiencing. For example, a higher rating on Extraversion may suggest that the
rater finds the spouse to be too dominant, overly involved with friends and
acquaintances, overcontrolling, and taking too many risks. A lower rating on
Extraversion may suggest that the rater finds the partner to be unaffectionate,
passive, unassertive, and not communicating in the relationship. Significant
differences on the facet scales may suggest more specific issues and areas of
contention.
Piedmont (1998) provided data validating the usefulness of COA for
identifying specific marital conflicts. Table 19.1 presents some of the potential
issues that may arise when raters perceive their partners as being higher or lower
on each of the personality domains. For example, when the partner is rated
higher on Neuroticism than he or she rates himself or herself, the rater may
perceive the partner as immature, jealous, and complaining. If the rater sees the
partner as lower on Neuroticism than he or she rates himself or herself, the rater
may see the partner as being emotionally bland and as expressing too much
emotional control. Significant differences on each personality domain are
associated with specific types of perceived issues and points of conflict. The
more dissatisfaction the rater is experiencing in the relationship, the more
discrepancies there will be. The NEO PI 3 scoring program does calculate a
profile agreement coefficient (PAC) when a rating and self-report are combined
into a single report. This coefficient is similar in nature to a correlation
coefficient in that it ranges in value from zero to one. Unlike the correlation
coefficient, the PAC assesses agreement in terms of both pattern and magnitude
(see McCrae, 2008, for a review of this statistic). The program also provides an
interpretive context for evaluating the magnitude of this coefficient (i.e., denotes
whether the level of agreement is below average, average, or above average).
It should be kept in mind that distortions in perception are not always
bidirectional. It is possible that only one partner in a relationship is experiencing
distress, and that person’s PAC value should be low. Yet, the partner may have a
higher level of marital satisfaction and evidence a higher PAC. Thus, COA can
be useful in identifying which individual is having a problem and the
corresponding issues that are at stake.
Step 3: Cross-Over Analysis: Identifying Areas of Compatibility

Although discrepancies between a rating and a self-report may indicate
problems that the rater is experiencing in the relationship, an opposite approach
may be helpful for identifying areas of compatibility. Specifically, a self-report
can be compared to the rating of that individual’s partner. To the extent that an
individual perceives his or her partner as having qualities that can satisfy the
rater’s needs, then the rater should experience higher levels of relationship
satisfaction. Whether the partner actually self-reports having those qualities is
not the issue. Rather, it is the person’s belief that his or her partner has those
qualities that is important. We believe that there are certain personality
combinations that would support enhanced feelings of compatibility in a
relationship, and these patterns are provided in Table 19.2. For now, these
patterns represent a set of hypotheses that is in need of empirical testing. They
have emerged from our work with couples both clinically and empirically. The
patterns are slightly different for men and women, reflecting our belief that
gender roles have an influence on the expectations individuals bring to their
relationship.
Table 19.1. Potential Relationship Problems Experienced by Raters Who Perceive

Their Partners Higher or Lower on the FFM Personality Domains
Potential Problemsa
Personality Rated Higher on Domain Rated Lower on Domain
Domain
Neuroticism Feels inadequate, complains a lot, Too calm, emotionally bland, too much
immature, jealous emotional control
Extraversion Always wants to “party,” hyperactive, Loner, aloof, has few or no friends, does
shows little interest in my projects not communicate
Openness Snobbish, nonconformist, overly Uncultured, rigid, lacks emotional depth
analytical
Agreeableness Unable to set limits, easily manipulated, Conceited, arrogant, selfish, cruel,
gullible confrontational
Conscientiousness Compulsive, miserly, too regimented, Self-centered, pleasure seeking, lazy,
expects too much from others spends money too easily, unreliable
a Items taken from the Couples Critical Incidents Check List (Piedmont & Piedmont, 1996).
Column 1 in Table 19.2 breaks down the self-report score on each domain into
three categories: low (T-scores below 45), average (T-scores between 45 and
55), and high (T-scores above 55). Column 2 presents those ratings a husband
has of his wife that would lead to higher levels of compatibility, and column 3
presents those ratings a wife has of her husband that are related to higher levels
of compatibility for her. For example, if a husband reports being low on
Extraversion, then according to column 2 he would be happiest if he perceives
his wife as being high on Extraversion. If she were rated low on Extraversion,
this would represent a lack of compatibility. If a wife rated herself as being low
on Extraversion, perceiving her husband as high on Extraversion would similarly
indicate an area of compatibility, although if she rated her husband low on
Extraversion, this would not necessarily be problematic as hypothesized for the
husband. Although the patterns are similar for both genders, there are some
subtle differences. For example, men low to average on Agreeableness would
feel more compatibility with their spouses if they were high on Agreeableness.
However, men scoring high on Agreeableness feel more comfortable with wives
who are low on that dimension. For women, regardless of their own self-rated
levels of Agreeableness, they feel most compatible with men who are high on
that dimension. Women who rate their husbands as low on Agreeableness feel
lower levels of satisfaction.
The value of this approach is that it allows for an assessment of what may be
working in a relationship. Despite any conflict that may be present, it is
important to be equally aware of potential areas of satisfaction and support that
may also be operating. These dynamics may be potential therapeutic resources
that can be drawn upon in the course of treatment. However, it is not currently
known how these sources of compatibility and contention operate relative to
each other. Can a higher level of compatibility moderate the adverse impact of
conflict? Regardless of levels of conflict, does a relationship need some level of
compatibility to remain viable? These are questions for future research to
address. Although there is currently no index that can quantify the level of
compatibility, clinicians will need to apply their own judgment in applying the
hypotheses found in Table 19.2. Nonetheless, COA can help provide a way to
conceptualize the complex dynamics of attraction and conflict in a manageable
way. As noted in the previous section, levels of compatibility are not necessary
bidirectional. Patterns may exist that would provide a sense of satisfaction for
one partner but not the other. The following section will present a single case of
a couple who were in marital counseling and indicate how the COA can be
applied in a therapeutically meaningful manner.
Table 19.2. Cross-Over Analysis Patterns Associated with Increased Levels of

Compatibility
Self-Rated Personality Level Husband’s Rating of Wife Wife’s Rating of Husband
Neuroticism
Low Low N Average N
Average Low N Low N
High Low N Low N (high bad)
Extraversion
Low High E (low E bad) High E
Average High E High or Low E (average bad)
High High E (low E bad) High or Low E (average bad)
Openness
Low High O Average to High O (low bad)
Average High O High O
High Average O (low bad) Average O (low or high bad)
Agreeableness
Low High A Average to High A (low bad)
Average High A High A (low bad)
High Low A High A (low or average bad)
Conscientiousness
Low High C Average to High C (low bad)
Average Average to High C (low bad) Average to High C (low bad)
High High C (low bad) Average to High C (low bad)
Case Study: Leigh and Ed G.

Ed is a 39-year-old white male with a college education who currently works
as an Assistant Pastor in a conservative Christian church, having worked in a
secular job for 15 years after attending a Bible college. He learned carpentry
skills from his father, who was a self-educated man and an entrepreneur who
built up his own business until it financially dissolved in the 1990s and he lost
everything, including the house in which they lived. Ed said that he was not
close to his father who worked all the time and on weekends watched television
and drank. Ed described his mother as a “good woman who loves me, but she
can’t show it.” He described her as “loud and heavy handed” and as a “ball of
nerves, frustrated in her marriage and she took it out on me.” As the oldest child
with one sister 2 years his junior, Ed felt as if he had to fend for himself with
little nurture being given. Only in retrospect did he learn that his difficulties with
math and reading in school were from his dyslexia. He reported some kind of a
“nervous breakdown” in high school. Ed says that what he learned in his family
system about marriage was what “not to do,” having observed an affectionless
marriage with lots of anger and resentment (his mother always slept on the
couch). His parents divorced when he was in college and he said he always
feared the prospect of divorce and dysfunction in marriage. Ed said that he
learned to cope with adversity in his family system by blocking out any stress,
learning to be overresponsible, learning to “just take the hits,” including thinking
of himself as stupid, wrong, or worthless, and learning to be outgoing to attract
friends even though on the inside he knew he was a different kind of person.
Leigh is a 37-year-old white female with a Bachelor’s degree in Religious
Education who currently works part-time as a paralegal in a local law office and
has primary responsibility for their three children ages 9, 6, and 3 years. Leigh’s
father was a dentist in her early years, but when she was about 10 years old he
became the pastor of a church. After her parent’s divorce (when she was dating
Ed) her father returned to his profession as a dentist. Leigh described her father
as loving, kind, and strong, but also said that he was depressed and on
medication much of her life. She said that she was “close to him,” but admits
that he was never around. Leigh’s mother, who died of cancer in 2007, was a
“drama-queen.” Leigh said that “Every major life event for me was ‘bombed’ by
my mother’s drama.” As the oldest child with a younger brother (by 15 months)
and sister (by 4 years), Leigh described herself as the “stable one” in the family.
She learned to cope with the frustrations in her life by going inward, being self-
contained, and learning to rely only on herself. She said that what she learned
about marriage from her family was that you have to be committed even if you
are not happy and “God blesses that.” She said that she learned that marriage is
hard work and the person just has to be obedient.
The couple has been married for 13 years. They began dating in college and
dated for 4.5 years before they married. They admit to having conflict in their
relationship from the beginning, but it became worse they said after the birth of
their first child 9 years ago and has included times of physical violence
perpetrated by both Ed and Leigh. Ed said that he decided to marry Leigh
because during a difficult time in college when he was asked to leave the school
“because I could not keep my mouth shut” she was “my friend and was on my
team, giving attention to me and being the one person who was not against me. I
developed a relationship and fell in love with her, but I was not ‘head over heels’
in love.” Leigh said that she married Ed because she liked his ideas and he
became her friend. She appreciated his desire to follow God and be used by God.
She said, “I felt the Lord was matching me with him to support and encourage
him and, even though it was easier just to rely on myself and I knew that
marriage would be a lot of work, the Lord said to ‘be with Ed.’ It was a logical
decision for me. I decided analytically. I decided that he was ‘my family’ and I
did not want him to leave.” For the past 7 years, the couple has lived in Leigh’s
father’s house for financial reasons, with her father having moved to the
basement.
Within the year prior to their coming to counseling, both Ed and Leigh had
called individually to talk about setting up an appointment for marital counseling
and to give their reasons for why it would be hard to come and why their spouse
might not choose to come. The couple came to therapy after an incident in which
Leigh pushed Ed and he blew up saying that he no longer cared what happened
and that there was nothing he could do anyway. The senior pastor of their church
was called in to mediate and required them to get counseling since in their
conservative Christian church Ed would lose his job if they divorced. On her
intake sheet where it asked for the reason for seeking counseling at this time,
Leigh wrote, “Marital issues—my husband is angry and reached a level that can
no longer allow for functioning or civility.” Ed wrote nothing in response to this
question on his intake sheet. The couple attended eight counseling sessions
between the months of July 2013 and January 2014, often canceling scheduled
sessions. Several assessments were given to the couple early in an attempt to
develop a working formulation of the couple’s style of interaction. Although
they were given the COA forms to fill out earlier, they delayed in returning the
forms and the COA was not scored until October 31, 2013. Each person also
completed the Couples Critical Incidents Check List (CCICL; Piedmont &
Piedmont, 1996), a form that contains over 130 specific marital problems that
have been linked with high and low scores on each of the FFM personality
domains. Individuals check those behaviors they find problematic in the
relationship (see Piedmont, 1998, for more information about the scale and how
it relates to COA scores on the NEO).
Essentially a Level 1 Assessment, the CCICL identifies specific personal
qualities of the partner that are causing distress for the rater. These ratings can be
usefully compared to the observer ratings of personality to determine whether
the identified problem characteristics are consistent with the personality
expectations that the rater holds of the partner. Discrepancies between these two
sets of ratings provide a basis for exploring with the rater the reasons why he or
she maintains a belief about the partner that is not substantiated by experiences.
In addition, the individual elements of the measure identify very specific issues
that are causing conflict in the relationship for each person (see Table 19.1).
These items can be used as a point of departure in treatment for addressing areas
of conflict.
Step 1: Comparing Self-Reports

Figure 19.1 presents a comparison of Ed’s and Leigh’s self-report profiles. As
can be seen, there are areas of similarities as well as areas of differences between
the two profiles. The agreement coefficient for these two protocols is .40,
indicating a low level of convergences across the two profiles.
On Neuroticism, both score similarly overall, but Ed scores higher on the facet
scales of Self-Consciousness and Impulsiveness, whereas Leigh scores higher on
Anxiety and Hostility. Both acknowledge on-going feelings of sadness and
helplessness. Thus, Ed’s negative emotionality surrounds feelings of personal
inadequacy whereas Leigh’s negative feelings center on anxiety and worry.
Although both are high on Neuroticism, suggesting a very emotionally volatile
combination, it is interesting to point out that both do score low on
Vulnerability. This may suggest that each member has the ability to manage his
or her personal distress in adaptive ways. Combined with higher scores on
Conscientiousness for the couple, it may be possible that they are able to manage
their emotional weaknesses appropriately (i.e., in ways that may enable them to
provide emotional succor to the other in their times of need).
The Extraversion domain presents some interesting areas of disagreement.
Leigh is much less sociable and charismatic than Ed, who clearly prefers being
with others and whose personal enthusiasm may attract many social contacts.
This may create some stress in their lives as Ed will seek to identify and develop
relationships, whereas his wife will prefer that they spend more time together
and alone, including being away from him. Given Leigh’s high level of
Assertiveness, she may be very forthcoming in her ability to set limits on the
number and types of social interactions in which she may be willing to engage.
Thus, Leigh may find herself feeling anxious and uncomfortable by her
husband’s on-going desires to reach out into a larger social world. Ed, on the
other hand, may find his wife manipulative and confrontational surrounding his
social needs.
On Openness, there are significant differences between them. Ed is very high
on this domain (and on five of the six facet scales) whereas Leigh is low on the
facets of Fantasy, Feelings, and Values. Both share similar scores on Ideas. The
contrasts on this domain are quite clear. Ed is very flexible, creative, inquisitive,
empathic, and intellectually complex. Leigh, on the other hand, prefers more
structure and focus. She is not very empathic and has a reasonable desire for
structure and clarity. One area in clear need of examination is the Values facet.
Leigh has some very strongly held beliefs that help define her worldview. Ed
may be more flexible in terms of his own values. The extent to which Ed may
not conform to Leigh’s expectations (i.e., roles and duties of husbands and
wives, religious/political values) would be a source of contention and conflict
between the two. Although Leigh is high on Ideas, indicating a preference for
“big picture” thinking and intellectual creativity, her lower scores on Feelings
and Fantasy may provide a source of conflict for Ed. Although Leigh can
converse on any topic, her ability to emote and process deep and powerful
emotions will be limited. Ed may find that Leigh is unable to understand his
feelings and is emotionally superficial. He may also find that she is rigid,
dogmatic, and authoritarian on certain matters that are important to him.
Conversely, Leigh may not appreciate it when Ed questions her world views. Ed
scores high on Fantasy, indicating that he has a well-developed inner world in
which he may like to spend time. This would be something that Leigh may not
understand. She tends to be quite realistic in her views and may find it difficult
when Ed spends time in flights of fancy. It would be important to make sure that
Ed does not use this inner world as a place of refuge in times of conflict in the
marriage. He does not want to appear as if he is actually escaping from his
relationship.
The couple continues to evidence many differences on Agreeableness,
suggesting that they have some different views of others and their own sense of
care and compassion. An inspection of the facet scales for this domain indicates
fewer differences than one would be led to expect given the overall domain
scores. Perhaps the two most important discrepancies to consider is that Leigh is
much lower than Ed on Compliance and Tender-Mindedness. Leigh certainly
likes getting her way. Interpreting this score in light of her high Assertiveness
scale noted above, she may appear to be very confrontational and insistent in her
desire to get her way. Coupled with the lower Openness scores, Ed may find
Leigh to be uncompromising and unresponsive to his needs. Her lower Tender-
Mindedness score in conjunction with her low score on Openness to Feelings
may also create an impression of emotional detachment. Leigh may appear to be
cold, punitive, and unfeeling. As someone who is intuitive, empathic, and having
a well-developed inner world, Ed may find Leigh to be emotionally unavailable.
It is interesting to note their comparable low scores on Trust, which may suggest
a dynamic that may be complicating their relationship. Low scores on Trust
indicate a tendency to be wary of others and their intentions. It would be
necessary to explore how this motivation may be operating in their relationship
to impede their ability to be vulnerable.
Figure 19.1. Combined self-reports for Ed and Leigh.
Finally, on Conscientiousness the two indicate their highest level of similarity.

Both score high on this domain, and facet level scores are also very comparable
to each other. As noted in the discussion on Neuroticism, the high levels of Self-
Discipline and Deliberation can provide emotional resources for regulating
whatever negative emotionality each experiences. They both share high levels of
aspiration for themselves and work hard to reach their goals. It will be important,
however, to understand how their drive and ambition are being expressed in the
relationship. Are they pursuing similar goals? Are they competing against each
other? Given the differences between the two noted above, it is important to
consider how they manage situations in which their goals are divergent, such as
in the area of social engagements.
The self-report profiles provide insight into some of the potential motivational
conflicts this couple will have to deal with in their relationship. These will be
on-going issues, and how they respond to them will set the tempo and texture of
their marriage. It becomes important therefore to examine the cross-over profiles
of the couple to determine how these conflicts may have shaped their
perceptions of each other.
Step 2: Cross-Over Analysis
Analyzing Ed’s areas of conflict. Figure 19.2 presents the self-rated profile
of Leigh with Ed’s observer rating. The NEO PI-3 combined report indicated a
couple agreement coefficient of –.17, suggesting a very low level of
convergence between the two ratings. This suggests that Ed may be experiencing
high levels of marital distress. On the CCICL he rated his overall level of
dissatisfaction as a “7,” extreme dissatisfaction, very unhappy, and experiencing
much conflict. Overall, Ed rated Leigh significantly lower on Agreeableness and
Conscientiousness. An examination of the facet scales indicates significantly
higher ratings on Angry Hostility, Impulsiveness, Vulnerability, and Warmth.
Thus, Ed believes his wife is much more emotionally undercontrolled, hostile,
anxious, and unable to cope with stress. High ratings on Neuroticism indicate
that Ed sees Leigh as being emotionally selfish, self-absorbed, and unable to
understand his feelings and needs. Such ratings indicate that these behaviors are
found to be problematic by Ed. On the CCICL, he indicated that he found his
wife emotionally unstable, yells and screams frequently, can’t receive criticism,
is moody, and worries too much.
Concerning Extraversion, both ratings indicate that Leigh is low on this
domain. An examination of the facets indicates that Ed sees Leigh as being much
lower on Warmth, Activity, and Positive Emotions. This pattern indicates a
perception of Leigh as being cold, stoic, and unenthusiastic. Ed’s ratings on the
Interpersonal domain of the CCICL includes unaffectionate, does not
communicate, and is distant from others. The similar high ratings on
Assertiveness indicate a person who wishes to be in control and dominant. Ed’s
ratings of her on the CCICL confirm this: too dominant, always want to be the
boss, overcontrolling. This rating in conjunction with low ratings on Compliance
(A4) indicates that Leigh always wants to have things done her own way,
leaving little room for compromise or accommodation. Lacking the feelings of
warmth and acceptance, such forcefulness may be seen as aggressiveness and
hostility by others.
The lower ratings by Ed on Openness indicates that he sees Leigh as being
rigid, inflexible, and dogmatic. The very low ratings on Actions and Values
suggest a perception of Leigh as having a high need for structure and clarity. The
need for organization is accented by the low Values score, which may suggest
that Leigh is seen as having a clear agenda by which things must be done. Again,
Leigh is perceived as having little room for making changes or being
accommodating of the wishes of others. CCICL ratings include intolerant of
diversity, rigid, insensitive to the feelings of others, to set in his or her ways. In
conjunction with the findings discussed above, Ed may find that Leigh is a truly
frustrating person to deal with. Rigid and emotionally distant/cold, Leigh
demands that others conform to her wishes and beliefs. Little discussion is
tolerated and all appeals are dismissed. Ed may find Leigh’s intransience to be a
constant source of frustration in the relationship.
Figure 19.2. Ed’s COA profile.
Agreeableness shows distinctive differences between the self-reports and

observer reports. Ed sees Leigh as very low on Agreeableness, with ratings on
all facets in the low to very low range. With the exception of Tender-
Mindedness, Ed rates Leigh significantly lower on all the facets then she rates
herself. Clearly, Ed sees Leigh as very antagonistic, manipulative, deceptive, and
vengeful. Leigh is seen as capable of doing whatever it takes to get her way,
including physical and verbal intimidation. Ed’s ratings of her on the CCICL
confirm his very antagonistic perceptions: cruel, cynical, manipulative,
confrontational, selfish, always does things his or her way, physically abusive.
The very low rating on Modesty also suggests that Leigh is perceived as being
arrogant, condescending, and contemptuous of others. Other ratings of her on the
CCICL included arrogant, conceited, “stuck on” himself or herself.
Rounding out Ed’s perceptions of Leigh are significantly lower ratings on the
domain and facet scales of Conscientiousness. Whereas Leigh sees herself as
having drive, ambition, and competence, Ed’s ratings are all in the low to very
low range. Low ratings on Conscientiousness indicate a perception of the person
as being very self-oriented and pleasure-oriented. Such individuals are seen as
requiring immediate gratification of their needs and as unable to control their
own impulses well. They cannot be counted on to follow through on their
commitments and are willing to cut corners to get what they want. There is little
regard for the needs of others. Ed’s CCICL ratings included self-centered,
sloppy/messy, breaks confidences, unable to maintain emotional intimacy.
What emerges from this analysis is a perception of Leigh as being very much
centered in her own personal world of needs and expectations. She evidences
little understanding of and regard for the needs and feelings of others. Rigid,
dictatorial, and very much physically confrontational in style, Leigh is seen as
being immovable. Ed experiences little emotional warmth and intimacy in his
relationship with Leigh. He perceives Leigh as being a very
emotionally/psychologically strong woman, capable of firmly “pushing back”
against him when in conflict. Ed may feel intimidated by her strong sense of
self-confidence and personal power (high ratings on Assertiveness and low
ratings on Self-Consciousness, Compliance, and Modesty). The low profile
agreement coefficient indicates that Ed is experiencing a large amount of
emotional distress and conflict in this relationship. There appear to be many
areas of conflict, including Leigh’s perceived intransigence, physical acting-out,
and emotional detachment.
Analyzing Ed’s areas of compatibility. The next question to emerge is, “Are
there any aspects to these ratings that offer potential sources of strength?” Figure
19.3 presents Ed’s self-report scores alongside his ratings of Leigh in order to
examine areas of potential compatibility. We will use the hypotheses presented
in Table 19.2 as a guide for this analysis.
Ed scores himself high on Neuroticism and whereas Leigh is rated as having
an average score on the domain, she is rated as being higher than Ed on four of
the six facet scales for the domain. Thus, her rated levels of Neuroticism are not
in the low range and may be interpreted as being high. There is little evidence
here for compatibility. Considering Extraversion, Ed scores average and would
benefit from having a wife high on this domain. Leigh is rated as being low and
therefore does not offer a compatible profile. A similar pattern emerges for
Openness. With Ed scoring high on this domain, it is expected that a wife who is
average on this domain would be ideal. Leigh scores in the low range, which is
considered to be particularly toxic for a man high on Openness.
The Agreeableness domain seems to offer some level of compatibility. With
Ed scoring high and Leigh being rated low, this seems to be a positive
combination. A person scoring high on Agreeableness may find some comfort
and personal security in a partner who is low on this dimension. This may be an
issue to explore and develop; how Leigh provides a safe “space” for Ed. Her
sense of conviction and perceived personal power may provide some
compensation for and clarity to Ed’s lowered sense of self-esteem and
emotive/intuitive approach to life. However, the negative patterns on the other
domains may mitigate any positive effects generated by this combination.
Finally, another noncompatible combination is found for Conscientiousness.
A high score on this domain for Ed would hopefully find a similar high rating
for Leigh, but just the opposite is found. Overall, there appears to be little
evidence of compatibility of Ed with Leigh in these ratings. With little perceived
motivational incentive for coming together and a high level of experienced
distress, it will be important to explore what is keeping Ed in the relationship.
Perhaps it is his sense of dutifulness and the need to stay in the marriage “for the
kids.” It may also be possible that he may feel “safe” with Leigh; she is a very
strong psychological individual. Given his high self-ratings on Agreeableness,
Self-Consciousness, and Feelings, Ed may believe himself to be an emotionally
sensitive and vulnerable person who finds some psychological protection in his
relationship with Leigh.
Figure 19.3. Ed’s COA compatibility profile.
Analyzing Leigh’s areas of conflict. Figure 19.4 presents Ed’s self-report

along with Leigh’s rating of him. From the NEO scoring report, the congruence
coefficient is .11, suggesting a low level of agreement. Leigh appears to be
equally inaccurate in describing Ed’s personality. On the CCICL Leigh also rates
the level of dissatisfaction in her relationship at “7,” indicating extreme
dissatisfaction, very unhappy, and much conflict. There are a number of
significant (i.e., 15 point differences) differences between the two profiles (19 of
the 35 scales), indicating many areas in which Leigh may not clearly understand
Ed’s motivations.
Leigh rates Ed very high on Neuroticism, although Ed’s self-reported score is
also in the high range. Although accurately noting Ed’s high scores on
Depression and Self-Consciousness, Leigh also sees Ed as being high to very
high on Anxiety, Angry Hostility, and Vulnerability. Leigh perceives Ed as
being very emotionally labile and fragile. In addition, she believes that he is not
adept at coping with these negative feelings and acts them out. On the CCICL
Leigh indicated that she found Ed to be very needy and selected descriptors of
him that included immature, easily panics, feels inadequate, hostile attitude
toward others, moody.
On Extraversion, Leigh rates Ed significantly lower on Warmth and Positive
Emotions. Although Ed may see himself as a welcoming, engaging person
toward others, Leigh does not experience any charisma or approachability.
Overall, she rates Ed low on Extraversion; this suggests that in conjunction with
her high rating of him on Neuroticism, she believes that Ed has a poor sense of
both well-being and emotional adjustment. Any interpersonal motivations are
perceived to reflect Ed’s tendency to seek reassurance and succorance from
others. Leigh also rates Ed high on Assertiveness, and this may suggest that she
perceives his interpersonal style to be controlling and manipulative. On the
CCICL she indicated on the Interpersonal domain that she found Ed to be too
dominant, always wants to be the boss, overcontrolling, unaffectionate. Thus,
Ed’s interpersonal style is perceived as being very selfish and egoistic. Ed is
seen as actively reaching out to others in order to manipulate others into meeting
his needs. He seems to lack much interest in the needs and feelings of others.
Leigh does indicate more agreement with Ed’s ratings on the Openness
domain. She acknowledges his investment in his inner world (high ratings on
Fantasy, Aesthetics, and Feelings) and his focus on inner experiences. Less
agreement is seen with the facet scales of Ideas and Values, in which Leigh sees
Ed as much more dogmatic and rigid in his views. Leigh may experience Ed as
having a clear agenda about how their relationship must proceed, along with all
the ancillary expectations that arise from such a clear cut worldview. Leigh
appears to believe that there is little flexibility in Ed’s ability to accommodate
her needs. On the CCICL Leigh rated Ed as sees the world in black or white
terms, rigid, insensitive to the feelings of others. An interesting pattern of ratings
is found for Leigh. On the one hand, she sees her husband as being empathic,
emotionally sensitive, and having a well-developed inner life. On the other hand,
Ed evidences some rigidity and strong authoritarian attitudes. This contrast can
give rise to perceptions that Ed is selfish, manipulative, and egotistical. He can
well articulate his needs and how he believes they need to be addressed by
others. This clarity may create the impression that although he knows what he
wants, he is not considering the feelings/needs of others.
The Agreeableness domain provides the greatest differences between the two
sets of ratings, with Leigh seeing Ed as being much more antagonistic, selfish,
and manipulative than he sees himself. All of Leigh’s ratings are significantly
lower than the corresponding self-report and are in the low to very low range
(with the exception of Straightforwardness, which is on the cusp between low
and average with a T-score rating of 45). Leigh’s ratings on the CCICL clearly
capture her perceptions of Ed: cynical, stubborn, cruel, “stuck on” himself or
herself, selfish, manipulative, guarded, arrogant, conceited, confrontational,
suspicious/nontrusting, always does things his or her way. Such low ratings on
Agreeableness may also indicate that the partner acts out in ways that hurt or
injure the rater. Those low on Agreeableness are not beyond using physical force
to attain their goals. On the CCICL Leigh does note the following issues in their
relationship: physically abusive, mocks me in front of others, verbally abusive,
romantically insensitive, fails to take a stand on issues I think are important. It is
clear that the tensions in the relationship do spill over into overt (public)
behaviors. It would need to be explored how extensive the physical and verbal
abuse is.
Figure 19.4. Leigh’s COA profile.
Finally, Leigh rates Ed lower on all the facets of Conscientiousness with the
exception of Order, where he rated in the high range. Significantly lower ratings
are found for Competence, Dutifulness, Achievement Striving, and Self-
Discipline. This pattern indicates a perception of Ed as being very interested in
his own immediate gratifications. He is not ambitious or forward looking, and
can easily be persuaded to put off his work to pursue more pleasurable activities.
In conjunction with the low ratings on Agreeableness, these lower scores would
also suggest that Leigh experiences Ed as being exploitative in his actions, using
others as “things” for his own interest. Little consideration, respect, or mutuality
is recognized in this pattern of ratings. Leigh rates Ed on the CCICL as being
self-centered, unambitious, and unreliable. The high rating on Order may
continue to reflect Leigh’s perception of Ed as being rigid and structured. His
desire for order may work out of his need to control and contain his feelings of
inadequacy and anxiety. Giving form and strength to this need for order are the
perceived authoritarian attitudes that demand strict adherence to specific patterns
of behavior and action.
Overall, it is clear the Leigh is experiencing quite a bit of distress in the
relationship. She views Ed as an emotionally underregulated individual who
reaches out to others to provide him with emotional support and satisfaction. His
motives are perceived as being selfish and exploitative; he seems to show little
interest in or consideration for the needs and desires of others. Failure to comply
is met with both physical and verbal abuse. There is a perceived rigidity in Ed
that makes Leigh believe that compromise or change is not an option. However,
an examination of areas of compatibility may provide perspectives on potential
areas in which the relationship may have strengths. Figure 19.5 provides an
analysis of compatibility based on Leigh’s self-report scores and her ratings of
Ed. Using the hypotheses in Table 19.2, areas in which the relationship may
provide some positive connections will be examined.
Analyzing Leigh’s areas of compatibility. Concerning Neuroticism, Leigh’s
high self-score would require a partner who is low on this domain, which is not
the case with Ed who is rated high. Two individuals scoring high on Neuroticism
would generate much negative affect in the relationship. Without appropriate
psychological mechanisms to regulate these emotions, conflict can quickly
ensue. Thus, it is not surprising that there is much tension in this relationship.
Another lack of fit is found for Extraversion. Leigh’s low score on this
domain indicates that she would seek out someone who is high. Although both
have high scores on Assertiveness and Activity, the lower scores on the
remaining facets indicate that there are few positive emotions being experienced.
On a more positive note is the pattern observed on Openness. Leigh’s average
score on this domain indicates that she would seek out someone high, which is
the case here for the domain score. Ed being rated as high is compatible with
Leigh’s self-reported level. Ed exhibits much more comfort with his inner world
of experiences and feelings than does Leigh. He may be able to provide Leigh
with a language and structure for her own efforts at understanding her inner life
and emotions. However, Leigh’s ratings of Ed also indicate that he is more
stereotypical in his thinking and worldview.
On Agreeableness, Ed’s low rating is at odds with Leigh’s own average self-
reported score. Ed’s low Agreeableness is not compatible with Leigh, who may
find Ed’s need for autonomy and control in conflict with her own desires for the
same. Finally, on Conscientiousness there is nominal compatibility: Ed’s
average rating is consistent with Leigh’s high self-reported score. However, an
inspection of the facet scales indicates that Leigh rates Ed low on all the facets
except Order, on which he scores very high. It seems that the rating on this one
facet is what is moving the overall score into the average range. It would be
worth exploring how Ed’s level of orderliness can be used to build rapport with
Leigh, who also shares a high score on this facet.
Overall, it may be possible that Leigh’s attraction to Ed may form around
three dynamics. First, she does experience high levels of negative affect,
particularly around anxiety and depression. Ed’s facility with managing and
discussing emotions, especially negative feelings, may provide her with a means
for confronting, discussing, and managing these feelings. Leigh may find it
comforting to know that her negative feelings are not off-putting to Ed. A second
area of compatibility may be found in Ed’s high level of Assertiveness,
comparable to Leigh’s. When she gets “pushy” and demanding, she may find
comfort in knowing the Ed will “push back” and not retreat. Finally, Leigh sees
herself as very focused and driven. She has set goals for herself that she works
systematically toward attaining. Her competitive efforts may find support in
Ed’s perceived low scores on Values and Ideas, as well as his high score on
Order. He may provide Leigh with ancillary emotional support and
reinforcement for her own ambitions.
Figure 19.5. Leigh’s COA compatibility profile.
Interpreting the Cross-Over Analysis Findings

The most distinctive aspect of this couple’s protocol is the mutual lack of
perceptual accuracy in their ratings of each other (profile agreement coefficients
of –.17 and .11 for the husband and wife, respectively). This level of distortion
underscores the extreme amount of marital discord that is being experienced by
each partner. The patterns of these distortions are similar for each person. Each
sees the other as rigid, selfish, manipulative, self-involved, and abusive. Each
sees high levels of emotional dysregulation in the partner and that is creating
high levels of objective distress and negative affect. That each sees the other as
being rigid and locked into their own perspective may indicate that the couple
has reached an impasse in their relationship. Whatever discussions are occurring
between them, it may have reached that point at which they are unable to be
flexible; doing so may result in a “loss of face” in the relationship and indicate
weakness.
Another area of concern is the low self-report ratings on Trust for both
individuals. People low on trust are cynical and skeptical and assume that others
are dishonest and threatening. As such, they may keep others at an emotional
distance and have difficulties in establishing and maintaining intimacy with
others. That both members of this couple score so low may indicate a serious
issue in managing this couple in treatment. Given the high levels of
dissatisfaction, the husband and wife may find it difficult to find rapport. The
conflicts inherent to their relationship will most likely serve to heighten their
mutual distrust and make compromise difficult.
Based on their self-reported scores, it appears that Ed is the more emotionally
sensitive one. His profile indicates a sensitivity to inner feelings and emotions.
As noted in the compatibility analysis, Leigh may have found this an attractive
quality, because he could help her verbalize her own inner feelings. Ed also sees
himself as being an outgoing, sociable individual who can exude charm and
warmth. This is an interesting point given that Ed also scores himself high on
Self-Consciousness and Depression (also note the lower score on Activity).
Thus, despite the happy face he may present to others, inside he has issues
concerning his personal sense of worth and adequacy. It is not uncommon to find
that individuals with high scores on Self-Consciousness tend to be more socially
phobic, not wanting to draw attention to themselves. Such public scrutiny could
lead to ridicule, criticism, and rejection; these are all issues that are anathema to
individuals with low self-esteem. Thus, Leigh may be picking up on something
in Ed’s interpersonal behavior: He reaches out to people to glean emotional and
social succor from them. She may be identifying a manipulative quality
concerning his social interactions. Interestingly, it may be Ed’s emotional
vulnerabilities that lead him to be attracted to Leigh. Although she has her own
negative affect (depression and anxiety), she does not share Ed’s low self-
esteem. Leigh’s high level of assertiveness, social confidence, leadership
qualities (e.g., low scores on Compliance and high scores on Assertiveness), and
high aspiration levels may make her appear to Ed as a strong figure who can
psychologically protect him from social criticism and other threats to his sense
of worthiness.
However, those same perceived strengths in Leigh can also make him feel
intimidated as well. Leigh’s lower levels of Agreeableness, her desire to be in
control of all situations, and her drive and ambition, can all be threatening to Ed,
especially when Leigh is provoked or challenged. Leigh rates herself as having a
more closed experiential style, and perceives Ed similarly. As indicated earlier
when examining Leigh’s compatibility patterns, the fact that Leigh perceived Ed
to be as focused, directed, and experientially closed as herself may be a positive
resource for the couple. Leigh’s own personal energy and directiveness may be a
source of conflict in her larger social world. Others may bristle under the
strength of her focus and may find her more brusque style off-putting. However,
Leigh may believe, and rightly so, that Ed has much less of a problem with these
qualities than those in her larger social world. It may be reassuring to her to
know that whatever personal drives she has, Ed has the psychological
wherewithal not to be negatively affected by it. But Leigh may not realize just
how sensitive Ed is and that his open experiential style means that he does
acutely experience Leigh’s brusqueness in a personal manner. Although his own
levels of Assertiveness and Conscientiousness may help him “push back” when
needed, his low self-esteem and high openness to feelings will make him pay an
emotional price for these encounters. Ed’s low score on Activity further
indicates that he may not always have the emotional energy for these encounters
and could get overwhelmed by them. Leigh does not seem to recognize these
emotional vulnerabilities in Ed and may not recognize when Ed begins to shut
down.
When Ed becomes overwhelmed with Leigh’s emotional energy and
directiveness, two things may happen. First, Ed may just stop engaging in the
process and put up emotional guards against Leigh’s high energy levels. This
may make it seem as if he was becoming disinterested, uninvolved, and
dismissive. As Ed disengages, he may also become more rigid and stereotypical
in his thinking. If Ed finds the encounters as hurtful, he may just refuse to be
engaged in further discussions. Again, this may lead to him being perceived as
being rigid and selfish. Second, such powerful encounters, especially ones in
which Ed may have felt hurt by Leigh, may lead him to seek out emotional
reassurances and succorance from his partner. Leigh may not realize that Ed is
emotionally exhausted and seeking comfort and instead perceive these behaviors
as Ed trying to impose his own agenda on the situation. He may believe he is
reaching out emotionally to Leigh, but she may see this as his effort to use her
for his own emotional needs and reject his advances. The result of this process is
two people feeling hurt, used, and objectified.
Treatment would need to focus on outlining for Leigh just how emotionally
sensitive Ed is and how he depends on her for emotional strength. Although he
may have innate abilities to manage Leigh’s strong persona, both need to
become more aware of when Ed is becoming overwhelmed and the need for the
situation to be defused. Both individuals ought to become more aware of the
emotional vulnerabilities each possesses. Leigh may need to become more in
touch with her feelings and how she may be perceived by others. Both may need
to find more nuanced ways of communicating with each other. Helping to clarify
their perceptions of each other may help to improve their levels of marital
satisfaction. The therapist may need to exercise a large amount of diplomacy in
helping each of these individuals to become dislodged from their current
intransigence. To emotionally defend themselves, these two people have created
emotional bunkers to protect themselves from attacks by the other. Ed will need
to appreciate Leigh’s more individualistic orientation to life. She is an introvert
who prefers being by herself. She is not comfortable with groups or
interpersonal circles. Whether this is a consequence of earlier family of origin
issues would need to be established. Nonetheless, Leigh establishes very thick
boundaries between herself and others. No doubt that stress and interpersonal
crisis will serve only to exacerbate these issues. Ed will need to find ways of
respecting the personal space Leigh wants while simultaneously finding ways of
constructively engaging her emotionally. This will need to be a learning process
for both of them.
Therapeutic Epilogue
Although the previous analysis was generated “blind” by the first author by
looking only at the COA data, it provides a good deal of accuracy with regards
to the actual therapeutic experience. Differences in scores on the NEO PI-3
actually presented the possibility of success of the marriage. For instance,
Leigh’s strong psychological character seen in her high level of assertiveness,
leadership qualities, social confidence, and high aspiration levels was a part of
why Ed married her at a time when she was the only person on “his team” and
was a strong advocate for him when others were against him. Part of the therapy
worked to help him see how this was still the case as Leigh demonstrated
strength in dealing with the family while also working outside of the home. She
was also an advocate for him during difficulties at his current job and also had
the potential to understand his mixed motives for putting on a happy face and
seeing himself as warm and caring for others. This, in fact, was a style of coping
that he learned early in life as he would escape his family dysfunction by forcing
himself to be outgoing to attract friends, when on the inside he felt worthless and
inadequate personally and socially. This was his attempt to “put the negative
away” and it may have contributed to what he described as a “nervous
breakdown” in his teen years. Similarly, Ed’s openness to new ideas and new
ways of thinking were a part of why Leigh married him. She was attracted to his
ideas that had core values similar to her own but were new and exciting to her.
In the marriage, Ed had the potential to draw Leigh out of her rigid and “one
right way” kind of thinking. His own sensitivity to inner feelings was a potential
gift in the marriage not only to understand and share his own inner world, but
also to guide Leigh into understanding and sharing her inner world. In fact, at
some level she wanted his leadership in this area. The articulation and
affirmation of these positive potentials were a focus of the couple’s therapy as
they were invited to notice and to reinforce these positives in the other. The high
scores on Consciousness were a positive for this couple, which sporadically gave
them a determined effort to work on the relationship, often driving 2 hours each
way to come to therapy. Their low scores on the Vulnerability facet were also
reflected in their periodic willingness to open up to one another during the
therapy hour in a revealing way. These were attempts on the part of Ed and
Leigh to manage their distress in adaptive ways, although their “default” coping
mechanism of retreat from the other and retreat from therapy often prevailed as a
maladaptive way to manage their distress.
The most striking data in the COA analysis, which prevailed in the couple’s
therapy, is in the couple coefficient agreement score of –.17 for Ed and .11 for
Leigh. These scores indicate that the images and beliefs about the other person
constructed in the mind of the rater are far different from the other’s own
experience of self. In the case of Ed and Leigh, these constructed images are
usually more negative and appear to inhibit the rater’s ability to see certain
personality characteristics in a positive light with good potential for the
marriage. For instance, Ed’s higher rating of Leigh on Neuroticism reflect his
image of her as being emotionally unstable, needy, and out of control with little
energy to pay attention to or attend to his needs. He was not able to see her as
someone who needed his help and to see her as a hurting person who could
benefit from some of his self-determined Tender-Mindedness. His lower rating
of Leigh on Extraversion reflects his image of her as cold and distant, which
blocked his ability to see the times during therapy when she was making
intentional efforts to move toward him emotionally and physically. Ed’s lower
rating of Leigh on Openness contributed to his seeing her as more rigid than she
really was and left him unable to affirm her for the consistency and stability that
she often provided the family. His low assessment of her Conscientiousness
caused her to comment on her disorganization around the house and blocked him
from seeing her concentrated efforts to manage the home and work outside the
home. Similarly, Leigh’s low rating of Ed on Extraversion is indicative of her
inability to see his efforts of being positive toward her and contributed to the
misinterpretation of his efforts at socialization as avoidance of family
responsibilities and as an attempt to be somebody that he was not. Ed reported
that “she sees me as damaged goods because of the family in which I grew up.”
Leigh’s low rating of Ed on Agreeableness reflects her rejection of his attempts
to understand her inner world in the times that he tenderly tried to approach her
for emotional intimacy. It also led to her judgment of his pastoral outreach to
others as fake and simply a diversion from the family so that he would look
good. Leigh’s low rating of Ed on Conscientiousness was indicative of a way of
seeing him that made her blind to the major efforts that he made to fix up a
house that he did not own and to clean up after her father. These factors and
others contributed to a predominant image that they similarly held of each other
and that they revealed in a poignant therapy session on August 23, 2013—
namely, that the other person was an “abuser.” More difficult for them was to
acknowledge their own abusive tendencies and to establish an image of the other
person as someone with positive characteristics who was trying to call for help
with their deepest needs.
It is not that there were no events in the marriage or actions on the part of the
other that led to these constructed images that emerge from the rater in the COA
analysis. However, it may also be the case that Ed and Leigh brought to the
marriage interpretive schemas already formed in the nonnurturing environments
in which they were raised, or formed from genetic predispositions encoded in the
brain. When they were presented with the COA data and the possibility that they
were constructing an overly negative image of the other person, they defaulted to
the safety of retreat and chose not to continue in therapy. This is accurately
described in the COA analysis above as a retreat to their emotional bunkers to
protect themselves from perceived attacks by the other, or others, unable to trust
the possibility of a differently constructed image of their selves, the other, or the
world.
Conclusions
Couple therapy is complex work, involving interactions at multiple levels. The
dynamic process of two people interacting can be difficult to track and organize.
Personality assessment provides a tool for capturing specific aspects of this
process in ways that promote understanding and encourage dialogue. The NEO
PI-3 is an ideal tool for use within couple therapy. Its validated self- and
observer rating forms provide a tremendous amount of information about the
personal styles of the couple, both separately and together. The personality
interpretations provide clear and nonreactive information about the relationship
that enables both members to step back and see themselves in a way that
promotes them taking a “second look” at their style of relating. The use of the
cross-over analysis paradigm generates a tremendous amount of relevant
material. The essence of COA is identifying those points of discrepancy among
the various ratings. The areas of disagreement identify where important points of
conflict exist. They also help to identify the underlying issues in the rater that
may be fueling the conflict. Thus, COA provides the clinician with areas of
engagement with the clients in an effort to understand the psychological
dynamics underlying these discrepancies.
Although trait based, FFM scores can be easily integrated with a broad variety
of theories for interpreting the expectations behind the ratings of the partner.
Psychodynamic, behavioral, gestalt, interpersonal, or humanistic interpretations
can easily be fit to the perceptions evidenced in the NEO PI-3 protocol. There is
no doubt that this type of personality information can serve as the platform for
the generation of clinical hypotheses. The CCICL can be a useful complement to
this process. It can identify those specific behaviors that are creating problems
within the relationship. The items on the CCICL have been selected on the basis
of both their relevance for capturing salient aspects of conflict in a relationship
and their relevance to high and low scores on each of the personality dimensions.
For example, partners of individuals high on Neuroticism may find their mate
whiny, obsessive, or moody. Those seen as being low on Neuroticism may be
perceived as too calm, emotionally bland, or unconcerned about the impressions
made on others. Usually, the more problems that are checked, the more distress
there is in the relationship. The CCICL can provide a very specific place to start
in discussing issues among the couple.
Perhaps the most useful application of the COA would be with premarital
couples. Individuals exploring the possibilities of a committed relationship could
greatly benefit from this intimate examination of each other’s personal
preferences and expectations. Individuals can come to examine aspects of their
own personalities as well as the implications these qualities may have for their
on-going relationship.
As noted earlier, the hypotheses we presented were intended for heterosexual
couples, demonstrating the need for research to determine whether these
expectations would hold for gay and lesbian couples as well. Greater
examination of the personological implications of discrepancies between self-
ratings and observer ratings needs to be done. The CCICL was a first effort at
trying to link specific relationship issues to particular incongruities on the NEO
PI-3 profile. Much more work needs to be done in empirically developing these
relationships, especially as they relate to the facet scales. A new dimension
added to the COA process was an assessment of compatibility. Rather than
simply focusing on areas of distress and conflict, it seems appropriate to also
identify positive dimensions of the relationship. Individuals seek others who can
complement them in ways that provide gratification for basic needs. In some
instances we seek individuals like ourselves (e.g., those scoring high on
Extraversion wanting high levels of Extraversion in our partners), whereas at
other times we seek someone different (e.g., those low on Conscientiousness
seeking someone average to high on the dimension). Ultimately compatibility
may be a more complex process than we have hypothesized. The patterns
outlined in Table 19.2 provide a set of expectations that is in need of further
testing and analysis. Nonetheless, these dynamics can serve as potential
resources to be drawn upon over the course of treatment.
Finally, one area that was not touched on in this article but offers a new
dimension of analysis concerns a comparison of the two observer ratings. Do
areas of convergence and divergence have implications for relationship stability
and satisfaction? What new insight would a consideration of the observer ratings
provide over what is already gleaned from the COA? Although the ratings reflect
perceptions held of the partner, it would be interesting to see how these
perceptions, and their related expectations, may be serving to create conflict
and/or build compatibility. In the end, it will be necessary to derive algorithms
that are capable of efficiently linking all of this information together in ways that
will allow both useful analysis of a couple’s intimacy patterns and assessment of
the viability of the relationship.
References
Buss, D. M. (1984). Marital assortment for personality dispositions: Assessment with
three different data sources. Behavior Genetics, 14, 111–123.
Buss, D. M. (1991). Conflict in married couples: Personality predictors of anger and
upset. Journal of Personality, 59, 663–688.
Byrne, M., Carr, A., & Clark, M. (2004). The efficacy of behavioral couple therapy and
emotionally focused therapy for couple distress. Contemporary Family Therapy, 26,
361–387.
Christensen, A., Atkins, D. C., Baucom, B., & Yi, J. (2010). Marital status and
satisfaction five years following randomized clinical trial comparing traditional versus
integrative behavioral couple therapy. Journal of Consulting and Clinical Psychology,
78, 225–235.
Cookerly, J. R. (1980). Does marital therapy do any lasting good? Journal of Marital and
Family Therapy, 6, 393–397.
Costa, P. T., Jr., & McCrae, R. R. (1992). NEO PI-R Professional manual. Odessa, FL:
Gaunt, R. (2006). Couple similarity and marital satisfaction: Are similar spouses happier?
Journal of Personality, 74, 1401–1420. doi:10.1111/j.1467-6494.2006.00414.x.
Hahlweg, K., Baucom, D. H., Grawe-Gerber, M., & Snyder, D. K. (2009). Strengthening
couples and families: Dissemination of interventions for the treatment and prevention
of couple distress. In K. Hahlweg, M. Grawe-Gerber, & D. H. Baucom (Eds.),
Enhancing couples: The shape of couple therapy to come (pp. 3–30). Gottingen,
Germany: Hogrefe.
Jacobson, N. S., & Addis, M. E. (1993). Research on couples and couple therapy: What
do we know? Where are we going? Journal of Consulting and Clinical Psychology, 61,
85–93.
Jacobson, N. S., & Christensen, A. (1996). Acceptance and change in couple therapy.
New York, NY: W.W. Norton & Company.
Jacobson, N. S., Schmaling, K. B., & Holtzworth-Munroe, A. (1987). Component
analysis of behavioral marital therapy: 2-year follow-up and prediction relapse. Journal
of Marital and Family Therapy, 13, 187–195.
Keller, M. C., Thiessen, D., & Young, R. K. (1996). Mate assortment in dating and
married couples. Personality and Individual Differences, 21, 217–221.
Linehan, M. (1993). Cognitive behavioral treatment of borderline personality disorders.
New York, NY: Guilford.
Locke, H. J., & Wallace, K. M. (1959). Short marital-adjustment and prediction tests:
Their reliability and validity. Marriage and Family Living, 21, 251–255.
McCrae, R. R. (1982). Consensual validation of personality traits: Evidence from self-
reports and ratings. Journal of Personality and Social Psychology, 43, 292–303.
McCrae, R. R. (1994). The counterpoint of personality assessment: Self-reports and
observer ratings. Assessment, 1, 151–164.
McCrae, R. R. (2008). A note on some measures of profile agreement. Journal of
Personality Assessment, 90, 105–109. doi:10.1080/00223890701845104.
McCrae, R. R., & Costa, P. T., Jr. (2010). NEO inventories, professional manual. Lutz,
Megargee, E. I. (1972). The California Personality Inventory handbook. San Francisco,
CA: Jossey-Bass.
Olson, D. H., Fournier, D. G., & Druckman, J. M. (1987). PREPARE/ENRICH
counselors manual (rev. ed.). Minneapolis, MN: PREPARE-ENRICH, Inc.
applications. New York, NY: Plenum Press.
Piedmont, R. L. (2005). Understanding personality and its assessment from a trait
perspective. In B. T. Erford (Ed.), A counselor’s guide to assessment (1st ed., pp. 98–
101). New York, NY: Houghton-Mifflin/Lakaska Press.
Piedmont, R. L., & Piedmont, R. I. (1996). Couples Critical Incidents Check List.
Baltimore, MD: Author.
Schinka, J. A., Kinder, B., & Kremer, T. (1997). Research validity scales for the NEO-PI-
R: Development and initial validation. Journal of Personality Assessment, 68, 127–
138.
Snyder, D. K., Wills, R. M., & Grady-Fletcher, A. (1991). Long-term effectiveness of
behavioral versus insight-oriented marital therapy: A 4-year follow-up study. Journal
of Consulting and Clinical Psychology, 59, 138–141.
Taylor, R. M., & Morrison, L. P. (1984). Taylor-Johnson Temperament Analysis manual.
Los Angeles, CA: Psychological Publications.
Five Factor Model and Personality Disorder
Thomas A. Widiger, Whitney L. Gore, Cristina Crego, Stephanie L. Rojas, and Joshua R.
Oltmanns
Abstract
The purpose of this chapter is to provide an overview of the relationship of the
Five Factor Model (FFM) to personality disorder. The FFM has traditionally been
viewed as a dimensional model of normal personality structure. However, it
should probably be viewed as a dimensional model of general personality
structure, including maladaptive as well as adaptive personality traits. Discussed
herein is the empirical support for the coverage of personality disorders within the
FFM; the ability of the FFM to explain the convergence and divergence among
personality disorder scales; the relationship of the FFM to the DSM-5 dimensional
trait model; the empirical support for maladaptivity within both poles of each
FFM domain (focusing in particular on agreeableness, conscientiousness, and
openness); and the development of scales for the assessment of maladaptive
variants of the FFM.
Key Words: Five Factor Model, trait, personality, personality disorder, DSM-IV,
DSM-5
One of the strengths of the Five Factor Model (FFM) is its robustness (John,
Naumann, & Soto, 2008). “Personality psychology has been long beset by a
chaotic plethora of personality constructs that sometimes differ in label while
measuring nearly the same thing, and sometimes have the same label while
measuring very different things” (Funder, 2001, p. 200). As expressed by Funder
(2001), “The use of five broad traits as a common currency for personality
psychology has been an important counterforce to this Tower of Babel” (p. 200).
The FFM has indeed been used quite effectively as a basis for comparing,
contrasting, and integrating seemingly diverse sets of personality trait scales
(Goldberg, 1993; McCrae & Costa, 2003; Ozer & Reise, 1994). “One of the
great strengths of the Big Five taxonomy is that it can capture, at a broad level of
abstraction, the commonalities among most of the existing systems of
personality traits, thus providing an integrative descriptive model” (John et al.,
2008, p. 139).
The FFM is comparably robust in its coverage of abnormal as well as normal
personality functioning (Clark, 2007; Widiger & Costa, 1994). The purpose of
this chapter is to provide the empirical support for the coverage of personality
disorder and dysfunction within the FFM; the ability of the FFM to explain the
convergence and divergence among personality disorder scales; the relationship
of the FFM to the DSM-5 dimensional trait model; the theoretical and empirical
support for maladaptivity within both poles of each FFM domain (focusing in
particular on agreeableness, conscientiousness, and openness); and the
development of scales for the assessment of maladaptive variants of the FFM.
FFM Coverage of Personality Disorders

Costa, McCrae, and their colleagues conducted a series of studies in the 1980s
indicating how the FFM, as assessed by the NEO Personality Inventory (Costa &
McCrae, 1992), can account for constructs contained within alternative models
of personality (McCrae & Costa, 2003; see also the chapter by Costa and
McCrae). O’Connor (2002) conducted integrative factor analyses of previously
published findings from approximately 75 studies involving FFM scales along
with the scales of 28 commonly used self-report inventories of personality (see
also the chapter by O’Connor). He concluded that “the factor structures that exist
in the scales of many popular inventories can be closely replicated using data
derived solely from the scale associations with the FFM” (O’Connor, 2002, p.
198). O’Connor (2002) further suggested that “the basic dimensions that exist in
other personality inventories can thus be considered ‘well captured’ by the
FFM” (p. 198).
The research conducted by O’Connor (2002), McCrae and Costa (2003), and
others is traditionally considered to be studies of normal personality. However,
much of this research is directly relevant to the coverage of maladaptive
personality functioning, as most of the instruments and scales they investigated
have been and continue to be used within clinical populations to assess
personality dysfunction (Widiger, Costa, Gore, & Crego, 2013). For example,
McCrae, Costa, and Busch (1986) demonstrated how the 100 items within the
California Q-Set (CQS; Block, 2008) are well understood from the perspective
of the FFM. The CQS items were developed by successive panels of
psychodynamically oriented clinicians seeking a common language. A factor
analysis of the complete set of items yielded five factors that aligned closely
with the FFM. An inspection of the CQS items illustrates its pertinence to
abnormal, as well as normal, personality functioning. The neuroticism factor
contrasted CQS items such as “thin-skinned,” “extrapunitive,” and “brittle ego
defenses” (with “socially poised” and “calm, relaxed”); introversion contrasted
“avoids close relationships” and “emotionally bland” (with “self-dramatizing,”
“talkative,” and “behaves assertively”); openness contrasted “unusual thought
processes,” “engages in fantasy, daydreams,” and “rebellious noncomforming”
(with “moralistic,” “uncomfortable with complexities,” and “favors conservative
values”); antagonism contrasted “basically distrustful,” “expresses hostility
directly,” and “critical, skeptical” (with “behaves in a giving way” and “warm,
compassionate”); and, finally, low conscientiousness contrasted “self-indulgent”
and “unable to delay gratification” (with “dependable, responsible,” and “has a
high aspiration level”). Support for their interpretation of these factors was
obtained from convergent and discriminant validity correlations with self and
peer assessments of the FFM. In sum, the results of their study demonstrated a
close correspondence of a sophisticated psychodynamic nomenclature that
concerned fundamental components of personality disturbance with the FFM.
The CQS “represents a distillation of clinical insights, and the fact that very
similar factors can be found in it provides striking support for the five factor
model” (McCrae et al., 1986, p. 442).
Mullins-Sweatt and Widiger (2007) reported similar results with the Shedler–
Westen Assessment Procedure-200 (SWAP-200), a psychodynamically oriented
clinician Q-sort for personality disorder pathology comparable to the CQS
(Shedler & Westen, 2004b). They had persons with significant personality
problems described in terms of the SWAP-200 and the FFM. They reported a
close convergence of the FFM and SWAP-200 descriptions. SWAP-200
Emotional Dysregulation and Dysphoria aligned with FFM neuroticism;
Schizoid Orientation aligned with introversion; Obsessionality aligned with
conscientiousness; and Hostility, Psychopathy, and Narcissism aligned with
antagonism (consistent with expectations, Psychopathy also correlated with low
conscientiousness and Narcissism correlated with extraversion). The only
SWAP-200 scales that did not have large effect size relationships with the FFM
were Dissociation, Oedipal Conflict, and Sexual Conflict. However, these scales
included many items that do not appear to be strongly related to personality
disorder or personality. For example, the SWAP-200 Sexual Conflict scale
includes items such as “Experiences a specific sexual dysfunction during sexual
intercourse or attempts at intercourse (e.g., inhibited orgasm or vaginismus in
women, impotence or premature ejaculation in men)” and “Tends to see sexual
experiences as somehow revolting or disgusting” (Shedler & Westen, 2004a, p.
1750). SWAP-200 Sexual Conflict is perhaps best understood to be a measure of
sexual dysfunction rather than a measure of personality dysfunction.
Shedler and Westen (2004a) used a subset of SWAP-200 items to create a
SWAP-based measure of the FFM with the intention of demonstrating “that the
five factor model is not sufficiently comprehensive” (p. 1753). They felt that
only 30% of the SWAP-200 items had any relevance to the FFM. McCrae,
Lockenhoff, and Costa (2005), however, coded 93% of the SWAP-200 items in
terms of the FFM, suggesting that all but a small minority of the SWAP-200 can
be understood from the perspective of the FFM.
Mullins-Sweatt and Widiger (2008) compared the convergent (and
discriminant) validity of the Shedler and Westen and McCrae et al. SWAP-200
scales (administered by clinicians describing their patients) using the NEO
Personality Inventory-Revised (NEO PI-R; Costa & McCrae, 1992) as the FFM
criterion measure. There was no difference in convergent validity for the
respective Extraversion scales, but the McCrae et al. Conscientiousness scale
had significantly better convergent validity than the Shedler and Westen
Conscientiousness scale, and the McCrae et al. Neuroticism, Agreeableness, and
Openness scales had substantially higher convergent validity. The convergent
validity coefficients for the McCrae et al. scales using 93% of the SWAP-200
item set were .74 for Neuroticism, .72 for Extraversion, .51 for Openness, .77 for
Agreeableness, and .76 for Conscientiousness.
A considerable body of research indicates that the FFM successfully accounts
for the features and symptoms of the personality disorders included within the
American Psychiatric Association’s (APA) Diagnostic and Statistical Manual of
Mental Disorders (DSM-5; American Psychiatric Association, 2013) (e.g.,
Aluja, Cuevas, Garcia, & Garcia, 2007; Bagby, Costa, Widiger, Ryder, &
Marshall, 2005; Bagby, Marshall, & Georgiades, 2005; Ball, Tennen, Poling,
Kranzler, & Rounsaville, 1997; Bastiaansen, Rossi, & De Fruyt, 2013;
Bastiaansen, Rossi, Schotte, & De Fruyt, 2011; Blais, 1997; Chai et al., 2012;
Costa & McCrae, 1990; De Fruyt, De Clercq, van de Wiele, & Van Heeringen,
2006; Dyce & O’Connor, 1998; Few et al., 2010; Furnham, 2014; Furnham &
Crump, 2005; Huprich, 2003b; Madsen, Parsons, & Grubin, 2006; Nederstrom
& Furnham, 2012; Nestadt et al., 2008; O’Connor, 2005; Oluf & Furnham,
2015; Quirk, Christiansen, Wagner, & McNulty, 2003; Ramanaiah, Rielage, &
Cheng, 2002; Reynolds & Clark, 2001; Rolland & de Fruyt, 2003; Rossier &
Rigozzi, 2008; Saulsman & Page, 2004; Soldz, Budman, Demby, & Merry,
1993; Stepp, Trull, Burr, Wolftenstein, & Vieth, 2005; Trull, 1992; Wang et al.,
2003; Wiggins & Pincus, 1989; Yeung, Lyons, Waternaux, Faraone, & Tsuang,
1993). Widiger, Samuel, Mullins-Sweat, Gore, and Crego (2012) summarize the
research across the personality disorders. Reviews of the research concerning
individual personality disorders have been provided for the borderline (Trull &
Brown, 2013), the schizotypal (Edmundson & Kwapil, 2013), the antisocial
(Derefinko & Lynam, 2013), the dependent (Gore & Pincus, 2013), and the
narcissistic (Campbell & Miller, 2013) personality disorders. Additional
maladaptive personality functioning not recognized within the APA diagnostic
manual is also covered, including, for instance, the traits of psychopathy
(Borroni, Somma, Andershed, Maffei, & Fossati, 2014; Crego & Widiger, 2014;
Derefinko & Lynam, 2013; Gaughan, Miller, Pryor, & Lynam, 2009; Lilienfeld
et al., 2015; Lynam & Miller, 2015; Lynam & Widiger, 2007), successful
psychopathy (Mullins-Sweatt, Glover, Derefinko, Miller, & Widiger, 2010),
Cleckley psychopathy (Crego & Widiger, 2015a), alexithymia (Luminet, Bagby,
Wagner, Taylor, & Parker, 1999; Taylor & Bagby, 2013; Zimmerman, Rossier,
de Stadelhofen, & Gaillard, 2005), depressive personality disorder (Bagby,
Watson, & Ryder, 2013; Huprich, 2000, 2003a; Vachon, Sellbom, Ryder, Miller,
& Bagby, 2009), and prejudice (Flynn, 2005). Livesley (2001) concluded on the
basis of his review of the FFM-personality disorder research that “all categorical
diagnoses of DSM can be accommodated within the five factor framework” (p.
24). Markon, Krueger, and Watson (2005) conducted meta-analytic and
exploratory hierarchical factor analyses of numerous measures of normal and
abnormal personality functioning, and consistently obtained a five factor
solution that they indicated “strongly resembles the Big Five factor structure
commonly described in the literature, including neuroticism, agreeableness,
extraversion, conscientiousness, and openness factors” (p. 144). Clark (2007)
concluded that “the five factor model of personality is widely accepted as
representing the higher-order structure of both normal and abnormal personality
traits” (p. 246).
Samuel, Simms, Clark, Livesley, and Widiger (2010) demonstrated through
item response theory analysis that the maladaptive personality trait scales of the
Dimensional Assessment of Personality Pathology-Basic Questionnaire (DAPP-
BQ; Livesley & Jackson, 2009) and the Schedule for Nonadaptive and Adaptive
Personality (SNAP; Clark, 1993) lie along the same latent traits as those
assessed by the NEO PI-R (Costa & McCrae, 1992). The primary distinction is
that the DAPP-BQ and SNAP scales have relatively greater fidelity (or
coverage) for the assessment of the (maladaptively) extreme variants of FFM
traits, whereas the NEO PI-R has relatively greater fidelity for the more normal
variants. In addition, what was most evident from this study was that there is
considerably more overlap among the scales than uniqueness, due in part to the
fact that the NEO PI-R assesses a substantial range of maladaptivity with respect
to neuroticism, introversion, low openness, antagonism, and low
conscientiousness.
Samuel, Carroll, Rounsaville, and Ball (2013) focused their IRT results
specifically on borderline personality disorder. They indicated that the borderline
symptoms (e.g., recurrent suicidality) fall along the same latent trait as FFM
neuroticism and have relatively greater fidelity for the (maladaptively) extreme
variants of neuroticism, whereas the NEO PI-R has relatively greater fidelity for
the more normal variants. Stepp et al. (2012) integrated items from the NEO PI-
R, the Tridimensional Personality Questionnaire (TCI; Cloninger, 2000), and the
SNAP (Clark, 1993) in a confirmatory factor analysis that documented the
presence of a common five factor model that was closely aligned with the FFM.
Using IRT, they selected the optimal subset of items from each instrument, with
NEO PI-R items covering the lower range of each domain and the SNAP and/or
TCI items covering the more extreme, maladaptive range.
Personality Disorder Scales and the FFM

There has been quite a bit of research using the FFM to effectively
differentiate among different forms of psychopathology (e.g., Bagby et al., 1997;
Bienvenu et al., 2001; see also the chapter by Bagby, Uliaszek, Gralnick, and Al-
Dajani), including the personality disorders (Bagby et al., 2005; Miller, Bagby,
Pilkonis, Reynolds, & Lynam, 2005). Each personality disorder in fact has a
relatively distinctive FFM profile. Lynam and Widiger (2001) and Samuel and
Widiger (2004) asked researchers and clinicians (respectively) to describe a
prototypic case of each personality disorder in terms of the FFM. The
descriptions included strikingly different profiles for the histrionic (high
extraversion) versus the schizoid (high introversion); dependent (high
agreeableness) versus antisocial (high antagonism); and obsessive–compulsive
(high conscientiousness) versus antisocial (low conscientiousness). The avoidant
and schizoid do share introversion, but the former is also characterized by
neuroticism whereas schizoid is not. The avoidant and dependent share
neuroticism, but the latter is also characterized by agreeableness. The schizoid
and schizotypal share introversion, but the latter is also characterized by
neuroticism and openness.
There should not, of course, be entirely distinct FFM profiles for each DSM-
IV-TR personality disorder because these disorders overlap and obtain
considerable diagnostic cooccurrence (Clark, 2007). Lynam and Widiger (2001)
in fact demonstrated that the extent to which the DSM-IV-TR personality
disorders cooccur is explained by the FFM traits they share. The “overlap among
FFM profiles reproduced well the covariation obtained for the schizoid,
schizotypal, antisocial, borderline, histrionic, narcissistic, avoidant, and
compulsive PDs aggregated across several sets of studies” (Lynam & Widiger,
2001, p. 410). The only personality disorder whose diagnostic cooccurrence
could not be explained well was the dependent, precisely because its FFM
description was considerably more differentiated from other personality
disorders than would be suggested by its extensive diagnostic cooccurrence.
Samuel and Widiger (2008b) conducted a meta-analysis of research relating
FFM scales to the DSM-IV personality disorders. The results confirmed the
distinctions suggested by Lynam and Widiger (2001) and Samuel and Widiger
(2004). The empirically based profiles correlated with expert-consensus profiles
from a low of .60 for the dependent personality disorder to a high of .92 for the
obsessive–compulsive. One of the more important results from their meta-
analysis though was strong instrument effects, particularly for the relationship of
conscientiousness with OCPD, openness with the schizotypal personality
disorder, and extraversion with the histrionic.
Indeed, the FFM has also been very useful in providing meaningful
differentiations among scales purportedly assessing the same personality
disorder (e.g., Paulhus & Williams, 2002; Ruiz, Pincus, & Schinka, 2008;
Trobst, Ayearst, & Salekin, 2004), including more specifically the antisocial
(Costa & McCrae, 1990; Decuyper, De Pauw, De Fruyt, De Bolle, & De Clercq,
2009; Gudonis, Miller, Miller, & Lynam, 2008; Hicklin & Widiger, 2005;
Seibert, Miller, Few, Zeichner, & Lynam, 2011), dependent (Lowe et al., 2009;
McBride, Zuroff, Bagby, & Bacchiochi, 2006; Mongrain, 1993; Pincus &
Gurtman, 1995; Zuroff, 1994), narcissistic (Miller & Campbell, 2008; Samuel &
Widiger, 2008a), histrionic (Gore, Tomiatti, & Widiger, 2011), and obsessive–
compulsive (Samuel & Widiger, 2010).
For example, Costa and McCrae (1990) compared different editions of the
Millon Clinical Multiaxial Inventory (MCMI; Millon, Millon, Davis, &
Grossman, 2009) with respect to the FFM. They indicated how the early version
of the MCMI antisocial personality disorder scale was saturated with high
neuroticism that was removed from subsequent versions, which, in turn,
increased the representation of low conscientiousness. Samuel and Widiger
(2010) compared, from the perspective of the FFM, eight different DSM-IV-TR
obsessive–compulsive personality disorder (OCPD) measures. The Minnesota
Multiphasic Personality Inventory (MMPI; Hathaway & McKinley, 1943)
assessment of OCPD (Somwaru & Ben-Porath, 1995), in stark contrast to the
others, was confined largely to neuroticism, with little to no representation of
conscientiousness, likely due to the fact that the MMPI-2 item pool in general
lacks much representation of FFM conscientiousness (Trull, Useda, Costa, &
McCrae, 1995). The SNAP (Clark, 1993) provided approximately equal weight
to its coverage of neuroticism and conscientiousness, whereas the MCMI-III was
heavily saturated with items from conscientiousness. Even more striking for the
MCMI-III was its negative correlation with neuroticism, whereas all other
OCPD measures correlated positively with neuroticism. From this perspective, it
is not surprising that the MCMI-III OCPD scale obtains weak convergent
validity with other measures of OCPD (Miller, Few, & Widiger, 2012; Widiger
& Boyd, 2009).
Gore et al. (2011) indicated how different measures of histrionic personality
traits can be understood from the perspective of the FFM. The MCMI-III and
MMPI-2 assessments of histrionic personality disorder (HPD) correlated
negatively with neuroticism, whereas in striking contrast the HPD scales from
the Coolidge Axis II Inventory (CATI; Coolidge & Merwin, 1992), the SNAP,
the OMNI Personality Inventory-IV (OMNI-IV; Loranger, 2001), the PDQ-IV,
and the WISPI (Klein et al., 1993) all correlated positively with neuroticism.
Furthermore, the MCMI-III and MMPI-2 HPD scales correlated substantially
with extraversion (above .70), whereas the PDQ-IV, OMNI-IV, and WISPI were
uncorrelated with extraversion (the SNAP and CATI though obtained
moderately high positive correlations). The PDQ-IV, OMNI-IV, and WISPI
HPD scales correlated negatively with conscientiousness, whereas the MMPI-2
and MCMI-III were not.
The term “impulsivity” is used by a wide variety of researchers, yet
alternative assessments of impulsivity can produce strikingly different results.
Whiteside and Lynam (2001) compared alternative measures of “impulsivity”
from the perspective of the FFM, leading to the useful distinction of negative
urgency (a facet of neuroticism), lack of premeditation (low deliberation), lack
of perseverance (low self-discipline), and sensation seeking (a facet of
extraversion). They demonstrated how existing measures of “impulsivity” vary
considerably with respect to which variant is being assessed, as well as the social
and clinical importance of the four different variants. Considerable amounts of
subsequent validation and extension of these FFM variants have since been
published (e.g., Lynam, Miller, Miller, Bournovalova, & Lejuez, 2011; Ruiz et
al., 2008; Sharma, Markon, & Clark, 2013; Whiteside, Lynam, Miller, &
Reynolds, 2005). Cyders and Smith (2008) and Smith et al. (2007) have also
extended this FFM-based model of “impulsivity” to include positive urgency, a
facet of extraversion.
One of the more heavily researched personality disorder constructs is the
“dark triad,” combining key traits from psychopathy, narcissism, and
Machiavellianism into one “super” or “meta” evil syndrome (Paulhus &
Williams, 2002). Furnham, Richards, Rangel, and Jones (2014) demonstrated
through a meta-analysis of dark triad FFM research that the three dark triad
constructs can be largely understood as maladaptive variants of FFM
antagonism, albeit they can also be distinguished with Machiavellianism and
psychopathy including aspects of low conscientiousness and neuroticism, and
psychopathy involving facets of extraversion. Indeed, what is shared among the
dark triad constructs and what is distinct for each of them can all be well
accounted for in terms of the FFM domains and facets (DeShong, Grant, &
Mullins-Sweatt, 2015; O’Boyle, Forsyth, Banks, Story, & White, 2015).
Samuel and Widiger (2008a) compared and contrasted five alternative
measures of narcissism from the perspective of the FFM. Among their findings
was that the SNAP (Clark, 1993) and PDQ-IV (Bagby & Farvolden, 2004) were
confined largely to aspects of antagonism with no relationship to neuroticism or
extraversion. In stark contrast, the MMPI-2 assessment of NPD was unrelated to
antagonism and was evenly weighted in its coverage of neuroticism and
extraversion. The MCMI-III fell in between, including a negative relationship
with neuroticism and positive relationships with antagonism and extraversion.
Similar findings were reported by Miller and Campbell (2008). The distinction
between grandiose and vulnerable narcissism has also been well understood in
terms of the FFM (Campbell & Miller, 2013), with both variants involving
antagonism (and extraversion), but vulnerable narcissism including traits of high
neuroticism and grandiose narcissism either being unrelated or negatively
correlated with neuroticism (Glover, Miller, Lynam, Crego, & Widiger, 2012;
Miller et al., 2013).
Finally, Crego and Widiger (2014) contrasted alternative conceptualizations
and assessments of antisocial personality disorder (ASPD) and psychopathy
from the perspective of the FFM. The more traditional measures of ASPD
(assessed by the PDQ-R) and psychopathy (assessed by the Self-Report
Psychopathy—Version III [SRP-III]; Paulhus, Neumann, & Hare, 2015) were
confined to the domains of antagonism and low conscientiousness (or
disinhibition). In stark contrast, the more recently developed measures and
models of psychopathy, provided by the Psychopathic Personality Inventory-
Revised (PPI-R; Lilienfeld & Widows, 2005), the Triachic Model of
Psychopathy (Patrick, Fowles, & Krueger, 2009), and the Elemental
Psychopathy Assessment (EPA; Lynam et al., 2011) all include components of
low neuroticism (e.g., fearless and invulnerability) and high extraversion
(boldness and dominance) that converge highly with one another but not with the
traditional models. Their findings are consistent with those of Lynam et al.
(2011), Few, Miller, and Lynam (2013), and Poy et al. (2014).
In sum, research has indicated that the personality disorders are well
understood in terms of the FFM (Clark, 2007; Livesley, 2001; Widiger et al.,
2012) as well as distinguished (Lynam & Widiger, 2001; Samuel & Widiger,
2004, 2008). The FFM has also been very effective in explaining the diagnostic
cooccurrence among the personality disorders (Lynam & Widiger, 2001).
Finally, FFM research has also been effective in explaining the lack of adequate
convergence among alternative measures of the same personality disorder
construct (Widiger, Costa, Gore, & Crego, 2013).
DSM-5 Section III and the FFM

A major step toward a conceptualization of the DSM-IV-TR personality
disorders from the perspective of the FFM occurred with DSM-5 (American
Psychiatric Association, 2013), the current edition of the American Psychiatric
Association diagnostic manual. The American Psychiatric Association (2013)
DSM-5 includes a dimensional model of maladaptive traits consisting of five
broad domains of negative affectivity, detachment, psychoticism, antagonism,
and disinhibition. These five broad domains are aligned with the FFM. As
expressed in DSM-5, “these five broad domains are maladaptive variants of the
five domains of the extensively validated and replicated personality model
known as the ‘Big Five’ or Five Factor Model of personality (FFM)” (American
Psychiatric Association, 2013, p. 773). As affirmed by Krueger and Markon
(2014, p. 480), “these domains can be understood as maladaptive extremes of the
five factor model (FFM) that has usefully framed extensive research in the field
of personality and individual differences (Widiger & Costa, 2012, 2013).”
However, this alignment might also be somewhat precarious or uncertain.
First, there was (and continues to be) substantial opposition to the inclusion of
any dimensional trait model in DSM-5 (e.g., Clarkin & Huprich, 2011;
Gunderson, 2010; Shedler et al., 2010). This opposition was certainly a major
factor in the eventual decision to relegate the dimensional trait model to an
appendix of DSM-5. In addition, there also appeared to be some ambivalence
within the work group itself, if not in fact opposition, to an alignment with the
FFM. Even persons who have had no history of advocating for or working from
the perspective of the FFM perceived a reluctant attitude toward the FFM within
the DSM-5 work group (e.g., Blashfield & Reynolds, 2012; Lilienfeld, Watts, &
Smith, 2012; Livesley, 2012).
For example, the final literature review in support of the dimensional trait
model (i.e., American Psychiatric Association, 2012) made only a passing
reference to the extensive body of FFM personality disorder research
summarized within this chapter (Blashfield & Reynolds, 2012; Widiger, 2013).
In addition, the initial proposal was for a newly developed six-factor model, for
which an alignment with two of the five domains of the FFM (i.e.,
conscientiousness and openness) was explicitly rejected (i.e., Clark & Krueger,
2010). It was also suggested that the DSM-5 trait model could not be aligned
with the FFM because the FFM was equivalent to one particular, copyrighted
measure, the NEO PI-R of Costa and McCrae (1992). “A potential conflict of
interest exists if the DSM-5 articulates a trait model that is isomorphic with a
model that is assessed by a specific inventory sold by a specific test publisher”
(Krueger et al., 2011, p. 326; see also Krueger, 2013). However, proposals for an
alignment with the FFM had never suggested that the model be equivalent to the
NEO PI-R domains and facets. Indeed, a proposal for DSM-5 provided at a
preliminary DSM-5 preparatory conference that was aligned explicitly with the
FFM had no such measure-specific association or copyright concerns (i.e.,
Widiger & Simonsen, 2005). In fact, if there is a dimensional trait model that is
assessed by alternative instruments, it is the FFM (e.g., De Raad & Perugini,
2002; Samuel, 2013; see also the chapter by Simms, Williams, and Simms).
There are a number of commonly used measures of the FFM that are freely
available, the most notable of which is the IPIP-NEO (Goldberg et al., 2006).
In contrast, there has never been any reluctance to align the DSM-5 trait model
with the Psychopathology-5 (PSY-5; Harkness, McNulty, & Ben-Porath, 1995;
see American Psychiatric Association, 2013, p. 773) that is in fact “assessed by a
specific inventory sold by a specific test publisher” (Krueger et al., 2011, p.
326). There exists only one measure for the PSY-5, which is a specific,
copyrighted inventory, the Minnesota Multiphasic Personality Inventory-2
(MMPI-2) sold by a specific test publisher (University of Minnesota Press).
Although there is an IPIP-NEO that is freely available to all researchers and
clinicians to use, there is no equivalent “IPIP-MMPI-2.”
Nevertheless, in the end, the decision was to align the dimensional trait model
with the FFM (Krueger & Markon, 2014), including an explicit recognition of
this alignment within the diagnostic manual itself (American Psychiatric
Association, 2013, p. 773). The dimensional trait proposal did not receive final
approval for inclusion in the main body of the text, but it was at least included
within Section III of DSM-5 for “emerging measures and models” (American
Psychiatric Association, 2013, p. 728). There is likely to remain a preference
among some to align the DSM-5 dimensional trait model with the MMPI-2 PSY-
5, rather than with the FFM, but perhaps the validity and utility of an alignment
with the FFM will become more apparent over time. Empirical support for an
alignment of the DSM-5 trait model with the FFM has indeed been indicated in a
number of studies, although the alignment of DSM-5 psychoticism with
openness has been problematic (e.g., Crego, Gore, Rojas, & Widiger, 2015; De
Fruyt et al., 2013; Gore & Widiger, 2013; Griffin & Samuel, 2014; Thomas et
al., 2012; Wright & Simms, 2014).
FFM Openness and DSM-5 Psychoticism

The initial position of the DSM-5 Personality and Personality Disorders Work
Group was that there was no alignment of FFM openness with schizotypal
thinking (e.g., Krueger et al., 2011). As stated within the first DSM-5 website
post, “only the ‘social and interpersonal deficits’ of schizotypal personality
disorder, and not the ‘cognitive or perceptual distortions and eccentricities of
behavior’ are tapped by FFM traits” (Clark & Krueger, 2010, para. 2).
Parsimony and consistency would suggest that there would or should be some
form of an alignment of FFM openness with some form of maladaptive
personality functioning. It is fundamentally inconsistent as well as
nonparsimonious to suggest that four domains of normal personality have an
abnormal variant but for no apparent reason one domain (i.e., openness) does
not, and that four domains of abnormal personality have a normal variant but for
no apparent reason one domain (i.e., oddity or psychoticism) does not. It would
seem more compelling that there would be a maladaptive variant of each pole of
the FFM (Nettle, 2006a, 2006b).
Odd, peculiar, and/or aberrant beliefs would appear to be the natural
expectation for this domain (see also the chapter by Sutin). An association of
open-mindedness with schizotypal and/or paranormal beliefs has been reported
in a number of studies (e.g., Miller & Tal, 2007; Nettle, 2006b; Nettle & Clegg,
2006; Thalbourne, 2000). On the other hand, the initial position of the DSM-5
work group was to a degree understandable because the empirical relationship of
FFM openness with the DSM-IV personality disorders has consistently been the
weakest, including the relationship with schizotypal thinking (Bastiaansen,
Rossi, Schotte, & De Fruyt, 2011; Samuel & Widiger, 2008b; Saulsman & Page,
2004; Watson, Clark, & Chmielewski, 2008).
Watson et al. (2008), for instance, conducted factor analyses in which they
separated openness from maladaptive traits, including schizotypal cognitions
(albeit see the chapter by Wright for the suggestion that this may have been a
statistical artifact). Other factor analytic studies have supported an alignment of
schizotypal thinking with openness (e.g., Camisa et al., 2005; Kwapil, Barrantes-
Vidal, & Silvia, 2008; Wiggins & Pincus, 1989), including more recent studies
by DSM-5 work group members and other researchers concerning the
relationship of FFM openness with DSM-5 psychoticism (e.g., De Fruyt et al.,
2013; Gore & Widiger, 2013; Griffin & Samuel, 2014; Thomas et al., 2012).
Stepp et al. (2012) integrated items from the NEO PI-R, the TCI (Cloninger,
2000), and the SNAP (Clark, 1993) in a confirmatory factor analysis that
documented the presence of a common five factor model that was closely
aligned with the FFM, including openness. Using IRT, they selected the optimal
subset of items from each instrument. They reported that items from the NEO
PI-R scales assessing openness to ideas, fantasy, and aesthetics defined the
normal range of their unconventionality dimension, whereas the SNAP scale
assessing eccentric perceptions (along with TCI self-transcendence) defined the
abnormal range.
There are reasons though to expect that the relationship of openness with
schizotypal thinking (and DSM-5 Section III psychoticism in particular) will not
always be confirmed. One reason for a relatively weak relationship of openness
with schizotypal personality disorder (STPD) is that STPD includes indicators of
both high and low openness. Studies have indicated that schizotypal
symptomatology can have opposite relationships to different facets of openness.
Kwapil et al. (2008) and Ross, Lutz, and Bailey (2002) reported that the positive
symptoms of schizotypia (e.g., magical ideation and perceptual aberrations)
correlated positively with FFM openness (a finding replicated by Asail,
Sugimori, Bando, & Tanno, 2011), whereas negative symptoms of schizotypia
(e.g., physical anhedonia) correlated negatively. To the extent that an assessment
of STPD includes both positive and negative components, a correlation with
FFM openness might not appear as they may cancel each other out.
More important than the complexity of STPD, however, is how both
constructs, openness and psychoticism, are conceptualized and assessed (Ashton
& Lee, 2012; Chmielewski, Bagby, Markon, Ring, & Ryder, 2014; Gore &
Widiger, 2013; Samuel & Widiger, 2008b). Haigler and Widiger (2001)
demonstrated empirically that when NEO PI-R Openness items are revised to
assess maladaptive variants of the same openness content, correlations with
schizotypy emerged, but still only at a marginal level. The NEO PI-R Openness
scale was originally constructed prior to any knowledge of Costa or McCrae
regarding the lexical Big Five (see the chapters by De Raad and Mlačić, by
Costa and McCrae, and by Wright). Costa and McCrae (1980) began with just a
three-factor model, assessed by the NEO Inventory (e.g., McCrae & Costa,
1983). They conceptualized openness at that time to represent ideal personality
traits such as self-actualization, an open mind, and self-realization, citing
humanism papers and texts by Coan (1974), Rokeach (1960), and Rogers (1961).
Soon after the development of the NEO Inventory, Costa and McCrae became
aware of the Big Five and they extended their model and instrument to include
the domains of agreeableness and conscientiousness. However, they did not
revise the scales for neuroticism, extraversion, or openness. This does not appear
to have been a significant problem for the assessment of extraversion, but they
subsequently acknowledged that NEO PI-R Openness did not align as well with
lexical Big Five intellect (McCrae, 1990).
In more recent years, a number of alternative models (as well as measures) of
FFM and/or Big Five openness have been developed (see the chapter by Sutin)
for which a clear relationship of schizotypal thinking with openness is much
more readily apparent. For example, Piedmont, Sherman, Sherman, Dy-Liacco,
and Williams (2009) developed scales to assess maladaptive variants of both
high and low FFM openness. One of their scales for high openness is Odd and
Eccentric, which correlates significantly with schizotypal personality disorder
and various measures of aberrant perceptions and paranormal beliefs (Piedmont,
Sherman, & Sherman, 2012). Edmundson et al. (2011) constructed a measure of
schizotypal personality traits from the perspective of the FFM, and indicated that
its scales assessing aberrant perceptions, aberrant ideas, and odd, eccentric
behavior converged well with alternative measures of FFM openness. DeYoung,
Quilty, and Peterson (2007) developed the Big Five Aspects Scale (BFAS),
which distinguishes between openness and intellect within a domain of
“openness/intellect” (see also the chapter by Allen and DeYoung). DeYoung,
Grazioplene, and Peterson (2012) suggest that BFAS Openness will be
associated with schizotypal thinking but BFAS Intellect will not (see also
Chmielewski et al., 2014). Van Kampen (2012) includes within his 5-
Dimensional Personality Test (5DPT) an Absorption scale, which aligns
explicitly with FFM openness and assesses dissociative absorption and positive
symptoms of schizotypy. The HEXACO-Personality Inventory (HEXACO-PI;
Lee & Ashton, 2004) includes an Openness to Experience scale that corresponds
conceptually and empirically with FFM openness. This HEXACO-PI scale
includes four facet scales, one of which is titled Unconventionality, which
assesses the disposition to be eccentric, weird, peculiar, odd, and strange.
Tellegen similarly includes an unconventionality domain within his dimensional
trait model that is aligned explicitly with FFM openness, the scale for which
contains items that assess normal openness (e.g., curious, inquisitive,
imaginative, and creative) as well as items that concern attributes such as having
ideas or beliefs that have little basis within reality, dwelling upon fantasies, or
often engaging in activities that are bizarre, deviant, or aberrant (Tellegen &
Waller, 1987). Tellegen (1993) in fact suggested years ago with respect to his
conceptualization of openness that “markers of this type are clinically
suggestive, particularly of schizotypal personality disorder” (p. 126).
The strength of the alignment can also be similarly affected by the
conceptualization and assessment of schizotypal thinking, oddity, and/or
psychoticism. For example, DSM-5 psychoticism may go beyond schizotypal
magical and aberrant thinking to include overt psychotic delusions that are
apparent in schizophrenia but not in STPD. The assessment of DSM-5
psychoticism within the Personality Inventory for DSM-5 (PID-5; Krueger et al.,
2012), which is the official measure for the DSM-5 dimensional trait model,
includes psychotic symptoms that perhaps lie outside of general personality
structure (e.g., “Sometimes I feel ‘controlled’ by thoughts that belong to
someone else,” and “Sometimes I think someone else is removing thoughts from
my head”). Items that suggest Schneiderian delusions (Schneider, 1959), such as
thought control and thought broadcasting, are perhaps best understood as part of
a schizophrenic psychotic disorder rather than reflecting the magical thinking
and perceptual confusions that would be evident in persons who are just odd,
peculiar, and/or eccentric in a schizotypic manner (Kwapil & Barrantes-Vidal,
2012).
Bipolarity of Maladaptive Personality Structure

There are some differences between the DSM-5 dimensional trait model and
the FFM of maladaptive personality functioning (Trull, 2012). The FFM of
personality disorder, as it is typically presented, includes maladaptivity at both
poles of all five domains (e.g., Lynam & Widiger, 2001; Miller, 2012; Samuel,
2011; Trull, 2012; Widiger & Trull, 2007). The DSM-5 dimensional trait model
does indeed recognize some of this “bipolarity of maladaptivity,” but to a much
more limited degree (Clark & Krueger, 2010; Krueger et al., 2011; see also the
chapter by Simms, Williams, and Simms).
Some dimensional models of maladaptive personality functioning are entirely
unipolar in the maladaptivity of their structure. One example of such a model is
the Dimensional Assessment of Personality Pathology by Livesley (Livesley &
Jackson, 2009), consisting of the four domains of emotional dysregulation,
dissocial behavior, inhibition, and compulsivity. Traits that are considered to be
opposite or low on these domains do not involve any maladaptive functioning
(i.e., “unipolar” in their coverage of maladaptivity). However, in contrast, most
other dimensional trait models and/or measures include at least some degree of
bipolar maladaptivity, with some more so than others.
For example, the SNAP by Clark (1993) includes considerable bipolarity. The
12 scales of the SNAP are organized with respect to three domains: negative
affectivity (or negative temperament), positive affectivity (or positive
temperament), and constraint versus disinhibition. The Exhibitionism and
Entitlement scales load positively on the positive affectivity domain (which
aligns with FFM extraversion), whereas the Detachment scale loads negatively.
Similarly, Propriety and Workaholism load positively on the constraint factor
(which aligns with FFM conscientiousness), whereas the Impulsivity scale loads
negatively. This SNAP bipolarity has been replicated in many subsequent factor
analytic studies (e.g., Markon et al., 2005; Simms & Clark, 2005, 2006; Watson
et al., 2008). The DSM-5 trait model does not include exhibitionism, entitlement,
propriety, or workaholism. The DSM-5 trait model though does include rigid
perfectionism, as a trait opposite to disinhibition (i.e., a maladaptive variant of
conscientiousness).
There is a comparable bipolarity in the Computerized Adaptive Test of
Personality Disorder (CAT-PD; Simms et al., 2011; see also the chapter by
Simms, Williams, and Simms). The CAT-PD has three scales that load
negatively on disinhibition (i.e., Perfectionism, Rigidity, and Workaholism,
which aligns with FFM conscientiousness), whereas Irresponsibility, Non-
planfulness, and Non-perseverance load positively (Wright & Simms, 2014).
The CAT-PD also has a scale for maladaptive extraversion: Exhibitionism,
which loads negatively on the detachment factor (which aligns with FFM
introversion).
In sum, existing measures and models do include some representation of
maladaptive variants of the domains of extraversion and conscientiousness,
including the CAT-PD (Wright & Simms, 2014) and the SNAP (Simms & Clark,
2005, 2006). Nevertheless, questions have been raised with respect to the
inclusion of any such representation (see the chapters by Wright and by Simms,
Williams, and Simms). The extent of this representation is likely to be a
continued point of contention and disagreement.
Pettersson and colleagues suggest that the maladaptivity of both poles of all
five domains emerges more clearly after variance is first removed due to an
evaluation bias (Pettersson, Turkheimer, Horn, & Menatti, 2012; Pettersson et
al., 2014). “We suggest that there is some degree of bipolarity in most, if not all,
traits in terms of both their adaptive and their maladaptive qualities” (Pettersson
et al., 2014, p. 444). In Pettersson et al. (2012), traits such as overbearing and
wild became apparent for extraversion; workaholic and uptight for
conscientiousness; and gullible, timid, and submissive for agreeableness.
The variance that Pettersson and colleagues (2014) refer to as constituting an
evaluative method factor may also reflect, for the most part, the extent of
maladaptive dysfunction. There will be a very high correlation between extent of
positive versus negative evaluation and extent of adaptive versus maladaptive
functioning. Pettersson et al. (2014) indicated how persons will endorse items
that are opposite to one another, such as perfectionism that leads to job failure
and negligence that leads to job failure. They suggest that this reflects a
contaminating method variance of evaluation (e.g., a disposition to endorse
negative, undesirable traits irrespective of the content). However, it may also
simply reflect the fact that there will be a strong positive correlation of one form
of maladaptive behavior with other maladaptive behaviors (again regardless of
content). For example, persons may describe themselves as being both
perfectionistic and irresponsible, largely because they both result in a tendency
to do jobs poorly. There is considerable discussion in the current
psychopathology literature regarding a “p-factor,” or a general factor of
maladaptivity that contributes to a nonspecific covariation (comorbidity) across
widely diverse forms of mental disorder (Caspi et al., 2014). This p-factor is
evident as well within the field of personality disorder (Wright et al., 2012).
Personality disorders that clearly involve behaviors opposite to one another (e.g.,
dependent and antisocial or avoidant and histrionic) can be comorbid (diagnosed
in the same person), due largely to the fact that they involve comparable levels
or manner of dysfunction.
It would seem straightforward to suggest that gullibility is a maladaptive
variant of being trusting, selflessness is an extreme variant of altruism,
subservience is an extreme form of compliance, workaholism is a maladaptive
variant of achievement-striving, and perfectionism is a maladaptive variant of
competence (Nettle, 2006a). Yet, it would also not be surprising to find a lack of
a positive correlation (for instance) between adaptive trust and maladaptive
gullibility, or between adaptive compliance and maladaptive subservience.
Adaptive traits will naturally correlate negatively with maladaptive traits, even
when they concern the same trait. It is then perhaps all the more compelling that
researchers have in fact been able to demonstrate positive correlations between
adaptive and maladaptive variants of the same trait (Haigler & Widiger, 2001),
as the natural tendency for there to be a negative correlation has to be overcome.
Maladaptive agreeableness and extraversion have been well recognized when
these FFM domains are considered from the perspective of the interpersonal
circumplex (IPC). FFM agreeableness and extraversion are readily understood as
approximately 45 degree rotations of the IPC dimensions of agency and
communion (McCrae & Costa, 1989; Wiggins & Pincus, 2002). This has not
received any significant dispute. And it is also well established that there are
maladaptive variants of all eight octants of the IPC (Pincus & Hopwood, 2012),
including the locations occupied by agreeableness (as well as its polar opposite
antagonism) and extraversion (as well as its polar opposite introversion). There
are even well-established measures of the maladaptive variants for every octant
of the circumplex, such as the Inventory of Interpersonal Problems (Horowitz,
Alden, Wiggins, & Pincus, 2000). Wright et al. (2012) reviewed the DSM-5
dimensional trait model from the perspective of the IPC and noted the
inadequate representation of maladaptive variants of the gregarious–extraverted,
warm–agreeable, and even the unassuming–ingenuous octants. “Pathologically
warm traits are not extensively represented in the DSM-5 trait model, relative to
other interpersonal domains” (Wright et al., 2012, p. 271). In sum, the presence
of the maladaptive variants of the lower right and upper right octants of the IPC
have not been in dispute, and these are the precise locations of FFM
agreeableness and FFM extraversion, respectively. To suggest that there are no
meaningful maladaptive variants of extraversion and agreeableness would be to
neglect the considerable body of IPC personality disorder literature and research
(see Pincus & Hopwood, 2012).
Coker, Samuel, and Widiger (2002) conducted a lexical study on the presence
and extent of socially undesirable, maladaptive traits within the English
language. They coded each of the 1,710 trait terms within the language with
respect to their undesirability and then considered their location within the FFM
(Goldberg, 1993). It was apparent that there are considerably more desirable
terms for low neuroticism than for high neuroticism, for extraversion than for
introversion, for high rather than low openness, for agreeableness than for
antagonism, and for high conscientiousness than for low conscientiousness.
Nevertheless, there were still many undesirable, maladaptive trait terms for
agreeableness, extraversion, openness, conscientiousness, and even for low
neuroticism. In fact, 43% of the extraversion traits were considered to be
undesirable. Table 20.1 provides the percentages for each pole.
Table 20.2 provides illustrative examples for all 10 poles, based on the
findings of Coker et al. (2002).
Table 20.1 Percent of Maladaptive (Undesirable) Trait Terms within High and Low
Poles of the Big Five
Big Five Domain
Emotional Surgency Intellect Agreeableness Conscientiousness
Instability (Extraversion) (Openness)
(Neuroticism)
High 60% 43% 26% 14% 20%
Low 8% 89% 94% 97% 96%
Note: Data adapted from Coker, Samuel, and Widiger (2002).
Table 20.2 Maladaptive (Undesirable) Trait Terms within High and Low Poles of
the Big Five
Big Five Domain
Emotional Surgency Intellect Agreeableness Conscientiousness
Instability (Extraversion) (Openness)
(Neuroticism)
High Defensive Blustery Overindulgent Deceivable Overbookish
Moody Exaggerative Rebellious Dependent Overcautious
Hypersensitive Flaunty Unconformable Ingratiating Stringent
Self-Destructive Showy Unconventional Transparent Tight
Low Conscienceless Aloof Dogmatic Deceitful Careless
Emotionless Humorless Prejudiced Heartless Disorderly
Inexcitable Reclusive Unimaginative Treacherous Heedless
Inhuman Somber Unreflective Violent Reckless
Note: Data obtained from Coker, Samuel, and Widiger (2002).
For extraversion there was long-winded, blustery, showy, flaunty, and

exaggerative; for conscientiousness there was overbookish, overcautious,
leisureless, stringent, and tight; and for agreeableness there was deceivable,
dependent, soft-shelled, and ingratiating. These three sets of traits are quite
suggestive of the histrionic, obsessive–compulsive, and dependent personality
disorders (respectively) that are defined in large part by the domains of
extraversion, conscientiousness, and agreeableness, respectively (Lynam &
Widiger, 2001; Saulsman & Page, 2004; Samuel & Widiger, 2004, 2008).
Boudreaux, Piedmont, Sherman, and Ozer (2013) developed the Multi-
Context Problems Checklist as a measure of personality-related problems in
living, particularly with respect to employment and work. In the course of this
effort they identified “specific personality-related problems associated with each
pole of each trait of the five factor model of personality” (p. 62). It was not
difficult to identify problems associated with high neuroticism, introversion, low
openness, antagonism, and low conscientiousness. However, they did indeed
identify problems associated with high extraversion (e.g., needing to be in
control, imposing opinions and beliefs on the partner), high openness (e.g., an
overactive imagination), high agreeableness (not being able to say “no” and
being too conforming), and high conscientiousness (e.g., pushing too hard to
excel and being perfectionistic).
Nettle (2006a) provided an evolutionary model for understanding the
emergence of the factors of the FFM. “Each of the Big Five dimensions of
human personality can be seen as the result of a trade-off between different
fitness costs and benefits” (Nettle, 2006a, p. 622). This helps explain the
genetics of individual differences. “As there is no unconditionally optimal value
of these trade-offs, it is to be expected that genetic diversity will be retained in
the population” (p. 622). It is apparent across evolutionary models of
psychopathology, personality, and personality disorder that there are both costs
and benefits for what is generally considered normal and abnormal personality
functioning (Crespi, 2014; Del Giudice & Del Giudice, 2014).
Nettle (2006a), for example, suggested that extraversion, although largely
adaptive with respect to exploration, activity, and sexual pursuit, also carries
with it risk-taking and maladaptive sensation-seeking. “People falling at this end
of the continuum are more likely to be sexually promiscuous, emotionally
intrusive, and engage in excessive self-disclosure and thrill-seeking behaviors”
(see also the chapter by Wilt and Revelle). The benefits of conscientious self-
control, orderliness, and achievement-striving are self-evident, but Nettle
(2006a) suggested that this domain of personality can also have significant costs,
as in perfectionism and missed opportunities (due to excessive constraint).
Agreeableness is generally quite desirable, but “very high agreeableness, if it led
to an excessive attention to the needs and interests of others, or excessive
trusting, would be detrimental to fitness” (Nettle, 2006a, p. 627). Openness is a
divergent cognitive style that seeks novelty, creativity, and complexity. “Though
such a cognitive style might appear purely beneficial, it is conceptually very
similar to components of schizotypy” (Nettle, 2006a, p. 626). “The unusual
thinking style characteristic of openness can lead to nonveridical ideas about the
world, from supernatural or paranormal belief systems to the frank break with
reality” (Nettle, 2006a, p. 627).
Even low levels of neuroticism can be maladaptive, contributing to a failure to
avoid hazards and anticipate negative outcomes (Nettle, 2006a). It can perhaps
be difficult for some to recognize or appreciate the potentially maladaptive
nature of low neuroticism. Certainly emotional stability is quite adaptive and
desirable. However, imagine being incapable of feeling anxious. Neuroticism
exists as a universal trait in part because it does have certain benefits for
adaptive functioning (Crespi, 2014; Del Giudice & Del Giudice, 2014). If
neuroticism does have adaptive strengths, then clearly its absence would at times
be experienced as maladaptive. For example, anxiety is a useful trait for
anticipating negative outcomes and risks. The absence of an ability to feel
anxious would be analogous to the inability to feel physical pain, as in the case
of congenital analgesia, a very debilitating and life-threatening disease. Persons
who are abnormally low in anxiousness are unlikely to avoid dangerous
activities or to respond to cues of social and physical harm. Explored in the
follow sections will be the empirical support for maladaptive conscientiousness
and maladaptive agreeableness.
Maladaptive Conscientiousness
It is self-evident that low conscientiousness is associated with maladaptive
personality functioning. However, there is also research demonstrating that high
levels of conscientiousness can also be problematic, interpreted as another
instance of the “too-much-of-a-good-thing” phenomenon (Pierce & Aguinis,
2013). Indeed, there have been a number of studies demonstrating a curvilinear
relationship of conscientiousness with problematic outcomes, indicating that
both low and high levels of conscientiousness are associated with decreased
performance (Carter et al., 2014, 2015; Le, Oh, Robbins, Ilies, Holland, &
Westrick, 2011; Pierce & Aguinis, 2013; Whetzel et al., 2010).
Studies have also indicated a relationship of FFM conscientiousness with the
OCPD symptomatology (e.g., Bastiaansen, Rossi, Schotte, & De Fruyt, 2011;
Clark & Livesley, 2002; Clark, Livesley, Schroeder, & Irish, 1996; Clark,
Vorhies, & McEwen, 2002; Markon et al., 2005; Saulsman & Page, 2004;
Schroeder, Wormworth, & Livesley, 1992). In their meta-analysis of FFM-
OCPD research Saulsman and Page (2004) had reported a relatively small (albeit
still significant) association of OCPD with FFM conscientiousness. They
concluded that “those [personality disorders] particularly characterised by [FFM]
orderliness show positive associations with conscientiousness (e.g., obsessive-
compulsive)” (p. 1075). O’Connor (2005), in his meta-analysis, had concluded
that OCPD aligns well with conscientiousness (obtaining a loading of .72 on the
respective factor), replicating two earlier meta-analytic studies by O’Connor and
colleagues that also clearly aligned compulsivity with conscientiousness (i.e.,
O’Connor, 2002; O’Connor & Dyce, 1998). Samuel and Widiger (2008b)
similarly concluded in their meta-analysis that “a predominant finding of the
studies included within this meta-analysis was a positive correlation of FFM
conscientiousness facets with obsessive-compulsivity personality disorder” (p.
12).
The relationship of the full syndrome of DSM-IV OCPD with
conscientiousness though will not always be strong in part because OCPD is a
heterogeneous personality disorder that includes more than just facets of
conscientiousness. Joint factor analyses of the FFM with the more specific
components of OCPD, such as compulsivity (assessed, for example, by the
DAPP-BQ; Livesley & Jackson, 2009) and workaholism and propriety (assessed
by the SNAP; Clark, 1993), have provided clearer and consistent evidence for a
strong association with conscientiousness in factor analytic studies by DSM-5
work group members (e.g., Clark et al., 1996, 2002; Markon et al., 2005;
Schroeder et al., 1992). In reviewing their models together, Clark and Livesley
(2002) had in fact concluded that “compulsivity (conventionality-rigidity)
undoubtedly tapped conscientiousness” (p. 167). Livesley from the beginning of
the presentation of his dimensional trait model aligned his domain of
compulsivity with FFM conscientiousness. “Compulsivity resembles the
conscientiousness domain of the 5-factor approach” (Livesley, Jang, & Vernon,
1998, p. 945). In an early draft of the dimensional trait model for DSM-5,
Krueger, Skodol, Livesley, Shrout, and Huang (2008) had included “orderliness”
and “conscientiousness” as facets within the domain of compulsivity.
Samuel and Widiger (2011) explored the relationship of conscientiousness
with compulsivity using six alternative measures of conscientiousness, seven
alternative measures of OCPD, and three scales assessing specific components
of OCPD (i.e., Compulsivity, Workaholism, and Propriety). They reported a
robust relationship between DAPP-BQ Compulsivity and all six measures of
conscientiousness. SNAP Workaholism and Propriety similarly related strongly
with FFM conscientiousness, consistent with prior research that has related
measures of perfectionism with FFM conscientiousness (e.g., Stoeber, Otto, &
Dalbert, 2009). The relationship weakened somewhat with the broader measures
of OCPD, likely due again to its inclusion of some components of personality
beyond conscientiousness, such as high neuroticism and low openness (Lynam
& Widiger, 2001; Samuel & Widiger, 2008b, 2010).
Samuel et al. (2012) developed a measure of OCPD from the perspective of
the FFM, the Five Factor Obsessive-Compulsive Inventory (FFOCI), which
includes six scales that were hypothesized to assess maladaptive variants of
conscientiousness: Perfectionism, Fastidiousness, Punctiliousness, Doggedness,
Workaholism, and Ruminative Deliberation. They reported that all six scales
were moderately to strongly related to conscientiousness (as well as with
measures of OCPD). Crego, Samuel, and Widiger (2015) replicated and
extended their findings, including additional measures of conscientiousness,
such as the Dependability scale from the Inventory of Personal Characteristics
(Tellegen & Waller, 1987); Activity from the Zuckerman–Kuhlman–Aluja
Personality Questionnaire (Aluja, Kuhlman, & Zuckerman, 2010);
Conscientiousness from the International Personality Item Pool-NEO (IPIP-
NEO; Goldberg et al., 2006); and Orderliness from the 5DPT (van Kampen,
2012). All six FFOCI maladaptive conscientiousness scales related robustly with
all four alternative measures of conscientiousness. For example, the correlations
with IPIP-NEO Conscientiousness ranged from .52 (for Ruminative
Deliberation) to .70 (for Perfectionism). Their correlations with any one of the
other four domains of the FFM were never higher than .26.
It is perhaps not really that difficult to understand that compulsivity, and
thereby many of the traits of OCPD, are maladaptive variants of FFM
conscientiousness. The essential feature of OCPD is “a preoccupation with
orderliness, perfectionism, and mental and interpersonal control” (American
Psychiatric Association, 2013, p. 678), including within its diagnostic criteria
traits such as perfectionism, preoccupation with order and organization,
workaholism, and, quite explicitly, overconscientiousness. Similarly, FFM
conscientiousness includes facets such as order, discipline, achievement-striving,
and deliberation (Costa & McCrae, 1992). It would seem apparent that
maladaptive and/or extreme variants of order, discipline, achievement-striving,
and deliberation would be the OCPD traits of perfectionism, preoccupation with
order and organization, workaholism, and overconscientiousness.
In sum, it is well accepted that low conscientiousness, characterized by
disinhibition, laxness, irresponsibility, negligence, and rashness, includes
maladaptive personality traits, and aligns closely with the domain of
disinhibition within the DSM-5 dimensional trait model (De Fruyt et al., 2013;
Gore & Widiger, 2013; Griffin & Samuel, 2014; Thomas et al., 2012; Wright et
al., 2012). However, it also appears to be apparent that high conscientiousness
can also be associated with maladaptive personality traits, such as perfectionism,
propriety, fastidiousness, punctiliousness, doggedness, workaholism, and
ruminative deliberation, which together can be identified as a domain of
compulsivity (Crego, Samuel, et al., 2015; Livesley et al., 1998; Samuel & Gore,
2012). This domain of maladaptive personality functioning is not included
within the DSM-5 dimensional trait model, but it is one of the four fundamental
domains of maladaptive personality functioning included within Livesley’s
DAPP (Livesley & Jackson, 2009) and the dimensional trait model proposed for
ICD-11 (Tyrer et al., 2011).
Maladaptive Agreeableness
Agreeableness includes traits such as compliant, trusting, modest, and
altruistic. It is evident that it is typically preferable to be characteristically
agreeable than characteristically antagonistic. However, it would also seem
evident that some persons are overly compliant, excessively trusting,
inordinately modest, and/or excessively altruistic. Indeed, there are quite a few
maladaptive trait terms within high agreeableness: gullibility, self-effacement,
subservience, submissive, docile, servile, clinging, defenseless, selfless, and
acquiescent (Coker et al., 2002). In fact, these are all traits that are quite
suggestive of a dependent personality disorder. Leary (1957) devoted a chapter
of his seminal IPC text to “the dependent personality” (p. 292), located in the
lower-right (“docile-dependent”) octant of the IPC, which is the precise location
of FFM agreeableness (Gore & Pincus, 2013; Pincus & Hopwood, 2012). In its
less severe form it was said to involve a “poignant or trustful conformity”; in its
more severe form it was said to involve a “helpless dependency.”
Quite a few studies have indeed confirmed a close relationship between
dependency and the agreeableness octant of the IPC (e.g., Morey, 1985; Sim &
Romney, 1990). For example, Soldz, Budman, Demby, and Merry (1993)
reported that dependent personality disorder (DPD) was empirically located
within the lower-right “exploitable” octant of the IPC. Trobst, Ayearst, and
Salekin (2004) considered four alternative measures of DPD, and again located it
within the lower-right quadrant, spread along the octants of unassured–
submissive, unassuming–ingenuous, and warm–agreeable, depending upon
which measure of DPD was used. Finally, Smith, Hilsenroth, and Bornstein
(2009) likewise reported that the closer patients resembled the prototypic case of
DPD, the more they described their relational style as affiliative–submissive,
nonassertive, and overly accommodating, all located again within the lower-right
quadrant of the IPC.
Research concerned directly with the relationship of dependency and/or DPD
with FFM agreeableness, however, has not always provided strong support,
despite the consistent findings when agreeableness is understood to be a variant
of the IPC. Some self-report inventory studies have confirmed the relationship
between FFM agreeableness and DPD (e.g., Bagby et al., 2001; Mongrain, 1993;
Wiggins & Pincus, 1989; Zuroff, 1994). However, many self-report inventory
studies have failed to obtain a strong relationship, as indicated in meta-analyses
by Bornstein and Cecero (2000), Miller and Lynam (2008), Saulsman and Page
(2004), and Samuel and Widiger (2008b). Samuel and Widiger (2008b) though
did indicate a significant effect of the instrument on the strength of this
relationship.
Haigler and Widiger (2001) reported that the NEO PI-R (Costa & McCrae,
1992), the most commonly used measure in FFM research, is heavily
imbalanced in its assessment of adaptive and maladaptive variants of high versus
low conscientiousness, neuroticism, extraversion, openness, and most relevant
here, agreeableness. They indicated that 83% of the NEO PI-R Agreeableness
items were assessing for adaptive rather than maladaptive agreeableness. Four
meta-analyses of FFM dependency have been conducted. Seventeen of the 18
studies considered by Bornstein and Cecero (2000), 11 of the 15 samples
considered by Saulsman and Page (2004), all of the studies considered by Miller
and Lynam (2008), and all of the studies considered by Samuel and Widiger
(2008b) used either the NEO PI-R or an instrument closely modeled after the
NEO PI-R. Haigler and Widiger (2001) experimentally altered the NEO PI-R by
inserting words in the test items to change the direction of the maladaptivity
without otherwise changing the content of the items. For example, the NEO PI–
R Altruism items “I try to be courteous to everyone I meet,” “Some people think
of me as cold and calculating” (reverse keyed), “I think of myself as a charitable
person,” “Some people think I’m selfish and egotistical” (reverse keyed), and “I
go out of my way to help others if I can” (Costa & McCrae, 1992, p. 72) all
describe behavior for which it would be preferable (or adaptive) to endorse the
item in the altruistic direction. The experimentally altered versions were “I am
overly courteous to everyone I meet,” “I can be cold and calculating when it’s
necessary,” “I am so charitable that I give more than I can afford,” “Most people
think that I take good care of my own needs,” and “I have sacrificed my own
needs to help others” (respectively). NEO PI-R Agreeableness, prior to the
manipulation, had correlations of .04, .17, and .04 with three independent
measures of DPD. The correlations increased to .57, .66, and .45 (respectively)
with the experimentally altered version.
Lowe, Edmundson, and Widiger (2009) replicated and extended the findings
of Haigler and Widiger (2001), including three measures of DPD, eight
measures of trait dependency, and two measures of dependency from alternative
dimensional models of personality disorder (i.e., Diffidence from the DAPP-BQ
and Dependency from the SNAP). They found that only 8% of the 13 measures
of trait dependency and/or DPD in the student sample and only 15% of the 13
scales within the clinical sample correlated significantly with NEO PI-R
Agreeableness. When the experimentally altered version of the NEO PI-R
Agreeableness scale was administered, 77% of the 13 trait dependency or DPD
measures correlated significantly with Agreeableness in the student sample and
92% of the 13 scales correlated significantly in the clinical sample.
Dependent personality traits are included within the domain of neuroticism (or
negative affectivity) within some alternative dimensional trait models and
measures, such as the PID-5 (Krueger et al., 2012), SNAP (Clark, 1993), and
HEXACO PI-R (Lee & Ashton, 2004). For the PID-5 and SNAP this is due in
part to the absence of an opportunity for dependent traits to be included within
agreeableness, given the absence of this domain within the respective measure
and model. Harlan and Clark (1999) and Linde et al. (2013) though did fail to
replicate the location of Dependency within SNAP Negative Affectivity in their
effort to replicate the factor analytic structure of the SNAP. However, even when
a measure of agreeableness has been included within a respective factor analysis,
dependent traits have at times still loaded within the domain of neuroticism (e.g.,
Clark et al., 1996, 2002; Lee & Ashton, 2004).
DPD does appear to be largely a mixture of the traits of neuroticism and
agreeableness. Blais (1997) had clinicians assess their patients with respect to
DPD and the FFM. He reported an essentially equal relationship of DPD to both
neuroticism and agreeableness. Similar results were obtained by Sprock (2002)
and Mullins-Sweatt and Widiger (2007). Lynam and Widiger (2001) asked
researchers to describe prototypic cases of DPD in terms of the domains and
facets of the FFM. DPD included facets from both agreeableness (i.e., trust,
compliance, and modesty) and neuroticism (i.e., anxiousness and self-
consciousness). Samuel and Widiger (2004) surveyed clinicians in a similar
fashion and again reported facets of both agreeableness (i.e., compliance and
modesty) and neuroticism (i.e., anxiousness, depressiveness, self-consciousness,
and vulnerability).
Submission and subservience are clearly a manner of interpersonal relatedness
(Leary, 1957). However, DPD also includes feelings of neediness, insecurity,
vulnerability, and helplessness. Self-report items that are said to be concerned
with the interpersonal components of dependency (e.g., submissiveness) will at
times conflate the assessment of the submissive, meek, and subservient manner
of interpersonal relatedness with associated feelings of insecurity, helplessness,
despondence, and dysphoria. Just as antagonistic behavior will often be
accompanied by feelings of angry hostility (from neuroticism), submissive and
dependent behavior will often be accompanied by feelings of anxiousness,
insecurity, and helplessness. In other words, the extent to which a measure of
dependency or submission aligns with agreeableness versus neuroticism may
depend on the extent to which the measure is saturated with the feelings of
inadequacy along with the submissive manner of interpersonal relatedness.
Conclusions
In sum, a considerable body of theory and existing research is consistent with
the proposal that there are maladaptive variants of all 10 poles of all five
domains of the FFM. The distribution of this maladaptivity is sorely
disproportionate (e.g., there is considerably more maladaptivity for antagonistic
traits than for agreeable traits and considerably more adaptivity for agreeable
traits than for antagonistic traits), but that does not belie the existence of some
degree of socially and clinically important maladaptivity within both poles.
Working against the validation of this bipolarity, however, is the simple fact that
measures of adaptive functioning will typically (if not invariably) correlate
negatively with measures of maladaptive functioning, and measures of
maladaptive functioning will often correlate positively with one another
(regardless of the content). It will be important for future research addressing
this bipolarity to recognize that a maladaptive variant of a normal trait may in
fact correlate negatively with that trait due simply to this p-factor. However,
there are measures in which this bipolarity has been explicitly incorporated,
which is discussed further in the next section.
Maladaptive FFM Scales

A common criticism of the FFM, with respect to its coverage of personality
disorders, has been that the primary FFM measures have lacked adequate fidelity
for the assessment of its maladaptive variants (e.g., Krueger et al., 2011;
Reynolds & Clark, 2001). This concern no longer applies, as the field is now
awash with measures to assess maladaptive variants of the FFM (Samuel, 2013;
see also the chapter by Simms, Williams, and Simms).
However, it is important to appreciate that the NEO PI-R (Costa & McCrae,
1992) is not well understood as simply a measure of normal, adaptive
personality functioning. It also provides an assessment of maladaptive variants
of high neuroticism, low extraversion, low openness, high antagonism, and low
conscientiousness (Haigler & Widiger, 2001) and these are the poles of the FFM
that are primarily involved for most (but not all) of the DSM-IV-TR personality
disorders (Lynam & Widiger, 2001). It is largely for this reason that the NEO PI-
R has in fact been used quite effectively for the assessment of many of the
personality disorders (Bastiaansen, Rossi, & De Fruyt, 2013; Miller, 2012). This
fact alone, that the NEO PI-R, a measure of normal personality, provides a
measure of (for instance) borderline, antisocial, and avoidant personality
disorder that is comparable to direct measures of these personality disorders, is
itself a testament to the validity of understanding these personality disorders
from the perspective of the FFM (Miller, 2013). The exceptions tend to be for
the assessment of the dependent, obsessive–compulsive, schizotypal, and
histrionic personality disorders (Miller, 2012) as these include significant
components of maladaptive agreeableness, conscientiousness, openness, and
extraversion, respectively (Lynam & Widiger, 2001; Samuel & Widiger, 2004)
and the NEO PI-R is very limited in its coverage of the maladaptive variants of
these poles of the FFM (Haigler & Widiger, 2001). In addition, the NEO PI-R is
also limited in its coverage of all possible maladaptive variants of the other
poles. For example, absent from its assessment of neuroticism is affective
instability (Miller & Pilkonis, 2006) as well as the full range of helpless,
vulnerable self-destructiveness, and angry rage (Mullins-Sweatt et al., 2012).
The Structured Interview for the Five Factor Model (SIFFM; Trull et al.,
1998) was modeled after the NEO PI-R (Costa & McCrae, 1992) although it was
developed in part to increase the coverage of maladaptive variants of low
neuroticism, high extraversion, high openness, high agreeableness, and high
consciousness. However, it has not been used that often, given perhaps the
preference within general personality research for self-report inventories relative
to structured interviews. In addition, the SIFFM does not in any case provide as
much coverage of the maladaptive variants of the poles of the FFM as would be
needed for an adequate assessment of all the DSM-IV (now DSM-5) personality
disorders (Bagby et al., 2005; Trull, Widiger, & Burr, 2001).
De Clercq, De Fruyt, Van Leeuwen, and Mervielde (2006) modeled the
development of the Dimensional Personality Symptom Item Pool (DIPSI) after
the SIFFM. They constructed items to assess maladaptive variants of the facet
scales within the Hierarchical Personality Inventory for Children (HiPIC;
Mervielde & De Fruyt, 2002). The HiPIC provides an assessment of FFM traits
within children and adolescents; however, it is confined largely to the normal
range, comparable to the NEO PI-R. The DIPSI provides an assessment of the
maladaptive variants of HiPIC items and scales. Together they provide an
integrative assessment of normal and abnormal personality functioning in
children and adolescents from the perspective of the FFM (De Clercq & De
Fruyt, 2003, 2012; see also the chapter by De Fruyt, De Clercq, and De Bolle).
The original version of the DIPSI did not include an assessment of maladaptive
variants of high HiPIC Imagination. However, a revised version of the DIPSI
now includes maladaptive variants of imagination (De Fruyt & De Clercq,
2014).
An additional measure of maladaptive variants of the FFM is provided by the
Personality Inventory for DSM-5 (PID-5; Krueger, Derringer, Markon, Watson,
& Skodol, 2012). The PID-5 is a self-report measure of the dimensional trait
model of DSM-5, which in turn is said to be coordinated with the FFM (see
earlier section). However, a potential limitation of the PID-5 is that it is confined
largely to just five of the 10 poles of the FFM. In this regard the PID-5 is well
aligned with the NEO PI-R. Both the PID-5 and the NEO PI-R are confined
largely to maladaptive variants of high neuroticism, low extraversion, low
agreeableness, and low conscientiousness, with little or no representation of
maladaptive variants of low neuroticism, high extraversion, high agreeableness,
or high conscientiousness. However, unlike the NEO PI-R, the PID-5 assesses
for maladaptive high openness, rather than maladaptive low openness,
contributing to the weaker convergence of PID-5 Psychoticism with NEO PI-R
Openness (Gore & Widiger, 2013).
As noted earlier, the PID-5 does include some limited coverage of the
bipolarity. For example, there is one trait scale for maladaptively high
conscientiousness (i.e., Rigid Perfectionism). It also includes Restricted
Affectivity as a maladaptive variant of low neuroticism (Krueger et al., 2012),
but this scale is perhaps more accurately placed within introversion rather than
low neuroticism (Watson, Stasik, Ro, & Clark, 2013; Widiger, Costa, &
McCrae, 2013). Nevertheless, beyond these two scales there is little
representation of maladaptively low neuroticism, extraversion, low openness,
agreeableness, or conscientiousness, which limits its ability to cover significant
traits of some personality disorders, such as the obsessive–compulsive (Crego,
Samuel, et al., 2015) and psychopathy (Crego & Widiger, 2014). Because the
model does not include maladaptively low neuroticism, DSM-5 is unable to
assess psychopathic fearlessness and glib charm directly (Lynam & Widiger,
2007). However, in DSM-5 it is suggested to key the scale for anxiousness
negatively to assess for fearlessness (American Psychiatric Association, 2013, p.
765). Anxiousness will indeed correlate negatively with fearlessness, but a scale
for the assessment of anxiousness might not actually be able to assess effectively
for or distinguish between levels of fearlessness (Crego & Widiger, 2014).
Simms et al. (2011) developed the Computerized Adaptive Test-Personality
Disorder (CAT-PD; see also the chapter by Simms, Williams, and Simms). They
initially worked from the set of five domains identified by Widiger and
Simonsen’s (2005) integration of alternative dimensional trait models, including
the FFM: emotional dysregulation versus stability, extroversion versus
introversion, antagonism versus compliance, constraint versus impulsivity, and
unconventionality versus closedness to experience (see Simms et al., 2011, for a
description of the rationale for and construction of the CAT-PD). They
established an initial list of 53 candidate traits organized into five corresponding
domains. Simms et al. (2011) then developed draft items, working from 2,413
items within the International Personality Item Pool (IPIP), many of which were
written to assess domains and facets of the FFM (Goldberg et al., 2006).
Through factor, item response theory, and content analyses, the item pool was
eventually reduced for a “static” version (CAT-PD-SF; Wright & Simms, 2014)
to 212 items assessing 33 traits organized into five domains (i.e., negative
emotionality, detachment, antagonism, disconstraint, and psychoticism). The
final version of the instrument includes 33 scales, such as Manipulativeness
(from antagonism), Romantic Disinterest (from introversion), Exhibitionism
(from extraversion), and Perfectionism (from conscientiousness). Initial
validation suggests that its scales align well with the FFM, albeit not surprisingly
the convergence of the schizotypal thinking scales with FFM openness was
relatively weaker (Wright & Simms, 2014).
The scales of the CAT-PD are very similar to those of the PID-5. In fact, all
but three of the PID-5 scales are included in the CAT-PD. The CAT-PD has
more coverage, in that it includes 33 scales, relative to the 25 of the PID-5,
although the CAT-PD does not appear to have scales comparable to the PID-5
Attention-Seeking, Perseveration, or Distractability. The PID-5, in turn, does not
appear to have scales comparable to the CAT-PD Cognitive Problems,
Domineering, Exhibitionism, Fantasy Proness, Health Anxiety, Rudeness, Self-
Harm, Norm-Violation, or Workaholism. Some of this reflects not only the
greater number of scales within the CAT-PD, but also its inclusion of
maladaptive variants of extraversion (e.g., Exhibitionism) and conscientiousness
(e.g., Workaholism).
There is also a Clinician Rating Form for an assessment of the DSM-5
dimensional trait model on the basis of an unstructured clinical interview
(American Psychiatric Association, 2010). This measure is no longer posted on
the DSM-5 website, however, and so may not be readily accessed. There are
other abbreviated measures of the FFM, such as the Five Factor Model Rating
Form (FFMRF; Mullins-Sweatt, Jamerson, Samuel, Olson, & Widiger, 2006).
The FFMRF is a one-page rating form that includes an assessment of all the 30
NEO PI-R facets of the FFM. The FFMRF has been used in a number of studies
as a clinician rating form and as a self-report inventory (Samuel, Mullins-Sweatt,
& Widiger, 2013).
A limitation of the FFMRF is that it does not provide adequate coverage of the
maladaptive variants of the FFM. Rojas and Widiger (2014) therefore developed
the Five Factor Form (FFF). The FFF is again a one-page rating form that can be
used by clinicians to describe their patients or as a self-report. It includes all 30
NEO PI-R FFM facets and, unique to this measure, maladaptive variants for all
60 poles of all 30 facets. For example, for the facet of modesty (from
agreeableness), a score of 5 indicates “self-effacing, self-denigrating,” 4 is
“humble, modest, unassuming,” 3 is neutral, 2 is “confident, self-assured,” and 1
is “boastful, vain, pretentious, arrogant.” For the facet of achievement-striving
(from conscientiousness), a score of 5 indicates “workaholic, acclaim-seeking,”
4 is “purposeful, diligent, ambitious,” 3 is neutral, 2 is “carefree, content,” and 1
is “aimless, shiftless, desultory.” The FFF is comparable in structure and format
to the Sliderbar Inventory (Pettersson et al., 2014), which also includes
maladaptive variants at both poles of each item. A few of the FFF items are in
fact aligned well with respective Sliderbar items. For example, one
“aggressive/submissive” Sliderbar Inventory item contrasts “I get mad easily and
often get into fights” (maladaptive) and “I stand up for myself if someone has
done me wrong” (adaptive) at one pole, with “I rarely lose my temper”
(adaptive) and “I am a meek person” (maladaptive) at the other pole. A
comparable FFF item contrasts “combative, aggressive” and “critical, contrary”
at one pole with “cooperative, obedient, and deferential” and “yielding,
submissive, meek” at the other pole. However, a distinction between the two
measures is that the Sliderbar Inventory was constructed to assess opposite
components of the DSM-IV-TR personality disorders, whereas the FFF was
constructed to assess the opposite poles of the FFM. Rojas and Widiger (2014)
provided empirical support for the validity of the FFF as a measure of the FFM.
There are also eight self-report inventories constructed to assess the DSM-IV-
R (now DSM-5) personality disorders from the perspective of the FFM,
including psychopathy (Lynam et al., 2011), schizotypal personality disorder
(Edmundson et al., 2011), and histrionic (Tomiatti, Gore, Lynam, Miller, &
Widiger, 2012). A special section of the Journal of Personality Assessment
(Widiger, Lynam, Miller, & Oltmanns, 2012) was devoted to the initial
validation studies for five additional measures: the Five Factor Borderline
Inventory (FFBI; Mullins-Sweatt et al., 2012), the Five Factor Avoidant
Assessment (FFAvA; Lynam et al., 2012), the Five Factor Dependency
Inventory (FFDI; Gore, Presnall, Lynam, Miller, & Widiger, 2012), the Five
Factor Narcissism Inventory (FFNI; Glover, Miller, Lynam, Crego, & Widiger,
2012), and the Five Factor Obsessive Compulsive Inventory (FFOCI; Samuel et
al., 2012). Each of these instruments was constructed by first identifying which
facets of the FFM (as provided within the NEO PI-R) appear to be most relevant
for each respective personality disorder. The source for this information included
researchers’ descriptions of each respective personality disorder in terms of the
FFM (i.e., Lynam & Widiger, 2001), clinicians’ descriptions of each personality
disorder (i.e., Samuel & Widiger, 2004), and FFM-personality disorder research
(e.g., Samuel & Widiger, 2008b). Scales were then constructed to assess the
maladaptive variants of each facet that were specific to each personality disorder
(e.g., Perfectionism, Workaholism, Punctiliousness, and Doggedness as
maladaptive variants of conscientiousness for the FFOCI). The scales were
subsequently validated by demonstrating convergence with both their respective
parent FFM facet scales and alternative measures of the respective personality
disorder. Finally, each of the measures was shown to have incremental validity
over alternative measures of these personality disorders. Additional validation
studies have been published concerning the EPA (Miller et al., 2011; Wilson,
Miller, Zeichner, Lynam, & Widiger, 2011), the FFNI (Miller, Few et al., 2013;
Miller, Gentile, & Campbell, 2013; Miller et al., 2015), the FFOCI (Crego,
Samuel, & Widiger, 2015), the FFDI (Gore & Widiger, 2015), and the FFBI
(DeShong, Lengel, Sauer-Zavala, O’Meara, & Mullins-Sweatt, 2015). It is
important to note that factor analysis of Five Factor Model Personality Disorder
(FFMPD) scales will not necessarily yield five factors. Each instrument has a
very disproportionate representation of the FFM domains. However, each scale
is appropriately described as an FFM-based measure because each was
conceptually derived from an FFM domain and there has been consistently
strong support for the convergent and discriminant validity of the alignment of
each scale with its respective FFM domain.
The approach taken in the construction of these scales, to disambiguate DSM-
IV-TR personality disorders in terms of FFM facets, helps to ensure that all of
the maladaptive personality traits included within a respective DSM-IV-TR
personality disorder are adequately covered. It is evident that there is still
considerable interest in these personality syndromes (Mullins-Sweatt, Bernstein,
& Widiger, 2012; Shedler et al., 2010). These FFM personality disorder scales
therefore provide a bridge, or a means of translation, between the DSM-IV-TR
(now DSM-5) and the FFM.
If future research with these FFM personality disorder scales is confined to
just their total scores, these scales will recreate many of the problems that are
currently present for the existing categories (e.g., heterogeneity of membership
and diagnostic overlap), but a strength of these measures relative to many other
DSM-IV self-report inventories is that these instruments can also be broken
down into their subscales, thereby dismantling the heterogeneous syndromes into
more distinctive component parts. It is evident, for example, that clinicians,
when treating a personality disorder, do not address the entire personality
structure with each intervention (Paris, 2006). They focus instead on underlying
components, such as the dysregulated anger, the fragility, or the oppositionality
of persons diagnosed with borderline personality disorder. These components are
assessed independently and specifically by the scales of the FFBI (Mullins-
Sweatt et al., 2012), providing considerably greater utility in clinical practice
than the more global measures of borderline personality disorder (Mullins-
Sweatt & Lengel, 2012).
The coverage provided by the FFMPD scales is substantial, with 99 total
scales. Therefore lack of adequate coverage is unlikely to be an issue for the
FFMPD scales. However, there are clearly far too many scales to include within
any one particular study. To facilitate usage, abbreviated versions have been
validated for the EPA (Lynam et al., 2013), the FFNI (Sherman et al., 2015), and
the FFBI (DeShong, Mullins-Sweatt, Miller, Widiger, & Lynam, 2015), and
others are being developed as well.
The FFMPD scales were constructed in part to allow a respective personality
disorder to be assessed from the perspective of the FFM (Lynam, 2012).
However, they can also be understood as “measures to assess maladaptive
variants of the five factor model” (Widiger et al., 2012, p. 450). It is anticipated
that researchers and clinicians will want to select subsets of FFMPD scales for
their particular interest. For example, a clinician or researcher might be
interested in assessing only for maladaptive variants of agreeableness, and would
thereby confine the test administration to the FFDI Gullibility, Selflessness,
Subservience, and Self-Effacing scales. Alternatively, the clinician or researcher
might wish to consider only maladaptive variants of extraversion, including, for
instance, the EPA Dominance and Thrill-Seeking scales (Lynam et al., 2011),
the FFNI Exhibitionism and Authoritativeness scales (Glover et al., 2012),
and/or the FFHI Attention-Seeking, Social Butterfly, and Flirtatiousness scales
(Tomiatti et al., 2012).
Table 20.3 provides scales from the PID-5 (Krueger et al., 2012), CAT-PD
(Simms et al., 2011), and FFMPD (Widiger et al., 2012) for each of the 10 poles
of the five domains of the FFM. Note that the PID-5 FFM locations are based on
the original placements provided by Krueger et al. (2011, 2012); some locations
may have since changed (American Psychiatric Association, 2013). In addition,
not all of the FFMPD scales are included. Crego and Widiger (2015b) report the
results of joint factor analyses of the PID-5, CAT-PD, and 36 of the FFMPD
scales, addressing their convergent and discriminant validity.
An advantage of the CAT-PD, PID-5, and FFMPD scales is their conceptual
and empirical coordination with the FFM. The PID-5 and FFMPD scales also
include algorithms to assess a respective personality disorder, with a
considerable body of research supporting these algorithms (Krueger & Markon,
2014; Widiger et al., 2012). These scoring algorithms allow researchers and
clinicians to relate their findings for a respective personality disorder to the
FFM. Considering personality disorders from the perspective of the FFM is
useful in the development of a more integrative understanding of normal and
abnormal personality (Clark, 2007; Widiger & Trull, 2007). To the extent that
disorders of personality are understood as maladaptive variants of FFM
personality structure, it is possible to bring to these personality disorders the
extensive construct validity research concerning the genetics, childhood
antecedents, course, universality, and positive and negative life outcomes
identified for the FFM (Allik, 2005; Allik, Realo, & McCrae, 2013; Caspi et al.,
2005; Mervielde et al., 2005; Roberts & DelVecchio, 2000; Yamagata et al.,
2006).
Conclusions
The FFM of personality disorder provides a reasonably comprehensive
integration of normal and abnormal personality within a common hierarchical
structure. Advantages of the FFM of personality disorder include the provision
of precise, individualized descriptions of the personality structure of each
patient, the inclusion of homogeneous trait constructs that will have more
specific treatment implications, and the inclusion of normal, adaptive personality
traits that will provide a richer, fuller, and more appreciative description of each
patient (Widiger et al., 2012). The FFM of personality disorder resolves the
many fundamental limitations of the categorical model (e.g., heterogeneity
within diagnoses, inadequate coverage, lack of consistent diagnostic thresholds,
and excessive diagnostic cooccurrence), and brings to the nomenclature a wealth
of knowledge concerning the origins, childhood antecedents, stability, and
universality of the dispositions that underlie personality disorder. It is indeed
apparent that DSM-5 is becoming much closer to the FFM through the inclusion
of a five-domain dimensional model that aligns closely with the five domains of
the FFM, and through an emphasis on FFM traits for the diagnosis of each
respective personality disorder type being retained and/or deleted.
Table 20.3 Scales from Three Measures of Maladaptive Variants of the Five Factor
Model
Maladaptive Trait Measure
Five Factor Model Pole of PID-51 CAT-PD FFMPD2
Domain Domain
Neuroticism High Anxiousness Affective Affective
Lability Dysregulation
Emotional Lability Anger Dysregulated
Anger
Hostility Anxiousness Helplessness
Perseveration Cognitive Need for
Problems Admiration
Separation Insecurity Depressiveness Pessimism
Submissiveness Health Anxiety Self Disturbance
Mistrust Separation
Insecurity
Relationship Shamefulness
Insecurity
Self-harm Social
Discomfort
Submissiveness Urgency
Low Restricted Affectivity Indifference
Invulnerability
Unconcern
Extraversion High Exhibitionism Attention-
Seeking
Dominance
Intimacy-Seeking
Exhibitionism
Thrill-Seeking
Low Anhedonia Anhedonia Detached
Coldness
Depressivity Emotional Joylessness
Detachment
Intimacy Avoidance Romantic Risk Aversion
Disinterest
Suspiciousness Social Social Anhedonia
Withdrawal
Withdrawal Social Isolation
Unassertive
Openness High Eccentricity Unusual Beliefs Aberrant Ideas
Cognitive–Perceptual Unusual Aberrant
Dysregulation Experiences Perceptions
Unusual Beliefs and Fantasy Fantasies
Experiences Proneness
Peculiarity Odd and
Eccentric
Low Dogmatic
Constricted
Agreeableness High Gullibility
Self-Effacing
Selflessness
Subservience
Timorous
Low Attention Seeking Callousness Arrogance
Callousness Domineering Callousness
Deceitfulness Grandiosity Entitlement
Grandiosity Hostile Exploitative
Aggression
Manipulativeness Manipulativeness Manipulative
Norm Violation Oppositional
Rudeness Suspiciousness
Conscientiousness High Rigid Perfectionism Perfectionism Acclaim-Seeking
Rigidity Doggedness
Workaholism Fastidiousness
Perfectionism
Punctiliousness
Ruminative
Deliberation
Workaholism
Low Distractibility Irresponsibility Disorderliness
Impulsivity Non-Planfulness Impersistence
Irresponsibility Non- Impressionistic
Perseverance Thinking
Risk Taking Risk Taking Ineptitude
Negligence
Rashness
Note: PID-5, Personality Disorder Interview for DSM-5 (Krueger et al., 2012); CAT-PD, Computerized
Adaptive Test of Personality Disorder (Simms et al., 2011; Wright & Simms, 2014); FFMPD, Five Factor
Model Personality Disorder scales (Lynam, 2012; Widiger et al., 2012); Cog-Per, Cognitive–Perceptual;
Exp, Experiences.
1 Location of some PID-5 scales have shifted with DSM-5 (American Psychiatric Association, 2013).
2 Not all of the FFMPD scales (Widiger et al., 2012) are included.
Acknowledgments
The first author of this paper is an author or co-author of instruments
considered in this chapter. However, the author receives no royalties or financial
benefits from any of these instruments.
References
Allik, J. (2005). Personality dimensions across cultures. Journal of Personality Disorders,
19, 212–232.
Allik, J., Realo, A., & McCrae, R. R. (2013). Universality of the five factor model of
Aluja, A., Cuevas, L., Garcia, L. F., & Garcia, O. (2007). Predictions of the MCMI-III
disorders from NEO PI-R domains and facets: Comparison between American and
Spanish samples. International Journal of Clinical and Health Psychology, 7, 307–321.
Aluja, A., Kuhlman, M., & Zuckerman, M. (2010). Development of the Zuckerman-
Kuhlman-Aluja Personality Questionnaire (ZKA-P): A factor/facet version of the
Zuckerman-Kuhlman Personality Questionnaire (ZKPQ). Journal of Personality
Assessment, 92(5), 416–431.
American Psychiatric Association. (2010). DSM-5 clinicians’ personality trait rating
form. Washington, DC: American Psychiatric Association.
American Psychiatric Association. (2012). Rationale for the proposed changes to the
personality disorders classification in DSM-5. Unpublished manuscript. Washington,
DC: American Psychiatric Association.
disorders (5th ed.). Washington, DC: American Psychiatric Association.
Asail, T., Sugimori, E., Bando, N., & Tanno, Y. (2011). The hierarchic structure in
schizotypy and the five factor model of personality. Psychiatry Research, 185, 78–83.
Ashton, M. C., & Lee, K. (2012). Oddity, schizotypy/dissociation, and personality.
Bagby, R. M., Bindseil, K. D., Rector, N. A., Schuller, D. R., Joffe, R. J., Young, L. T.,
… Cooke, R. G. (1997). The relationship between the five factor model of personality
and unipolar depression, bipolar disorder, and schizophrenia. Psychiatry Research, 70,
83–94.
Bagby, R. M., Costa, P. T., Widiger, T. A., Ryder, A. G., & Marshall, M. (2005). DSM-
IV personality disorders and the five factor model of personality: A multi-method
examination of domain- and facet-level predictions. European Journal of Personality,
19, 307–324.
Bagby, R. M., & Farvolden, P. (2004). The Personality Diagnostic Questionnaire-4
(PDQ-4). In M. J. Hilsenroth, D. L. Segal, & M. Hersen (Eds.), Comprehensive
handbook of psychological assessment, Volume 2. Personality assessment (pp. 122–
133). New York, NY: John Wiley.
Bagby, R. M., Gilchrist, E. J., Rector, N. A., Dickens, S. E., Joffe, R. T., Levitt, A., …
Kennedy, S. H. (2001). The stability and validity of the sociotropy and autonomy
personality dimensions as measured by the Revised Personal Style Inventory. Cognitive
Therapy and Research, 25, 765–779.
Bagby, R. M., Marshall, M., & Georgiades, S. (2005). Dimensional personality traits and
the prediction of DSM-IV personality disorder symptom counts in a nonclinical
sample. Journal of Personality Disorders, 19, 53–67.
Bagby, R. M., Watson, C., & Ryder, A. G. (2013). Depressive personality disorder and
the five factor model. In T. A. Widiger & P. T. Costa (Eds.), Personality disorders and
the five factor model of personality (3rd. ed., pp. 179–192). Washington, DC:
Ball, S. A., Tennen, H., Poling, J. C., Kranzler, H. R., & Rounsaville, B. J. (1997).
Personality, temperament, and character dimensions and the DSM-IV personality
disorders in substance abusers. Journal of Abnormal Psychology, 106, 545–553.
Bastiaansen, L., Rossi, G., & De Fruyt, F. (2013). Comparing five sets of Five Factor
Model personality disorder counts in a heterogeneous sample of psychiatric patients.
Bastiaansen, L., Rossi, G., Schotte, C., & De Fruyt, F. (2011). The structure of
personality disorders: Comparing the DSM-IV-TR Axis II classification with the five
factor model framework using structural equation modeling. Journal of Personality
Disorders, 25, 378–396.
Bienvenu, O. J., Nestadt, G., Samuels, J. F., Costa, P. T., Howard, W. T., & Eaton, W. W.
(2001). Phobic, panic, and major depressive disorders and the five factor model of
personality. Journal of Nervous and Mental Disease, 189, 154–161.
Blais, M. A. (1997). Clinician ratings of the five factor model of personality and the
DSM-IV personality disorders. Journal of Nervous and Mental Disease, 185, 388–394.
Blashfield, R. K., & Reynolds, S. M. (2012). An invisible college view of the DSM-5
personality disorder classification. Journal of Personality Disorders, 26, 821–829.
Block, J. (2008). The Q-sort in character appraisal: Encoding subjective impressions of
persons quantitatively. Washington, DC: American Psychological Association.
Bornstein, R. F., & Cecero, J. J. (2000). Deconstructing dependency in a five factor
world: A meta-analytic review. Journal of Personality Assessment, 74, 324–343.
Borroni, S., Somma, A., Andershed, H., Maffei, C., & Fossati, A. (2014). Psychopathy
dimensions, Big Five traits, and dispositional aggression in adolescence: Issues of
gender consistency. Personality and Individual Differences, 66, 199–203.
Boudreaux, M. J., Piedmont, R. L., Sherman, M. F., & Ozer, D. J. (2013). Identifying
personality-related problems in living: The multi-context problems checklist. Journal
of Personality Assessment, 95, 62–73.
Camisa, K. M., Bockbrader, M. A., Lysaker, P., Rae, L. L., Brenner, C. A., & O’Donnell,
B. F. (2005). Personality traits in schizophrenia and related personality disorders.
Psychiatry Research, 133, 23–33.
Campbell, W. K., & Miller, J. D. (2013). Narcissistic personality disorder and the five
factor model: Delineating narcissistic personality disorder, grandiose narcissism, and
vulnerable narcissism. In T. A. Widiger & P. T. Costa (Eds.), Personality disorders
and the five factor model (3rd ed., pp. 133–146). Washington, DC: American
Carter, N. T., Dalal, D. K., Boyce, A. S., O’Connell, M. S., Kung, M. C., & Delgado, K.
M. (2014). Uncovering curvilinear relationships between conscientiousness and job
performance: How theoretically appropriate measurement makes an empirical
difference. Journal of Applied Psychology, 99, 564–586.
Carter, N. T., Guan, L., Maples, J. L., Williamson, R. L., & Miller, J. D. (2015). The
downsides of extreme conscientiousness for psychological well-being: The role of
obsessive compulsive tendencies. Journal of Personality.
Caspi, A., Houts, R. M., Belsky, D. W., Goldman-Mellor, S. J., Harrington, H., Israel, S.,
… Moffitt, T. E. (2014). The p factor one general psychopathology factor in the
Chai, H., Xu, S., Zhu, J., Chen, W., Xu, Y., He, W., & Wang, W. (2012). Further
evidence for the fifth higher trait of personality pathology: A correlation study using
normal and disordered personality measures. Psychiatry Research, 200, 444–449.
Clark, L. A. (1993). Manual for the Schedule for Nonadaptive and Adaptive Personality
(SNAP). Minneapolis, MN: University of Minnesota Press.
Clark, L. A., & Krueger, R. F. (2010, February 10). Rationale for a six-domain trait
dimensional diagnostic system for personality disorder. Retrieved from
http://www.dsm5.org/ProposedRevisions/Pages/RationaleforaSix-
DomainTraitDimensionalDiagnosticSystemforPersonalityDisorder.aspx.
Clark, L. A., & Livesley, W. J. (2002). Two approaches to identifying the dimensions of
personality disorder: Convergence on the five factor model. In P. T. Costa & T. A.
Widiger (Eds.), Personality disorders and the five factor model of personality (2nd ed.,
pp. 161–176). Washington, DC: American Psychological Association.
Clark, L. A., Livesley, W. J., Schroeder, M. L., & Irish, S. L. (1996). Convergence of two
systems for assessing personality disorder. Psychological Assessment, 8, 294–303.
Clark, L. A., Vorhies, L., & McEwen, J. L. (2002). Personality disorder symptomatology
from the five factor perspective. In P. T. Costa & T. A. Widiger (Eds.), Personality
disorders and the five factor model of personality (2nd ed., pp. 125–160). Washington,
Clarkin, J. F., & Huprich, S. K. (2011). Do DSM-5 personality disorder proposals meet
criteria for clinical utility? Journal of Personality Disorders, 25, 192–205.
Cloninger, C. R. (2000). A practical way to diagnose personality disorder: A proposal.
Journal of Personality Disorders, 14, 98–108.
Press.
Coker, L. A., Samuel, D. B., & Widiger, T. A. (2002). Maladaptive personality
functioning within the Big Five and the FFM. Journal of Personality Disorders, 16,
385–401.
Coolidge, F. L., & Merwin, M. M. (1992). Reliability and validity of the Coolidge Axis II
Inventory: A new inventory for the assessment of personality disorders. Journal of
Costa, P. T., & McCrae, R. R. (1980). Still stable after all these years: Personality as a
key to some issues in adulthood and old age. In P. B. Baltes & O. G. Brim (Eds.), Life
span development and behavior (Vol. 3). New York, NY: Academic Press.
Costa, P. T., & McCrae, R. R. (1990). Personality disorders and the five factor model of
personality. Journal of Personality Disorders, 4, 362–371.
Costa, P. T., & McCrae, R. R. (1992). Revised NEO Personality Inventory (NEO PI-R)
and NEO Five Factor Inventory (NEO-FFI) professional manual. Odessa, FL:
Crego, C., Gore, W. L., Rojas, S. L., & Widiger, T. A. (2015). The discriminant (and
convergent) validity of the Personality Inventory for DSM-5. Personality Disorders:
Crego, C., Samuel, D. B., & Widiger, T. A. (2015). The FFOCI and other measures and
models of OCPD. Assessment, 22, 135–151.
Crego, C., & Widiger, T. A. (2014). Psychopathy, DSM-5, and a caution. Personality
Disorders: Theory, Research, and Treatment, 5, 335–347.
Crego, C., & Widiger, T. A. (2015a). Cleckley psychopathy: Revisited. Journal of
Abnormal Psychology.
Crego, C., & Widiger, T. A. (2015b). Convergent and discriminant validity of alternative
measures of maladaptive personality traits. Psychological Assessment.
Crespi, B. (2014). An evolutionary framework for psychological maladaptations.
Psychological Inquiry, 25, 322–324.
Cyders, M. A., & Smith, G. T. (2008). Emotion-based dispositions to rash action: Positive
and negative urgency. Psychological Bulletin, 134, 807–828.
De Clercq, B., & De Fruyt, F. (2003). Personality disorder symptoms in adolescence: A
five factor model perspective. Journal of Personality Disorders, 17, 269–292.
De Clercq, B., & De Fruyt, F. (2012). A five factor model framework for understanding
childhood personality disorder antecedents. Journal of Personality, 80, 1533–1563.
perspective for DSM-V. Journal of Abnormal Psychology, 115(4), 639–657.
De Fruyt, F., De Clerq, B., De Bolle, M., Willie, B., Markon, K. E., & Krueger, R. F.
(2013). General and maladaptive traits in a five factor framework for DSM-5 in a
university student sample. Assessment, 20, 295–307.
De Fruyt, F., De Clercq, B. J., van de Wiele, L., & Van Heeringen, K. (2006). The
validity of Cloninger’s psychobiological model versus the five factor model to predict
DSM-IV personality disorders in a heterogeneous psychiatric sample: Domain, facet,
and residualized facet descriptions. Journal of Personality, 74, 479–510.
Hogrefe & Huber.
Decuyper, M., De Pauw, S., De Fruyt, F., De Bolle, M., & De Clercq, B. J. (2009). A
meta-analysis of psychopathy-, antisocial PD- and FFM associations. European
Del Giudice, M., & Del Giudice, M. (2014). An evolutionary life history framework for
psychopathology. Psychological Inquiry, 25, 261–300.
Derefinko, K., & Lynam, D. R. (2013). Psychopathy from the perspective of the five
factor model of personality (3rd. ed., pp. 103–118). Washington, DC: American
DeShong, H. L., Grant, D. M., & Mullins-Sweatt, S. N. (2015). Comparing models of
counterproductive workplace behaviors: The Five Factor Model and the Dark Triad.
DeShong, H. L., Lengel, G. J., Sauer-Zavala, S. E., O’Meara, M., & Mullins-Sweatt, S.
N. (2015). Construct validity of the Five Factor Borderline Inventory. Assessment, 22,
319–331.
DeShong, H. L., Mullins-Sweatt, S. N., Miller, J. D., Widiger, T. A., & Lynam, D. R.
(2015). Development of a short form of the Five Factor Borderline Inventory.
aspects of the big five. Journal of Personality and Social Psychology, 93(5), 880–896.
Dyce, J. A., & O’Connor, B. P. (1998). Personality disorders and the five factor model: A
test of the facet-level predictions. Journal of Personality Disorders, 12, 31–45.
Edmundson, M., & Kwapil, T. R. (2013). A five factor model perspective of schizotypal
personality disorder. In T. A. Widiger & P. T. Costa (Eds.), Personality disorders and
the five factor model (3rd ed., pp. 147–162). Washington, DC: American Psychological
Association.
Edmundson, M., Lynam, D. R., Miller, J. D., Gore, W. L., & Widiger, T. A. (2011). A
five factor measure of schizotypal personality traits. Assessment, 18, 321–334.
Few, L. R., Miller, J. D., & Lynam, D. R. (2013). An examination of the factor structure
of the Elemental Psychopathy Assessment. Personality Disorders: Theory, Research,
and Treatment, 4, 247–253.
Few, L. R., Miller, J. D., Morse, J. Q., Yaggi, K. E., Reynolds, S. K., & Pilkonis, P. A.
(2010). Examining the reliability and validity of clinician ratings on the Five Factor
Model score sheet. Assessment, 17, 440–453.
Flynn, F. J. (2005). Having an open mind: The impact of openness to experience on
interracial attitudes and impression formation. Journal of Personality and Social
Furnham, A. (2014). A bright side, facet analysis of histrionic personality disorder: The
relationship between the HDS colourful factor and the NEO-PI-R facets in a large adult
sample. The Journal of Social Psychology, 154, 527–536.
Furnham, A., & Crump, J. (2005). Personality traits, types, and disorders: An examination
of the relationship between three self-report measures. European Journal of
Furnham, A., Richards, S., Rangel, L., & Jones, D. N. (2014). Measuring malevolence:
Quantitative issues surrounding the Dark Triad of personality. Personality and
Gaughan, E. T., Miller, J. D., Pryor, L. R., & Lynam, D. R. (2009). Comparing two
alternative models of general personality in the assessment of psychopathy: A test of
the NEO PI-R and the MPQ. Journal of Personality, 77, 965–996.
Glover, N., Miller, J. D., Lynam, D. R., Crego, C., & Widiger, T. A. (2012). The Five
Factor Narcissism Inventory: A five factor measure of narcissistic personality traits.
Goldberg, L. R., Johnson, J. A., Eber, H. W., Hogan, R., Ashton, M. C., Cloninger, C., &
Gough, H. G. (2006). The international personality item pool and the future of public-
domain personality measures. Journal of Research in Personality, 40(1), 84–96.
Gore, W. L., & Pincus, A. L. (2013). Dependency and the five factor model. In T. A.
Widiger & P. T. Costa (Eds.), Personality disorders and the five factor model (3rd ed.,
Gore, W. L., Presnall, J., Lynam, D. R., Miller, J. D., & Widiger, T. A. (2012). A five
factor measure of dependent personality traits. Journal of Personality Assessment, 94,
488–499.
Gore, W. L., Tomiatti, M., & Widiger, T. A. (2011). The home for histrionism.
Personality and Mental Health, 5, 57–72.
Gore, W. L., & Widiger, T. A. (2013). The DSM-5 dimensional trait model and five
Gore, W. L., & Widiger, T. A. (2015). Assessment of dependency by the FFDI:
Comparisons to the PID-5 and maladaptive agreeableness. Personality and Mental
Health, 9, 258–276.
Griffin, S. A., & Samuel, D. B. (2014). A closer look at the lower-order structure of the
Personality Inventory for DSM-5: Comparison with the Five Factor Model. Personality
Gudonis, L. C., Miller, D. J., Miller, J. D., & Lynam, D. R. (2008). Conceptualizing
personality disorders from a general model of personality functioning: Antisocial
personality disorder and the five factor model. Personality and Mental Health, 2, 249–
264.
Gunderson, J. G. (2010). Commentary on “Personality traits and the classification of
mental disorders: Toward a more complete integration in DSM-5 and an empirical
model of psychopathology.” Personality Disorders: Theory, Research, and Treatment,
1, 119–122.
Assessment, 7(1), 104.
Harlan, E., & Clark, L. A. (1999). Short forms of the Schedule for Nonadaptive and
Adaptive Personality for self- and collateral ratings: Development, reliability, and
validity. Assessment, 6, 131–145.
Hicklin, J., & Widiger, T. A. (2005). Similarities and differences among antisocial and
psychopathic personality inventories from the perspective of general personality
functioning. European Journal of Personality, 19, 325–342.
Horowitz, L. M., Alden, L. E., Wiggins, J. S., & Pincus, A. L. (2000). IIP-64/IIP-32
professional manual. San Antonio, TX: The Psychological Corporation.
Huprich, S. K. (2000). Describing depressive personality analogues and dysthymics on
the NEO-Personality Inventory-Revised. Journal of Clinical Psychology, 56, 1521–
1534.
Huprich, S. K. (2003a). Evaluating facet-level predictions and construct validity of
depressive personality disorder. Journal of Personality Disorders, 17, 219–232.
Huprich, S. K. (2003b). Evaluating NEO personality inventory-revised personality
profiles in veterans with personality disorders. Journal of Personality Disorders, 17,
33–44.
Klein, M. H., Benjamin, L. S., Rosenfeld, R., Treece, C., Husted, J., & Greist, J. H.
(1993). The Wisconsin Personality Disorders Inventory: I. Development, reliability,
and validity. Journal of Personality Disorders, 7, 285–303.
Krueger, R. F. (2013). Promoting the vanguard. Personality Disorders: Theory, Research,
and Treatment, 4, 367.
Krueger, R. F., Derringer, J., Markon, K. F., Watson, D., & Skodol, A. E. (2012). Initial
Krueger R. F., Eaton, N. R., Derringer, J., Markon, K. E., Watson, D., & Skodol, A. E.
(2011). Personality in DSM-5: Helping delineate personality disorder content and
framing the meta-structure. Journal of Personality Assessment, 93, 325–331.
Krueger, R. F., & Markon, K. E. (2014). The role of the DSM-5 personality trait model in
moving toward a quantitative and empirically based approach to classifying personality
and psychopathology. Annual Review of Clinical Psychology, 10, 477–501.
Krueger, R. F., Skodol, A. E., Livesley, W. J., Shrout, P. E., & Huang, Y. (2008).
Synthesizing dimensional and categorical approaches to personality disorders: Refining
the research agenda for DSM-IV Axis II. In J. E. Helzer, H. C. Kraemer, R. F. Krueger,
H-U. Wittchen, P. J. Sirovatka, & D. A. Regier (Eds.), Dimensional approaches to
diagnostic classification. Refining the research agenda for DSM-V (pp. 85–100).
Washington, DC: American Psychiatric Association.
Kwapil, T. R., & Barrantes-Vidal, N. (2012). Schizotypal personality disorder: An
integrative review. In T. A. Widiger (Ed.), The Oxford handbook of personality
disorders (pp. 437–476). New York, NY: Oxford University Press.
Kwapil, T. R., Barrantes-Vidal, N., & Silvia, P. J. (2008). The dimensional structure of
the Wisconsin Schizotypy Scales: Factor identification and construct validity.
Schizophrenia Bulletin, 34, 444–457.
Le, H., Oh, I. S., Robbins, S. B., Ilies, R., Holland, E., & Westrick, P. (2011). Too much
of a good thing: Curvilinear relationships between personality traits and job
performance. Journal of Applied Psychology, 96, 113–133.
Lilienfeld, S. O., Watts, A. L., Francis Smith, S., Berg, J. M., & Latzman, R. D. (2015).
Psychopathy deconstructed and reconstructed: Identifying and assembling the
personality building blocks of Cleckley’s chimera. Journal of Personality, 83, 593–
610.
Lilienfeld, S. O., Watts, A. L., & Smith, S. F. (2012). The DSM revision as a social
psychological process: A commentary on Blashfield and Reynolds. Journal of
Lilienfeld, S. O., & Widows, M. R. (2005). Psychopathic personality inventory—Revised
professional manual. Lutz, FL: Psychological Assessment Resources, Inc.
Linde, J. A., Stringer, D., Simms, L. J., & Clark, L. A. (2013). The Schedule for
Nonadaptive and Adaptive Personality for Youth (SNAP-Y): A new measure for
assessing adolescent personality and personality pathology. Assessment, 20, 387–404.
Livesley, W. J. (2001). Conceptual and taxonomic issues. In W. J. Livesley (Ed.),
Handbook of personality disorders. Theory, research, and treatment (pp. 3–38). New
Livesley, W. J. (2012). Tradition versus empiricism in the current DSM-5 proposal for
revising the classification of personality disorders. Criminal Behavior and Mental
Health, 22, 81–90.
Livesley, W. J., & Jackson, D. (2009). Manual for the Dimensional Assessment of
Personality Pathology—Basic Questionnaire. Port Huron, MI: Sigma Press.
Livesley, W. J., Jang, K. L., & Vernon, P. A. (1998). Phenotypic and genetic structure of
traits delineating personality disorder. Archives of General Psychiatry, 55(10), 941–
948.
Loranger, A. W. (2001). OMNI personality inventories. Professional manual. Odessa,
Lowe, J. R., Edmundson, M., & Widiger, T. A. (2009). Assessment of dependency,
agreeableness, and their relationship. Psychological Assessment, 21, 543–553.
Luminet, O., Bagby, R. M., Wagner, H., Taylor, G. J., & Parker, J. D. A. (1999). The
relationship between alexithymia and the five factor model of personality: A facet level
analysis. Journal of Personality Assessment, 73, 345–358.
Lynam, D. R. (2012). Assessment of maladaptive variants of five factor model traits.
Lynam, D. R., Gaughan, E. T., Miller, J. D., Miller, D. J., Mullins-Sweatt, S., & Widiger,
T. A. (2011). Assessing the basic traits associated with psychopathy: Development and
validation of the Elemental Psychopathy Assessment. Psychological Assessment, 23,
108–124.
Lynam D. R., Loehr, A., Miller, J. D., & Widiger, T. A. (2012). A five factor measure of
avoidant personality: The FFAvA. Journal of Personality Assessment, 94, 466–474.
Lynam, D. R., & Miller, J. D. (2015). Psychopathy from a basic trait perspective: The
utility of a Five Factor Model approach. Journal of Personality, 83, 611–626.
Lynam, D. R., Miller, J. D., Miller, D. J., Bournovalova, M. A., & Lejuez, C. W. (2011).
Testing the relations between impulsivity-related traits, suicidality and non-suicidal
self-injury: A test of the incremental validity of the UPPS model. Personality
Lynam, D. R., Sherman, E. D., Samuel, D. B., Miller, J. D., Few, L. R., & Widiger, T. A.
(2013). Development of a short form of the Elemental Psychopathy Assessment.
Lynam, D. R., & Widiger, T. A. (2007). Using a general model of personality to identify
the basic elements of psychopathy. Journal of Personality Disorders, 21, 160–178.
Madsen, L., Parsons, S., & Grubin, D. (2006). The relationship between the five factor
model and DSM personality disorders in a sample of child molesters. Personality and
McBride, C., Zuroff, D. C., Bagby, R. M., & Bacchiochi, J. (2006). Depressive
Experiences Questionnaire: Does it measure maladaptive and adaptive forms of
dependency? Social Behavior and Personality, 34, 1–16.
McCrae, R. R., & Costa, P. T. (1983). Joint factors in self-reports and ratings:
Neuroticism, Extraversion, and Openness to Experience. Personality and Individual
McCrae, R. R., & Costa, P. T. (1989). The structure of interpersonal traits: Wiggins’
circumplex and the five factor model. Journal of Personality and Social Psychology,
56, 586–595.
McCrae, R. R., & Costa, P. T. (2003). Personality in adulthood. A five factor theory
McCrae, R. R., Costa, P. T., & Busch, C. M. (1986). Evaluating comprehensiveness in
personality systems: The California Q-Set and the five factor model. Journal of
McCrae, R. R., Lockenhoff, C. E., & Costa, P. T., Jr. (2005). A step toward DSM-V:
Cataloguing personality-related problems in living. European Journal of Personality,
19, 269–286.
personality and developmental psychopathology as childhood antecedents of
personality disorders. Journal of Personality Disorders, 19, 171–201.
Mervielde, I., & De Fruyt, F. (2002). Assessing children’s traits with the Hierarchical
Personality Inventory for Children. In B. De Raad & M. Perugini (Eds.), Big Five
assessment (pp. 129–146). Seattle, WA: Hogrefe and Huber.
Miller, G. F., & Tal, I. R. (2007). Schizotypy versus openness and intelligence as
predictors of creativity. Schizophrenia Research, 93(1), 317–324.
Miller, J. D. (2012). Five Factor model personality disorder prototypes: A review of their
development, validity, and comparison with alternative approaches. Journal of
Miller, J. D. (2013). Prototype matching and the five factor model: Capturing the DSM-
IV personality disorders. In T. A. Widiger & P. T. Costa (Eds.), Personality disorders
and the five factor model of personality (3rd ed., pp. 249–268). Washington, DC:
Miller, J. D., Bagby, R. M., Pilkonis, P. A., Reynolds, S. K., & Lynam, D. R. (2005). A
simplified technique for scoring the DSM-IV personality disorders with the five factor
model. Assessment, 12, 404–415.
Miller, J. D., & Campbell, W. K. (2008). Comparing clinical and social-personality
conceptualizations of narcissism. Journal of Personality, 76, 449–476.
Miller, J. D., Few, L. R., & Widiger, T. A. (2012). Assessment of personality disorders
and related traits: Bridging DSM-IV-TR and DSM-5. In T. A. Widiger (Ed.), Oxford
handbook of personality disorder (pp. 108–140). New York, NY: Oxford University
Press.
Miller, J. D., Few, L. R., Wilson, L., Gentile, B., Widiger, T. A., MacKillop, J., &
Campbell, W. K. (2013). The Five Factor Narcissism Inventory (FFNI): A test of the
convergent, discriminant, and incremental validity of FFNI scores in clinical and
community samples. Psychological Assessment, 25, 748–758.
Miller, J. D., Gaughan, E. T., Maples, J., Gentile, B., Lynam, D. R., & Widiger, T. A.
(2011). Examining the construct validity of the Elemental Psychopathy Assessment.
Miller, J. D., Gentile, B., & Campbell, W. K. (2013). A test of the construct validity of
the Five Factor Narcissism Inventory. Journal of Personality Assessment, 95, 377–387.
Miller, J. D., & Lynam, D. R. (2008). Dependent personality disorder: Comparing an
expert generated and empirically derived five factor model personality disorder count.
Miller, J. D., McCain, J. L., Few, L. R., Crego, C., Lynam, D. R., Widiger, T. A., &
Campbell, W. K. (2015). Thinking structurally: An examination of the Five Factor
Narcissism Inventory and its components. Journal of Personality Disorders, 24, 1–18.
Millon, T., Millon, C., Davis, R., & Grossman, S. (2009). MCMI-III manual (4th ed.).
Minneapolis, MN: Pearson.
Miller, J. D., & Pilkonis, P. A. (2006). Neuroticism and affective instability: The same or
different? American Journal of Psychiatry, 163, 839–845.
Mongrain, M. (1993). Dependency and self-criticism located within the five factor model
of personality. Personality and Individual Differences, 15, 455–462.
Morey, L. C. (1985). An empirical comparison of interpersonal and DSM-III personality
disorders: Their derivation and correlates. Journal of Personality Assessment, 49, 245–
251.
Mullins-Sweatt, S. N., Bernstein, D. P., & Widiger, T. A. (2012). Retention or deletion of
personality disorder diagnoses for DSM-5: An expert consensus approach. Journal of
Mullins-Sweatt, S. N., Edmundson, M., Sauer-Zavala, S., Lynam, D. R., Miller, J. D., &
Widiger, T. A. (2012). five factor measure of borderline personality traits. Journal of
Mullins-Sweatt, S. N., Glover, N., Derefinko, K. J., Miller, J. D., & Widiger, T. A.
(2010). The search for the successful psychopath. Journal of Research in Personality,
44, 554–558.
Mullins-Sweatt, S. N., Jamerson, J. E., Samuel, S. B., Olson, D. R., & Widiger, T. A.
(2006). Psychometric properties of an abbreviated instrument for the assessment of the
five factor model. Assessment, 13, 119–137.
Mullins-Sweatt, S. N., & Lengel, G. J. (2012). Clinical utility of the five factor model of
Mullins-Sweatt, S. N., & Widiger, T. A. (2007). The Shedler-Westen Assessment
Procedure from the perspective of general personality structure. Journal of Abnormal
Mullins-Sweatt, S. N., & Widiger, T. A. (2008). Comparative validity of the Shedler-
Westen Assessment Procedure. Psychological Assessment, 20, 183–188.
Nederström, M., & Furnham, A. (2012). The relationship between the FFM and
personality disorders in a personnel selection sample. Scandinavian Journal of
Nestadt, G., Costa, P. T., Hsu, F-C., Samuels, J., Bienvenu, O. J., & Eaton, W. W. (2008).
The relationship between the five factor model and latent Diagnostic and Statistical
Manual of Mental Disorders, fourth edition personality disorder dimensions.
Comprehensive Psychiatry, 49, 98–105.
Nettle, D. (2006a). The evolution of personality variation in humans and other animals.
Nettle, D. (2006b). Schizotypy and mental health amongst poets, visual artists, and
mathematicians. Journal of Research in Personality, 40, 876–890.
Nettle, D., & Clegg, H. (2006). Schizotypy, creativity and mating success in humans.
Proceedings of the Royal Society B: Biological Sciences, 273(1586), 611–615.
O’Boyle, E. H., Forsyth, D. R., Banks, G. C., Story, P. A., & White, C. D. (2015). A
meta-analytic test of redundancy and relative importance of the Dark Triad and Five
Factor Model of personality. Journal of Personality, 83, 644–664.
O’Connor, B. P. (2002). A quantitative review of the comprehensiveness of the five
O’Connor, B. P., & Dyce, J. A. (1998). A test of models of personality disorder
configuration. Journal of Abnormal Psychology, 107, 3–16.
Oluf, G. A., & Furnham, A. (2015). The relationship between bright- and dark-side
personality traits. Personality and Individual Differences, 87, 206–211.
Ozer, D. J., & Reise, S. P. (1994). Personality Assessment. Annual Review of Psychology,
45, 357–388.
Paris, J. (2006). Personality disorders: psychiatry’s stepchildren come of age. Invited
address at the 159th Annual Meeting of the American Psychiatric Association, Toronto,
Canada.
Patrick, C. J., Fowles, D. C., & Krueger, R. F. (2009). Triarchic conceptualization of
psychopathy: Developmental origins of disinhibition, boldness, and meanness.
Development and Psychopathology, 21, 913–938.
Paulhus, D. L., Neumann, C. S., & Hare, R. D. (2015). Manual for the Self-report
Psychopathy scale. Toronto, Canada: Multi-Health Systems.
Paulhus, D. L., & Williams, K. M. (2002). The dark triad of personality: Narcissism,
machiavellianism, and psychopathy. Journal of Research in Personality, 36, 556–563.
traits? Psychological Assessment, 26, 433–446.
Pettersson, E., Turkheimer, E., Horn, E. E., & Menatti, A. R. (2012). The general factor
of personality and evaluation. European Journal of Personality, 26, 292–302.
Piedmont, R. L., Sherman, M. F., & Sherman, N. C. (2012). Maladaptively high and low
openness: The case for experiential permeability. Journal of Personality, 80, 1641–
1668.
(2009). Using the five factor model to identify a new personality disorder domain: The
case for experiential permeability. Journal of Personality and Social Psychology, 96,
1245–1258.
Pierce, J. R., & Aguinis, H. (2013). The too-much-of-a-good-thing effect in management.
Journal of Management, 39, 313–338.
Pincus, A. L., & Gurtman, M. B. (1995). The three faces of interpersonal dependency:
Structural analysis of self-report dependency measures. Journal of Personality and
Pincus, A. L., & Hopwood, C. J. (2012). A contemporary interpersonal model of
personality pathology and personality disorder. In T. A. Widiger (Ed.), The Oxford
Press.
Poy, R., Seggara, P., Esteller, A., Lopez, R., & Molto, J. (2014). FFM description of the
triarchic conceptualization of psychopathy in men and women. Psychological
Quirk, S. W., Christiansen, N. D., Wagner, S. H., & McNulty, J. L. (2003). On the
usefulness of measures of normal personality for clinical assessment: Evidence of the
incremental validity of the Revised NEO Personality Inventory. Psychological
Ramanaiah, N. V., Rielage, J. K., & Cheng, Y. (2002). Cloninger’s Temperament and
Character Inventory and the NEO Five Factor Inventory. Psychological Reports, 90,
1059–1063.
Reynolds, S. K., & Clark, L. A. (2001). Predicting dimensions of personality disorder
from domains and facets of the five factor model. Journal of Personality, 69, 199–222.
Rogers, C. R. (1961). On becoming a person: A therapist’s view of psychotherapy.
Boston, MA: Houghton-Mifflin.
Rokeach, M. (1960). The open and closed mind. New York, NY: Basic Books.
Rojas, S. L., & Widiger, T. A. (2014). The convergent and discriminant validity of the
Five Factor Form. Assessment, 21, 143–157.
Rolland, J-P., & de Fruyt, F. (2003). The validity of FFM personality dimensions and
maladaptive traits to predict negative affect at work. European Journal of Personality,
17, 101–121.
Ross, S. R., Lutz, C. J., & Bailey, S. E. (2002). Positive and negative symptoms of
schizotypy and the five factor model: A domain and facet level analysis. Journal of
Rossier, J., & Rigozzi, C. (2008). Personality disorders and the five factor model among
French speakers in Africa and Europe. Canadian Journal of Psychiatry, 53, 534–544.
Ruiz, M. A., Pincus, A. L., & Schinka, J. A. (2008). Externalizing pathology and the five
factor model: A meta-analysis of personality traits associated with antisocial
personality disorder, substance use disorder, and their co-occurrence. Journal of
Samuel, D. B. (2011). Assessing personality in the DSM-5: The utility of bipolar
constructs. Journal of Personality Assessment, 93, 390–397.
Samuel, D. B. (2013). Assessing the five factor model of personality disorder. In T. A.
Widiger & P. T. Costa (Eds.), Personality disorders and the five factor model of
personality (3rd ed., pp. 221–232). Washington, DC: American Psychological
Association.
Samuel, D. B., Carroll, K. M., Rounsaville, B. J., & Ball, S. A. (2013). Personality
disorders as maladaptive, extreme variants of normal personality: Borderline
personality disorder and neuroticism in a substance using sample. Journal of
Samuel, D. B., & Gore, W. L. (2012). Maladaptive variants of conscientiousness and
agreeableness. Journal of Personality, 80, 1669–1696.
Samuel, D. B., Mullins-Sweatt, S. N., & Widiger, T. A. (2013). An investigation of the
factor structure and convergent and discriminant validity of the Five Factor Model
Rating Form. Assessment, 20, 24–35.
Samuel, D. B., Riddell, A. D. B., Lynam, D. R., Miller, J. D., & Widiger, T. A. (2012). A
five factor measure of obsessive-compulsive personality traits. Journal of Personality
Samuel, D. B., Simms, L. J., Clark, L. A., Livesley, W. J., & Widiger, T. A. (2010). An
item response theory integration of normal and abnormal personality scales.
Personality Disorders: Theory, Research, and Treatment, 1, 5–21.
Samuel, D. B., & Widiger, T. A. (2004). Clinicians’ descriptions of prototypic personality
disorders. Journal of Personality Disorders, 18, 286–308.
Samuel, D. B., & Widiger, T. A. (2008a). Convergence of narcissism measures from the
perspective of general personality functioning. Assessment, 15, 364–374.
Samuel, D. B., & Widiger, T. A. (2008b). A meta-analytic review of the relationships
analysis. Clinical Psychology Review, 28, 1326–1342.
Samuel, D. B., & Widiger, T. A. (2010). A comparison of obsessive-compulsive
personality disorder scales. Journal of Personality Assessment, 92, 232–240.
Samuel, D. B., & Widiger, T. A. (2011). Conscientiousness and obsessive-compulsive
personality disorder. Personality Disorders: Theory, Research, and Treatment, 2, 161–
174.
1085.
Schneider, K. (1959). Clinical psychopathology. New York, NY: Grune & Stratton.
Schroeder, M. L., Wormworth, J. A., & Livesley, W. J. (1992). Dimensions of personality
disorder and their relationship to the Big Five dimensions of personality. Psychological
Seibert, L. A., Miller, J. D., Few, L. R., Zeichner, A., & Lynam, D. R. (2011). An
examination of the structure of self-report psychopathy measures and their relations
with general traits and externalizing behaviors. Personality Disorders: Theory,
Research, and Treatment, 2, 193–208.
Sharma, L., Markon, K. E., & Clark, L. A. (2013). Toward a theory of distinct types of
“impulsive” behaviors: A meta-analysis of self-report and behavioral measures.
Psychology Bulletin, 140, 374–408.
Shedler, J., Beck, A., Fonagy, P., Gabbard, G. O., Gunderson, J. G., Kernberg, O., …
Westen, D. (2010). Personality disorders in DSM-5. American Journal of Psychiatry,
167, 1027–1028.
Shedler, J., & Westen, D. (2004a). Dimensions of personality pathology: An alternative to
the Five Factor Model. American Journal of Psychiatry, 161, 1743–1754.
Shedler, J., & Westen, D. (2004b). Refining personality disorder diagnosis: Integrating
science and practice. American Journal of Psychiatry, 161(8), 1350–1365.
Sherman, E. D., Miller, J. D., Few, L. R., Campbell, W. K., Widiger, T. A., Crego, C., &
Lynam, D. R. (2015). Development of a Short Form of the Five Factor Narcissism
Inventory: The FFNI-SF. Psychological Assessment, 27, 1110–1116.
Sim, J. P., & Romney, D. M. (1990). The relationship between a circumplex model of
interpersonal behaviors and personality disorders. Journal of Personality Disorders, 4,
329–341.
Simms, L. J., & Clark, L. A. (2005). Validation of a computerized adaptive version of the
Schedule for Nonadaptive and Adaptive Personality (SNAP). Psychological
Simms, L. J., & Clark, L. A. (2006). The Schedule for Nonadaptive and Adaptive
Personality (SNAP): A dimensional measure of traits relevant to personality and
personality pathology. In S. Strack (Ed.), Differentiating normal and abnormal
personality (2nd ed., pp. 431–450). New York, NY: Springer.
CAT-PD project. Journal of Personality Assessment, 93, 380–389.
Smith, G. T., Fischer, S., Cyders, M. A., Annus, A. M., Spillane, N. S., & McCarthy, D.
M. (2007). On the validity and utility of discriminating among impulsivity-like traits.
Smith, S. W., Hilsenroth, M. J., & Bornstein, R. F. (2009). Convergent validity of the
SWAP-200 dependency scales. Journal of Nervous and Mental Disease, 197, 613–618.
Soldz, S., Budman, S., Demby, A., & Merry, J. (1993). Representation of personality
disorders in circumplex and five factor space: Explorations with a clinical sample.
Somwaru, D. P., & Ben-Porath, Y. S. (1995). Development and reliability of MMPI-2
based personality disorder scales. Paper presented at the 30th Annual Workshop and
Symposium on Recent Developments in Use of the MMPI-2 & MMPI-A. St.
Petersburg Beach, Florida.
Sprock, J. (2002). A comparative study of the dimensions and facets of the five factor
model in the diagnosis of cases of personality disorder. Journal of Personality
Disorders, 16, 402–423.
Stepp, S. D., Trull, T. J., Burr, R. M., Wolftenstein, M., & Vieth, A. Z. (2005).
Incremental validity of the Structured Interview for the Five Factor Model of
personality (SIFFM). European Journal of Personality, 19, 343–357.
Stepp, S. D., Yu, L., Miller, J. D., Hallquist, M. N., Trull, T. J., & Pilkonis, P. A. (2012).
Integrating competing dimensional models of personality: Linking the SNAP, TCI, and
NEO using item response theory. Personality Disorders: Theory, Research, and
Treatment, 3, 107–126.
Stoeber, J., Otto, K., & Dalbert, C. (2009). Perfectionism and the Big Five:
Conscientiousness predicts longitudinal increases in self-oriented perfectionism.
Taylor, G. J., & Bagby, R. M. (2013). Alexithymia and the five factor model of
factor model of personality (pp. 193–208). Washington, DC: American Psychological
Association.
Tellegen, A. (1993). Folk concepts and psychological concepts of personality and
personality disorder. Psychological Inquiry, 4, 122–130.
Development of the Multidimensional Personality Questionnaire. Unpublished
manuscript, Minneapolis, Minnesota.
Thalbourne, M. A. (2000). Transliminality and creativity. The Journal of Creative
Behavior, 34, 193–202.
Thomas, K. M., Yalch, M. M., Krueger, R. F., Wright, A. G. C., Markon, K. E., &
Hopwood, C. J. (2012). The convergent structure of DSM-5 personality trait facets and
Five Factor Model trait domains. Assessment, 12, 308–311.
Tomiatti, M., Gore, W. L., Lynam, D. R., Miller, J. D., & Widiger, T. A. (2012). A five
factor measure of histrionic personality traits. In A. M. Columbus (Ed.), Advances in
psychology research (Vol. 87, pp. 113–138). Hauppauge, NY: Nova Science
Publishers.
Trobst, K. K., Ayearst, L. E., & Salekin, R. T. (2004). Where is the personality in
personality disorder assessment? A comparison across four sets of personality disorder
scales. Multivariate Behavioral Research, 39, 231–271.
Trull, T. J. (1992). DSM-III-R personality disorders and the five factor model of
Trull, T. J. (2012). The five factor model of personality disorder and DSM-5. Journal of
Trull, T. J., & Brown, W. C. (2013). Borderline personality disorder: A five factor model
perspective. In T. A. Widiger & P. T. Costa (Eds.), Personality disorders and the five
factor model (3rd ed., pp. 118–132). Washington, DC: American Psychological
Association.
Trull, T. J., Useda, J. D., Costa, P. T., & McCrae, R. R. (1995). Comparison of the
MMPI-2 Personality Psychopathology Five (PSY-5), the NEO-PI, and the NEO-PI-R.
Trull, T. J., Widiger, T. A., & Burr, R. (2001). A structured interview for the assessment
of the five factor model of personality: Facet-level relations to the Axis II personality
disorders. Journal of Personality, 69, 175–198.
Trull, T. J., Widiger, T. A., Useda, J. D., Holcomb, J., Doan, B-T., Axelrod, S. R., …
Gershuny, B. S. (1998). A structured interview for the assessment of the five factor
model of personality. Psychological Assessment, 10, 229–240.
Tyrer, P., Crawford, M., Mulder, R., Blashfield, R., Farnam, A., Fossati, A., … Reed, G.
M. (2011). The rationale for the reclassification of personality disorder in the 11th
revision of the international classification of diseases (ICD-11). Personality and Mental
Health, 5, 246–259.
Vachon, D. D., Sellbom, M., Ryder, A. G., Miller, J. D., & Bagby, R. M. (2009). A five
factor model description of depressive personality disorder. Journal of Personality
Disorders, 23, 447–465.
Van Kampen, D. (2012). The 5-Dimensional Personality Test (5DPT): Relationships with
two lexically based instruments and the validation of the Absorption scale. Journal of
Wang, W., Hu, L., Mu, L., Chen, D., Song, Q., Zhou, M., … He, C. (2003). Functioning
styles of personality disorders and five factor normal personality traits: A correlation
study in Chinese students. BMC Psychiatry, 3(1), 11.
relevant to psychopathology: II. Further articulation of a comprehensive unified trait
Watson, D., Stasik, S. M., Ro, E., & Clark, L. A. (2013). Integrating normal and
pathological personality: Relating the DSM-5 trait-dimensional model to general traits
of personality. Assessment, 20, 312–326.
Whetzel, D. L., McDaniel, M. A., Yost, A. P., & Kim, N. (2010). Linearity of
personality–performance relationships: A large-scale examination. International
Journal of Selection and Assessment, 18, 310–320.
Whiteside, S. P., & Lynam, D. R. (2001). The five factor model and impulsivity: Using a
structural model of personality to understand impulsivity. Personality and Individual
Whiteside, S. P., Lynam, D. R., Miller, J. D., & Reynolds, S. K. (2005). Validation of the
UPPS impulsive behavior scale: A four-factor model of impulsivity. European Journal
Widiger, T. A. (2013). DSM-5 personality disorders: A postmortem and future look.
Widiger, T. A., & Boyd, S. (2009). Personality disorders assessment instruments. In J. N.
Butcher (Ed.), Oxford handbook of personality assessment (pp. 336–363). New York,
Widiger, T. A., & Costa, P. T. (1994). Personality and personality disorders. Journal of
Widiger, T. A., & Costa, P. T. (2012). Integrating normal and abnormal personality
structure: The Five Factor Model. Journal of Personality, 80, 1471–1506.
Widiger, T. A., & Costa, P. T. (Eds.). (2013). Personality disorders and the five factor
model of personality (3rd ed.). Washington, DC: American Psychological Association.
Widiger, T. A., Costa, P. T., Gore, W. L., & Crego, C. (2013). Five factor model
personality disorder research. In T. A. Widiger & P. T. Costa (Eds.), Personality
disorders and the five factor model of personality (3rd ed., pp. 75–100). Washington,
Widiger, T. A., Costa, P. T., & McCrae R. R. (2013). Diagnosis of personality disorder
using the five factor model and the proposed DSM-5. In T. A. Widiger & P. T. Costa
(Eds.), Personality disorders and the five factor model of personality (3rd ed., pp. 285–
maladaptive variants of the five factor model. Journal of Personality Assessment, 94,
450–455.
Integrating normal and abnormal personality structure: The five factor model. In T. A.
disorder: Finding a common ground. Journal of Personality Disorders, 19, 110–130.
Wilson, L., Miller, J. D., Zeichner, A., Lynam, D. R., & Widiger, T. A. (2011). An
examination of the validity of the Elemental Psychopathy Assessment: Relations with
other psychopathy measures, aggression, and externalizing behaviors. Journal of
Psychopathology and Behavioral Assessment, 33, 315–322.
Wiggins, J. S., & Pincus, A. L. (1989). Conceptions of personality disorders and
dimensions of personality. Psychological Assessment, 1, 305–316.
Wiggins, J. S., & Pincus, A. L. (2002). Personality structure and the structure of
personality disorders. In P. T. Costa & T. A. Widiger (Eds.), Personality disorders and
the five factor model of personality (2nd ed., pp. 103–124). Washington, DC: American
Wright, A. G. C., & Simms, L J. (2014). On the structure of personality disorder traits:
Conjoint analysis of the CAT-PD, PID-5, and NEO PI-3 trait models. Personality
traits. Journal of Abnormal Psychology, 121, 951–957.
Yamagata, S., Suzuki, A., Ando, J., One, Y., Kijima, N., Yoshimura, K., … Jang, K. L.
Yeung, A. S., Lyons, M. J., Waternaux, C. M., Faraone, S. V., & Tsuang, M. T. (1993).
The relationship between DSM-III personality disorders and the five factor model of
personality. Comprehensive Psychiatry, 34, 227–234.
Zimmerman, G., Rossier, J., de Stadelhofen, F. M., & Gaillard, F. (2005). Alexithymia
assessment and relations with dimensions of personality. European Journal of
Zuroff, D. C. (1994). Depressive personality styles and the five factor model of
personality. Journal of Personality Assessment, 63, 453–472.
Axis I Disorders
R. Michael Bagby, Amanda A. Uliaszek, Tara M. Gralnick, and Nadia Al-Dajani
Abstract
The purpose of this chapter is to summarize and discuss the complex relationship
between Five Factor Model (FFM) personality traits and clinical (Axis I)
psychopathology, including depressive, bipolar, anxiety, obsessive–compulsive,
eating, schizophrenia and psychotic, trauma and stress-related, and substance use
disorders. Considered herein will be the alternative forms of relationship,
including vulnerability, common cause, pathoplasty, complication/scar, and
spectrum. This chapter will highlight the necessity for well-designed, longitudinal
studies aimed at elucidating the complex relationships between the FFM and
clinical disorders. Consistent research supports Neuroticism as a vulnerability
factor to certain disorders, even sharing genetic etiology. However, there are also
important contributions for each of the other four domains. The majority of this
research is in the area of mood and anxiety disorders. Expanding these studies to
include other forms of psychopathology could help identify common personality
vulnerabilities to psychopathology, as well as unique predictors of certain
constellations of symptoms.
Key Words: Axis I psychopathology, depression, anxiety, substance use, eating
disorder, trauma, vulnerability, common cause, pathoplasty, spectrum
There is an impressive literature examining the relationships between the Five

Factor Model (FFM) of personality and clinical disorders (e.g., Andersen &
Bienvenu, 2011; Kotov, Gamez, Schmidt, & Watson, 2010), including several
review articles summarizing these results. Many of these relationships are well
established; for example, the relationship between Neuroticism and major
depressive disorder (e.g., Bagby, Quilty, & Ryder, 2008) and between low
Extraversion and social anxiety disorder (e.g., Kotov, Gamez, Schmidt, &
Watson, 2010). In this article we aim not only to review this vast literature, but
to summarize these results in terms of theoretical etiological models (see Figure
21.1) with the hope of furthering our understanding of such relationships beyond
simple correlational findings. We aim to replicate the structure that other authors
have taken when describing the relationships between personality and
psychopathology in general (Widiger & Smith, 2008) and personality and
depression in particular (Bagby, Quilty, & Ryder, 2008). Because these models
are in no way mutually exclusive, we aim to describe research supporting each
of these models for all clinical disorders discussed. However, where no evidence
exists, those sections have been removed from the respective disorder section.
In each of the following sections focused on groups of clinical disorders as
described in the Diagnostic and Statistical Manual of Mental Disorders (DSM-
5), we will begin with a subsection on cross-sectional relations. Whereas
longitudinal studies examining the dynamic relationships between personality
and clinical disorders might be ideal to test the models described below, it is first
necessary to establish correlational relationships between constructs. The
majority of research for a given disorder often consists of these findings. These
important initial steps will be described in each section below.
Figure 21.1. Etiological models of the association between personality and depression.
The first etiological model explored is the vulnerability model. In this model,
premorbid personality traits contribute to the onset of a clinical disorder.
Personality traits serve as a diathesis to the occurrence of one or more disorders.
Research examining this model must be longitudinal in nature, assessing
personality before the onset of a disorder (often in childhood or adolescence) and
determine through regression or path analysis that personality predicts the
disorder.
The second etiological model is the common cause model. In this model, both
clinical disorders and personality traits are independent constructs that are
caused by a shared diathesis. In this case, personality and psychopathology do
not need to be assessed in a particular temporal sequence. Most frequently, the
common causes examined are genetic or biological in nature.
The third etiological model is the pathoplastic model. This model posits that
personality traits may not necessarily cause the onset of a disorder, but instead
affect how a disorder is expressed. This relates to the severity, symptomatology,
course, and/or prognosis of the disorder. In this case, the presence of certain
personality traits may either ameliorate or exacerbate the distress or impairment
associated with a disorder.
A fourth set of etiological models includes the complication and scar models.
In these cases, the clinical disorder has an effect on personality traits (in essence,
the opposite of the vulnerability model). In the case of a complication model, the
disorder changes personality only during the course of the disorder. This is
ideally assessed by examining personality before the onset of a disorder, again
during the course of the disorder, and then after remission of the disorder. The
complication model is supported when the design yields a definitive change in
personality during the course of the disorder that returns to baseline after
remittance. The scar model is assessed using the same methodology, but in this
case personality never returns to baseline levels.
A final etiological model to be explored is the spectrum model. Here we
assume an underlying dimension that cuts across both personality and
psychopathology, ranging from normal processes (personality traits) to mild,
moderate, and severe pathological processes (clinical disorder). Evidence
supporting this model will demonstrate through taxometric analyses that
constructs are best described as belonging to a single dimension, rather than
belonging to separate classes.
Depressive Disorders
Cross-Sectional Relations
An extensive literature has accumulated examining the association between FFM
personality domains and depressive disorders. In this regard, the Kotov et al.
(2010) meta-analysis indicated that individuals with major depressive disorder
(MDD) possess heightened levels of Neuroticism (N) and decreased levels of
Conscientiousness (C) compared to healthy controls, as well as more robust
effects of this nature found in dysthymic individuals. Because Extraversion (E)
has demonstrated inconsistent associations with MDD across studies, it was not
as highly related to the disorder in the context of the meta-analysis. Moreover,
results did not support consistent relationships between MDD status and
Agreeableness (A) or Openness (O) (Kotov et al., 2010).
Though high N, low C, and low E are common features of many
psychological disorders (Kotov et al., 2010), high N and low E have been
associated with MDD irrespective of comorbid diagnoses (Spinhoven, van der
Does, Ormel, Zitman, & Penninx, 2013). In addition, N has been shown to be
elevated in individuals with premenstrual dysphoric disorder compared to
healthy controls (Adewuya, Loto, & Adewumi, 2008) and in atypically
depressed patients compared to patients with other depressive subtypes (Chopra
et al., 2005). O also has played a role in differentiating depressive subtypes, with
elevations in this dimension found in individuals with seasonal affective disorder
compared to individuals with nonseasonal depression (Bagby, Schuller, Levitt,
Joffe, & Harkness, 1996). Importantly, although the aforementioned findings
contribute to a preliminary understanding of the personality–MDD relationship,
their cross-sectional designs do not allow the directionality of associations to be
inferred.
Vulnerability Model
Prospective longitudinal studies of personality in never-depressed individuals
provide an effective means of testing the vulnerability model. As such, this
model would be supported by evidence demonstrating that personality
dimensions predict the onset of depressive disorders (Klein, Kotov, & Bufferd,
2011). In this regard, De Graaf, Bijl, Ravelli, Smit, and Vollebergh (2002)
conducted a prospective epidemiological study investigating the determinants of
the first lifetime incidence of depressive disorders in a large sample of Dutch
adults. Individuals high in N (as measured by the Groningse Ne Questionnaire;
Ormel, 1980) were more likely to experience a first lifetime incidence of a
depressive disorder over a 12-month time period, irrespective of negative life
events. In addition to the main effect of N, there was also an interaction between
N and life events, wherein individuals high in N were more likely to have a
depressogenic response to a negative life event. Uliaszek et al. (2010) extended
this research by demonstrating that N (assessed as a composite of two FFM
measures and one non-FFM measure) partially accounted for the longitudinal
relationship between interpersonal chronic life stress and depressive disorders in
adolescents. Such findings suggest that N in combination with stressful life
events may confer a risk for depression, supporting the diathesis–stress model.
Findings also support the stress generation model (Hammen, 1991) by
suggesting that N increases vulnerability toward depression, which in turn
increases interpersonal chronic life stress (Uliaszek et al., 2010).
To our knowledge, no study to date has prospectively investigated the effect
of E, C, A, or O on the onset of a first lifetime incidence of depressive disorders.
Weiss et al. (2009) conducted a prospective study examining whether the FFM
domains predicted the incidence of depressive episodes in a sample of older
adults. Individuals were categorized on NEO style graphs (Costa & McCrae,
1998), which plotted each possible combination of two FFM domains. The
graphs specified whether individuals were high or low on the first and second
domains for each trait pairing. High N and low C were found to predict a major
or minor depressive episode over a 22-week time period, suggesting that low
impulse control may play a role in depression onset. Moreover, high N interacted
with low E (denoting a gloomy pessimistic style), with high E (denoting an
overly emotional style), and with high O (denoting a hypersensitive style) to
confer a risk for a major depressive episode. Low C also interacted with low E
(denoting a lethargic style), with high E (denoting a fun-loving style), with low
A (denoting an undistinguished style), with low O (denoting a reluctant
scholarly style), and with high O (denoting a dreamy style) to serve as risk
factors for major or minor depression. Although this investigation provides
novel insight into the role of personality traits and their combinations in the
development of depressive episodes in older adults, the study design is not an
ideal test of the vulnerability model as it did not specifically address first
lifetime depressive episodes. As such, it may be that previous episodes of
depression have influenced personality scores at baseline (supporting the
complication/scar model) or personality influenced the chronicity of depression
across the lifespan (supporting the pathoplasty model).
In light of consistent cross-sectional evidence demonstrating that N is
heightened during MDD (Kotov et al., 2010), it is reasonable to question
whether baseline N scores are inflated by subthreshold depressive symptoms. In
this regard, whereas Uliaszek et al. (2010) did account for baseline depressive
symptoms, De Graaf et al. (2002) did not. In a subsequent study using the same
sample as De Graaf et al. (2002), Ormel, Oldehinkel, and Vollebergh (2004)
noted that although baseline N scores may conceivably reflect subthreshold
depressive symptoms, this is unlikely given that heightened N was observed in
their sample more than 1 year prior to the onset of the first lifetime episodes of
major depression. The authors further suggested that since subthreshold
depressive symptoms were present long before and after the onset of major
depressive episodes, it may be warranted to consider such symptoms as
expressions of a deeper-rooted personality vulnerability.
This “state trait issue” (Costa, Bagby, Herbst, & McCrea, 2005) has been the
subject of contentious debate in the field, with many researchers asserting that
personality scores are systematically confounded by depressive symptoms and
therefore have no meaning when assessed in the presence of depression
(Hirschfeld, Klerman, Clayton, & Keller, 1983; Joffe & Regan, 1988; Liebowitz
Stallone, Dunner & Fieve, 1979; Wetzler, Kahn, Cahn, van Praag, & Asnis,
1990). This perspective implies that many tests of the etiological models
described in this chapter may yield inflated and biased results. Costa et al.
(2005) empirically addressed this issue by demonstrating that the reliability,
validity, and factor structure of the Revised NEO Personality Inventory (NEO
PI-R; Costa & McCrae, 1998) in acutely depressed patients were maintained
following successful antidepressant medication (ADM) treatment. They
interpreted the findings as suggesting that personality levels assessed during
episodes of major depression should be regarded as accurate assessments of an
individual’s existing condition instead of as meaningless distortions. Uliaszek et
al. (2009) examined whether the predictive relationship between N and
depression was artificially inflated by content overlap (specifically in the
depression facet of the FFM). Structural equation modeling techniques did
reveal a slight inflation between the depression facet and symptoms of
depression; however, the primary relationship was driven by general N variance
suggesting that results supporting a predictive relationship between N and
depression may be only mildly inflated.
Common Cause Model
Whereas evidence supporting the vulnerability model suggests that personality
features influence the development of depressive disorders, the common cause
model is supported by evidence suggesting that the same etiological features that
predispose individuals to develop certain personality features predispose
individuals to develop depressive disorders (Bagby, Quilty, & Ryder, 2008). To
this end, in light of findings suggesting that both MDD (Kendler, Gatz, Gardner,
& Pedersen, 2006; Sullivan, Neale, & Kendler, 2000) and personality variation
(Loehlin, 1992; Loehlin, McCrae, Costa, & John 1998) are genetically rooted,
Kendler and Myers (2010) investigated the extent to which genetic factors
account for the personality–MDD relationship in a large-scale twin study. They
found evidence suggesting that E, A, and O are minimally and negatively
associated with genetic risk for MDD, whereas high N and low C exhibit
moderate genetic associations with the lifetime onset of the disorder. These
results directly support the common cause model.
Considering that a number of earlier prospective twin studies have suggested
that N, as defined by models other than the FFM, is linked to MDD through a
shared genetic diathesis (Fanous, Neale, Aggen, & Kendler, 2007; Kendler,
Gatz, Gardner, & Pedersen, 2006; Kendler, Neale, Kessler, Heath, & Eaves,
1993), evidence has accumulated elucidating the nature of this shared genetic
factor in the context of FFM defined N. Though the serotonin transporter gene
polymorphism is a frequently studied candidate gene underlying personality
constructs and psychopathological outcomes (for a review see Margoob &
Mushtaq, 2011, as well as the chapter by Jarnecke and South), evidence is mixed
regarding its association with depressive disorders (Margoob & Mushtaq, 2011)
and N (Takano et al., 2007; Terracciano, Löckenhoff, Crum, Bienvenu, & Costa,
2008). Another gene variation, the brain-derived neurotrophic factor val66met
polymorphism, has been independently found to be associated with elevated
depressive symptoms and N (Gatt et al., 2007; Sen et al., 2003). Given that this
gene variation is implicated in decreased hippocampal volume (Bueller et al.,
2006; Nemoto et al., 2006; Pezawas et al., 2004; Szeszko et al., 2005), Joffe et
al. (2009) investigated its relationship with FFM personality traits, depressive
symptoms, and gray matter volume of the hippocampus (see also the chapter by
Allen and DeYoung). Increased N and depressive symptoms were related to
lower hippocampal volume only in individuals who carried a certain allele of the
gene variation (met carriers).
Variations in the cannabinoid receptor 1 gene have also been shown to confer
a risk for depressive symptoms (particularly in the context of recent stressful life
events), with additional associations found between this gene and FFM
personality traits (positive associations for N and negative associations for A;
Juhasz et al., 2009). They suggest that the cannabinoid receptor 1 gene may be a
marker of vulnerability to depression through a high N and low A phenotype
(Juhasz et al., 2009). In this regard, it is commonly theorized that personality
serves as an intermediate phenotype between biological markers and
psychopathological outcomes, a perspective that concurrently supports the
vulnerability and common cause models. Importantly, prospective genetic
research is needed to test this perspective.
Pathoplasty Model
Although the vulnerability and the common-cause models are supported by
investigations describing the role of personality features in the development of
depressive disorders, the pathoplasty model is supported by evidence suggesting
that personality features predict the severity, chronicity, and prognosis of such
disorders (Bagby, Quilty, & Ryder, 2008; Klein et al., 2011). To this end,
Vrshek-Schallhorn, Czarlinski, Mineka, Zinbarg, and Craske (2011) investigated
the role of N and E (as assessed by composite scores including both FFM and
non-FFM measures) in prospectively predicting suicide ideation in adolescents
and young adults experiencing depressive episodes. Low baseline E was
associated with suicidal ideation across a 3-month time period in males only.
Results suggest that males low in E may be less inclined to seek social support
during times of distress compared to females, placing them at particular risk for
suicidal ideation (Vrshek-Schallhorn et al., 2011).
In another prospective investigation, Spinhoven, de Rooij, Heiser, Smit, and
Penninx (2012) examined the prognostic value of FFM traits in predicting
changes in comorbidity patterns of emotional disorders. Individuals with
depressive and/or anxiety disorders who were low in N or high in C were more
likely to transition to a more favorable comorbidity class over the course of 2
years (analyses corrected for baseline symptom severity and other FFM
personality dimensions). These findings are in line with evidence from
prospective investigations implicating N in the relapse/recurrence of depressive
episodes (Ormel, et al., 2004) and implicating C in improved follow-up
outcomes in inpatients with MDD who have been discharged (Anderson &
McLean, 1997). To this end, Spinhoven et al. (2012) suggested that C may be
particularly instrumental in facilitating the successful implementation of
problem-solving tasks associated with positive outcomes. These conclusions are
in line with prospective studies supporting the role of C in predicting return to
work following an MDD-related absence (Hees, Koeter, & Schene, 2012) and in
help-seeking engagement over the long-term course of the disorder (Schomerus
et al., 2013).
Importantly, a number of treatment studies have demonstrated that FFM
personality dimensions predict treatment responsiveness in patients with MDD
(e.g., Bagby, Joffe, Parker, Kalemba, & Harkness, 1995; Bagby, Levitan,
Kennedy, Levitt, & Joffe, 1999; Bagby, Quilty, et al., 2008; Canuto et al., 2009;
Du, Bakish, Ravindran, & Hrdina, 2002; Quilty et al., 2008), though the specific
dimensions related to treatment responsiveness have been shown to vary across
studies. It is of note that a majority of these investigations examined the role of
personality in the context of a treatment regimen involving ADM (e.g., Bagby et
al., 1995, 1999; Canuto et al., 2009; Du et al., 2002; Quilty et al., 2008).
Moreover, one investigation (Blom et al., 2007) did not support the predictive
utility of any FFM dimension across ADM as well as psychotherapy for MDD,
with only severity of depressive episode, duration of depressive episode, and use
of medical services (to a lesser degree) predictive of treatment outcomes.
A general evaluation of the literature supports all five factors as holding
predictive utility in the context of treatment for MDD (Bagby et al., 1995, 1999;
Bagby, Quilty, & Ryder, 2008; Canuto et al., 2009; Du et al., 2002; Quilty et al.,
2008), with N most consistently predicting poor responsiveness (Bagby et al.,
2008; Canuto et al., 2009; Quilty et al., 2008). For example, Canuto et al. (2009)
investigated the role of FFM personality dimensions in predicting the response
to day hospital treatment (consisting of a combination of group therapy,
individual therapy, ADM, as well as family and network meetings) in older
adults with either MDD or a depressive episode of bipolar disorder. N was the
only domain predictive of treatment outcomes, indicating that this domain may
be associated with reduced or slower improvement in depressive symptoms over
the course treatment.
Bagby, Quilty, et al. (2008) examined the role FFM personality dimensions in
a differential treatment response to either cognitive–behavioral therapy (CBT) or
ADM for MDD. Patients with low levels of O showed a reduced response to
both treatments. Moreover, patients with high levels of N showed an improved
response to ADM compared to CBT. They suggested that O, reflecting a need to
contemplate experience and a tendency to maintain a flexible position toward
novel ideas (McCrae & Costa, 1997), may interfere with the necessity of
accepting the treatment rationale (Bagby, Quilty, et al., 2008). In addition,
patients with high levels of N may lack the regulative capacity needed for
effective engagement in CBT. As such, they may be more responsive to ADM
because this treatment targets affective symptoms associated with MDD without
relying on the patient’s attentional capacity (Knutson, Burgdorf, & Panksepp,
1998). Results highlight the importance of considering differential influences of
treatment types when examining associations between personality and treatment
response (Bagby, Quilty, et al., 2008).
It is of note that in an earlier investigation, Bagby et al. (1995) did not find
that N predicted the response to ADM after controlling for baseline depression
severity. Such findings are consistent with those of Petersen et al. (2002), who
also reported a null relationship between N and responsiveness to ADM in
patients with MDD. Instead, Bagby et al. (1995) found that low E predicted poor
treatment outcomes. They suggested that although N appears to have stronger
relations to the nature of MDD (regarding affect and behavior), low E in the
context of treatment may potentiate social withdrawal, which is an essential
behavioral component of MDD (Bagby et al., 1995). Conversely, in a
prospective examination of the relationship between FFM dimensions and ADM
adherence in patients with MDD, high E predicted reduced treatment compliance
(Cohen, Boggio, & Fregni, 2009). These findings were interpreted as indicating
that patients high in E may be too engaged in other activities to prioritize their
medication.
Du et al. (2002) found A to be the only domain predictive of the response to
ADM, with high A positively related to treatment outcomes. Their analyses
indicated that pretreatment A accounted for 18% of the variance in posttreatment
depression while controlling for baseline depressive symptoms and the five
domains. In this regard, agreeable patients may be more inclined to believe in
the possibility of improvement, potentially increasing responsiveness (Du et al.,
2002). In light of theoretical accounts proposing that A is related to treatment
outcomes by way of its affect on therapeutic alliance (Feeley, De Rubeis, &
Gelfand, 1999; Klein et al., 2003; Orlinsky, Grawe, & Parks, 1994), Kushner,
Quilty, Uliaszek, McBride, and Bagby (2015) examined the role of therapeutic
alliance on the relationship between personality and depressive symptoms
following ADM or psychotherapy for MDD. Mediational analyses supported the
notion that A is associated with treatment outcome by way therapeutic alliance,
irrespective of treatment modality.
In a related study, Quilty et al. (2008) examined the association between FFM
personality domains and treatment response following a combined ADM and
psychotherapy regimen for MDD. Patients were randomly assigned to the ADM
condition and deliberately designated to receive either supportive, cognitive–
behavioral, or psychodynamic interventions. Analyses that accounted for shared
variance across domains indicated that low N and high C were uniquely
predictive of treatment response. Additional analyses indicated that C interacted
with E in the prediction of treatment response, such that patients high in C were
more likely to respond to treatment when they were also high in E (Quilty et al.,
2008). It was suggested that high C and E may reflect a particularly useful
combination of characteristics required for treatment gains, perhaps facilitating
the ease with which patients foster a therapeutic relationship, maintain
involvement in therapy sessions, and/or fulfill treatment requirements (Quilty et
al., 2008).
Overall, studies examining the predictive capacity of personality in the
expression of depressive disorders across time and treatments demonstrate the
clinical utility of the FFM domains, though mixed findings suggest that the
relationship may be largely determined by the research design and statistical
analyses employed. Importantly, although the pathoplasticity model is generally
evaluated by way of longitudinal studies examining the associations between
personality traits and clinical features of depressive disorders, an alternative
explanation of findings from these investigations is that personality traits reflect
markers of an etiologically separate subgroup of depressive disorders, rather
than having a contributory impact on the expression of depression (Klein et al.,
2011). Multiwave follow-up investigations may provide a means of evaluating
this possibility. In this regard, if personality directly predicts disorder course
(instead of predicting a latent disorder class), changes in personality would be
expected to influence consequent changes in clinical features (Klein et al., 2011).
Complication/Scar Model
The complication/scar model describes the potential causal influence of
depressive disorders on personality. This general model is therefore supported
by evidence suggesting that depressive disorders predict changes in personality
features. More specifically, the complication model is supported by evidence
suggesting that personality is restored to premorbid levels following remission,
whereas the scar model is supported by evidence suggesting that state depression
induces lasting personality change, irrespective of remission status.
Importantly, although FFM personality features have originally been
described as remaining stable in adulthood (McCrae & Costa, 1999; Terracciano
et al., 2008), evidence suggests that they are amenable to gradual change over
time (Roberts, Walton, & Viechtbauer, 2006; Srivastava, John, Gosling, &
Potter, 2003;) and to acute change in the advent of depressive disorders (with
changes generally up to one standard deviation in magnitude; e.g., Costa et al.,
2005; De Fruyt, van Leeuwen, Bagby, Rolland, & Rouillon, 2006; Santor,
Bagby, & Joffe, 1997).1 Supporting the concomitancy of change in depressive
symptoms and N, findings from the multiwave prospective investigation of
Ormel et al. (2004) on a large community sample indicated that mean-level N
scores increase during major depressive episodes. Given that N returned to
premorbid levels after remission, findings support the complication model rather
than the scar model. Ormel et al. (2004) cautioned against using short time
intervals in between assessments when conducting prospective research of this
nature, as this increases the risk of erroneously perceiving scar effects that
reflect residual MDD symptoms likely to resolve in time (Ormel et al., 2004).
Weber et al. (2010) found further support for the complication model in their
study examining the stability of personality traits, cognitive processes, and brain
volumes in older adults with and without early-onset depression across a 2-year
time period. Although cognitive performance and volumetric magnetic
resonance imaging (MRI) results were preserved in remitted individuals,
heightened N scores observed in this group at baseline were shown to decrease
to levels comparable to healthy controls at follow-up. The findings of Karsten et
al. (2012) also indicated that recovery from depressive disorders is associated
with decreases in N over a 2-year period, though they reported additional
associations with E and C. These three personality dimensions also showed
change in the opposite direction in the advent of a depressive disorder, providing
additional evidence in support of the complication model. It is of note that
although results indicated that personality scores of recovered individuals
assessed at follow-up were less than one-half standard deviation above
normative scores (Hoekstra, Ormel, & De Fruyt, 1996), personality scores in this
group continued to differ from individuals without a previous depressive
disorder (with heightened levels of N and decreased levels of E and C
maintained). Moreover, although both depressive and anxiety disorders
(corrected for one another) were associated with changes in N, depressive
symptoms demonstrated stronger associations with changes in E and C
compared to anxiety symptoms (Karsten et al., 2012).
Similarly, treatment studies have indicated that N decreases and E increases in
response to successful ADM for MDD (Bagby et al., 1995, 1999; Costa et al.,
2005; De Fruyt et al., 2006; Du et al., 2002; Santor et al., 1997; Tang et al.,
2009) and seasonal affective disorder (Jain, Blais, Otto, Hirshfeld, & Sachs,
1999). Such changes have generally exhibited modest associations with
depressive symptom change, suggesting meaningful shifts in personality rather
than temporary self-reports (Costa et al., 2005; De Fruyt et al., 2006; Santor et
al., 1997). Changes in N and E have also been observed following a treatment
involving ADM alone, psychotherapy alone, or a combination of both modalities
in patients with depressive disorders (Griens, Jonker, Spinhoven, & Blom,
2002). It is of note that whereas most studies have not included a placebo control
group, Tang et al. (2009) found a greater personality change associated with
ADM than with placebo.
Whereas changes in N and E are fairly consistent across treatment studies,
some evidence suggests that C (Costa et al., 2005; De Fruyt et al., 2006), A (De
Fruyt et al., 2006), and O (Costa et al., 2005; De Fruyt et al., 2006) increase in
response to ADM for MDD. Shifts in all five personality dimensions were also
observed following a combined ADM and psychotherapy treatment regimen in
patients with the disorder (Quilty et al., 2008). The aforementioned findings
support the synchrony of change in depressive symptoms and FFM dimensions,
countering the commonly held misconception that after the age of 30 years
personality “has set like plaster and will never soften again” (James, as cited in
Costa & McCrae, 1994, p. 21). Although findings support the complication/scar
model, studies assessing the stability of personality following treatment are
needed in order to tease apart complication and scar effects.
Spectrum Model
Although the aforementioned four etiological models imply that personality is
qualitatively distinct from depressive disorders, the spectrum model posits that
depressive disorders denote an extreme manifestation of personality features
(Bagby, Quilty, & Ryder, 2008). Given that relevant personality features and
depressive disorders are posited to lie on the same continuum, the spectrum
model is supported by findings indicating that the personality–depression
relationship is reasonably specific (Klein et al., 2011). As such, the
aforementioned evidence specifying common etiological factors underlying N
and depressive symptoms can also be regarded as supporting the spectrum model
(e.g., Joffe et al., 2009; Juhasz et al., 2009). However, although cross-sectional
(Kotov et al., 2010) and longitudinal (e.g., Karsten et al., 2012; Ormel et al.,
2004) evidence consistently suggests that individuals with MDD possess
heightened levels of N (in both acute and remitted phases of the disorder), this
domain is elevated in almost all psychological disorders (see below). Although
findings support some overlap between the genotypic and phenotypic features
underlying N and depressive disorders, evidence does not support the distinctive
expression of extreme N as a manifestation of depression.
Importantly, the applicably of the spectrum model to MDD is deterred by the
fact that MDD is more episodic in nature compared to personality features. In
this regard, this model may be most applicable to chronic or subclinical
manifestations of depression (Klein et al., 2011). In support of this perspective,
N has been more strongly associated with dysthymia compared to MDD (Kotov
et al., 2010). Moreover, the potentially overlapping depression facet of N only
minimally explained the association between depressive symptoms and N, with a
general N factor largely accounting for this association (Uliaszek et al., 2009).
Although results suggest that the depression facet of N likely falls on the same
spectrum as depressive disorders, they directly counter the spectrum model by
suggesting that content overlap minimally accounts for the N–depressive
symptom relationship. Overall, N and depression appear to be qualitatively,
rather than quantitatively, distinct.
Bipolar Disorders
Several studies have examined the association between FFM and bipolar
disorder. Bagby, Young, et al. (1996) found that, in comparison to individuals
with unipolar depression, patients with bipolar disorder scored significantly
higher on O and higher on a facet of E (i.e., positive feelings). Contrary to these
findings, Furukawa et al. (1998) found no differences between bipolar patients,
healthy controls, and unipolar patients on any of the FFM domains. This aberrant
finding is likely due to a small sample size (n = 8) and lack of power to detect a
meaningful difference between groups. Additionally, comparisons between
individuals with seasonal affective disorder and individuals with bipolar disorder
revealed that the latter scored significantly higher on N and significantly lower
on E, O, and C (Jain et al., 1999). In another study that compared FFM scores
across a series of disorders, individuals with bipolar disorder were found to
exhibit higher scores on O in comparison to all other disorders investigated, and
higher scores on E in comparison to all other disorders investigated with the
exception of pathological gambling and psychotic disorders (Uliaszek, Al-
Dajani, & Bagby, 2014). Although most studies investigate bipolar disorder as a
categorical construct, Quilty, Sellbom, Tackett, and Bagby (2009) were
interested in examining personality correlates of a unidimensional construct of
bipolarity, in comparison with a two-dimensional construct of mania and
depression. When investigated as one dimension, bipolarity was associated with
high N and low A. When investigated as two dimensions, however, depression
was associated with high N and low E, whereas mania was associated with high
N, high E, and low A. Such findings illustrate that personality is
characteristically distinct during a depressive episode compared to a manic
episode, which may support the complication model. This hypothesis can be
tested through prospective designs in future research.
Common Cause Model

To date, there is only one study that examined an underlying common cause
between personality traits as described by the FFM and bipolar disorder.
Probands diagnosed with bipolar disorder and their siblings were recruited for
this study. Siblings had to meet the criteria for bipolar disorder or
schizoaffective bipolar type. Three distinct phenotypes were investigated:
narrow, which consisted of subjects with full manic episodes and patients with a
DSM-IV diagnosis of either schizoaffective disorder or bipolar I; intermediate,
which consisted of the same subjects as above plus those with bipolar II; and
broad, which consisted of the same diagnoses as above plus individuals with
recurrent depression (i.e., a general affective disorder phenotype). The influence
of personality traits based on genetics and based on the environment was
investigated. Low levels of E and the narrow phenotype were related via shared
genetic material, whereas low A, low C, high N, and high O were related to this
phenotype through environmental factors. The same pattern was observed for the
intermediate phenotype. For the broad phenotype, low levels of A, C, and E and
high levels of N were related through shared genetic material, whereas low
levels of A and C and high levels of N were also related through environmental
factors (Hare et al., 2012). It seems that the genes shared by E and a bipolar
diagnosis are vulnerability factors for the development of this disorder,
regardless of the breadth of the investigated phenotype.
Pathoplasty Model
The expression and frequency of the disparate affective states in bipolar
disorder have been linked to several outcomes, including suicidal risk, treatment
nonadherence, and morbidity rates (see Goodwin & Jamison, 2007). In an effort
to elucidate the reasons behind mixed states (i.e., depression and mania
simultaneously present) and rapid cycling (i.e., four or more episodes/switches
from one pole to the next in a given year) in bipolar disorder, Koszewska and
Rybakowski (2008) investigated associations between personality traits and
these two forms of bipolar disorder. In patients with mixed state episodes, N was
significantly elevated in comparison to patients without mixed episodes. On the
other hand, no differences were found between patients with rapid cycling versus
those without rapid cycling. Based on these findings, high rates of N may
predispose individuals to mixed episodes, whereas personality traits are not
related to rapid cycling. It should be noted that this was a cross-sectional study
and that the interpretation of these data should be viewed with caution.
Similarly, Kim, Joo, Kim, Lim, and Kim (2011) examined the total number of
hospitalizations for depression, mania, and mixed episodes and their relationship
to personality traits. Additionally, they were interested in examining whether
personality characteristics are related to patients who experience affective
switches (e.g., depression to mania) without a euthymic period versus those who
do experience a euthymic period. Overall, the number of hospitalizations for
depression was related to high N, low E, and low O. Patients who did not
experience a euthymic state between affective switches showed elevated levels
of N in comparison to patients who did experience this euthymic state. No other
relationships were found between personality and affective state. Again, these
results are cross-sectional and should be interpreted with caution.
In considering high suicide risk for individuals with bipolar disorder (Angst,
Stassen, Clayton, & Angst, 2002), risk factors related to suicidal ideation in
bipolar disorder may provide useful information for prevention strategies. Allen
et al. (2005) were interested in examining correlates of suicidal ideation in
patients with bipolar disorder who have either attempted suicide in the past or
have never attempted suicide. High rates of O and low E were related to suicidal
ideation in individuals who have previously attempted suicide, whereas high
rates of N were related to suicidal ideation in individuals who have never
attempted suicide. Similarly, the role of lipid profiles and personality traits in
suicidal history in bipolar patients has also been investigated. Previous research
has shown that lipid profiles, particularly low n-3 polyunsaturated fatty acids,
are related to past suicide attempts (Huan et al., 2004), and fluctuations in n-3
intake are related to violent suicides (De Vriese, Christophe, & Maes, 2004).
Generally, specific lipid profiles of interest were found to be positively related to
O and A and to past suicide attempts. This suggests that manipulation of lipid
profiles through diet may alter personality profiles and reduce the risk of suicide
in patients at risk (Evans et al., 2012). As mentioned previously, these two
studies are cross-sectional in nature, and should therefore be interpreted with
caution. Marangell et al. (2006) conducted a 2-year prospective study with 1,556
patients with bipolar disorder in order to elucidate nonoverlapping relationships
between a series of variables (e.g., past suicide history, the NEO PI-R
Neuroticism subscale, and frequency of depression) and suicide
attempts/completions. Of note, only past history of suicide attempts uniquely
predicted prospective suicide attempts/completions, with no relationship found
between N and subsequent suicidal behavior.
Anxiety Disorders
The meta-analysis of Kotov et al. (2010) indicated that individuals with a
diagnosis of social anxiety disorder (SAD), generalized anxiety disorder (GAD),
panic disorder, and agoraphobia possess heightened levels of N and decreased
levels of E and C compared to healthy controls. A specific phobia diagnosis,
however, did not follow this pattern, with personality profiles in this group
remaining within one standard deviation of healthy controls. Such findings
suggest that specific phobias may have an externally rooted etiology that may be
more influenced by conditioning than by personality dysfunction.
Given that the majority of psychological disorders have been found to be
associated high N and low C (with many also associated with low E; Kotov et
al., 2010), Spinhoven et al. (2013) investigated whether comparable personality
profiles across disorders can be explained by disorder comorbidity. Their
investigation of patients with SAD, panic disorder, MDD, and healthy controls
partially supported this perspective. Individuals with a panic disorder diagnosis
had heightened levels of N at baseline, without exhibiting significant differences
in E and C compared to controls. On the other hand, individuals with a pure
panic disorder diagnosis had reduced levels of N and heightened levels of E at
baseline compared to individuals with SAD and MDD. Individuals with SAD
and MDD exhibited a baseline profile of high N and low E, irrespective of
comorbid disorders.
Rosellini and Brown (2011) examined the latent structure of the NEO Five
Factor Inventory (NEO FFI; Costa & McCrae, 1994) in relation to anxiety
(SAD, GAD, and panic disorder/agoraphobia) and depressive disorders in a large
clinical sample. Structural equation modeling suggested that a five factor
solution fit the structure of the data. N was positively associated with SAD and
GAD, E was negatively associated with SAD and panic disorder/agoraphobia, C
was positively associated with GAD, and O was negatively associated with panic
disorder/agoraphobia. The relationship between O and panic
disorder/agoraphobia may suggest that decreased curiosity and heightened
conservativeness influence the situational avoidance associated with a fear of
experiencing a panic attack. In addition, the heightened levels of C found to be
associated with GAD are consistent with the clinical picture of the disorder,
possibly denoting perfectionist tendencies (Brown & Barlow, 2009) resulting
from intolerance of uncertainty (Dugas, Gagnon, Ladouceur, & Freeston, 1998).
Vulnerability Model
Barlow (2000, 2004) posited the triple vulnerability model of emotional
disorders, suggesting that a general biological vulnerability (i.e.,
temperament/personality variables of N and E), general psychological
vulnerability (i.e., a perception of control over stress and emotions), and
disorder-specific psychological vulnerability (e.g., intolerance of uncertainty for
GAD; anxiety sensitivity for panic disorder) may contribute to the etiology of
mood and anxiety disorders. Brown and Naragon-Gainey (2013) tested this
structural equation model in a large clinical sample, focusing on SAD and GAD
among the anxiety disorders. Findings indicated that the general biological
vulnerability dimensions of N and E exhibited the strongest effects on such
diagnoses. Consistent with Rosellini and Brown (2011), increased N was
associated with both disorders (with the largest effects found for GAD), and
decreased E was associated with SAD. E also demonstrated a direct positive
relationship with GAD, although zero-order associations were not significant.
Such relationships between personality characteristics and anxiety disorder
diagnoses remained significant after general and disorder-specific psychological
vulnerability factors were added to the model. Moreover, general psychological
vulnerability factors contributed to the prediction of GAD, but not SAD,
suggesting that this posited universal factor may not cut across disorders.
Proposed disorder-specific vulnerability factors also did not exert unique effects
on their respective diagnoses, with intolerance of uncertainty contributing to the
prediction of SAD but not GAD. Findings nonetheless provided some support
for the vulnerability model, suggesting that personality factors interact with
cognitive factors to predict GAD and SAD. Although Rosellini and Brown
(2011) employed mediational analyses on a large clinical sample, the study’s
cross-sectional design limits its capacity to directly evaluate the vulnerability
model.
In this regard, Hong and Paunonen (2011) investigated the role of FFM
personality traits and affective-cognitive vulnerabilities (e.g., anxiety sensitivity,
intolerance of uncertainty, and social-phobic inferential style) on a range of
psychopathological symptoms (including anxiety, worry, and social anxiety) in a
sample of undergraduates using a longitudinal experience-sampling method
(across a 1-month time period). Results indicated that affective–cognitive
vulnerabilities mediated the relationship between personality and anxiety-related
symptoms. Specifically, individuals with high levels of N and (to a lesser extent)
low levels of E were found to be at greater risk of developing anxiety-related
symptoms, with such a risk likely amplified by negative interpretations of bodily
sensations of anxiety, ambiguous situations, and negative self-perceptions in
social contexts. It is of note that only N conferred a risk for depression, with this
relationship mediated by a depressogenic inferential style, dysfunctional
attitudes, and ruminative tendencies.
Relatedly, Uliaszek et al. (2010) found that both N and E partially accounted
for the longitudinal association between interpersonal chronic life stress and
SAD in adolescents. Given that only N accounted for this association in the
context of MDD, such findings also highlight potential similarities and
differences in the etiological role of personality across SAD and MDD.
Watanabe, Nakao, Tokuyama, and Takeda’s (2005) 5-year cohort study of
factors contributing to the onset of panic attacks similarly broadened our
understanding of the role of personality and life events in the development of
anxiety-related symptoms. Specifically, they found evidence supporting the
predictive role of both N and recent stressful life events on the first onset of
panic attacks in their sample of white-collar Japanese workers.
Common-Cause Model
In support of the common-cause model, cross-sectional associations have been
found between hypothalamic–pituitary–adrenal (HPA) axis functioning and N
(Ormel et al., 2013), as well as HPA functioning and anxiety disorders
(Kirschbaum & Helhammer, 2000; Adam et al., 2008). Given that HPA
dysfunction is commonly believed to serve as a risk factor for MDD (Ehlert,
Gaab, & Heinrichs, 2001; Herbert, 2013; Thase et al., 2002), Adam et al. (2014)
investigated whether the cortisol awakening response (one feature of the daily
rhythm of cortisol) prospectively predicted the first onset of specific anxiety
disorders in a 6-year longitudinal study. Even after controlling for MDD and
other variables, an increased cortical awakening response predicted anxiety
disorder onset, with SAD diagnoses largely driving this finding. In support of the
vulnerability model, baseline N also approached significance in predicting
anxiety disorder onset, although the study did not report whether N was related
to HPA axis functioning. Evidence that HPA axis functioning predicted both N
and anxiety disorders would be direct support for the common-cause model.
In addition, a single nucleotide polymorphism in the COMT gene [which
results in a methionine (Met) to valine (Val) substitution in the codon 158
(COMT Val158Met) polymorphism] has been found to be related to anxiety
(e.g., Enoch, Xu, Ferro, Harris, & Goldman, 2003. However, cross-sectional
relations between this gene variant and FFM personality are inconsistent in the
literature (Harris et al., 2005; Hoth et al. 2006; see also the chapter by Jarnecke
and South). In this regard, Lehto, Akkermann, Parik, Veidebaum, and Harro
(2013) investigated the relationship between COMT Val158Met polymorphism
and personality traits in a birth cohort sample of white participants (assessed at
ages 15, 18, and 25 years). Findings indicated that the COMT Val158Met
polymorphism predicted N in females, and C decreased over time in individuals
with a Val/Val genotype. This study therefore supports the common cause model
when taking into account previous research establishing a relationship between
this gene variant and anxiety disorders. With that said, COMT Val158Met
polymorphism was not associated with anxiety disorders in this investigation
(Lehto et al., 2013).
Pathoplasty Model
Carrera et al. (2006) investigated the relationship between personality traits
and the expression and course of panic disorder. Specifically, they examined
whether personality traits were associated with agoraphobia, panic disorder
severity, and response to pharmacological treatment. Individuals with
agoraphobia demonstrated lower levels of E compared to healthy controls,
suggesting that low levels of E may contribute to the development of
agoraphobia in individuals with panic disorder. Personality, however, was
unrelated to panic disorder severity and treatment response.
Smits et al. (2013) investigated whether personality traits influence
responsiveness to group CBT for SAD and/or moderated effects of
pharmacotherapy (d-cycloserine) in augmenting this treatment. Personality did
not directly influence outcomes of group CBT for SAD. However, the addition
of pharmacotherapy improved outcomes only in patients with low levels of C
and high levels of A. It was noted that findings related to C may have a
biological explanation. Specifically, low C has been found to be associated with
the brain-derived neurotrophic factor Val66Met genotype Met-allele (Hiio et al.,
2011); this gene variant has been found to be associated with deficits in
extinction learning (Soliman et al., 2010), and impairments in extinction learning
have been found to be countered by the administration of d-cycloserine in mice
(Yu et al., 2009). Smits et al. (2013) did not provide a biological explanation for
the observed moderator effect of A, but noted that high A likely also serves as a
marker for an impaired capacity for extinction learning (e.g., Harcourt, Kirkby,
Daniels, & Montgomery, 1998). Such findings support the pathoplasty model by
suggesting that personality influences the course of anxiety disorders.
Glinski and Page (2010) investigated whether personality traits change
following group CBT for SAD, and found N that decreased and E increased
following treatment. Importantly, although SAD symptoms diminished
following treatment, analyses did not control for symptom severity or report
associations between changes in symptoms and changes in personality. As such,
it may be that changes in N and E capture changes in SAD symptoms, as
opposed to reflecting independent changes in personality that manifest on
account of symptom change.
To this end, Karsten et al. (2012) examined state effects of anxiety disorders
(specifically, SAD, GAD, panic disorder, and agoraphobia without panic) and
depressive disorders on personality over the course of 2 years. Individuals were
divided into four groups: (1) disorder absent at baseline and follow-up, (2)
occurrence of a disorder, (3) recovery from a disorder, and (4) disorder present
at baseline and follow-up. Changes in N were found to be associated with the
occurrence and recovery of anxiety disorders. In addition, changes in E and C
were only marginally related to the occurrence and recovery of anxiety
disorders, with stronger associations of this nature associated with depressive
disorders. Findings suggest that fluctuations in anxiety disorders likely influence
levels of N, but fluctuations in depressive disorders may have a greater influence
on levels of E and C as compared to anxiety disorders. It is of note, however,
that all associations demonstrated at most small effect sizes, suggesting a limited
(but existing) capacity for personality change in individuals with anxiety
disorders.
Using the same sample as Karsten et al. (2012), Spinhoven, Penelo, de Rooij,
Penninx, and Ormel (2014) further studied the association between changes in N
and changes in anxiety and depressive disorders by including an additional
assessment 2 years following the original 2-year longitudinal investigation.
Results from structural equation modeling revealed that the associations between
anxiety and depressive disorders and N were likely explained by three elements.
The first consisted of a strong association between the stable components of
psychopathology and the stable components of N; the second consisted of a
modest concurrent association between changes in psychopathology and changes
in N; and the third consisted of small to modest lagged effects of
psychopathology on changes in N. Moreover, premorbid and postmorbid N
levels did not significantly differ following the occurrence of an anxiety or
depressive disorder, providing evidence countering the scar model. Because the
occurrence of a disorder influenced morbid levels of N, findings support the
complication model.
Spectrum Model
Cross-sectional investigations examining unique associations between FFM
facets and anxiety disorders provide evidence countering the spectrum model.
As an example, Rector, Bagby, Huta, and Ayearst (2012) found that individuals
with SAD possessed heightened levels of self-consciousness and decreased
levels of assertiveness compared to individuals with a diagnosis of other mood
and anxiety disorders. With only one facet of N and one facet of E
demonstrating unique associations with SAD, findings counter the notion that
the personality domains in the FFM are merely extreme manifestations of the
disorder.
Additionally, cross-sectional investigations examining symptom-level
associations between personality domains and anxiety disorders provide some
evidence opposing the spectrum model. To this end, Watson and Naragon-
Gainey (2014) found that N was strongly correlated with anxious mood and
worry, was moderately correlated with panic, symptoms of social phobia, and
checking, and was modestly correlated with agoraphobia and specific phobia.
Moreover, E was negatively associated with symptoms of SAD. C, A, and O
provided limited predictive information beyond that of N. Although N findings
demonstrated specificity at the symptom level (reflecting a degree of
overlapping content with anxiety disorders), it is important to consider that this
domain was differentially related to various symptoms of certain anxiety
disorders. For example, although anxious mood, panic, and agoraphobia are
symptoms of panic disorder with agoraphobia, N was highly related to anxious
mood, moderately related to panic, and modestly related to agoraphobia. As
such, findings suggest that anxiety disorders, each of which is composed of
clusters of symptoms, may be distinct constructs from N.
Obsessive–Compulsive Disorders (OCD)

Several studies have examined the cross-sectional relationships between the
FFM and obsessive–compulsive disorder (OCD; La Salle-Ricci et al., 2006;
Rector, Hood, Richter, & Bagby, 2002; Rosellini & Brown, 2011; Watson &
Naragon-Gainey, 2014). Rector et al. (2002) compared relationships between
FFM and patients with OCD versus those with depression. Higher scores on E,
A, and C and lower scores on N differentiated individuals with OCD from those
with depression. Furthermore, Kotov et al. (2010) have recently synthesized
some of these findings in a meta-analysis that investigated associations between
the FFM and anxiety, depressive, and substance use disorders. In relation to
OCD, they found strong significant relationships between OCD and N, low E,
and low C.
Vulnerability Model
Brown and Naragon-Gainey (2013) conducted a series of structural regression
models to test the hypothesis that N and E were general vulnerability factors in
the development of depression, OCD, SAD, and GAD. N was found to predict
all the disorders, whereas E was not significantly related to GAD or OCD.
However, they point out that there were suppression effects, such that holding N
constant resulted in a positive correlation between E and these disorders. It is
important to mention that these data were cross-sectional in nature. Therefore, it
is possible that these findings provide evidence for other models, such as the scar
model.
Spectrum Model
In an effort to elucidate differential relationships between FFM traits and
several mood and anxiety disorders, Rector et al. (2012) investigated personality
trait profiles of individuals with MDD, posttraumatic stress disorder (PTSD),
SAD, panic disorder (with or without agoraphobia), and OCD. Overall,
differential relationships were found at the facet level of the FFM. OCD
emerged as having only one unique association in comparison to the other
disorders, with an elevation on the order facet of C. As such, OCD may
represent an extreme expression of order, as defined by the NEO PI-R facets. In
another investigation, Rector, Richter, and Bagby (2005) found differential
relationships between specific symptoms of OCD and NEO PI-R facets: low
openness to ideas was related to higher obsession severity, whereas low
openness to actions was related to higher compulsive severity. Both facets are
subsumed by the O domain. Based on these results, it can be hypothesized that
symptoms of obsession and compulsions are extreme (low) variants of the above
O facets. Rees, Anderson, and Egan (2006) also investigated unique associations
between NEO PI-R facets and individuals with OCD in comparison to
individuals with anxiety/depressive disorders. Lower scores on actions (O facet),
competence (C facet), and self-discipline (C facet) were differentially related to
OCD. Again, this may suggest that the facets above, along with OCD, fall on a
continuum of severity. It is important to mention that the above studies are all
cross-sectional in nature. Therefore, the possibility that the above FFM facets
fall on a continuum of OCD traits is speculative. Researchers can provide
clarification by obtaining data from longitudinal studies, which would parse out
the relationships between the FFM and OCD. For example, it is possible that the
above facets represent a vulnerability to developing OCD. Future research
should attempt to parse out these relationships more clearly.
Trauma and Stressor-Related Disorders

Given that a diagnosis of trauma and stressor-related disorders necessarily
involves a history of exposure to a traumatic event, a central focus of personality
and trauma research is to determine whether personality successfully predicts
trauma reactions or if traumatic experiences tend to overpower any personality
influence (Jakšić, Brajković, Ivezić, Topić, & Jakovljević, 2012; Johnson &
Thompson, 2008). Importantly, marked individual differences in trauma
reactions suggest that traumatic experiences are unlikely to be the sole
etiological factor involved in the development of this class of disorders. To this
end, although exposure to traumatic events is relatively common, only between
8% and 20% of individuals develop (for instance) PTSD following traumatic
exposure (Kessler, Sonnega, Bromet, Hughes, & Nelson, 1995). Most of the
cross-sectional studies examining the personality–trauma relationship have
therefore aimed to ascertain whether personality traits play a key role in
differentiating individuals who develop a trauma reaction following a traumatic
event from those who do not.
In the Jakšić et al. (2012) review of the literature, N was the only FFM trait
that was consistently found to relate to PTSD at a cross-sectional level. In an
attempt to delineate the influence of trauma exposure and personality in the
development of PTSD, Talbert, Braswell, Albrecht, Hyer, and Boudewyns
(1993) divided Vietnam veterans with combat-related PTSD into three groups
related to their degree of trauma exposure. Although there were no observed
differences in the personality profiles of each trauma-exposed group, PTSD
patients demonstrated heightened levels of N and decreased levels of A
compared to the general population. N was also associated with PTSD symptoms
among veterans of Iraq and/or Afghanistan (Clark & Owens, 2012), emergency
management professionals (La Fauci-Schutt & Marotta, 2011), and heart attack
survivors (Chung, Berger, Jones, & Rudd, 2006). Although findings support an
influence of personality in trauma reactions, they either imply that N confers a
risk for PTSD following a traumatic event (supporting the vulnerability model)
or that experiencing PTSD following a traumatic event increases N (supporting
the complication/scar model). As always, prospective designs are needed in
order to tease apart such possibilities.
Vulnerability Model
To our knowledge, no study to date has directly examined the effect FFM
personality dimensions have on the development of PTSD using a prospective
design. Nightingale and Williams’ (2000) posttrauma longitudinal study,
however, found evidence supporting an indirect role of personality in the
development of PTSD. In their investigation examining the impact of attitudes
about emotional expression and personality on PTSD development 1 week and 6
weeks following a road traffic accident, they found that negative attitudes about
emotional expression prospectively predicted PTSD and was also negatively
associated with E, A, and O. Though the causal influence of personality in
developing attitudes about emotional expression and PTSD was not directly
evaluated, findings shed light on possible mechanisms by which personality may
confer vulnerability to trauma reactions.
The distinct requirement of a traumatic experience for a PTSD diagnosis also
provides a unique opportunity to examine personality resilience in the absence of
trauma reactions. In this regard, Yuan et al. (2011) prospectively assessed factors
that protect against PTSD symptom development in police officers. Although
white race, lower exposure to critical incidents, greater assumptions of
benevolence of the world, and greater social adjustment during academy training
predicted lower PTSD symptoms after 2 years of employment, the FFM traits
were not associated with protection against PTSD symptom development.
Common Cause Model

Though no study to date has directly investigated the associations between
biological markers of PTSD and FFM dimensions, associations of this nature
have been indirectly examined using physiological proxies of PTSD symptoms.
Specifically, given that it has been shown that the process of consolidating
extinction learning is disrupted in individuals with PTSD (Wessa & Flor, 2007)
and rapid eye movement (REM) sleep deprivation is known to impair this
process (Silvestri, 2005), Spoormaker et al. (2010) investigated associations
between REM sleep disturbances, failure to consolidate extinction learning,
recall of fear extinction, and personality in healthy individuals undergoing a
conditioning/extinction task followed by a recall of the extinction task.
Functional magnetic resonance imaging and skin conductance response
measurements were taken during each task, and participants were instructed to
sleep for 90 minutes between conditioning/extinction and recall sessions.
Individuals who did not experience REM sleep demonstrated a decreased
capacity to habituate to extinction learning and a heightened physiological
response to the extinguished stimulus. They also exhibited lower levels of A and
O scores compared to individuals who experienced REM sleep. These findings
suggest that REM sleep deprivation may serve as a shared etiological factor
underlying PTSD symptoms and personality.
Pathoplasty Model
N has not only been implicated as a potential vulnerability factor in the
development of PTSD, but has also been shown to play a role in the expression
of PTSD symptoms. In this regard, Kuijer, Marshall, and Bishop (2014)
prospectively investigated the relationship between proposed pretrauma
variables (N, optimism, self-control, and depression) and PTSD symptoms 1
month and 3 months following two large-scale earthquakes in New Zealand. Not
only did heightened N, heightened pretrauma depression, reduced self-control,
and reduced optimism predict postearthquake PTSD symptoms (while
controlling for perceived life threat and objective trauma severity), all the
aforementioned pretrauma variables with the exception of optimism also
predicted increased PTSD symptoms over time. Such results suggest that N
likely influences both the etiology and course of PTSD.
Importantly, earthquake-related hassles also mediated the relationship
between N and postearthquake PTSD symptoms and PTSD symptom change
over time (Kuijer et al., 2014), supporting earlier findings suggesting that
individuals high in N likely experience stressors as more impairing than
individuals low in this trait (e.g., Suls & Martin, 2005). Moreover, when the
relative contribution of each pretrauma factor was examined, N robustly
predicted PTSD symptoms whereas optimism no longer served as a significant
predictor (Kuijer et al., 2014). Such results suggest that the effects of optimism
on adaptation following traumatic experiences may primarily be manifested
because of negative associations between optimism and N. It is of note,
however, that such conclusions are countered by cross-sectional findings
examining personality predictors of posttraumatic growth. In this regard, Wang,
Wang, Wang, Wu, and Liu (2013) found that N was unrelated to posttraumatic
growth in accidentally injured patients, with O instead demonstrating
associations with this experience.
In a related longitudinal study, Lee, Sudom, and Zamorski (2013) examined
factors contributing to psychological resilience in postdeployment Canadian
military personnel. They found that military personnel who reported higher
baseline levels of emotional stability and C demonstrated improved
postdeployment mental health outcomes (as assessed using the Mental Health
Component Summary; Ware & Sherbourne, 1992). Moreover, high levels of A
strengthened the association between combat experiences and postdeployment
mental health difficulties. It was suggested that agreeable individuals who hold
trust in others and in the world are more likely to have their world views
challenged by difficult experiences. Although similar explanations have been
provided to account for the influence of secure attachment on trauma reactions
(Kanninen, Punamäki, & Qouta, 2003), Clark and Owens’ (2012) cross-sectional
investigation found evidence suggesting that secure attachment interacts with
high C to instead protect against PTSD symptom severity. Such findings
highlight the complex role of personality factors in reactivity and resilience
following traumatic experiences.
To our knowledge, one study to date has prospectively investigated the
associations between personality traits and response to treatment for PTSD. In
this regard, van Emmerik, Kamphuis, Noordhof, and Emmelkamp (2011)
hypothesized that high levels of N would predict dropout in trauma-focused
CBT, and that low levels of O, A, and C would be associated with decreased
treatment responsiveness. Although personality traits did not directly influence
treatment outcomes or dropout, lower baseline scores on O were more strongly
associated with posttreatment PTSD symptoms in patients receiving the active
treatment compared to patients in the wait-list control condition. Expectedly,
heightened baseline PTSD symptoms strongly predicted PTSD symptoms at
posttreatment. These findings may indicate that patients with higher levels of O
are particularly amenable to exposure and cognitive restructuring exercises,
though baseline PTSD symptoms have been shown to influence treatment
outcomes more directly than personality traits.
Although no study to date has prospectively investigated personality change
following PTSD, Kamphuis, Emmelkamp, and Bartak (2003) compared
personality profiles of females experiencing PTSD symptoms following
postintimate stalking with those of female undergraduates. Lower O was
associated with PTSD symptoms in women who have endured postintimate
stalking, suggesting that women’s personalities may adapt to become more
guarded following such an experience. Although findings lend some support to
the complication/scar model, longitudinal designs are needed in order to directly
evaluate the model and to ultimately tease apart scar and complication effects.
Schizophrenia and Psychotic Disorders

Studies examining the relationship between schizophrenia, psychosis, and the
FFM are limited and tend to be cross-sectional. For this reason, cross-sectional
studies that may fit into the other models (e.g., the pathoplasty model) are placed
in those sections, despite their methodology. Please keep in mind that only when
otherwise stated, these studies are cross-sectional and their placement in the
models is hypothetical and should be considered with caution. Generally,
findings investigating the relationship between the FFM and schizophrenia tend
to find rates of high N and low C (for a review see Andersen & Bienvenu, 2011).
Comparisons between patients with schizophrenia, unipolar depression, or
bipolar depression on the FFM revealed that patients differed on E, A, and O.
Closer examination of these differences showed that patients with bipolar
depression scored higher than those with unipolar depression on a facet of E,
with no differences between schizophrenia and bipolar depression. Additionally,
patients with schizophrenia scored significantly lower on a facet of O in
comparison to patients with bipolar depression, and significantly lower on
another facet of O in comparison to unipolar and bipolar patients. These
relationships were no longer significant after controlling for demographic
variables (e.g., years of education, socioeconomic status). Finally, patients with
unipolar depression scored significantly higher on A than those with
schizophrenia, with no differences between bipolar and schizophrenia on this
domain (Bagby et al., 1997). When compared to healthy individuals, patients
with schizophrenia scored higher on N, and lower on E, O, A, and C (Camisa et
al., 2005), with another study showing higher rates of N and lower rates of C in
schizophrenia, in comparison to healthy controls (Gurrera, Nestor, & O’Donnell,
2000). Interestingly, individuals with psychotic disorders were found to have
low N scores in comparison to individuals with other psychiatric conditions,
with the exception of pathological gambling, PTSD, and somatoform disorders
(Uliaszek et al., 2014). As such, it may be that individuals with psychotic
illnesses exhibit higher scores on N in comparison to healthy individuals,
whereas their scores on N are significantly lower than individuals with other
psychiatric conditions (e.g., bipolar disorder, panic disorder, and social phobia).
Vulnerability and Scar Models

To investigate whether personality is a vulnerability factor for the onset of
psychosis, Krabbendam et al. (2002) examined a large sample of individuals
with no history of psychosis at baseline, and then 1 and 3 years following initial
assessment. They utilized the Gronigen Neuroticism Scale (Ormel, 1980) to
assess for the predictive value of N in psychosis onset. High rates of N at
baseline predicted psychotic symptoms at the 3-year follow-up, with higher rates
of N exhibiting a stronger association with the presence of psychotic symptoms.
Of note, the association between N and the presence of psychosis did not change
after adjusting for self-esteem, depression, and anxiety levels.
Due to the nature of the current literature, we could not tease apart findings
that were in support of the vulnerability model from those that were in support of
the scar model. These two models can be separated from one another only by
examining personality before and after illness onset to ascertain if there were any
changes. However, the following studies do provide some evidence for the
stability of personality in patients with schizophrenia after illness onset.
Individuals with first-onset psychosis were examined at baseline and then at 3-
month follow-up after undergoing CBT. In comparison to healthy individuals,
those with first onset psychosis had higher N, O, and A and lower E and C
scores. Although symptoms improved within the 3-month follow-up, personality
traits were stable, suggesting that these patients either had a vulnerable
personality style that was present before illness onset or that the presence of
schizophrenia altered personality (Beauchamp, Lecomte, Lecomte, Leclerc, &
Corbiere, 2006). In another study, individuals with schizophrenia were assessed
at intake and at a 6-month follow-up period. Low scores on C and high scores on
N were observed in patients, in comparison to healthy controls. Additionally,
personality remained stable during the 6-month period. Although positive
psychotic symptoms had no bearing on personality stability, negative symptoms
were related to personality stability and were considered “trait-like,” again
providing evidence for either the vulnerability or the scar models (Kentros et al.,
1997).
Pathoplasty Model
Studies examining neuropsychological deficits associated with psychotic
disorders have attempted to shed light on the moderating role of personality
traits through cross-sectional designs. Although these studies are summarized
under the pathoplasty model, we acknowledge that longitudinal studies are
needed. It has been consistently found that error-related negativity (ERN) is
blunted in schizophrenia (for a review see O’Donnell, Salisbury, Niznikiewicz,
Brenner, & Vohs, 2011), reflecting impairments in error detection. In a recent
study, blunted ERN was found across an array of psychotic disorders, illustrating
that it is not specific to schizophrenia. Of note, higher rates of N were associated
with increased ERN after controlling for extraneous variables. Blunted ERN was
also associated with unemployment and two or more hospitalizations during the
early phase of the illness (Foti, Kotov, Bromet, & Hajcak, 2012). The findings
from this study illustrate that higher rates of N may actually protect against
blunted ERN, which is associated with impaired functioning. Along these lines,
severity of delusions was positively associated with rates of N in a sample of
individuals with either schizophrenia or schizoaffective disorder, again
illustrating a relationship between personality and symptom severity (Lysakar,
Lancaster, Nees, & Davis, 2003). In addition, measures of visual attention in
individuals with schizophrenia were found to be negatively related to N and A,
whereas measures of executive function were negatively related to both C and N.
Furthermore, neuropsychological performance in both healthy controls and
patients was found to be related to personality, with N showing the most
consistent relationships across groups and with A and C showing the most
divergent relationships (Gurrera, Nestor, O’Donnell, Rosenberg, & McCarley,
2005).
As previously mentioned, personality has been shown to moderate impairment
in individuals with schizophrenia. Low N is negatively related to the number of
social contacts and employment, and higher rates of E, O, and A are positively
related to the number of social contacts (Kentros et al., 1997). Similarly, higher
rates of social interaction were found to be related to higher rates of A, whereas
capacity for intimacy was related to higher rates of both A and O in a sample of
patients with schizophrenia or schizoaffective disorder (Lysaker & Davis, 2004).
Couture, Lecomte, and Leclerc (2007) also investigated the association between
personality traits and social functioning in a sample of individuals diagnosed
with a schizophrenia spectrum disorder. High A was negatively associated with
greater social and independent living skills, whereas it was positively associated
with more inappropriate community behavior. Johansen, Melle, Iversen, and
Hestad (2013) examined personality traits related to interpersonal problems and
those related to therapeutic alliance. High N and low E, A, and C were all related
to interpersonal problems, whereas high A and low N were related to a stronger
therapeutic alliance.
In addition to impairments in interpersonal functioning, the relationship
between personality, symptomology, and treatment outcomes has been
investigated in this population. For example, associations between personality
and suicidal behaviors in patients with schizophrenia followed for 5 years show
that high C was found to protect against suicide attempts/completions (Pillmann,
Balzuweit, Haring, Blöink, & Marneros, 2003). Considering the importance of
treatment adherence in this high-risk population, Lecomte et al. (2008)
investigated correlates of medication adherence and service management in
individuals with early psychosis; they found that those with higher A scores had
poorer medication adherence, and those with low N and high A scores had
poorer service management. In line with these findings, Beauchamp, Lecomte,
Lecomte, Leclerc, and Corbiere (2011) examined the relationship between
behavioral change during treatment and personality in a sample of individuals
with early psychosis. Active coping strategies were linked to higher rates of E,
whereas passive coping strategies were linked to higher rates of N. Additionally,
personality profiles based on the FFM revealed that active coping strategies were
most closely linked to a personality profile with high O scores and another
profile with high E scores. In addition, Beauchamp et al. (2013) investigated the
relationship between personality traits, coping strategies, and symptom measures
for individuals with a first episode psychosis. Individuals were randomized into
three groups: cognitive–behavioral therapy, skills training for symptom
management, and a wait-list control. High rates of C were linked to therapeutic
changes in active coping strategies in individuals in the cognitive–behavioral
therapy group, high E was related to changes in active coping strategies in the
skills training group, and high rates of O were related to changes in active
coping strategies in the control group. Such findings illustrate the need to
examine personality prior to treatment in order to ensure that patients are placed
in a treatment program that is conducive to their personality profile and is likely
to engender positive change.
Gleeson, Rawlings, Jackson, and McGorry (2005) prospectively examined the
association between personality profile and symptom recurrence in first-episode
psychosis patients. Recurrence was predicted by high N and A and premorbid
conditions. After controlling for other factors, the relationship between
recurrence and A remained significant, although this was not the case for N.
Based on these findings, it is possible that high A improves the prognosis for
those experiencing first-episode psychosis.
Spectrum Model
Studies investigating the spectrum hypothesis in schizophrenia are scant. In
this section, a study examining personality traits of patients, their siblings, and
healthy controls will be summarized, followed by a brief summary of the current
debate on the relationship between O and psychosis (see also the chapter by
Widiger, Gore, Crego, Rojas, and Oltmanns). Boyette et al. (2013) examined the
relationship between FFM personality traits and individuals with schizophrenia,
their siblings, and healthy controls. Individuals with schizophrenia reported the
highest levels of N, followed by siblings, followed by healthy controls.
Additionally, high N and O levels were related to increased levels of subclinical
psychotic symptoms in all the groups assessed, but levels of O were not related
to clinical symptoms. Based on these findings, levels of N seem to increase with
familial risk of psychosis, whereas levels of O do not follow a dimensional
pattern. In considering this finding, we turn next to the debate concerning the
relationship of O with symptoms of psychosis.
To date, the relationship between psychotic symptoms and the O domain has
been inconsistent, with some studies finding a positive relationship (Ross, Lutz,
& Bailley, 2002), others finding a negative relationship (Gurrera et al., 2000),
and still others finding no relationship between these two variables (Bagby et al.,
1997). These inconsistencies have resulted in a debate in the current literature as
to whether there is a psychopathological variant of O and whether normative
personality can capture psychotic symptoms. Although the majority of studies
cited above showed relationships between high N, low C, and psychotic
disorders, there is still criticism that these traits do not fully capture positive
symptoms of psychosis (e.g., hallucinations and delusions). Proponents of the
use of O in describing psychotic symptoms suggest that O must be
reconceptualized to include psychopathological characteristics (for a review see
Chmielewski, Bagby, Markon, Ring, & Ryder, 2014). Peculiarity is one such
trait that has been shown to be related to O (Ross, et al., 2002) and has been
shown to be related to psychoticism (Tackett, Silberschmidt, Krueger, &
Sponheim, 2008). Additionally, De Young et al. (2011) established associations
between O and increased dopamine release, a biological marker implicated in
psychotic disorders (for a review see Davis, Kahn, Ko, & Davidson, 1991). On
the other hand, researchers opposed to the use of O have suggested the inclusion
of a new trait, most commonly labeled psychoticism (e.g., Harkness, McNulty,
& Ben-Porath, 1995; Krueger, Derringer, Markon, Watson, & Skodol, 2012).
Eating Disorders
A number of researchers have examined the relationships between eating
disorders and several different personality traits (for review, see Cassin & von
Ranson, 2005). Podar, Hannus, and Allik (1999) examined personality traits in
three distinct groups: patients with eating disorders, individuals in a weight-
reduction program, and healthy controls. The eating-disordered group scored
significantly higher than the two latter groups on N, and significantly lower on E
and C. The weight-reduction group scored similarly to controls on all domains
except for O, on which they scored significantly lower than controls (and the
same as the eating-disordered group).
Vulnerability Model
In a 2-year longitudinal study, the personality traits (as measured by
Goldberg’s unipolar Big-Five Markers; Goldberg, 1992) of participants with a
lifetime eating disorder, those with a first onset eating disorder (at the 2-year
follow-up), and healthy individuals were assessed. Personality profiles for
individuals with a lifetime eating disorder resembled those with first-onset eating
disorder before the onset of the disorder, with low scores on A and C and high
scores on O. Of note, individuals with a lifetime eating disorder also scored
lower on emotional stability (i.e., the opposite pole of N). This suggests that
individuals with a similar personality profile may be more vulnerable to
developing an eating disorder, and that developing an eating disorder may
decrease the levels of emotional stability, thereby partially supporting the
scar/complication model (Ghaderi & Scott, 2000).
Pathoplasty Model
In an effort to examine differences and similarities between subtypes of
anorexia nervosa (AN), Bollen and Wojciechowski (2004) examined the
personality profiles of individuals with AN restricting subtype and AN binge-
eating/purging subtype. Although high scores on N characterized both groups,
the former was also related to higher rates of C and A. This may illustrate the
role that personality plays in symptom expression (Widiger & Smith, 2008).
Additionally, personality profiles of female patients with an eating disorder have
been found to be related to severity of psychopathology and specific behavioral
characteristics. Cluster analyses of FFM scales revealed three personality
profiles: (1) resilient/high functioning (i.e., no elevations), (2)
undercontrolled/emotionally dysregulated (i.e., high N scores, low A and C
scores), and (3) overcontrolled/constricted (i.e., high N and C scores, low O
scores). Although these personality profiles were not related to eating disorder
subtype, they were related to disorder comorbidity (resilient individuals
exhibited less comorbidity than others) and those falling into the undercontrolled
cluster exhibited more impulsive traits and behaviors (Claes et al., 2006). This
illustrates the role that personality may play in disorder manifestation. It is
important to note that both studies are cross-sectional and interpretation of the
results as supporting the pathoplasty model should be taken with caution.
Substance Use Disorders

A meta-analysis examining the relationship between alcohol-related problems
and the FFM found that low C, low A, and high N were most associated with
alcohol-related problems (Malouff, Thorsteinsson, Rooke, & Schutte, 2007). In
addition, relationships between substance use and other health-related problems
(i.e., overall health and elevated body mass index) reveal that all three problems
are characterized by low C. Of note, high N was related to greater severity of
health-related problems, whereas high E was differentially associated with
substance use (Atherton, Robins, Rentfrow, & Lamb, 2014). Another disorder
that commonly cooccurs with substance-related issues is antisocial personality
disorder (Grant et al., 2004). Ruiz, Pincus, and Schinka (2008) were interested in
examining personality as a common factor between substance use disorders and
antisocial personality disorder. A meta-analysis was conducted and revealed that
both disorders were characterized by low C, low A, and high impulsivity.
Impulsivity was defined as a constellation of facet scores that cuts across FFM
domains (i.e., NEO PI-R Impulsivity, Excitement Seeking, low Deliberation, and
low Self-Discipline). However, the NEO PI-R denotes impulsivity specifically
as a facet of N. Of note, individuals with antisocial personality disorder scored
significantly lower on N than those with a substance use disorder or with a
comorbid substance use disorder and antisocial personality disorder. It is
possible that high N is a vulnerability factor for developing a substance use
disorder, whereas low N is a vulnerability factor for developing antisocial
personality disorder, in the presence of the personality profile that describes both
disorders. This hypothesis can be tested through a longitudinal design that
investigates personality before and after illness onset.
Vulnerability and Scar/Complication Models

To date, longitudinal studies that examine the vulnerability or
scar/complication models do not exist. The study that is placed in this section is
a cross-sectional study that may be interpreted as supporting either model.
Hopwood et al. (2007) examined the relationship between personality and past
substance use disorders, those with a current substance use disorder, and patient
controls who never met the criteria for a substance use disorder. Of note, the past
and current group differed only on a facet of E (i.e., excitement-seeking), such
that individuals who never endorsed criteria for a substance use disorder scored
the lowest, those with a current diagnosis scored the highest, and those with a
past diagnosis fell in between these two groups. Otherwise, the past and current
group had elevated scores on facets of N, and lower scores on A and C, in
comparison to the control group. It is possible that N, A, and C are either
vulnerability factors for developing a substance use disorder or that development
of the disorder results in this personality profile, which is forever altered (i.e.,
supporting the scar hypothesis). The facet that differentiated the past and current
groups supports the complication model, such that disorder presence results in
elevation in excitement-seeking. It is not possible to ascertain whether this study
supports the scar or vulnerability hypothesis, since patients were not examined
prior to the first onset of the illness.
Pathoplasty Model
Although most studies examining the relationship between the course of
substance-related problems and personality are cross-sectional, these studies
seem to support the pathoplasty model by identifying differential relationships
between heavy and moderate users, and by identifying differences in substance
use characteristics between personality profiles. For instance, high E was found
to be associated with number of drinks in the past month in a university sample,
whereas high N was associated with alcohol-related problems. Low C was
associated with the number of drinks in the past month and alcohol-related
problems. Additionally, high N and low C were related to coping–anxiety
drinking motivation, only high N was related to coping–depression drinking
motivation, and high E and low C were related to enhancement drinking
motivation (Mezquita, Stewart, & Ruiperez, 2010). Similarly, low C was found
to be related to elevated levels of weekly alcohol use in a university sample
(Clark et al., 2012). Walton and Roberts (2004) investigated differential
relationships between heavy, moderate, and nondrug/alcohol users and
personality traits, as measured by Goldberg’s (1999) measure of the Big Five.
Heavy users scored lower than the other groups on C and A, whereas abstainers
scored lower on E in comparison to the other groups. Finally, ter Bogt, Engels,
and Dubas (2006) investigated differences between a nationally representative
sample and a sample of party (i.e., rave) goers on personality traits, as measured
by a Dutch version of Goldberg’s Big Five questionnaire (Gerris et al., 1998).
Party goers were divided into three groups: non-3,4-
methylenedioxymethamphetamine (MDMA) users, light users (two pills per
occasion), and heavy users (three or more pills per occasion). Both the
representative sample and non-MDMA party goers scored lower on E than light
and heavy users. Additionally, non-MDMA party goers scored low on A in
comparison to the representative sample, but were not significantly different on
A than light/heavy users. All three party goer groups scored lower on C than the
representative sample, indicating a relationship between attending rave parties
and levels of C. Non-MDMA party goers scored the lowest on emotional
stability (i.e., the opposite pole of N), whereas the representative sample scored
the highest on this domain. Light and heavy users did not differ from each other
on emotional stability.
Differences in substance of choice and the relationship to personality have
also been investigated. In one study, cigarette smokers had low C and high N
levels in comparison to never smokers. Although heroine/cocaine users exhibited
the same pattern, the scores were more extreme than those of cigarette smokers.
Marijuana users had high O levels and low C and A levels, with average levels
of N (Terracciano et al., 2008). In a similar study, three personality profiles were
obtained through cluster analyses and their relationships to substance abuse
characteristics (i.e., substance of choice, coping strategies, and
depressive/impulsive/hostile tendencies) were investigated. The first profile,
characterized by normal N levels, elevated E levels, and slightly lower A and C
scores (although higher than the other two groups), was characterized by alcohol
or cocaine substance use, with a large percentage of individuals using only one
of these substances. This group had the highest percentage of cocaine-only users
in comparison to the other groups. This group was also characterized by high
planful problem solving and positive reappraisal coping strategies, and low
escape–avoidance coping strategies. Finally, this group also exhibited lower
depressive, impulsive, and hostile tendencies. Group three, which was
characterized by the highest levels of N and the lowest levels of E, A, and C,
consisted of polysubstance users (i.e., alcohol and cocaine; alcohol, cocaine, and
another drug). This group had a relationship to the above variables that was the
reverse of that found in the first profile. Finally, group 2, which was
characterized by high N and low A and C, consisted of alcohol and cocaine
users, with the largest percentage of alcohol only users. This group also fell in
between groups 1 and 3 on all the aforementioned variables (Quirk &
McCormick, 1998).
Summary
To ease interpretation, a table summarizing findings (see Table 21.1) has been
provided. This table emphasizes research findings supporting the given models,
as well as the personality traits typically implicated; it leaves absent models that
have been tested only with correlational data when these data are insufficient to
test the necessary parameters of the model. Of course, the most prolific findings
are found in the category of cross-sectional relations; several of the disorders of
interest have amassed such a body of literature in regard to cross-sectional
relationships with the FFM that meta-analyses exist to combine the data. All
disorders are characterized by high N, lending further support to the notion that
N is a general risk factor for psychological distress and/or psychopathology.
Several of the other disorders were also categorized by low C and A—
highlighting cognitive difficulties with organization and responsibility, as well as
interpersonal difficulties apparent in those with a psychiatric disorder.
Because pathoplastic models can often be assessed with cross-sectional data,
there exists substantial research examining how personality traits affect the
expression of psychopathology. Most often, personality traits are shown to be
related to severity and treatment outcome. In several instances, high N is shown
to negatively affect treatment outcome and is associated with an increase in the
severity of disorder expression. This is the case for some mood and anxiety
disorders as well as psychotic and substance disorders. E and A often serve as
protective factors, improving treatment outcome or decreasing the severity of a
disorder. This resilience may come from increasing social support and help-
seeking behavior. For psychotic disorders, results were mixed regarding A; some
research pointed to it as a protective factor, whereas other research showed it to
be associated with pathological interpersonal behaviors. As a whole, these
studies support research emphasizing the value of personality assessment at
pretreatment stages to identify potential difficulties and likely prognosis.
In the limited longitudinal studies that have examined personality traits before
the initial onset of a disorder, N was a common predictor of depressive, anxiety,
and psychotic disorders. Some research examining biological and genetic
markers of disorders have identified the shared etiology of N for mood and
anxiety disorders, as well as low C for mood disorders. These cases, supporting a
common cause model, highlight the complex, intersecting relationships between
personality and psychopathology. There also exists limited evidence for high N
as a complication to depressive and anxiety disorders, with studies examining N
at preonset and postremission finding that N returns to premorbid rates after
remission. Finally, much less research has been dedicated to examining the
spectrum model of personality traits and clinical disorders compared to this
etiological model for personality disorders. The only data approximating some
support for this hypothesis involved the relationship between O and psychosis.
However, these findings are controversial, with some researchers suggesting the
psychosis cannot be captured by normal personality traits.
Table 21.1 General Summary of Results Linking Etiological Models to the Classes of
Psychopathology. Blank Boxes Indicate No Research, Null Findings, or Only
Speculative Findings (i.e., Cross-Sectional Results Supporting Vulnerability,
Common Cause, Complication/Scar, or Spectrum Models).
Cross- Vulnerability Common Pathoplastic Complication/Scar Spectrum
Sectional Model Cause Model Model Model
Relations Model
Depressive N; low E, N N; low C N; low E, O, N; low E, O, C
Disorders C C, A
Bipolar N, E, O; N; low E, N; low E, O

Disorders low A C, A
Anxiety N; low E, N; low E N low C, A N
Disorders O, C
Obsessive- N; low E,
Compulsive C
Disorders
Trauma and N N; low O
Stressor-
Related
Disorders
Schizophrenia N; low E, N N, A; low E, O
and Psychotic O, C, A O, C, A
Disorders
Eating N; low E, O; low A, C
Disorders C
Substance N; low A, N, E; low A,
Use Disorders C C
Note: N = neuroticism; E = extraversion; O = openness to experience; C = conscientiousness; A =
agreeableness.
In sum, this chapter highlights the necessity for well-designed, longitudinal

studies aimed at elucidating the complex relationships between the FFM and
clinical disorders. Consistent research supports N as a vulnerability factor to
certain disorders, even sharing genetic etiology. However, the majority of this
research is in the area of mood and anxiety disorders. Expanding these studies to
include other forms of psychopathology could help identify common personality
vulnerabilities to psychopathology, as well as unique predictors of certain
constellations of symptoms.
Notes
1. When interpreting findings relevant to personality change, it is necessary to consider
the distinction between absolute stability and relative stability (De Fruyt et al., 2006;
Santor, Bagby, & Joffe, 1997). In this regard, though mean levels of personality
scores have been shown to significantly change in the advent of MDD (i.e., absolute
stability), individuals’ relative positions to others with respect to their personality
scores tend to be maintained (i.e., relative stability; De Fruyt et al., 2006). Zinbarg,
Uliaszek, and Adler (2008) interpreted findings supporting absolute change in
personality in the advent of MDD as reflecting a variable component of personality
and findings supporting the relative stability of personality as reflecting a component
of personality that naturally evolves throughout adulthood (Fraley & Roberts, 2005).
References
Adam, E. K., Sutton, J. M., Doane, L. D., & Mineka, S. (2008). Incorporating
hypothalamic-pituitary-adrenal axis measures into preventive interventions for
adolescent depression: Are we there yet? Development and Psychopathology, 20, 975–
1001.
Adam, E. K., Vrshek-Schallhorn, S., Kendall, A. D., Mineka, S., Zinbarg, R. E., &
Craske, M. G. (2014). Prospective associations between the cortisol awakening
response and first onsets of anxiety disorders over a six-year follow-up—2013 Curt
Richter Award Winner. Psychoneuroendocrinology, 44, 47–59.
Adewuya, A. O., Loto, O. M., & Adewumi, T. A. (2008). Premenstrual dysphoric
disorder amongst Nigerian university students: Prevalence, comorbid conditions, and
correlates. Archives of Women’s Mental Health, 11(1), 13–18.
Allen, M. H., Chessick, C. A., Miklowitz, D. J., Goldberg, J. F., Wisniewski, S. R.,
Miyahara, S., … Sachs, G. S. (2005). Contributors to suicidal ideation among bipolar
patients with and without a history of suicide attempts. Suicide and Life-Threatening
Behavior, 35(6), 671–680.
Anderson, K. W., & McLean, P. D. (1997). Conscientiousness in depression: Tendencies,
predictive utility, and longitudinal stability. Cognitive Therapy and Research, 21(2),
223–238.
Angst, F., Stassen, H. H., Clayton, P. J., & Angst, J. (2002). Mortality of patients with
mood disorders: Follow-up over 34–38 years. Journal of Affective Disorders, 68(2–3),
167–181.
Bagby, R. M., Bindseil, K. D., Schuller, D. R., Rector, N. A., Young, L. T., Cooke, R. G.,
… Joffe, R. T. (1997). Relationship between the five factor model of personality and
unipolar, bipolar and schizophrenic patients. Psychiatry Research, 70(2), 83–94.
Bagby, R. M., Joffe, R. T., Parker, J. D., Kalemba, V., & Harkness, K. L. (1995). Major
depression and the five factor model of personality. Journal of Personality Disorders,
9(3), 224–234.
Bagby, R. M., Levitan, R. D., Kennedy, S. H., Levitt, A. J., & Joffe, R. T. (1999).
Selective alteration of personality in response to noradrenergic and serotonergic
antidepressant medication in depressed sample: Evidence of non-specificity. Psychiatry
Research, 86(3), 211–216.
Bagby, R. M., Quilty, L. C., & Ryder, A. C. (2008). Personality and depression. The
Canadian Journal of Psychiatry, 53(1), 14–25.
Bagby, R. M., Quilty, L. C., Segal, Z. V., McBride, C. C., Kennedy, S. H., & Costa, P. T.
(2008). Personality and differential treatment response in major depression: A
randomized controlled trial comparing cognitive-behavioural therapy and
pharmacotherapy. Canadian Journal of Psychiatry, 53, 361–370.
Bagby, R. M., Schuller, D. R., Levitt, A. J., Joffe, R. T., & Harkness, K. L. (1996).
Seasonal and non-seasonal depression and the five factor model of personality. Journal
of Affective Disorders, 38(2), 89–95.
Bagby, R. M., Young, L. T., Schuller, D. R., Bindseil, K. D., Cooke, R. G., Dickens, S.
E., … Joffe, R. T. (1996). Bipolar disorder, unipolar depression and the five factor
model of personality. Journal of Affective Disorders, 41(1), 25–32.
Barlow, D. H. (2000). Unraveling the mysteries of anxiety and its disorders from the
perspective of emotion theory. American Psychologist, 55(11), 1247.
Barlow, D. H. (2004). Anxiety and its disorders: The nature and treatment of anxiety and
panic. New York, NY: The Guilford Press.
Beauchamp, M., Lecomte, T., Lecomte, C., Leclerc, C., & Corbière, M. (2006). Do
people with a first episode of psychosis differ in personality profiles? Schizophrenia
Research, 85(1–3), 162–167.
Beauchamp, M., Lecomte, T., Lecomte, C., Leclerc, C., & Corbière, M. (2011).
Personality traits in early psychosis: Relationship with symptom and coping treatment
outcomes. Early Intervention in Psychiatry, 5(1), 33–40.
Beauchamp, M., Lecomte, T., Lecomte, C., Leclerc, C., & Corbière, M. (2013). Do
personality traits matter when choosing a group therapy for early psychosis?
Psychology and Psychotherapy: Theory, Research and Practice, 86(1), 19–32.
Blom, M. B., Spinhoven, P., Hoffman, T., Jonker, K., Hoencamp, E., Haffmans, P. J., &
van Dyck, R. (2007). Severity and duration of depression, not personality factors,
predict short term outcome in the treatment of major depression. Journal of Affective
Disorders, 104(1), 119–126.
Bollen, E., & Wojciechowski, F. L. (2004). Anorexia nervosa subtypes and the big five
personality factors. European Eating Disorders Review, 12(2), 117–121.
Boyette, L., Korver-Nieberg, N., Verweij, K., Meijer, C., Dingemans, P., Cahn, W., & de
Haan, L. (2013). Associations between the five factor model personality traits and
psychotic experiences in patients with psychotic disorders, their siblings and controls.
Brown, T. A., & Barlow, D. H. (2009). A proposal for a dimensional classification system
based on the shared features of the DSM-IV anxiety and mood disorders: Implications
for assessment and treatment. Psychological Assessment, 21(3), 256.
Brown, T. A., & Naragon-Gainey, K. (2013). Evaluation of the unique and specific
contributions of dimensions of the triple vulnerability model to the prediction of DSM-
IV anxiety and mood disorder constructs. Behavior Therapy, 44(2), 277–292.
Bueller, J. A., Aftab, M., Sen, S., Gomez-Hassan, D., Burmeister, M., & Zubieta, J. K.
(2006). BDNF Val66Met allele is associated with reduced hippocampal volume in
healthy subjects. Biological Psychiatry, 59(9), 812–815.
Camisa, K. M., Bockbrader, M. A., Lysaker, P., Rae, L. L., Brenner, C. A., & O’Donnell,
B. F. (2005). Personality traits in schizophrenia and related personality disorders.
Canuto, A., Giannakopoulos, P., Meiler-Mititelu, C., Delaloye, C., Herrmann, F. R., &
Weber, K. (2009). Personality traits influence clinical outcome in day hospital-treated
elderly depressed patients. The American Journal of Geriatric Psychiatry, 17(4), 335–
343.
Carrera, M., Herran, A., Ramirez, M. L., Ayestaran, A., Sierra-Biddle, D., Hoyuela, F., …
Vázquez-Barquero, J. L. (2006). Personality traits in early phases of panic disorder:
Implications on the presence of agoraphobia, clinical severity and short-term outcome.
Acta Psychiatrica Scandinavica, 114(6), 417–425.
Cassin, S. E., & von Ranson, K. M. (2005). Personality and eating disorders: A decade in
review. Clinical Psychology Review, 25(7), 895–916.
Chopra, K. K., Bagby, R. M., Dickens, S., Kennedy, S. H., Ravindran, A., & Levitan, R.
D. (2005). A dimensional approach to personality in atypical depression. Psychiatry
Research, 134(2), 161–167.
Chung, M. C., Berger, Z., Jones, R., & Rudd, H. (2006). Posttraumatic stress disorder and
general health problems following myocardial infarction (Post-MI PTSD) among older
patients: The role of personality. International Journal of Geriatric Psychiatry, 21(12),
1163–1174.
Claes, L., Vandereycken, W., Luyten, P., Soenens, B., Pieters, G., & Vertommen, H.
(2006). Personality prototypes in eating disorders based on the big five model. Journal
of Personality Disorders, 20(4), 401–416.
Clark, A. A., & Owens, G. P. (2012). Attachment, personality characteristics, and
posttraumatic stress disorder in US veterans of Iraq and Afghanistan. Journal of
Traumatic Stress, 25(6), 657–664.
Clark, A., Tran, C., Weiss, A., Caselli, G., Nikčević, A. V., & Spada, M. M. (2012).
Personality and alcohol metacognitions as predictors of weekly levels of alcohol use in
binge drinking university students. Addictive Behaviors, 37(4), 537–540.
Cohen, R. B., Boggio, P. S., & Fregni, F. (2009). Risk factors for relapse after remission
with repetitive transcranial magnetic stimulation for the treatment of depression.
Depression and Anxiety, 26(7), 682–688.
Costa, P. T., Bagby, R. M., Herbst, J. H., & McCrae, R. R. (2005). Personality self-
Affective Disorders, 89(1), 45–55.
Costa, P. T., Jr., & McCrae, R. R. (1994). Set like plaster? Evidence for the stability of
adult personality. In T. F. Heatherton & J. L. Weinberger (Eds.), Can personality
change? (pp. 21–40). Washington, DC: American Psychological Association.
Costa, P. T., & McCrae, R. R. (1998). Manual supplement for the NEO-4. Odessa, FL:
Couture, S., Lecomte, T., & Leclerc, C. (2007). Personality characteristics and attachment
in first episode psychosis: Impact on social functioning. Journal of Nervous and Mental
Disease, 195(8), 631–639.
Davis, K. L., Kahn, R. S., Ko, G., & Davidson, M. (1991). Dopamine in schizophrenia: A
review and reconceptualization. The American Journal of Psychiatry, 148(11), 1474–
1486.
De Fruyt, F., van Leeuwen, K., Bagby, R. M., Rolland, J. P., & Rouillon, F. (2006).
Assessing and interpreting personality change and continuity in patients treated for
major depression. Psychological Assessment, 18(1), 71.
De Graaf, R., Bijl, R. V., Ravelli, A., Smit, F., & Vollebergh, W. A. M. (2002). Predictors
of first incidence of DSM-III-R psychiatric disorders in the general population:
Findings from the Netherlands Mental Health Survey and Incidence Study. Acta
Psychiatrica Scandinavica, 106(4), 303–313.
De Vriese, S. R., Christophe, A. B., & Maes, M. (2004). In humans, the seasonal
variation in poly-unsaturated fatty acids is related to the seasonal variation in violent
suicide and serotonergic markers of violent suicide. Prostaglandins, Leukotrienes, and
Essential Fatty Acids, 71(1), 13–18.
De Young, C. G., Cicchetti, D., Rogosch, F. A., Gray, J. R., Eastman, M., & Grigorenko,
dopamine system predicts Openness/Intellect. Journal of Research in Personality,
45(4), 364–371.
Du, L., Bakish, D., Ravindran, A. V., & Hrdina, P. D. (2002). Does fluoxetine influence
major depression by modifying five factor personality traits? Journal of Affective
Disorders, 71(1), 235–241.
Dugas, M. J., Gagnon, F., Ladouceur, R., & Freeston, M. H. (1998). Generalized anxiety
disorder: A preliminary test of a conceptual model. Behaviour Research and Therapy,
36(2), 215–226.
Ehlert, U., Gaab, J., & Heinrichs, M. (2001). Psychoneuroendocrinological contributions
to the etiology of depression, posttraumatic stress disorder, and stress-related bodily
disorders: The role of the hypothalamus-pituitary-adrenal axis. Biological Psychology,
57(1), 141–152.
Enoch, M. A., Xu, K., Ferro, E., Harris, C. R., & Goldman, D. (2003). Genetic origins of
anxiety in women: A role for a functional catechol-O-methyltransferase polymorphism.
Psychiatric Genetics, 13(1), 33–41.
Evans, S. J., Prossin, A. R., Harrington, G. J., Kamali, M., Ellingrod, V. L., Burant, C. F.,
& McInnis, M. G. (2012). Fats and factors: Lipid profiles associate with personality
factors and suicidal history in bipolar subjects. PLoS One, 7(1), e29297.
Fanous, A. H., Neale, M. C., Aggen, S. H., & Kendler, K. S. (2007). A longitudinal study
of personality and major depression in a population-based sample of male twins.
Psychological Medicine, 37(8), 1163–1172.
Feeley, M., De Rubeis, R. J., & Gelfand, L. A. (1999). The temporal relation of adherence
and alliance to symptom change in cognitive therapy for depression. Journal of
Consulting and Clinical Psychology, 67(4), 578.
Foti, D., Kotov, R., Bromet, E., & Hajcak, G. (2012). Beyond the broken error-related
negativity: Functional and diagnostic correlates of error processing in psychosis.
Fraley, R. C., & Roberts, B. W. (2005). Patterns of continuity: A dynamic model for
conceptualizing the stability of individual differences in psychological constructs
across the life course. Psychological Review, 112(1), 60.
Furukawa, T., Hori, S., Yoshida, S., Tsuji, M., Nakanishi, M., & Hamanaka, T. (1998).
Premorbid personality traits of patients with organic (ICD-10 F0), schizophrenic (F2),
mood (F3), and neurotic (F4) disorders according to the five factor model of
personality. Psychiatry Research, 78(3), 179–187.
Gatt, J. M., Clark, C. R., Kemp, A. H., Liddell, B. J., Dobson-Stone, C., Kuan, S. A., …
Williams, L. M. (2007). A genotype-endophenotype-phenotype path model of
depressed mood: Integrating cognitive and emotional markers. Journal of Integrative
Neuroscience, 6(1), 75–104.
Gerris, J. R. M., Houtmans, M. J. M., Kwaaitaal-Roosen, E. M. G., De Schipper, J. C.,
Vermulst, A. A., & Janssens, J. M. A. M. (1998). Parents, adolescents and young
adults in Dutch families: A longitudinal study. University of Nijmegen, Nijmegen, The
Netherlands: Institute of Family Studies.
Ghaderi, A., & Scott, B. (2000). The big five and eating disorders: A prospective study in
the general population. European Journal of Personality, 14(4), 311–323.
Gleeson, J. F., Rawlings, D., Jackson, H. J., & McGorry, P. D. (2005). Agreeableness and
neuroticism as predictors of relapse after first-episode psychosis: A prospective follow-
up study. The Journal of Nervous and Mental Disease, 193(3), 160–169.
Glinski, K., & Page, A. C. (2010). Modifiability of neuroticism, extraversion, and
agreeableness by group cognitive behaviour therapy for social anxiety disorder.
Behaviour Change, 27(01), 42–52.
Goldberg, L. R. (1992). The development of markers for the big-five factor structure.
measuring the lower level facets of several Five Factor Models. In I. Mervielde, I.
Goodwin, F. K., & Jamison, K. R. (2007). Manic-depressive illness: Bipolar disorders
and recurrent depression (2nd ed.). New York, NY: Oxford University Press.
Grant, B. F., Stinson, F. S., Dawson, D. A., Chou, S. P., Ruan, W. J., & Pickering, R. P.
(2004). Co-occurrence of 12-month alcohol and drug use disorders and personality
disorders in the United States: Results from the national epidemiologic survey on
alcohol and related conditions. Archives of General Psychiatry, 61(4), 361–368.
Griens, A. M. G. F., Jonker, K., Spinhoven, P. H., & Blom, M. B. J. (2002). The
influence of depressive state features on trait measurement. Journal of Affective
Disorders, 70(1), 95–99.
Gurrera, R. J., Nestor, P. G., & O’Donnell, B. F. (2000). Personality traits in
schizophrenia: Comparison with a community sample. Journal of Nervous and Mental
Disease, 188(1), 31–35.
Gurrera, R. J., Nestor, P. G., O’Donnell, B. F., Rosenberg, V., & McCarley, R. W.
(2005). Personality differences in schizophrenia are related to performance on
neuropsychological tasks. Journal of Nervous and Mental Disease, 193(11), 714–721.
Hammen, C. (1991). Generation of stress in the course of unipolar depression. Journal of
Abnormal Psychology, 100(4), 555.
Harcourt, L., Kirkby, K., Daniels, B., & Montgomery, I. (1998). The differential effect of
personality on computer-based treatment of agoraphobia. Comprehensive Psychiatry,
39(5), 303–307.
Hare, E., Contreras, J., Raventos, H., Flores, D., Jerez, A., Nicolini, H., … Escamilla, M.
(2012). Genetic structure of personality factors and bipolar disorder in families
segregating bipolar disorder. Journal of Affective Disorders, 136(3), 1027–1033.
Assessment, 7(1), 104.
Harris, S. E., Wright, A. F., Hayward, C., Starr, J. M., Whalley, L. J., & Deary, I. J.
(2005). The functional COMT polymorphism, Val158Met, is associated with logical
memory and the personality trait intellect/imagination in a cohort of healthy 79-year-
olds. Neuroscience Letters, 385(1), 1–6.
Hees, H. L., Koeter, M. W., & Schene, A. H. (2012). Predictors of long-term return to
work and symptom remission in sick-listed patients with major depression. The Journal
of Clinical Psychiatry, 73(8), e1048–e1055.
Herbert, J. (2013). Cortisol and depression: Three questions for psychiatry. Psychological
Medicine, 43(3), 449–469.
Hiio, K., Merenäkk, L., Nordquist, N., Parik, J., Oreland, L., Veidebaum, T., & Harro, J.
(2011). Effects of serotonin transporter promoter and BDNF Val66Met genotype on
personality traits in a population representative sample of adolescents. Psychiatric
Genetics, 21(5), 261–264.
Hirschfeld, R. M., Klerman, G. L., Clayton, P. J., & Keller, M. B. (1983). Personality and
depression: Empirical findings. Archives of General Psychiatry, 40(9), 993–998.
Hoekstra, H. A., Ormel, J., & De Fruyt, F. (1996). NEO persoonlijkheids vragenlijsten:
NEO-PI-R: NEO-FFI. Swets Test Services (STS).
Hong, R. Y., & Paunonen, S. V. (2011). Personality vulnerabilities to psychopathology:
Relations between trait structure and affective-cognitive processes. Journal of
Personality, 79(3), 527–562.
Hopwood, C. J., Morey, L. C., Skodol, A. E., Stout, R. L., Yen, S., Ansell, E. B., …
McGlashan, T. H. (2007). Five Factor model personality traits associated with alcohol-
related diagnoses in a clinical sample. Journal of Studies on Alcohol and Drugs, 68(3),
455–460.
Hoth, K. F., Paul, R. H., Williams, L. M., Dobson-Stone, C., Todd, E., Schofield, P. R.,
… Gordon, E. (2006). Associations between the COMT Val/Met polymorphism, early
life stress, and personality among healthy adults. Neuropsychiatric Disease and
Treatment, 2(2), 219.
Huan, M., Hamazaki, K., Sun, Y., Itomura, M., Liu, H., Kang, W., … Hamazaki, T.
(2004). Suicide attempt and n-3 fatty acid levels in red blood cells: A case control
study in china. Biological Psychiatry, 56(7), 490–496.
Jain, U., Blais, M. A., Otto, M. W., Hirshfeld, D. R., & Sachs, G. S. (1999). Five Factor
personality traits in patients with seasonal depression: Treatment effects and
comparisons with bipolar patients. Journal of Affective Disorders, 55(1), 51–54.
Jakšić, N., Brajković, L., Ivezić, E., Topić, R., & Jakovljević, M. (2012). The role of
personality traits in posttraumatic stress disorder (PTSD). Psychiatria Danubina, 24(3),
256–266.
Joffe, R. T., Gatt, J. M., Kemp, A. H., Grieve, S., Dobson-Stone, C., Kuan, S. A., …
Williams, L. M. (2009). Brain derived neurotrophic factor Val66Met polymorphism,
the five factor model of personality and hippocampal volume: Implications for
depressive illness. Human Brain Mapping, 30(4), 1246–1256.
Joffe, R. T., & Regan, J. J. (1988). Personality and depression. Journal of Psychiatric
Research, 22(4), 279–286.
Johansen, R., Melle, I., Iversen, V. C., & Hestad, K. (2013). Personality traits,
interpersonal problems and therapeutic alliance in early schizophrenia spectrum
disorders. Comprehensive Psychiatry, 54(8), 1169–1176.
Johnson, H., & Thompson, A. (2008). The development and maintenance of post-
traumatic stress disorder (PTSD) in civilian adult survivors of war trauma and torture:
A review. Clinical Psychology Review, 28(1), 36–47.
Juhasz, G., Chase, D., Pegg, E., Downey, D., Toth, Z. G., Stones, K., … Deakin, J. W.
(2009). CNR1 gene is associated with high neuroticism and low agreeableness and
interacts with recent negative life events to predict current depressive symptoms.
Neuropsychopharmacology, 34(8), 2019–2027.
Kamphuis, J. H., Emmelkamp, P. M., & Bartak, A. (2003). Individual differences in post-
traumatic stress following post-intimate stalking: Stalking severity and psychosocial
variables. British Journal of Clinical Psychology, 42(2), 145–156.
Kanninen, K., Punamäki, R. L., & Qouta, S. (2003). Personality and trauma: Adult
attachment and posttraumatic distress among former political prisoners. Peace and
Conflict: Journal of Peace Psychology, 9(2), 97.
Karsten, J., Penninx, B. W., Riese, H., Ormel, J., Nolen, W. A., & Hartman, C. A. (2012).
The state effect of depressive and anxiety disorders on big five personality traits.
Journal of Psychiatric Research, 46(5), 644–650.
depression: A Swedish longitudinal, population-based twin study. Archives of General
Psychiatry, 63(10), 1113–1120.
Kendler, K. S., & Myers, J. (2010). The genetic and environmental relationship between
major depression and the five factor model of personality. Psychological Medicine,
40(5), 801–806.
Kendler, K. S., Neale, M. C., Kessler, R. C., Heath, A. C., & Eaves, L. J. (1993). A
longitudinal twin study of personality and major depression in women. Archives of
Kentros, M., Smith, T. E., Hull, J., McKee, M., Terkelsen, K., & Capalbo, C. (1997).
Stability of personality traits in schizophrenia and schizoaffective disorder: A pilot
project. Journal of Nervous and Mental Disease, 185(9), 549–555.
Kessler, R. C., Sonnega, A., Bromet, E., Hughes, M., & Nelson, C. B. (1995).
Posttraumatic stress disorder in the National Comorbidity Survey. Archives of General
Psychiatry, 52(12), 1048–1060.
Kim, B., Joo, Y. H., Kim, S. Y., Lim, J., & Kim, E. O. (2011). Personality traits and
affective morbidity in patients with bipolar I disorder: The five factor model
perspective. Psychiatry Research, 185(1), 135–140.
Kirschbaum, C., & Helhammer, D. (2000). Salivary cortisol. In G. Fink (Ed.),
Encyclopedia of stress (pp. 379–383). San Diego, CA: Academic Press.
Klein, D. N., Kotov, R., & Bufferd, S. J. (2011). Personality and depression: Explanatory
models and review of the evidence. Annual Review of Clinical Psychology, 7, 269.
Klein, D. N., Schwartz, J. E., Santiago, N. J., Vivian, D., Vocisano, C., Castonguay, L.
G., … Keller, M. B. (2003). Therapeutic alliance in depression treatment: Controlling
for prior change and patient characteristics. Journal of Consulting and Clinical
Psychology, 71(6), 997.
Knutson, B., Burgdorf, J., & Panksepp, J. (1998). Anticipation of play elicits high-
frequency ultrasonic vocalizations in young rats. Journal of Comparative Psychology,
112(1), 65.
Koszewska, I., & Rybakowski, J. K. (2008). High neuroticism (measured by NEO-FFI) in
bipolar disorder is associated with mixed state but not with rapid cycling. Archives of
Psychiatry and Psychotherapy, 4, 21–25.
Bulletin, 136(5), 768.
Krabbendam, L., Janssen, I., Bak, M., Bijl, R. V., de Graaf, R., & van Os, J. (2002).
Neuroticism and low self-esteem as risk factors for psychosis. Social Psychiatry and
Psychiatric Epidemiology, 37(1), 1–6.
Krueger, R. F., Derringer, J., Markon, K. E., Watson, D., & Skodol, A. V. (2012). Initial
Psychological Medicine, 42(9), 1879.
Kuijer, R. G., Marshall, E. M., & Bishop, A. N. (2014). Prospective predictors of short-
term adjustment after the Canterbury earthquakes: Personality and depression.
Psychological Trauma: Theory, Research, Practice, and Policy, 6(4), 361.
Kushner, S. C., Quilty, L. C., Uliaszek, A., McBride, C., & Bagby, R. M. (2015).
Therapeutic alliance mediates the association between personality and treatment
outcome in patients with major depressive disorder. Unpublished manuscript.
La Fauci-Schutt, J. M., & Marotta, S. A. (2011). Personal and environmental predictors of
posttraumatic stress in emergency management professionals. Psychological Trauma:
Theory, Research, Practice, and Policy, 3(1), 8.
La Salle-Ricci, V. H., Arnkoff, D. B., Glass, C. R., Crawley, S. A., Ronquillo, J. G., &
Murphy, D. L. (2006). The hoarding dimension of OCD: Psychological comorbidity
and the five factor personality model. Behaviour Research and Therapy, 44(10), 1503–
1512.
Lecomte, T., Spidel, A., Leclerc, C., MacEwan, G. W., Greaves, C., & Bentall, R. P.
(2008). Predictors and profiles of treatment non-adherence and engagement in services
problems in early psychosis. Schizophrenia Research, 102(1), 295–302.
Lee, J. E., Sudom, K. A., & Zamorski, M. A. (2013). Longitudinal analysis of
psychological resilience and mental health in Canadian military personnel returning
from overseas deployment. Journal of Occupational Health Psychology, 18(3), 327.
Lehto, K., Akkermann, K., Parik, J., Veidebaum, T., & Harro, J. (2013). Effect of COMT
Val158Met polymorphism on personality traits and educational attainment in a
longitudinal population representative study. European Psychiatry, 28(8), 492–498.
Liebowitz, M. R., Stallone, F., Dunner, D. L., & Fieve, R. F. (1979). Personality features
of patients with primary affective disorder. Acta Psychiatrica Scandinavica, 60(2),
214–224.
Loehlin, J. C. (1992). Genes and environment in personality development. Thousand
Oaks, CA: Sage Publications.
Loehlin, J. C., McCrae, R. R., Costa, P. T., & John, O. P. (1998). Heritabilities of
common and measure-specific components of the Big Five personality factors. Journal
Lysaker, P. H., & Davis, L. W. (2004). Social function in schizophrenia and
schizoaffective disorder: Associations with personality, symptoms and neurocognition.
Health and Quality of Life Outcomes, 2(1), 15.
Lysaker, P. H., Lancaster, R. S., Nees, M. A., & Davis, L. W. (2003). Neuroticism and
visual memory impairments as predictors of the severity of delusions in schizophrenia.
Malouff, J. M., Thorsteinsson, E. B., Rooke, S. E., & Schutte, N. S. (2007). Alcohol
involvement and the five factor model of personality: A meta-analysis. Journal of Drug
Education, 37(3), 277–294.
Marangell, L. B., Bauer, M. S., Dennehy, E. B., Wisniewski, S. R., Allen, M. H.,
Miklowitz, D. J., … Thase, M. E. (2006). Prospective predictors of suicide and suicide
attempts in 1,556 patients with bipolar disorders followed for up to 2 years. Bipolar
Disorders, 8(5), 566–575.
Margoob, M. A., & Mushtaq, D. (2011). Serotonin transporter gene polymorphism and
psychiatric disorders: Is there a link? Indian Journal of Psychiatry, 53(4), 289.
McCrae, R. R., & Costa, P. T., Jr. (1997). Conceptions and correlates of openness to
experience. In R. Hogan, J. A. Johnson, & S. R. Briggs (Eds.), Handbook of
personality psychology (pp. 825–847). San Diego, CA: Academic Press.
McCrae, R. R., & Costa, P. T., Jr. (1999). A five factor theory of personality. In L. A.
Pervin & O. P. John (Eds.), Handbook of personality: Theory and research (Vol. 2, pp.
Mezquita, L., Stewart, S. H., & Ruipérez, M. (2010). Big-five personality domains predict
internal drinking motives in young adults. Personality and Individual Differences,
49(3), 240–245.
Nemoto, K., Ohnishi, T., Mori, T., Moriguchi, Y., Hashimoto, R., Asada, T., & Kunugi,
H. (2006). The Val66Met polymorphism of the brain-derived neurotrophic factor gene
affects age-related brain morphology. Neuroscience Letters, 397(1), 25–29.
Nightingale, J., & Williams, R. M. (2000). Attitudes to emotional expression and
personality in predicting post-traumatic stress disorder. British Journal of Clinical
Psychology, 39(3), 243–254.
O’Donnell, B. F., Salisbury, D. F., Niznikiewicz, M., Brenner, C., & Vohs, J. L. (2011).
Abnormalities of event-related potential components in schizophrenia. In S. J. Luck &
E. S. Kappenman (Eds.), Oxford handbook of event-related potential components (pp.
537–562). Oxford, England: Oxford University Press.
Orlinsky, D. E., Grawe, K., & Parks, B. K. (1994). Process and outcome in
psychotherapy: Noch einmal. In A. E. Bergin & S. L. Garfield (Eds.), Handbook of
psychotherapy and behavior change (4th ed., pp. 270–376). Oxford, England: John
Wiley & Sons.
Ormel, J. (1980). Moeite met leven of een moeilijk leven: Een vervolgonderzoek naar de
invloed van psychosociale belasting op het welbevinden van driehonderd
Nederlanders. Groningen, The Netherlands: Konstapel.
Ormel, J., Bastiaansen, A., Riese, H., Bos, E. H., Servaas, M., Ellenbogen, M., …
Aleman, A. (2013). The biological and psychological basis of neuroticism: Current
status and future directions. Neuroscience & Biobehavioral Reviews, 37(1), 59–72.
Ormel, J., Oldehinkel, A. J., & Vollebergh, W. (2004). Vulnerability before, during, and
after a major depressive episode: A 3-wave population-based study. Archives of
Petersen, T., Papakostas, G. I., Bottonari, K., Iacoviello, B., Alpert, J. E., Fava, M., &
Nierenberg, A. A. (2002). NEO-FFI factor scores as predictors of clinical response to
fluoxetine in depressed outpatients. Psychiatry Research, 109(1), 9–16.
Pezawas, L., Verchinski, B. A., Mattay, V. S., Callicott, J. H., Kolachana, B. S., Straub,
R. E., … Weinberger, D. R. (2004). The brain-derived neurotrophic factor val66met
polymorphism and variation in human cortical morphology. The Journal of
Neuroscience, 24(45), 10099–10102.
Pillmann, F., Balzuweit, S., Haring, A., Blöink, R., & Marneros, A. (2003). Suicidal
behavior in acute and transient psychotic disorders. Psychiatry Research, 117(3), 199–
209.
Podar, I., Hannus, A., & Allik, J. (1999). Personality and affectivity characteristics
associated with eating disorders: A comparison of eating disordered, weight-
preoccupied, and normal samples. Journal of Personality Assessment, 73(1), 133–147.
Quilty, L. C., De Fruyt, F., Rolland, J. P., Kennedy, S. H., Rouillon, P. F., & Bagby, R.
M. (2008). Dimensional personality traits and treatment outcome in patients with major
depressive disorder. Journal of Affective Disorders, 108(3), 241–250.
Quilty, L. C., Sellbom, M., Tackett, J. L., & Bagby, R. M. (2009). Personality trait
predictors of bipolar disorder symptoms. Psychiatry Research, 169(2), 159–163.
Quirk, S. W., & McCormick, R. A. (1998). Personality subtypes, coping styles, symptom
correlates, and substances of choice among a cohort of substance abusers. Assessment,
5(2), 157–169.
Rector, N. A., Bagby, R. M., Huta, V., & Ayearst, L. E. (2012). Examination of the trait
facets of the five factor model in discriminating specific mood and anxiety disorders.
Rector, N. A., Hood, K., Richter, M. A., & Bagby, R. M. (2002). Obsessive-compulsive
disorder and the five factor model of personality: Distinction and overlap with major
depressive disorder. Behaviour Research and Therapy, 40(10), 1205–1219.
Rector, N. A., Richter, M. A., & Bagby, R. M. (2005). The impact of personality on
symptom expression in obsessive-compulsive disorder. The Journal of Nervous and
Mental Disease, 193(4), 231–236.
Rees, C. S., Anderson, R. A., & Egan, S. J. (2006). Applying the five factor model of
personality to the exploration of the construct of risk-taking in obsessive-compulsive
disorder. Behavioural and Cognitive Psychotherapy, 34(1), 31–42.
Psychological Bulletin, 132(1), 1.
Rosellini, A. J., & Brown, T. A. (2011). The NEO Five Factor Inventory: Latent structure
and relationships with dimensions of anxiety and depressive disorders in a large
clinical sample. Assessment, 18(1), 27–38.
Ross, S. R., Lutz, C. J., & Bailley, S. E. (2002). Positive and negative symptoms of
schizotypy and the five factor model: A domain and facet level analysis. Journal of
Santor, D. A., Bagby, R. M., & Joffe, R. T. (1997). Evaluating stability and change in
personality and depression. Journal of Personality and Social Psychology, 73(6), 1354.
Schomerus, G., Appel, K., Meffert, P. J., Luppa, M., Andersen, R. M., Grabe, H. J., &
Baumeister, S. E. (2013). Personality-related factors as predictors of help-seeking for
depression: A population-based study applying the behavioral model of health services
use. Social Psychiatry and Psychiatric Epidemiology, 48(11), 1809–1817.
Sen, S., Nesse, R. M., Stoltenberg, S. F., Li, S., Gleiberman, L., Chakravarti, A., …
Burmeister, M. (2003). A BDNF coding variant is associated with the NEO personality
inventory domain neuroticism, a risk factor for depression. Neuropsychopharmacology
28, 397–401.
Silvestri, A. J. (2005). REM sleep deprivation affects extinction of cued but not
contextual fear conditioning. Physiology & Behavior, 84(3), 343–349.
Smits, J. A., Rosenfield, D., Otto, M. W., Powers, M. B., Hofmann, S. G., Telch, M. J.,
… Tart, C. D. (2013). D-Cycloserine enhancement of fear extinction is specific to
successful exposure sessions: Evidence from the treatment of height phobia. Biological
Psychiatry, 73(11), 1054–1058.
Soliman, F., Glatt, C. E., Bath, K. G., Levita, L., Jones, R. M., Pattwell, S. S., … Casey,
B. J. (2010). A genetic variant BDNF polymorphism alters extinction learning in both
mouse and human. Science, 327(5967), 863–866.
Spinhoven, P., de Rooij, M., Heiser, W., Smit, J. H., & Penninx, B. W. (2012).
Personality and changes in comorbidity patterns among anxiety and depressive
disorders. Journal of Abnormal Psychology, 121(4), 874.
Spinhoven, P., Penelo, E., de Rooij, M., Penninx, B. W., & Ormel, J. (2014). Reciprocal
effects of stable and temporary components of neuroticism and affective disorders:
Results of a longitudinal cohort study. Psychological Medicine, 44(2), 337–348.
Spinhoven, P., van der Does, W., Ormel, J., Zitman, F. G., & Penninx, B. W. (2013).
Confounding of Big Five personality assessments in emotional disorders by
comorbidity and current disorder. European Journal of Personality, 27(4), 389–397.
Spoormaker, V. I., Sturm, A., Andrade, K. C., Schröter, M. S., Goya-Maldonado, R.,
Holsboer, F., … Czisch, M. (2010). The neural correlates and temporal sequence of the
relationship between shock exposure, disturbed sleep and impaired consolidation of
fear extinction. Journal of Psychiatric Research, 44(16), 1121–1128.
Srivastava, S., John, O. P., Gosling, S. D., & Potter, J. (2003). Development of
personality in early and middle adulthood: Set like plaster or persistent change?
Journal of Personality and Social Psychology, 84(5), 1041.
Sullivan, P. F., Neale, M. C., & Kendler, K. S. (2000). Genetic epidemiology of major
depression: Review and meta-analysis. American Journal of Psychiatry, 157(10),
1552–1562.
Suls, J., & Martin, R. (2005). The daily life of the garden-variety neurotic: Reactivity,
stressor exposure, mood spillover, and maladaptive coping. Journal of Personality,
73(6), 1485–1510.
Szeszko, P. R., Lipsky, R., Mentschel, C., Robinson, D., Gunduz-Bruce, H., Sevy, S., …
Malhotra, A. K. (2005). Brain-derived neurotrophic factor val66met polymorphism and
volume of the hippocampal formation. Molecular Psychiatry, 10(7), 631–636.
dimensional model of personality disorder: Incorporating DSM cluster A
characteristics. Journal of Abnormal Psychology, 117(2), 454–459.
Takano, A., Arakawa, R., Hayashi, M., Takahashi, H., Ito, H., & Suhara, T. (2007).
Relationship between neuroticism personality trait and serotonin transporter binding.
Talbert, F. S., Braswell, L. C., Albrecht, I. W., Hyer, L. A., & Boudewyns, P. A. (1993).
NEO-PI profiles in PTSD as a function of trauma level. Journal of Clinical
Psychology, 49(5), 663–669.
Tang, T. Z., DeRubeis, R. J., Hollon, S. D., Amsterdam, J., Shelton, R., & Schalet, B.
(2009). Personality change during depression treatment: A placebo-controlled trial.
Archives of General Psychiatry, 66(12), 1322–1330.
Ter Bogt, Tom F. M., Engels, R. C. M. E., & Dubas, J. S. (2006). Party people:
Personality and MDMA use of house party visitors. Addictive Behaviors, 31(7), 1240–
1244.
Terracciano, A., Löckenhoff, C. E., Crum, R. M., Bienvenu, O. J., & Costa, P. T., Jr.
(2008). Five Factor model personality profiles of drug users. BMC Psychiatry, 8(1), 22.
Thase, M. E., Rush, A. J., Howland, R. H., Kornstein, S. G., Kocsis, J. H., Gelenberg, A.
J., … Harrison, W. (2002). Double-blind switch study of imipramine or sertraline
treatment of antidepressant-resistant chronic depression. Archives of General
Psychiatry, 59(3), 233–239.
Uliaszek, A. A., Al-Dajani, N., & Bagby, R. M. (2014). The relationship between
psychopathology and a hierarchical model of normal personality traits: Evidence from
a large psychiatric patient sample. Journal of Personality Disorders, 6, 1–16.
Uliaszek, A. A., Hauner, K. K., Zinbarg, R. E., Craske, M. G., Mineka, S., Griffith, J. W.,
& Rose, R. D. (2009). An examination of content overlap and disorder-specific
predictions in the associations of neuroticism with anxiety and depression. Journal of
Uliaszek, A. A., Zinbarg, R. E., Mineka, S., Craske, M. G., Sutton, J. M., Griffith, J. W.,
… Hammen, C. (2010). The role of neuroticism and extraversion in the stress-anxiety
and stress-depression relationships. Anxiety, Stress, & Coping, 23(4), 363–381.
van Emmerik, A. A., Kamphuis, J. H., Noordhof, A., & Emmelkamp, P. M. (2011). Catch
me if you can: Do the Five Factor Model personality traits moderate dropout and acute
treatment response in post-traumatic stress disorder patients? Psychotherapy and
Psychosomatics, 80(6), 386–388.
Vrshek-Schallhorn, S., Czarlinski, J., Mineka, S., Zinbarg, R. E., & Craske, M. (2011).
Prospective predictors of suicidal ideation during depressive episodes among older
adolescents and young adults. Personality and Individual Differences, 50(8), 1202–
1207.
Walton, K. E., & Roberts, B. W. (2004). On the relationship between substance use and
personality traits: Abstainers are not maladjusted. Journal of Research in Personality,
38(6), 515–535.
Wang, Y., Wang, H., Wang, J., Wu, J., & Liu, X. (2013). Prevalence and predictors of
posttraumatic growth in accidentally injured patients. Journal of Clinical Psychology in
Medical Settings, 20(1), 3–12.
Ware, J. E., Jr., & Sherbourne, C. D. (1992). The MOS 36-item short-form health survey
(SF-36): I. Conceptual framework and item selection. Medical Care, 30(6), 473–483.
Watanabe, A., Nakao, K., Tokuyama, M., & Takeda, M. (2005). Prediction of first
episode of panic attack among white-collar workers. Psychiatry and Clinical
Neurosciences, 59(2), 119–126.
Watson, D., & Naragon-Gainey, K. (2014). Personality, emotions, and the emotional
disorders. Clinical Psychological Science, 2(4), 422–442.
Weber, K., Giannakopoulos, P., Delaloye, C., De Bilbao, F., Moy, G., Moussa, A., …
Canuto, A. (2010). Volumetric MRI changes, cognition and personality traits in old age
depression. Journal of Affective Disorders, 124(3), 275–282.
Weiss, A., Sutin, A. R., Duberstein, P. R., Friedman, B., Bagby, R. M., & Costa, P. T.
(2009). The personality domains and styles of the five factor model are related to
incident depression in Medicare recipients aged 65 to 100. The American Journal of
Geriatric Psychiatry, 17(7), 591–601.
Wessa, M., & Flor, H. (2007). Failure of extinction of fear responses in posttraumatic
stress disorder: Evidence from second-order conditioning. The American Journal of
Psychiatry, 164(11), 1684–1692.
Wetzler, S., Kahn, R. S., Cahn, W., van Praag, H. M., & Asnis, G. M. (1990).
Psychological test characteristics of depressed and panic patients. Psychiatry Research,
31(2), 179–192.
Widiger, T. A., & Smith, G. T. (2008). Personality and psychopathology. In Handbook of
Press.
Yu, H., Wang, Y., Pattwell, S., Jing, D., Liu, T., Zhang, Y., … Chen, Z. Y. (2009).
Variant BDNF Val66Met polymorphism affects extinction of conditioned aversive
memory. The Journal of Neuroscience, 29(13), 4056–4064.
Yuan, C., Wang, Z., Inslicht, S. S., McCaslin, S. E., Metzler, T. J., Henn-Haase, C., …
Marmar, C. R. (2011). Protective factors for posttraumatic stress disorder symptoms in
a prospective study of police officers. Psychiatry Research, 188(1), 45–50.
Zinbarg, R. E., Uliaszek, A. A., & Adler, J. M. (2008). The role of personality in
psychotherapy for anxiety and depression. Journal of Personality, 76(6), 1649–1688.
The Five Factor Model of Personality and
Consequential Outcomes in Childhood and
Adolescence
Filip De Fruyt, Barbara De Clercq, and Marleen De Bolle
Abstract
The validity of the Five Factor Model (FFM) to describe personality differences
in childhood and adolescence is well established. Personality differences can be
reliably assessed in children and adolescents, and available research converges on
the validity of the FFM as the predominant model to provide a comprehensive and
manageable account of these notable differences. In addition, there is strong
agreement that personality traits in childhood/adolescence are related to a broad
range of short- and long-term consequential outcomes, underscoring their utility
in research and assessment. The aims of the present chapter are threefold: first, to
review FFM measures developed for children and adolescents; second, to discuss
parallels and dissimilarities obtained with adults; and finally to summarize the
significance and validity of the model in gaining an understanding of a broad
series of outcomes, including interpersonal relationships, psychopathology, health
and well-being, learning and learning outcomes, and long-term outcomes
manifested in adulthood.
Key Words: childhood, adolescence, measures, outcomes, relationships,
psychopathology, health, education
The validity of the Five Factor Model (FFM) in describing personality

differences in childhood and adolescence is now established. The developmental
roots of this research approach were formed together with the construction of
measures to assess the “Big Five”—or the FFM—in youth, measures that relied
on primary caregivers and teachers as knowledgeable informants of personality.
Eventually those assessment measures became supplemented with self-
descriptions of adolescents themselves. This evolving research domain
developed into a cumulative knowledge base that has been systematically
compared to findings obtained in adults, pointing to personality similarities as
well as differences across developmental stages of life (see also the chapter by
De Pauw). From the resulting evidence it is clear that personality differences can
be reliably assessed in children and adolescents. Furthermore, the research
demonstrates that the FFM is the model par excellence for providing a
comprehensive and manageable account of notable personality differences.
Finally, there is strong agreement that personality traits in childhood/adolescence
are related to a broad range of short- and long-term consequential behaviors,
underscoring the utility of those traits in research and assessment. The aim of the
present chapter is threefold: to review FFM measures developed for children and
adolescents, to discuss parallels and dissimilarities obtained in research with
adults, and finally to summarize the significance and validity of the model to
gain understanding of a broad series of outcomes, including interpersonal
relationships, psychopathology, health and well-being, learning and
implementation of learning, and long-term behaviors manifested in adulthood.
FFM Measures for Childhood and Adolescence

Although the study of individual differences in early childhood was mainly
the domain of temperament researchers through the first six decades of the
twentieth century, John Digman (1963) was among the first personality
psychologists to examine the structure of teacher ratings providing initial
evidence that the Big Five accounted for personality differences at much
younger ages than had previously been thought. Digman (1963) examined the
structure of teacher judgments of 102 first- and second-grade children on 39
scales using a Q-sort procedure and retained 11 factors labeled: superego
strength, hostility versus social integration, surgent excitement versus desurgent
apathy, autia (aesthetically sensitive and imaginative), general intelligence,
premsia versus harria (emotional distress), sex, social confidence, neuroticism
versus ego strength, an unidentified factor, and finally parental attitude:
harshness versus permissiveness. From hindsight, it is clear that Digman
probably retained too many factors, hampering replicability. In subsequent work,
Digman and Inouye (1986) analyzed teachers’ Q-sort ratings of 499 sixth-grade
children on 43 scales, this time extracting fewer factors (7), five of which were
clearly indicative of the Big Five—i.e., introversion-extraverion (surgency),
consideration and conscientiousness (agreeableness), will (to achieve),
emotional stability, and intellect (embracing also openness to experience). The
other two factors were an error factor loaded marginally by verbal expression
items and a doublet factor loaded by the items imaginative and esthetically
sensitive. The seminal work by Digman and his colleagues was followed by a
stream of top-down research studies in which measures primarily developed for
assessment with adults were used to assess the personality of younger age
groups, eventually after slight item modifications to make them more
appropriate.
John, Donahue, and Kentle (1991) developed the Big Five Inventory (BFI), a
44-item set assessing the Big Five. Given its fifth-grade reading level, the BFI
was also used to describe the personality of children from 10 years of age
onward (Soto, John, Gosling, & Potter, 2008), and it was used to study
personality development from age 10 to age 65 (Soto, John, Gosling, & Potter,
2011). Mervielde, Buyst, and De Fruyt (1995) used the Big 5 Bipolaire
Beoordelingschalen 25 (B5-BBS25; Mervielde, 1992) in their study of the
structure and validity of teacher personality descriptions of 4- to 12-year-olds to
predict grade point average (GPA). A four-factor structure, blending
conscientiousness with adjectives tapping into intellect/openness, supplemented
with the remaining Big Five factors, was found at kindergarten age, whereas the
full five factor structure emerged from primary school age onward, with
increasing predictive validity throughout primary school. Barbaranelli, Caprara,
Rabasca, and Pastorelli (2003) re-designed the adult Big Five Questionnaire
(BFQ; Caprara, Barbaranelli, Borgogni, & Perugini, 1993), expanding its
coverage with trait adjectives that were judged by parents and teachers as
suitable to denote personality differences observable in children and further
translated into short behaviorally oriented phrases, into the Big Five
Questionnaire—Children version (BFQ-C; Barbaranelli et al., 2003). They
analyzed descriptions provided by the children themselves, their teachers, and
their parents and demonstrated that the Big Five showed up in all rating sets.
Conscientiousness was a consistent predictor of academic achievement;
externalizing problems were associated with low conscientiousness and low
emotional stability, whereas low emotional stability was associated with
internalizing problems, as assessed with the Child Behavior Checklist (CBCL;
Achenbach, 1991). Finally, John, Caspi, Robins, Moffitt, and Stouthamer-
Loeber (1994) developed a set of markers for the FFM from the California Child
Q-set, and they also examined relationships with the internalizing and
externalizing syndrome scales of the CBCL.
Additional evidence for the validity of the FFM in childhood and adolescence
came from a bottom-up approach toward personality description, examining the
content, structure, and validity of parental free descriptions of children’s
personality conducted in multiple countries, including Belgium, China,
Germany, Greece, the Netherlands, Poland, and the United States (Kohnstamm,
Halverson, Mervielde, & Havill, 1998). This approach led to the development of
two inventories to assess higher- and lower-order traits in children, i.e. the
Hierarchical Personality Inventory for Children (HiPIC; Mervielde & De Fruyt,
1999; Mervielde, De Fruyt, & De Clercq, 2009) and the Inventory for Childhood
Individual Differences (ICID; Halverson et al., 2003). The HiPIC assesses five
higher-order factors: emotional stability, extraversion, imagination, benevolence,
and conscientiousness, and 18 lower-order traits, whereas the labels for the ICID
measure parallel the FFM nomenclature. Both bottom-up research lines
converged with the results from the top-down analyses proposing the FFM
domain traits as key dimensions to describe a youth’s personality. In addition,
this work contributed to establishing a lower-order structure of the FFM in
childhood, enabling more refined predictions. A comparison of the domain and
facet scales of both measures is provided in Table 22.1, which further lists the
domains and facets from a re-analysis of Digman’s original work (Goldberg,
2001), showing that there is also some convergence about what traits are salient
at the lower-order trait level in childhood. Tackett, Kushner, De Fruyt, and
Mervielde (2013) recently examined convergence among the HiPIC and the
short version of the ICID (Deal, Halverson, Martin, Victor, & Baker, 2007),
further examining their validity as predictors of internalizing and externalizing
psychopathology syndromes.
Age and Sex Differences

Age and sex differences have been well-documented for adults (Costa,
Terracciano, & McCrae, 2001; McCrae et al., 1999; McCrae & Terracciano,
2005; Roberts & DelVecchio, 2000) but were only recently systematically and
comprehensively examined for childhood and adolescence. Given the validity of
the FFM dimensions across developmental stages, the model enables
examination of age and sex differences at different time points across the life
cycle, both cross-sectionally and longitudinally. The research in adults has
demonstrated that some facets show opposite age or gender effects from the
factor they define—see, e.g., Roberts, Walton, & Viechtbauer, 2006;
Terracciano, McCrae, Brant, & Costa, 2005—indicating the need to supplement
domain-level research with a facet-level analysis to allow comprehensive
understanding of childhood and adolescent age and gender differences in
personality.
Age Differences
Both cross-sectional and longitudinal studies of adults generally agree in
showing declines in neuroticism, extraversion, and openness to experience, and
increases in agreeableness and conscientiousness from the mid-20s onward
(Roberts & DelVecchio, 2000). In childhood and adolescence, individuals
generally decrease in neuroticism, while increasing in agreeableness,
conscientiousness, and openness to experience (Branje, Van Lieshout, & Gerris,
2007; De Bolle et al., 2016; Klimstra, Hale, Raaijmakers, Branje, & Meeus,
2009). A somewhat divergent age trend for extraversion emerges from the
literature for childhood and adolescence, possibly due to divergent age trends for
the facets that make up extraversion (Roberts et al., 2006). More specifically,
declining age trends are observed for extraversion cross-sectionally and cross-
culturally (De Bolle et al., 2016; Soto et al., 2011), whereas other longitudinal
studies either could not detect age effects for extraversion or found increases
(Branje et al., 2007; Klimstra et al., 2009).
In childhood and adolescence, age effects of the facets generally follow those
of the domain they belong to in terms of direction, but some deviations from this
general pattern are observed, especially for extraversion and openness to
experience. For extraversion facets, young individuals around the world are
found to show increases in warmth but declines in activity and excitement-
seeking. Likewise for the openness domain, declining age trends are found
cross-culturally for fantasy and actions, while the remaining openness to
experience facets show increases (De Bolle et al., 2016).
Personality development in childhood and adolescence generally moves in the
direction of greater maturity and connects with general trends of personality
development observed later in life. The rate of change, however, overall appears
to be higher in childhood and adolescence than in adulthood, although there are
cultural differences in the rate of personality development for some personality
traits. The mechanisms behind these rate fluctuations are still the subject of
debate (De Bolle et al., 2016), with some researchers arguing that social
investments, such as earlier movement into adult roles, serve as agents of
accelerated change (Bleidorn et al., 2013).
Gender Differences
Contrary to the many popular books suggesting that gender differences in
personality can be expected to be large, studies in adults have shown that gender
differences are relatively modest in magnitude but consistent across studies
(Costa et al., 2001). Women generally score higher on all of the Big Five factors,
especially neuroticism and agreeableness. Overall, adult sex differences begin to
appear in adolescence and, with increasing age, sex differences found at younger
ages tend to develop—with respect to both direction and magnitude—toward
those observed for adulthood, with females scoring higher than males on most
traits. As in adulthood, sex differences in childhood and adolescence are modest
in magnitude (De Bolle et al., 2016; Klimstra et al., 2009; McCrae et al., 2002).
Divergent from this general trend is that adolescent girls score substantially
higher on conscientiousness (De Bolle et al., 2016; De Fruyt, Van Leeuwen, de
Bolle, & de Clercq, 2008), although boys and girls tend to converge on
conscientiousness as they move into adulthood. This mean score discrepancy,
specifically during adolescence, has been suggested as an explanation for girls’
higher academic achievement, especially in secondary education (De Fruyt et al.,
2008; Voyer & Voyer, 2014).
Table 22.1 Mid-level constructs of childhood personality

HiPIC (1) Digman (2) ICID (3)
Neuroticism Neuroticism
Anxiety Fearful/insecure
Negative emotionality
Self confidence Insecurity Shy
Resiliency
Extraversion Extraversion
Energy Activity level Activity level
Expressiveness
Optimism Positive emotionality
Shyness
Sociability Considerate/sociable
Self-assertion Openness
Imagination Imagination Openness
Creativity
Intellect Intellect
Curiosity
Benevolence Agreeableness
Altruism
Dominance Strong willed
Egocentrism
Compliance Antagonism Antagonism
Irritability
Hyperactive impulsivity
Conscientiousness Conscientiousness
Concentration Distractible
Persistence Perseverance
Orderliness Carefulness Organized
Achievement Achievement orientation
Mannerliness
Note.
(1) HiPIC: Hierarchical Personality Inventory for Children (Mervielde & De Fruyt, 1999);
(2) Mid-level constructs proposed by Goldberg (2001) from a re-analyses of Digman’s work;
(3) ICID: Inventory of Child Individual Differences (Halverson et al., 2003).
The literature on adults shows that women score higher on most personality
facets (i.e., feminine-typed facets), whereas men typically score higher on
assertiveness, excitement seeking, openness to ideas, and competence (McCrae,
Terracciano, & 78 Members of the Personality Profile of Cultures Project, 2005).
Before age 17, however, boys tend to score as high as girls on several feminine-
typed facets (i.e., self-consciousness, fantasy, actions, trust, straightforwardness,
and modesty; De Bolle et al., 2016). Furthermore, adolescent girls score higher
on the masculine-typed facet of assertiveness. From a review of the available
evidence, one can generally conclude that (1) far fewer significant facet-level
sex differences are found in childhood and early adolescence than in adulthood,
and (2) girls start to display higher levels of sex-typed personality traits at an
earlier age than boys (Costa et al., 2008; De Bolle et al., 2016; Soto et al., 2011).
Consequential Outcomes
Predictive validity of consequential outcomes is ultimately the most important
criterion to judge on the value of psychological constructs. In this regard, the
resurgence of trait psychology by the end of the twentieth century was to a large
extent attributable to numerous validity studies in different applied areas,
underscoring the utility and predictive validity of the trait concept, and the FFM
in particular (Woods, Lievens, De Fruyt, & Wille, 2013). Although the validity
of traits has been examined and reviewed primarily for adults (Ozer & Benet-
Martinez, 2006), a similar review of consequential outcomes can be compiled
for children and adolescents. For youth, the more short-term outcomes can
mainly be situated in the family, interpersonal, and school contexts (De Fruyt &
De Clercq, 2014), whereas long-term outcomes may include a much broader
range of settings, including also (mental) health and socioeconomic outcomes,
such as for example longevity (Bogg & Roberts, 2013), employability (De Fruyt
& Mervielde, 1999), and career attitudes and success (Wille, De Fruyt, & De
Clercq, 2013). The short-term consequential outcomes will be reviewed here.
Peer Interaction and Relationships

Extraversion and agreeableness are the two personality factors that are chiefly
considered in the context of interpersonal relations (see also the chapters by Wilt
and Revelle, and by Graziano and Tobin, respectively). Research has led to the
conclusion that these two dimensions are sufficient to describe most types of
social behavior (Wiggins & Broughton, 1991), with extraversion referring to
individual differences in the frequency of social interaction, whereas
agreeableness primarily taps into qualitative aspects of interpersonal tendencies
(Costa & McCrae, 1992a,b) representing the individual’s orientation towards
other people’s experiences, interests, and goals. Agreeableness is most closely
associated with processes and outcomes related to interpersonal conflict and
adjustment in children, whereas the extraversion dimension has been studied
frequently in the context of social anxiety and shyness. Less agreeable children
and adolescents report higher levels of day-to-day conflict and greater
acceptance of destructive conflict tactics; they also have poorer relationships
with peers and teachers (Jensen-Campbell, Gleason, Adams, & Malcolm, 2003).
These two interpersonal trait domains also play an important role in the study of
sociometric status and bullying.
Developmental psychologists interested in children’s social reputation often
use sociometric methods to assign individuals to social status categories,
including “popular,” “controversial,” “average,” “neglected,” and “rejected”
children (Coie & Dodge, 1983). Usually a peer nomination procedure is used, in
which members of a group nominate other group members whom they like or
dislike playing with. Popular children are those receiving many positive
nominations and few negative ones; “rejected” children receive many negative
and few positive votes; “controversial” kids receive many nominations of both
types, whereas “neglected” youths get few votes of either type. Average
individuals are usually defined as those who receive intermediate numbers of
positive and negative nominations. Although sociometric research has
considerably advanced knowledge of children’s interpersonal behavior and peer
relationships, Mervielde and De Fruyt (2000) argue that the nomination
procedure in itself does not shed much light on the reasons why children are
liked or disliked. Indeed, from a FFM perspective, there may be multiple reasons
why a child is nominated positively (e.g. “nominate the child you would like to
play with”), i.e. the child may be chosen because s/he is extraverted, or
agreeable, or reliable (conscientious), referring to very different personality traits
in terms of their FFM position.
Research by Newcomb, Bukowski, and Pattee (1993) showed that three global
categories of behavior—sociability, aggression, and withdrawal—are important
to distinguish among groups. This meta-analytic review showed that children in
each sociometric category had distinct behavioral repertoires that influence the
quality of their social relations. Peer-perceived sociability seems to be most
important for peer acceptance. Sociability implies being perceived as friendly
and considerate, but also as enthusiastic, intelligent, and imaginative.
Adolescents who are perceived as self-confident are also more accepted, and
those perceived as anxious, emotionally labile, insecure, and/or withdrawn are
likely to be rejected by their peers, just as aggressive and inattentive adolescents
are (Scholte, van Aken, & van Lieshout, 1997).
Bullying
Another form of interpersonal behavior that has received considerable
attention during the past 20 years is bullying at school. Bullying is a form of
verbal, physical, or social aggression involving repeated use of force against
peers over extended periods of time. It includes name-calling, threatening,
teasing, hitting, and exclusion (Olweus, 1993). Recent research has focused on
the larger peer group’s role in bullying situations, including peers’ social role
and personality traits. A number of studies specifically addressed the
associations between the Big Five and specific measures of
bullying/victimization (Bollmer, Harris, & Milich, 2006; Jensen-Campbell et al.,
2003; Tani, Greenman, Schneider, & Fregoso, 2003).
Salmivalli (1999) investigated peer group dynamics identifying six roles in
bully-victim situations: bully, victim, reinforcer of the bully, assistant of the
bully, defender of the victim, and outsider. Bullying can thus be perceived as a
group activity in which children might participate differently according to their
intrinsic personal characteristics. Tani et al. (2003) reduced these six categories
to four participant social role groups: pro-bullies, defenders of the victim,
outsiders, and victims. They then examined their FFM profiles, with the
following results:
Pro-bullies scored lower than defenders on agreeableness, reflecting their
preoccupation with their own goals and interests and a lack of sympathy for
other people’s suffering. This finding is consistent with literature showing that
bullies tend to use aggressive strategies for solving interpersonal problems (Slee,
1993), have a lack of empathy (Olweus, 1993), and manipulate and take
advantage of others (Sutton & Keogh, 2000). The lower agreeableness scores
may underlie their aggressive and manipulative tendencies. The pro-bullies also
score higher on extraversion, suggesting that children who bully others or
children who help the bullies are likely to assert themselves in social situations.
Defenders of the victim exhibited the highest agreeableness scores and
obtained lower neuroticism scores than pro-bullies and victims.
A mixed picture for the outsiders was observed. They scored lower than pro-
bullies and defenders on extraversion, lower than defenders on agreeableness,
higher than victims on conscientiousness, and lower than victims on neuroticism.
The low score on extraversion is in line with other studies that demonstrate that
outsiders tend to refrain from getting involved in bullying situations, because
low extraversion has been linked to reticence from social interactions.
Finally, Tani and colleagues (2003) found that victims have lower scores on
agreeableness (low A) than either defenders or outsiders, lower scores on
conscientiousness, and higher scores on neuroticism. This suggests that victims
are less equipped to build warm and trusting relationships (low A) and are less
purposeful and less strong-willed. The increased score on neuroticism is in line
with the literature, suggesting that victimized children have difficulty regulating
their emotions (Shields & Cicchetti, 2001), and this may act as a risk factor for
further victimization. An alternative explanation could be that victimization
leads children to be more inclined to look after and protect themselves (low A) at
the expense of sympathy of their peers, and that these behaviors invite
victimization from others (Tani et al., 2003). Bollmer and colleagues (2006)
investigated the associations between victimization/bullying in 99 children 10 to
13 years of age and reported that bullies were described by their parents as being
less agreeable and less conscientious. The prediction that bullies would also
score higher on extraversion was not supported. Victims scored significantly
lower on conscientiousness and higher on neuroticism.
More recently, De Bolle and Tackett (2013) adopted a person-centered
approach for the study of bullying/victimization and personality. Person-
centered approaches have an advantage in that they consider the constellation of
personality traits characterizing an individual. Using latent-class analysis these
investigators identified four groups: a resilient group (high on emotional stability
and the other FFM trait dimensions), a moderately scoring group, an
undercontrolled group (extraverted, but low on conscientiousness), and a mixed
group (low on emotional stability, extraversion, and imagination, but comparable
with the undercontrolled group for benevolence and conscientiousness).
Personality class membership significantly predicted the bully/victim
classification beyond gender. Children with a mixed or an undercontrolled
profile were four times more likely to be bully/victims than children assigned to
the moderate class, whereas resilient children were .61 times less likely than
children from the moderate class to be victims, rather than to be uninvolved in
bullying or victimization. In sum, these various studies clearly underscore that
children’s personality traits are related to how they function interpersonally at
school.
Learning Achievement and Outcomes

Educational, Learning, and Instruction (ELI) processes and outcomes have
been traditionally studied from the perspective of differences in cognitive ability,
but the past two decades have witnessed booming attention on noncognitive
measures, including interests, motivational factors, and personality traits.
Whereas interests have been mainly related to what people actually study (De
Fruyt & Mervielde, 1996), abilities, motivational factors, and personality have
been studied as predictors of performance in ELI contexts, with abilities
conceptually considered as more maximal indicators of performance, and
motivational and personality factors as more typical behaviors affecting ELI
outcomes. Strenze (2007) reported a corrected r = .56 between intelligence and
GPA, suggesting that almost a quarter of the variance in academic achievement
can be explained by intelligence scores. Today, there is abundant evidence that
personality traits considerably affect academic attainment, at least to a similar
extent as cognitive measures do.
Traits and study performance. von Stumm, Chamorro-Premuzic, and
Ackerman (2011) have described three different frameworks on how personality
and intelligence may affect school achievement tests and grades. First,
intelligence and personality may act independently on performance at school.
Second, personality may affect performance on intelligence tests, hence
affecting the performance level. For example, test anxiety may constrain an
individual’s test performance. Third and finally, personality traits may shape the
conditions under which persons will apply and invest their intelligence. von
Stumm and Ackerman (2013) meta-analytically examined the associations
between groups of investment traits, including social curiosity, abstract thinking,
intellectual thinking, absorption, openness, novelty-seeking and ambiguity, and
intelligence and found mostly positive associations (with an average estimate of
.30). Investment traits may have an impact on cognitive development and the
accumulation of knowledge and skills across the life course.
Although a series of specific traits have been examined with respect to
educational outcomes, such as procrastination (Steel, 2007), grit (Duckworth &
Seligman, 2006; Duckworth, Peterson, Matthews, & Kelly, 2007), and goal
setting-engagement (Bipp & van Dam, 2014), most work has been
accommodated in the framework of the FFM. Poropat (2009) meta-analytically
investigated the relations between the dimensions of the FFM and academic
performance. Some of the primary studies included in that meta-analytic
summary also reported correlations between intelligence and academic
performance, and were used as a point-of-reference. The sample weighted
correlation corrected for scale reliability between intelligence and academic
performance was .25, hence substantially lower than the .56 reported by Strenze
(2007). The coefficient estimated by Strenze was probably larger due to the
absence of range restriction, so the .25 reported by Poropat for intelligence
provides a benchmark to interpret the relative weight of the FFM traits. The
corrected correlations for the FFM scales and academic performance were .07,
.22, .02, –.01, and .12 for agreeableness, conscientiousness, emotional stability,
extraversion, and openness, respectively, with agreeableness, conscientiousness,
and openness as significant explanatory constructs. The correlation found for
conscientiousness almost equals that of intelligence. Academic level was found
to moderate the FFM-academic achievement association significantly, with the
largest coefficients found in primary education for both intelligence and all FFM
factors; declines from primary to secondary and tertiary level for intelligence,
agreeableness, emotional stability, and extraversion; and linear declines across
the three levels for openness. The correlation for conscientiousness did not
significantly alter across academic levels. The correlations in primary education
were .58, .30, .28, .20, .18, and .24 between academic achievement and,
respectively, intelligence, agreeableness, conscientiousness, emotional stability,
extraversion, and openness. Poropat (2009) further found that conscientiousness
added little to the prediction of tertiary GPA when partialled out for secondary
GPA, although it still performed slightly better than intelligence. In a meta-
analytic investigation of adult-rated child personality and academic performance
in primary education (Poropat, 2014a), corrected correlations of .43, .18, and .50
were reported for openness, emotional stability, and conscientiousness,
respectively, significantly outperforming the effects observed for self-ratings in
the case of openness and conscientiousness. These relationships were not
moderated by age or year of education (grades 1 to 7). The positive associations
between conscientiousness and openness on the one hand and academic
achievement on the other were also extended to other-rated personality (Poropat,
2014b).
A broad series of studies have directly related personality traits to education
performance outcomes, with the FFM conscientiousness, openness, and
neuroticism factors as key dimensions to describe achievement-relevant
personality (Briley, Domiteaux, & Tucker-Drob, 2014). Spengler and colleagues
(Spengler, Ludtke, Martin, & Brunner, 2013), following a large representative
sample of 15-year-old students and another sample of students in the ninth and
tenth grades, showed that conscientiousness better predicted grades (r = .15–.30),
whereas openness was more strongly associated with performance (.15–.32) on
math and reading items culled from the Programme for International Student
Assessment (PISA; Organisation for Economic Co-operation and Development,
2009).
Costa and McCrae (1998) acknowledged the prominence of both openness to
experience and conscientiousness for the ELI area, describing a learning
approach circumplex, formed by pairing these two FFM dimensions. The four
quadrants of this circumplex describe different learning approaches among
individuals. Students who score high on openness and conscientiousness are
called the “Good Students.” They like to learn and have the diligence and
aspiration level to excel. They are often creative in their problem-solving
approach and are likely to go as far academically as their gifts allow. By contrast
the “Reluctant Scholars” score low on both factors. They are described as
students who need special incentives to start learning, have problems
maintaining attention, and need help in organizing their work. Academic and
intellectual pursuits are not their strength. “By-the-bookers” are diligent,
methodical, organized, and abide by all the rules, but they lack imagination and
prefer step-by-step instructions. They excel at rote learning and have a strong
need for structure and closure. Finally, the “Dreamers” are attracted to new ideas
and can develop them with imaginative elaborations, but they may get lost in
fancy. They are less successful in completing their innovative projects and may
need help to stay focused. Dreamers are able to tolerate uncertainty and
ambiguity. The previous narrative descriptions of the quadrants certainly have
appeal, but to our knowledge, no study has explicitly examined the usefulness
and validity of this learning approach circumplex representation relying on FFM
traits.
Learning styles. Personality traits may also indirectly affect ELI outcomes,
for example via particular ways of learning that serve as mediating constructs
between traits and academic outcomes. Students differ in large part in the ways
they perceive and learn, and this variability has been represented in models on
learning styles or learning approaches. Learning style is mostly narrowly
conceived, often as a learning strategy or a combination of different learning
activities. When depth of information is emphasized, the concept “approach” is
used. There are a number of learning style instruments that are fairly similar to
one another (Furnham, 1996). Kolb’s theory (Kolb, 1976, 1984) received the
most attention in the literature, and served as a basis for Honey and Mumford’s
(1982) Learning Style Questionnaire (LSQ).
The LSQ has four scales: Activists, Reflectors, Theorists, and Pragmatists.
Activists get fully involved without bias in new experiences. They enjoy the here
and now, are open-minded, and are not skeptical, and this makes them
enthusiastic about anything new. Reflectors like to stand back, thinking about
experiences and considering these from many different perspectives. They
inform themselves, collect data, and reflect and anticipate before concluding and
acting. Theorists adapt and integrate observations into complex but logically
sound theories. They think problems through in a vertical, logical way and
assimilate distinct facts into coherent theories. They tend to be perfectionists and
like to analyze and synthesize. Pragmatists are keen on trying out ideas, theories,
and techniques to see if they work in practice. They positively search out new
ideas and take the first opportunity to experiment with applications. Entwistle
and colleagues (1995) developed an instrument for assessing learning styles, the
Approaches to Studying Inventory (ASI), which focuses on the level of
engagement or depth of processing applied during learning. The following
learning styles were identified: deep (intention to understand, relating ideas, use
of evidence, and active learning), surface (intention to reproduce, unrelated
memorizing, passive learning, and fear of failure), and strategic (study
organization, time management, alertness to assessment demands, and intention
to excel).
The narrative descriptions of the learning approach and style scales suggest
that they probably have variance in common with the FFM personality
dimensions, especially with conscientiousness and openness, and this has now
been well established in a series of studies (Diseth, 2003; Komarraju, Karau,
Schmeck, & Avdic, 2011). Komarraju et al. (2011) demonstrated that the FFM
traits together explain 14% of GPA variance, but learning styles explained an
additional 3%. They further provided evidence that synthesis-analytic and
elaborative processing learning styles mediated the relationship between
openness and academic performance. Likewise, Rosander and Bäckström (2012)
found that learning approaches increased the explained variance by 6% for girls
and 16% for boys, after controlling for intelligence and FFM traits. More
recently, Diseth (2013) demonstrated how the perception of the school
environment by the pupil and deep, surface, and strategic learning approaches
mediated the relationship between openness, neuroticism, and conscientiousness
and examination grades.
Psychopathology
Paralleling findings obtained in adults (Krueger, 2005), FFM traits are
strongly associated with a broad range of mental syndromes in childhood that
can be grouped under the umbrella of internalizing and externalizing disorders.
Recent work has demonstrated that these associations may be explained by four
different etiological models. Evidence in support of these models has been
comprehensively reviewed by De Bolle, Beyers, De Clercq, and De Fruyt (2012)
and Tackett (2006). A first model posits that personality may serve as a risk
factor for later psychopathology (vulnerability model), such as for instance
shown in a very recent study including 1,195 adolescents from the TRacking
Adolescents Individual Lives’ Survey (TRAILS) whose temperament at age 11
predicted internalizing and externalizing disorders at age 19 (Laceulle, Ormel,
Vollebergh, van Aken, & Nederhof, 2014). In addition, Laceulle et al. (2014)
showed that changes in temperament between age 11 and age 16 had a predictive
value for internalizing and externalizing psychopathology between age 16 and
age 19 above and beyond the effect of basal temperament scores. Second,
psychopathology may cause subsequent personality change (complication
model). Martin-Storey, Serbin, Stack, Ledingham, and Schwartzman (2012) for
instance recently reported that peer-rated aggression in childhood caused higher
levels of adult neuroticism. Self-perceived social withdrawal in childhood was in
addition associated with lower rates of conscientiousness in adulthood, whereas
peer perceived social withdrawal was related to lower levels of extraversion.
Furthermore, Klimstra, Akse, Hale, Raaijmakers, and Meeus (2010) found
evidence for the complication hypothesis asserting that adolescent problem
behavior affected personality traits. Third, the pathoplasty model assumes that
personality may shape the course and outcome of psychopathology. For
example, a recent study (de Haan, Deković, van den Akker, Stoltz, & Prinzie,
2013) indicated that changes in children’s self-reported personality between
childhood and adolescence (i.e. decreasing scores of all personality traits)
affected the scores on adjustment problems at age 17. More specifically, those
showing decreasing extraversion and imagination, as well as increasing
conscientiousness (referred to as the “overcontrollers”) across time (i.e. between
age 9 and age 17) had the highest scores on later internalizing psychopathology.
The “undercontrollers,” in contrast, displayed decreasing extraversion,
conscientiousness, and imagination across time and showed the highest rates of
externalizing behaviors at age 17. Also, children reporting an increasingly
extreme personality configuration across childhood and adolescence experienced
more internalizing and externalizing psychopathology in late adolescence, even
when controlling for previous levels of psychopathology and personality (van
den Akker et al., 2013). A fourth model, the spectrum model, assumes that
personality and psychopathology are associated because of shared underlying
etiological factors. It is closely related to the continuity model, referring to the
systematic phenotypic covariation of traits and psychopathology. Evidence for
the spectrum model in younger age groups was recently found by Martel,
Gremillion, Roberts, Zastrow, and Tackett (2014) examining longitudinal
relations between temperament traits and attention deficit hyperactivity disorder
(ADHD) symptoms. From a more general perspective, Tackett and colleagues
(2013) have underscored the existence of common genetic influences on a
general psychopathology factor and the negative emotionality trait in young twin
pairs, supporting the spectrum hypothesis in children and adolescents.
Throughout earlier reviews and more recent studies on these etiological
relations, it is remarkable that empirical evidence has been generated primarily
by relying on only a single model, without controlling for the effects of other
etiological associations. This methodological choice may have resulted in biased
findings and complicates the transparency of the conclusions culled from these
designs. From a more stringent perspective, De Bolle and colleagues (2012)
showed that the effects of the continuity model preponderate in the general trait-
psychopathology relation in childhood, with additional pathoplasty and
complication effects. These findings highlight the dimensional nature of traits
and psychopathology, suggesting that they primarily need to be understood as
continuous and related constructs, although they also represent both a unique set
of characteristics that appear to have a reciprocal influence from a more causally
oriented pathway.
The literature that speaks to consequential outcomes of personality in
childhood, however, is embedded within a vulnerability perspective on the trait-
psychopathology relationship. Beyond the evidence that mainly addresses more
traditional childhood outcomes such as anxiety, depression, or conduct
problems, more recent studies have demonstrated that early temperamental or
personality factors also have considerable value in understanding more specific
maladaptive outcomes, such as risk behavior in terms of alcohol or marijuana
use and risky driving behaviors. From a large representative sample, Dick and
colleagues (2013) prospectively studied the role of temperament prior to age 5
for understanding adolescent alcohol use and found that there are multiple
pathways through which early temperamental aspects lead to later alcohol use,
mediated by different childhood personality traits. Their finding moreover
endorses the mechanism of equifinality (Cicchetti & Rogosch, 1996), stating that
various early vulnerabilities may lead to a single outcome, hence indicating that
the dynamics of trait-outcome pathways cannot be understood within a one-to-
one predictor-outcome framework. Together with other studies (Conner,
Hellemann, Ritchie, & Noble, 2010; Masse & Tremblay, 1997), this study
demonstrated that childhood assessment at kindergarten age can be used for
preventive efforts with regard to substance use prior to the onset of drug
dependence. Focusing on risky driving behaviors, such as tailgating and
joyriding, and shoplifting, prospective empirical evidence (Begg & Langley,
2004) has shown that low conscientiousness significantly predicts these risky
behaviors and is an important characteristic to include in the development of
preventive programs for young drivers. From a personality × situation
perspective on early sexual risk behavior, Cooper (2010) concluded from a
prospective study of a large group of adolescents that some trait effects on risky
sexual behaviors, such as low-agency, high impulsivity, and high negative
affectivity, were invariant across both situational and relationship commitment
contexts. However, the results also pointed toward significant moderating effects
of traits on the relationship between specific contextual factors and sexual risk
behavior, indicating that personality predicted such behaviors more strongly in
contexts that were novel and ambiguous.
Pediatric Diseases
Personality traits are considered more and more in relation to the etiology,
management, and treatment of pediatric diseases. A recent meta-analytical
review showed for instance, that hronic fatigue syndrome in childhood can be
linked to excessive scores on conscientiousness (Lievesley, Rimes, & Chalder,
2014). Vollrath, Landolt, Gnehm, Laimbacher, and Sennhauser (2007)
empirically demonstrated that children with type 1 diabetes who showed high
scores on agreeableness and conscientiousness, and low scores on neuroticism,
maintained better glycemic control over time. Various studies also demonstrated
that the quality of life and perceived health of children with conditions such as
pediatric asthma (Lahaye, Van Broeck, Bodart, & Luminet, 2013), congenital
heart disease (Rassart et al., 2013), cancer survivors (De Clercq, De Fruyt, Koot,
& Benoit, 2004), and unintentional injuries (Vollrath & Landolt, 2005) can be
explained in part by Big Five personality traits, both from a cross-sectional and a
longitudinal perspective. Across these and other childhood studies, there are
striking similarities in terms of which childhood personality traits positively
predict the examined outcomes, with high conscientiousness and low
neuroticism as primary positive predictors.
Although these studies on personality and health-related issues all served the
pediatric field in terms of identifying significant child-related factors that play a
role in health behavior or disease adaptation, hence broadening the scope from
disease-related variables toward psychological constructs, most of these studies
suffer from various methodological biases that result from self-report and single-
informant designs. From this perspective, Friedman and Kern (2014) plead for
the use of multi-method assessments of personality and more objective measures
of health outcomes. In this regard, longevity can be considered as one of the
most reliable and valid measures of health, moreover representing the most vital
consequential outcome of human existence. From various longitudinal studies
that incorporated the assessment of childhood personality, it has been
convincingly shown that childhood conscientiousness plays a key role in
longevity (for a review see Friedman & Kern, 2014), both through a direct
pathway as through mediating influences on a number of core biopsychosocial
processes such as health behavior, risk behavior, and healthy interpersonal
environments.
Long-Term Outcomes
In his paper titled “the child is the father of the man,” Caspi (2000)
demonstrated that personality traits assessed in childhood and adolescence are
related to a broad range of consequential outcomes, including personality
development, mental health, forensic history, employment, and quality of
intimate and peer relationships. The reported prospective relationships derived
from a multi-assessment follow-up of the Dunedin birth cohort can be assumed
to be an underestimate of the FFM traits’ predictive potential, given that early
trait estimates were derived from ratings of a set of behavioral observations that
were not designed to assess the FFM. More recently, Kern et al. (2013) analyzed
caregiver ratings obtained from the Pittsburgh Youth Study when children were
between 12 and 13 years old and criterion ratings when they were in their mid-
twenties. These authors demonstrated similar findings, showing that the
agreeableness facets of compliance predicted better long-term adaptability in
terms of more schooling and less risk for unemployment, teenage fatherhood,
and crime, whereas the compassion facet was predictive of longer committed
relationships. The importance of early conscientiousness at age 16 for a range of
socioeconomic outcomes in adulthood, including salary, health, crime, and
savings behavior has been further documented by Prevoo and ter Weel (Prevoo
& ter Weel, 2013). More specific traits from the conscientiousness domain, such
as grit, describing engagement and perseverance for long-term objectives, were
demonstrated to be valid predictors of retention in such varied contexts as school
programs, the workplace, military special operations training courses, and
marriage (Eskreis-Winkler, Shulman, Beal, & Duckworth, 2014). The bottom-
line across different research is that personality indices of manageability in
childhood seem to predict how well people enter adulthood.
The World Health Organization (2014) recently launched a call for a stronger
focus on adolescent health. The top three causes of adolescent deaths across 109
countries were road traffic injuries, HIV/AIDS, and suicide, and the predominant
cause of illness and disability among adolescents aged 10 to 19 years was
depression (World Health Organization, 2014). All these conditions are directly
or indirectly strongly associated with children and adolescents’ personality traits,
with impulsiveness, self-control, and sensation-seeking related to unsafe sex in
youth and careless and/or drunk driving, whereas there is abundant evidence that
neuroticism is a vulnerability factor for depression and suicide. Given that traits
can be reliably assessed in childhood and adolescence, and that half of all people
developing a kind of mental disorder show an onset of symptoms in late
childhood, it is clear that professionals and policy makers need to pay more
attention to the developing personalities of individuals. The present chapter has
provided concrete suggestions on how these traits can be measured in childhood
and adolescence and what the potential of such measurements might be in
predicting outcomes.
References
Achenbach, T. M. (1991). Manual for the Child Behavior Checklist/4-18 and 1991
Profile. Burlington, VT: University of Vermont Department of Psychiatry.
Barbaranelli, C., Caprara, G. V., Rabasca, A., & Pastorelli, C. (2003). A questionnaire for
measuring the Big Five in late childhood. Personality and Individual Differences,
34(4), 645–664.
Begg, D. J., & Langley, J. D. (2004). Identifying predictors of persistent non-alcohol or
drug-related risky driving behaviours among a cohort of young adults. Accident
Analysis and Prevention, 36, 1067–1071.
Bipp, T., & van Dam, K. (2014). Extending hierarchical achievement motivation models:
The role of motivational needs for achievement goals and academic performance.
Bleidorn, W., Klimstra, T. A., Denissen, J. J. A., Rentfrow, P. J., Potter, J., & Gosling, S.
D. (2013). Personality maturation around the world: A cross-cultural examination of
social-investment theory. Psychological Science, 24(12), 2530–2540.
Bogg, T., & Roberts, B. W. (2013). The Case for conscientiousness: Evidence and
implications for a personality trait marker of health and longevity. Annals of
Behavioral Medicine, 45(3), 278–288.
Bollmer, J. M., Harris, M. J., & Milich, R. (2006). Reactions to bullying and peer
victimization: Narratives, physiological arousal, and personality. Journal of Research
in Personality, 40(5), 803–828.
Branje, S. J. T., Van Lieshout, C. F. M., & Gerris, J. R. M. (2007). Big Five personality
development in adolescence and adulthood. European Journal of Personality, 21(1),
45–62.
Briley, D. A., Domiteaux, M., & Tucker-Drob, E. M. (2014). Achievement-relevant
personality: Relations with the Big Five and validation of an efficient instrument.
Caprara, G. V., Barbaranelli, C., Borgogni, L., & Perugini, M. (1993). The big 5
questionnaire—a new questionnaire to assess the 5 factor model. Personality and
Caspi, A. (2000). The child is father of the man: Personality continuities from childhood
to adulthood. Journal of Personality and Social Psychology, 78(1), 158–172.
Cicchetti, D., & Rogosch, F. A. (1996). Equifinality and multifinality in developmental
psychopathology. Development and Psychopathology, 8, 597–600.
Coie, J. D., & Dodge, K. A. (1983). Continuities and changes in children’s social Status
—A five-year longitudinal study. Merrill-Palmer Quarterly-Journal of Developmental
Psychology, 29(3), 261–282.
Conner, B. T., Hellemann, G. S., Ritchie, T. L., & Noble, E. P. (2010). Genetic,
personality and environmental predictors of drug use in adolescents. Journal of
Substance Abuse Treatment, 38, 178–190.
Cooper, M. L. (2010). Toward a person × situation model of sexual risk-taking behaviors:
Illuminating the conditional effect of traits across sexual situations and relationship
contexts. Journal of Personality and Social Psychology, 98, 319–341.
Costa, P. T., & McCrae, R. R. (1992a). The five factor model of personality and its
relevance to personality disorders. Journal of Personality Disorders, 6(4), 343–359.
Costa, P. T., & McCrae, R. R. (1992b). Professional Manual: Revised NEO Personality
Inventory (NEO-PI-R) and NEO Five Factor-Inventory (NEO-FFI). Odessa, FL:
Costa, P. T., & McCrae, R. R. (1998). NEO-4. Professional Manual. Odessa, FL:
Costa, P. T. J., McCrae, R. R., & Martin, T. A. (2008). Incipient adult personality: The
NEO-PI-3 in middle-school-aged children. British Journal of Developmental
Costa, P. T., Terracciano, A., & McCrae, R. R. (2001). Gender differences in personality
traits across cultures: Robust and surprising findings. Journal of Personality and Social
Psychology, 81(2), 322–331.
psychopathology in referred and nonreferred children and adolescents: an investigation
121, 958–970.
De Bolle, M., De Fruyt, F., McCrae, R. R., Löckenhoff, C., Costa, P. T. Jr., Aguilar-
Vafaie, M., … Terracciano, A. (2016). Manuscript in preparation. Are adult-role
transitions key processes that drive personality maturation? A cross-cultural
investigation. Unpublished manuscript.
De Bolle, M., & Tackett, J. L. (2013). Anchoring bullying and victimization in children
within a Five Factor-Model-based person-centered framework. European Journal of
Personality, 27, 3, 280–289.
De Clercq, B., De Fruyt, F., Koot, H. M., & Benoit, Y. (2004). Quality of life in children
surviving cancer: A personality and multi-informant perspective. Journal of Pediatric
De Fruyt, F., & De Clercq, B. (2014). Childhood antecedents of personality disorders:
Towards an integrative developmental model. Annual Review of Clinical Psychology,
10, 449–476.
De Fruyt, F., & Mervielde, I. (1996). Personality and interests as predictors of educational
streaming and achievement. European Journal of Personality, 10(5), 405–425.
De Fruyt, F., & Mervielde, I. (1999). RIASEC types and Big Five traits as predictors of
employment status and nature of employment. Personnel Psychology, 52(3), 701–727.
De Fruyt, F., Van Leeuwen, K., de Bolle, M., & de Clercq, B. (2008). Sex differences in
school performance as a function of conscientiousness, imagination and the mediating
role of problem behaviour. European Journal of Personality, 22(3), 167–184.
de Haan, A., Deković, M., van den Akker, A. L., Stoltz, S. E. M. J., & Prinzie, P. (2013).
Developmental personality types from childhood to adolescence: Associations with
parenting and adjustment. Child Development, 84, 2015–2030.
Deal, J. E., Halverson, C. F., Martin, R. P., Victor, J., & Baker, S. (2007). The inventory
of children’s individual differences: Development and validation of a short version.
Dick, D. M., Aliev, F., Latendresse, S. J., Hickman, M., Heron, J., Macleod, J., …
Kendler, K. S. (2013). Adolescent alcohol use is predicted by childhood temperament
factors before age 5, with mediation through personality and peers. Alcoholism Clinical
and Experimental Research, 37, 2108–2117.
Digman, J. M. (1963). Principal dimensions of child personality as inferred from
teachers’ judgements. Child Development, 34(1), 43–60.
personality. Journal of Personality and Social Psychology, 50(1), 116–123.
Diseth, A. (2003). Personality and approaches to learning as predictors of academic
achievement. European Journal of Personality, 17(2), 143–155. doi: 10.1002/per.469
Diseth, A. (2013). Personality as an indirect predictor of academic achievement via
student course experience and approach to learning. Social Behavior and Personality,
41(8), 1297–1308.
Duckworth, A. L., Peterson, C., Matthews, M. D., & Kelly, D. R. (2007). Grit:
Perseverance and passion for long-term goals. Journal of Personality and Social
Psychology, 92(6), 1087–1101.
Duckworth, A. L., & Seligman, M. E. P. (2006). Self-discipline gives girls the edge:
Gender in self-discipline, grades, and achievement test scores. Journal of Educational
Psychology, 98(1), 198–208.
Entwistle, N. J., & Tait, H. (1995). The Revised Approaches to Studying Inventory.
Edinburgh, Scotland: Centre for Research on Learning and Instruction: University of
Edinburgh.
Eskreis-Winkler, L., Shulman, E. P., Beal, S. A., & Duckworth, A. L. (2014). The grit
effect: Predicting retention in the military, the workplace, school and marriage.
Frontiers in Psychology, 5.
Furnham, A. (1996). The FIRO-B, the learning style questionnaire, and the five factor
model. Journal of Social Behavior and Personality, 11(2), 285–299.
Goldberg, L. R. (2001). Analyses of Digman’s child-personality data: Derivation of Big-
Five factor scores from each of six samples. Journal of Personality, 69(5), 709–743.
Halverson, C. F., Havill, V. L., Deal, J., Baker, S. R., Victor, J. B., Pavlopoulos, V., …
Wen, L. (2003). Personality structure as derived from parental ratings of free
descriptions of children: The inventory of child individual differences. Journal of
Personality, 71(6), 995–1026.
Honey, P., & Mumford, A. (1982). The Manuals of Learning Styles. Maidenhead,
England: Honey Press.
Jensen-Campbell, L. A., Gleason, K. A., Adams, R., & Malcolm, K. T. (2003).
Interpersonal conflict, agreeableness, and personality development. Journal of
Personality, 71(6), 1059–1085.
John, O. P., Caspi, A., Robins, R. W., Moffitt, T. E., & Stouthamer-Loeber, M. (1994).
The “Little Five”: Exploring the nomological network of the Five Factor model of
personality in adolescent boys. Child Development, 65(1), 160–178.
Kern, M. L., Duckworth, A. L., Urzua, S. S., Loeber, R., Stouthamer-Loeber, M., &
Lynam, D. R. (2013). Do as you’re told! Facets of agreeableness and early adult
outcomes for inner-city boys. Journal of Research in Personality, 47(6), 795–799.
Klimstra, T. A., Akse, J., Hale, W. W., III, Raaijmakers, Q. A. W., & Meeus, W. H. J.
(2010). Longitudinal associations between personality traits and problem behavior
symptoms in adolescence. Journal of Research in Personality, 44, 273–284.
Klimstra, T. A., Hale, W. W., Raaijmakers, A. W., Branje, S. J. T., & Meeus, W. H. J.
(2009). Maturation of personality in adolescence. Journal of Personality and Social
Psychology, 96(4), 898–912. doi: 10.1037/a0014746
Kohnstamm, G. A., Halverson, C. F., Jr., Mervielde, I., & Havill, V. L. (1998). Parental
Descriptions of Child Personality. Developmental Antecedents of the Big Five?
Mahwah, NJ: Lawrence Erlbaum Associates.
Kolb, D. A. (1976). Learning Style Inventory: Technical Manual. Boston: MA: McBer.
Kolb, D. A. (1984). Experimental Learning. Englewood Cliffs, NJ: Prentice-Hall.
Komarraju, M., Karau, S. J., Schmeck, R. R., & Avdic, A. (2011). The Big Five
personality traits, learning styles, and academic achievement. Personality and
Krueger, R. F. (2005). Continuity of axes I and II: Toward a unified model of personality,
personality disorders, and clinical disorders. Journal of Personality Disorders, 19(3),
233–261.
Laceulle, O. M., Ormel, J., Vollebergh, W. A. M., van Aken, M. A. G., & Nederhof, E.
(2014). A test of the vulnerability model: Temperament and temperament change as
predictors of future mental disorders—the TRAILS study. Journal of Child Psychology
and Psychiatry, 55, 227–236.
Lahaye, M., Van Broeck, N., Bodart, E., & Luminet, O. (2013). Predicting quality of life
in pediatric asthma: the role of emotional competence and personality. Quality of Life
Research, 22, 907–916.
Lievesley, K., Rimes, K. A., & Chalder, T. (2014). A review of the predisposing,
precipitating and perpetuating factors in chronic fatigue syndrome in children and
adolescents. Clinical Psychology Review, 34, 233–248.
Martel, M. M., Gremillion, M. L., Roberts, B. A., Zastrow, B. L., & Tackett, J. L. (2014).
Longitudinal prediction of the one-year course of preschool ADHD symptoms:
Implications for models of temperament-ADHD associations. Personality and
Martin-Storey, A., Serbin, L. A., Stack, D. M., Ledingham, J. E., & Schwartzman, A. E.
(2012). Self and peer perceptions of childhood aggression, social withdrawal and
likeability predict adult personality factors: A prospective longitudinal study.
Masse, L. C., & Tremblay, R. E. (1997). Behavior of boys in kindergarten and the onset
of substance use during adolescence. Archives of General Psychiatry, 54, 62–68.
McCrae, R. R., & Terracciano, A. (2005). Universal features of personality traits from the
Psychology, 88(3), 547–561.
McCrae, R. R., Costa, P. T., de Lima, M. P., Simoes, A., Ostendorf, F., Angleitner, A., …
Piedmont, R. L. (1999). Age differences in personality across the adult life span:
Mervielde, I. (2002). Personality trait development from age 12 to age 18: longitudinal,
cross-sectional, and cross-cultural analyses. Journal of Personality and Social
Psychology, 83(6), 1456–1468.
McCrae, R. R., Terracciano, A., & 78 Members of the Personality Profile of Cultures
Data from 50 cultures. Journal of Personality and Social Psychology, 88(3), 547–561.
Mervielde, I. (1992). THE B5BBS-25—A Flemish set of bipolar markers for the Big-5
personality-factors. [Article]. Psychologica Belgica, 32(2), 195–210.
Inventory for Children (HiPIC). In I. Mervielde, I. Deary, F. De Fruyt & F. Ostendorf
(Eds.), Personality Psychology in Europe, Proceedings of the Eighth European
Conference on Personality Psychology (pp. 107–127). Tilburg, The Netherlands:
Mervielde, I., & De Fruyt, F. (2000). The Big Five personality factors as a model for the
structure of children’s peer nominations. European Journal of Personality, 14(2), 91–
106.
Mervielde, I., Buyst, V., & De Fruyt, F. (1995). The validity of the Big-5 as a model for
teachers’ ratings of individual-differences among children aged 4–12 years. Personality
Mervielde, I., De Fruyt, F., & De Clercq, B. (2009). Hiërarchische
Persoonlijkheidsvragenlijst voor Kinderen [Hierarchical Personality Inventory for
Children]: Handleiding. Amsterdam, the Netherlands: Hogrefe Publishers.
Newcomb, A. F., Bukowski, W. M., & Pattee, L. (1993). Childrens peer relations—A
metaanalytic review of popular, rejected, neglected, controversial, and average
sociometric status. Psychological Bulletin, 113(1), 99–128.
Organisation for Economic Co-operation and Development (OECD). (2009). PISA data
analysis manual. SPSS (2nd ed). Paris, France: OECD Publishing.
Olweus, D. (1993). Bullying at School: What We Know and What We Can Do.
Cambridge, MA: Blackwell.
outcomes. Annual Review of Psychology (Vol. 57, pp. 401–421). Palo Alto, CA:
Annual Reviews.
Poropat, A. E. (2009). A meta-analysis of the Five Factor model of personality and
academic performance. Psychological Bulletin, 135(2), 322–338.
Poropat, A. E. (2014a). A meta-analysis of adult-rated child personality and academic
performance in primary education. British Journal of Educational Psychology, 84(2),
239–252.
Poropat, A. E. (2014b). Other-rated personality and academic performance: Evidence and
implications. Learning and Individual Differences, 34, 24–32.
Prevoo, T., & ter Weel, B. (2013). The importance of early conscientiousness for socio-
economic outcomes: Evidence from the British Cohort Study IZA Discussion Paper.
Rassart, J., Luyckx, K., Goossens, E., Apers, S., Klimstra, T. A., & Moons, P. (2013).
Personality traits, quality of life and percieved health in adolescents with congenital
heart disease. Psychology & Health, 28, 319–335.
Rosander, P., & Backstrom, M. (2012). The unique contribution of learning approaches to
academic performance, after controlling for IQ and personality: Are there gender
differences? Learning and Individual Differences, 22(6), 820–826.
Salmivalli, C. (1999). Participant role approach to school bullying: implications for
interventions. Journal of Adolescence, 22(4), 453–459.
Scholte, R. H. J., van Aken, M. A. G., & van Lieshout, C. F. M. (1997). Adolescent
personality factors in self-ratings and peer nominations and their prediction of peer
acceptance and peer rejection. Journal of Personality Assessment, 69(3), 534–554.
Shields, A., & Cicchetti, D. (2001). Parental maltreatment and emotion dysregulation as
risk factors for bullying and victimization in middle childhood. Journal of Clinical
Child Psychology, 30(3), 349–363.
Slee, P. T. (1993). Bullying: A preliminary investigation of its nature and the efffects of
social cognition. Early Child Development and Care, 87, 47–57.
Soto, C. J., John, O. P., Gosling, S. D., & Potter, J. (2008). The developmental
psychometrics of big five self-reports: Acquiescence, factor structure, coherence, and
differentiation from ages 10 to 20. Journal of Personality and Social Psychology,
94(4), 718–737.
Spengler, M., Ludtke, O., Martin, R., & Brunner, M. (2013). Personality is related to
educational outcomes in late adolescence: Evidence from two large-scale achievement
studies. Journal of Research in Personality, 47(5), 613–625.
Steel, P. (2007). The nature of procrastination: A meta-analytic and theoretical review of
quintessential self-regulatory failure. Psychological Bulletin, 133(1), 65–94.
Strenze, T. (2007). Intelligence and socioeconomic status: A meta-analytic review of
longitudinal research. Intelligence, 35, 5, 401–426.
Sutton, J., & Keogh, E. (2000). Social competition in school: Relationships with bullying,
Machiavellianism and personality. British Journal of Educational Psychology, 70,
443–456.
Tackett, J. (2006). Evaluating models of the personality-psychopathology relationship in
children and adolescents. Clinical Psychology Review, 26, 584–599.
Tackett, J. L., Kushner, S. C., De Fruyt, F., & Mervielde, I. (2013). Delineating
temperament, and behavioral problems. Assessment, 20(6), 738–751.
Tackett, J. L., Lahey, B. B., Van Hulle, C., Waldman, I., Krueger, R. F., & Rathouz, P. J.
(2013). Common genetic influences on negative emotionality and a general
psychopathology factor in childhood and adolescence. Journal of Abnormal
Psychology, 122, 1142–1153.
Tani, F., Greenman, P. S., Schneider, B. H., & Fregoso, M. (2003). Bullying and the Big
Five—A study of childhood personality and participant roles in bullying incidents.
School Psychology International, 24(2), 131–146.
Terracciano, A., McCrae, R. R., Brant, L. J., & Costa, P. T. (2005). Hierarchical linear
modeling analyses of the NEO-PI-R scales in the Baltimore longitudinal study of
aging. Psychology and Aging, 20(3), 493–506.
van den Akker, A. L., Prinzie, P., Dekovic, M., De Haan, A. D., Asscher, J. J., &
Widiger, T. (2013). The development of personality extremity from childhood to
adolescence: relations to internalizing and externalizing problems. Journal of
Vollrath, M., & Landolt, M. A. (2005). Personality predicts quality of life in pediatric
patients with unintentional injuries: A 1-year follow-up study. Journal of Pediatric
Vollrath, M. E., Landolt, M. A., Gnehm, H. E., Laimbacher, J., & Sennhauser, F. H.
(2007). Child and parental personality are associated with glycaemic control in Type 1
diabetes. Diabetic Medicine, 24, 1028–1033.
von Stumm, S., Chamorro-Premuzic, T., & Ackerman, P. L. (2011). Re-visiting
intelligence–personality associations: Vindicating intellectual investment. In T.
Chamorro-Premuzic, S. von Stumm, & A. Furnham (eds.), Handbook of individual
differences (pp. 217–241). Chichester, England: Wiley-Blackwell.
von Stumm, S., & Ackerman, P. L. (2013). Investment and intellect: A review and meta-
analysis. Psychological Bulletin, 139, 4, 841–869.
Voyer, D., & Voyer, S. D. (2014). Gender differences in scholastic achievement: A meta-
analysis. Pyschological Bulletin, 140, 1174–1204.
Wiggins, J. S., & Broughton, R. (1991). A geometric taxonomy of personality-scales.
Wille, B., De Fruyt, F., & De Clercq, B. (2013). Expanding and reconceptualizing
aberrant personality at work: Validity of Five Factor model aberrant personality
tendencies to predict career outcomes. Personnel Psychology, 66(1), 173–223.
Woods, S. A., Lievens, F., De Fruyt, F., & Wille, B. (2013). Personality across working
life: The longitudinal and reciprocal influences of personality on work. Journal of
Organizational Behavior, 34, S7–S25. doi: 10.1002/job.1863
World Health Organization. (2014). WHO calls for stronger focus on adolescent health,
from http://www.who.int/mediacentre/news/releases/2014/focus-adolescent-health/en/
Clinical Utility of the Five Factor Model
Stephanie N. Mullins-Sweatt, Douglas B. Samuel, and Ashley C. Helle
Abstract
The purpose of this chapter is to discuss the clinical utility of the Five Factor
Model (FFM). This chapter will consider the clinical application of the FFM for
treatment in general, but its primary focus will be on the clinical utility of an FFM
of personality disorders. Discussed herein will be the three fundamental
components of clinical utility: ease of usage, communication, and treatment
planning. Empirical research concerning the clinical utility of the FFM also will
be considered in terms of the three components. Finally, research and examination
of clincians’ perspectives of the utilty of categorical and dimensional models of
personality will be discussed.
Key Words: clinical utility, treatment, Five Factor Model, personality disorder,
communication
A number of studies have shown the Five Factor Model (FFM) of personality
disorder (PD) to be a valid model for describing personality pathology (Widiger
& Mullins-Sweatt, 2009). A significant strength of the FFM, relative to all other
alternative dimensional models of personality and personality disorder, is the
presence of a large body of basic scientific research to support its validity as a
classification of personality, including childhood antecedents, temporal stability
across the lifespan, multivariate behavior genetic support for the personality
structure, molecular genetic support for neuroticism, and emic and etic cross-
cultural support (Mullins-Sweatt & Widiger, 2006; Widiger & Trull, 2007). As
acknowledged by the Chair of the American Psychiatric Association’s (APA)
Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5)
Personality and Personality Disorders Work Group, “similar construct validity
has been more elusive to attain with the current DSM-IV-TR personality disorder
categories” (Skodol et al., 2005, p. 1923).
It would have been a fundamental shift to the existing nomenclature to fully
integrate the psychiatric classification of personality disorder with the FFM
(Widiger & Trull, 2007). Nevertheless, the Personality and Personality Disorders
Work Group for DSM-5 recommended a significant revision to the personality
disorders section of DSM-IV-TR, proposing a hybrid categorical-dimensional
model that described PDs as combinations of core personality impairment (e.g.,
identity, self-direction) and maladaptive personality traits, the latter representing
“an extension of the Five Factor Model” (American Psychiatric Association,
2012, p. 7). Ultimately, however, the Board of Trustees of the American
Psychiatric Association did not endorse this proposal, instead placing the Work
Group’s model to DSM-5 Section III with other measures and models needing
additional research.
In addition to limited research on the proposed model, another barrier to such
an integration of personality disorders with dimensional models such as the FFM
is concerns regarding clinical utility. A special section of the Journal of
Abnormal Psychology was devoted to the discussion of shifting the entire DSM
to a dimensional model of classification. In his commentary, First (2005) argued
that clinical utility was “the most important obstacle standing in the way” (p.
561) of a dimensional model of personality replacing the diagnostic categories
within the next version of the DSM.
Interestingly though, the personality disorders section of the diagnostic
manual is perhaps the most fundamentally problematic set of diagnoses in terms
of clinical utility (Livesley, 2001). Verheul (2005) systematically reviewed
various components of clinical utility for categorical and dimensional models of
personality disorder and concluded that “overall, the categorical system has the
least evidence for clinical utility, especially with respect to coverage, reliability,
subtlety, and clinical decision-making” (p. 295). Problems with the categorical
diagnostic system have been well documented and include heterogeneity of
diagnostic membership, lack of precision in description, excessive diagnostic
cooccurrence, failure to lead to a specific diagnosis, reliance on the previous
“not otherwise specified” wastebasket diagnosis, and the unstable and arbitrary
diagnostic boundaries of the categories (Smith & Combs, 2010; Widiger &
Mullins-Sweatt, 2010). These issues significantly reduce the usefulness of
psychiatric diagnoses for clinicians. Therefore, it is not surprising that the
personality disorders are among the diagnostic categories of the DSM with
which clinicians have been least satisfied for many years (Bernstein et al., 2007;
Maser, Kaelber, & Weise, 1991). The purpose of this chapter is to discuss the
concerns and issues of clinical utility specific to the FFM of personality disorder.
Clinical Utility
The authors of the previous iterations of the Diagnostic and Statistical
Manual of Mental Disorders have suggested that clinical utility is the central and
fundamental driving force for constructing a diagnostic manual: “There is
unanimous agreement, even among those engaged in research, that the primary
purpose of DSM-IV is to facilitate clinical practice and communication”
(Frances et al., 1991, p. 410). This emphasis was again proclaimed in the first
paragraph of the introduction to the DSM-IV-TR: “Our highest priority has been
to provide a helpful guide to clinical practice” (American Psychiatric
Association, 2000, p. xxiii). Clinical utility was noted as a significant priority for
the DSM-5 Task Force, stating that “the DSM is above all a manual to be used
by clinicians, and changes made for DSM-V must be implementable in routine
specialty practices” (Kendler, Kupfer, Narrow, Phillips, & Fawcett, 2009, p. 1).
Finally, this priority was again reiterated in DSM-5 itself: “All of [our] efforts
were directed toward the goal of enhancing the clinical usefulness of DSM-5”
(American Psychiatric Association, 2013, p. 5).
Clinical utility has been defined as “the extent to which DSM assists clinical
decision makers in fulfilling the various clinical functions of a psychiatric
classification” (First et al., 2004, p. 947). Mullins-Sweatt and Widiger (2009)
reviewed existing models of clinical utility and suggested more specifically that
there are three primary components: ease of usage, treatment planning, and
communication. Each of these will be discussed as they pertain to the FFM
dimensional classification of personality disorder (Mullins-Sweatt, 2013;
Mullins-Sweatt & Lengel, 2012).
Ease of Usage
An often-expressed criticism of dimensional models of personality disorder is
that clinicians might find them too complex and cumbersome to use (First,
2005). First et al. (2004) suggested that one aspect of the “user friendliness” of a
diagnostic system includes the “length of time it takes to assess a particular
criteria set” (p. 949). This is certainly a valid concern (Clarkin & Huprich, 2011;
Widiger, 2011). A classification system that is too hard or complex to use will
clearly not be used, no matter how valid or informative it might be if it was in
fact used.
Ease of usage, of course, should not be the sole arbiter for the optimal
personality disorder classification. Perhaps the easiest form of personality
diagnosis is provided by narrative prototype matching. Prototype matching, in
which a one-to-three paragraph description of a prototypic case is subjectively
matched to your subjective impression of a patient, requires very little time or
effort. “Clinicians could make a complete Axis II diagnosis in 1 or 2 minutes”
(Westen, Shedler, & Bradley, 2006, p. 855) because there are no requirements to
assess each of the diagnostic criteria (or each of the traits) of a respective
personality syndrome. “To make a diagnosis, diagnosticians rate the overall
similarity or ‘match’ between a patient and the prototype using a 5-point rating
scale, considering the prototype as a whole rather than counting individual
symptoms” (Westen et al., 2006, p. 847).
Narrative prototype matching was in fact the initial proposal of the DSM-5
Personality Disorders Work Group, replacing the time-consuming task of
assessing diagnostic criterion sets with subjective impressions of a global match
to a narrative paragraph description (Skodol, 2012). However, the proposal was
eventually abandoned when it became apparent that the empirical support for its
reliability and validity was at best questionable, if not in fact weak (Pilkonis,
Hallquist, Morse, & Stepp, 2011; Widiger, 2011; Zimmerman, 2011).
A common criticism of the existing DSM-IV-TR (and now DSM-5) diagnostic
criterion sets though is that they are indeed impractical to use in general clinical
practice. The DSM-IV-TR diagnostic criterion sets were constructed by
researchers primarily for use in research (Frances, First, & Pincus, 1995). Each
DSM-5 personality disorder diagnosis is a complex constellation of a variety of
maladaptive personality traits (Clark, 2007; Lynam & Widiger, 2001). A
systematic assessment of these criterion sets typically requires approximately 2
hours (Widiger & Boyd, 2009), an amount of time that is unrealistic in general
clinical practice. It is not surprising then that clinicians routinely fail to assess
for all of a personality disorder’s respective criterion set (Garb, 2005), as it is
simply impractical for them to do so.
It has been suggested that it would take considerably less time to assess the 25
traits within the DSM-5 dimensional trait model than has been required for the
approximately 100 DSM-IV-TR diagnostic criteria (Skodol, 2012). Many of the
individual DSM-IV-TR diagnostic criteria are comparable to DSM-5 traits
(Lynam & Vachon, 2012), albeit the latter tend to be considerably broader in
their coverage (and hence fewer of them are perhaps needed to cover the same
ground). The DSM-IV-TR criterion sets trace their history to an effort to provide
behaviorally specific indicators of each syndrome (Frances, 1980; Spitzer,
Williams, & Skodol, 1980) and thereby need quite a few more to adequately
describe each syndrome.
This might not be the most accurate or fairest comparison though. The
personality descriptions for DSM-5 would typically be at the level of the 25
personality traits (many of which should perhaps have behaviorally specific
diagnostic criteria for their optimal assessment), whereas DSM-IV-TR
personality descriptions are at the level of the eight diagnostic categories.
Perhaps the five domains of the DSM-5 dimensional trait model are best
compared to the three DSM-IV-TR clusters, and the 25 traits compared to the
eight diagnostic categories, in which case the trait model is considerably more
complex than the categorical system. In any case, there are studies comparing
the DSM-IV-TR to the FFM and DSM-5 trait models with respect to ease of
usage that will be discussed later in this chapter.
The assessment of the sheer number of diagnostic criteria is further
complicated by the complexity of the constructs. The DSM-IV-TR categorical
syndromes are considerably more complicated to assess than any single, distinct
trait or domain. The purpose of the diagnostic manual is to help a clinician
identify which specific personality disorder is present, including a differential
diagnosis section to help when the patient appears to meet criteria for more than
one. However, different patients will have different and relatively unique
constellations of traits, typically possessing features of more than one
personality disorder (Clark, 2007; Widiger & Trull, 2007). Choosing just one
diagnosis can be considerably arbitrary; yet providing multiple diagnoses
significantly complicates the provision of a clear treatment plan. It does not help
that many of the criterion sets overlap (e.g., social avoidance is a feature of both
the schizoid and avoidant personality disorders; suspiciousness is a feature of
both the schizotypal and paranoid; anxiousness is a feature of both the avoidant
and dependent). In sum, clinicians are being asked to make differential
diagnoses among categories that are not, in fact, distinct and to identify one
specific personality disorder that optimally characterizes a patient’s maladaptive
personality functioning when many patients do not fit well into any one of the
options. Reflecting this, research has indicated that personality disorder not
otherwise specified (PDNOS) is often the most commonly diagnosed personality
disorder in previous diagnostic manuals (Verheul & Widiger, 2004; Zimmerman
et al., 2005). A dimensional trait model, in contrast, simply asks the clinician to
provide a profile description in terms of a set of relatively homogeneous and
distinct traits, with each profile relatively specific to each patient’s individual
personality structure.
Utilizing the FFM would require a clinician to learn a new classification
model. Few psychologists and even fewer psychiatrists are familiar with the
domains and facets of the FFM. However, the FFM structure is not difficult to
learn as it reproduces the naturally occurring structure within one’s language
(Ashton & Lee, 2001). The lexical research concerning the FFM documents
empirically that clinicians are already using the FFM when they think about a
person’s personality structure. In addition, the presence of one version of the
FFM within Section III of DSM-5 will also further familiarize clinicians,
including psychiatrists, with this structure. Research has even indicated that
sophisticated clinical constructs, included for example within the California Q-
set (McCrae, Costa, & Busch, 1986), the Shedler and Westen Assessment
Procedure-200 (Mullins-Sweatt & Widiger, 2007), and the DSM-IV-TR (and
current DSM-5) personality disorder syndromes (Samuel & Widiger, 2008;
Saulsman & Page, 2004), are all well understood and articulated in terms of the
FFM.
Communication
The primary purpose of an official diagnostic nomenclature is to provide a
common language of communication (Kendell, 1975). The impetus for the
development of DSM-I was the crippling confusion generated by the absence of
an authoritative, common language (Mullins-Sweatt & Widiger, 2009). Medical
centers, clinics, and clinicians were not using the same diagnoses, thereby
substantially hindering meaningful communication and consistency in care
(Widiger, 2008). “For a long time confusion reigned. Every self-respecting
alienist, and certainly every professor, had his own classification” (Kendell,
1975, p. 87).
Another recurring argument against shifting the American Psychiatric
Association diagnostic manual to a dimensional classification is the ease of
communicating the presence of diagnostic categories (First, 2005). A diagnostic
system should be useful for “communicating clinical information to
practitioners, patients and their families, and health care system administrators”
(First et al., 2004, p. 947). As expressed by the Chair of DSM-IV, “there is an
economy of communication and vividness of description in a categorical name
that may be lost in a dimensional profile” (Frances, 1993, p. 110).
It would indeed be clumsy and unmanageable for clinicians to list 20–50 traits
each time they wanted to describe a client to a colleague. However, the clinical
description can in fact be confined to just the five broad domains (e.g., emotional
instability, antagonism, disinhibition, detachment, and oddity), which will
provide quite a bit of differential treatment implications (Presnall, 2013; Widiger
& Presnall, 2013). Even at the trait level, in practice only a few key traits will
actually be needed to convey the central or primary concerns with respect to any
particular clinical decision (Bach, Markon, Simonsen, & Krueger, 2015; Widiger
& Presnall, 2013). Describing a patient in terms of three to five key traits will
likely be considerably more clear and informative than two to three diagnostic
categories, each of which only partially resembles the patient and refers to
symptoms and traits that are not in fact present.
The DSM-IV-TR diagnostic categories fail to provide sufficient information
for clinicians to make useful social and clinical decisions. This may be due, in
part, to the fact that people who share the same PD diagnosis can vary
substantially in what features of the respective disorder are present (Clark, 2007;
Trull & Durrett, 2005; Widiger & Trull, 2007). For example, there are 256
different combinations of criteria from which it is possible to receive the same
diagnosis of borderline personality disorder (Ellis, Abrams, & Abrams, 2009)
and it is even possible for two individuals to meet the DSM-IV-TR criteria for
borderline personality disorder yet have only one diagnostic feature in common.
The situation is even more bizarre for obsessive–compulsive PD as two patients
can obtain the diagnosis, but exhibit none of the same criteria.
Kraemer, Noda, and O’Hara (2004) argued that in the mental health
profession “a categorical diagnosis is necessary” (p. 21), also stating that
“clinicians who must decide whether to treat or not treat a patient, to hospitalize
or not, to treat a patient with a drug or with psychotherapy, or what type, must
inevitably use a categorical approach to diagnosis” (p. 12). Although seemingly
compelling, this is not in fact an accurate characterization of actual clinical
practice. In most clinical situations the decision is more dimensional than
categorical. Typically, there is a decision concerning a degree of medication
dosage, a frequency of therapy sessions, and even a level of hospitalization (e.g.,
day hospital, partial hospitalization, residential program, and traditional inpatient
hospitalization).
The simplicity of being able to use the same diagnostic category for all social
and clinical decisions is also offset by the inconsistency in the needs and
concerns of these different decisions. It is evident that the many clinical
decisions are not well informed by a uniform diagnostic threshold. Medication,
psychotherapy, disability, insurance coverage, and hospitalization are clinical
options that can imply very different levels of impairment. The current
diagnostic thresholds were not set to be optimal for any particular social or
clinical decision, and yet they are used to inform all of them (Regier & Narrow,
2002). A dimensional system has the flexibility to provide different thresholds
for different social and clinical decisions and would then be considerably more
useful for clinicians and more credible for social agencies than the current
categorical system. A more flexible (dimensional) classification could be
preferable to governmental, social, and professional agencies because it would
provide more reliable, valid, and explicitly defined bases for making these
important social and clinical decisions. Precisely for this reason, the authors of
DSM-5 included supplementary dimensional scales of functioning across the
diagnostic manual that can provide clinically useful information for predicting
behaviors and guiding clinical decisions, for treatment planning, and for
predicting the course of the disorder (American Psychiatric Association, 2013;
Helzer et al., 2008).
Communication in the public realm is also an important consideration for
clinical utility as the APA diagnostic manual is used to communicate
information to the general public concerning psychopathology. A recurring issue
for the diagnostic manual has been the stigmatization of a mental disorder
diagnosis. As expressed by Hinshaw and Stier (2008), “despite clear gains in
public knowledge related to mental illness over the past half-century, levels of
stigmatization as appraised by attitude surveys appear to have increased rather
than decreased in the United States” (p. 368). Stigma contributes to lower rates
of research funding, lower employment, poor housing, family burden, and
personal shame. The personality disorders have indeed been among the most
stigmatizing diagnoses (Aviram, Brodsky, & Stanley, 2006) and a barrier to
effective interventions (Fanaian, Lewis, & Grenyer, 2013). Personality disorders
are relatively unique in concerning ego-syntonic aspects of the self, or one’s
characteristic manner of thinking, feeling, behaving, and relating to others
throughout one’s adult life (Millon, 2011). In this regard, a personality disorder
diagnosis can be quite stigmatizing, such that the patient’s fundamental views of
the world, everyday behaviors and manner of interpersonal relatedness, as well
as his or her sense of self are considered to be a mental disorder.
Dimensional continuums of maladaptive psychological functioning are
associated with less stigmatizing attitudes, increased positive emotional reaction,
and less desire for social distance (Schmoerus, Matschinger, & Angermeyer,
2013). An integration of a classification of personality disorder with the
personality structure of the general population would similarly help offset some
of the stigmatization, as personality disorders would no longer be conceptualized
as qualitatively distinct from normal personality functioning. Instead, this
framework views PDs as maladaptive variants of the same personality traits that
are evident within all persons. The FFM of personality disorder provides a more
complete description of a person that recognizes and appreciates that the person
is more than just the personality disorder and that there are aspects to the self
that can be adaptive, even commendable, despite the presence of the maladaptive
personality traits. Some of these strengths may also be quite relevant to
treatment, such as openness to experience indicating an interest in mindfulness,
agreeableness indicating an engagement in group therapy, and conscientiousness
indicating a willingness and ability to adhere to the demands and rigor of
cognitive–behavioral therapy (Krueger & Eaton, 2010; Widiger & Mullins-
Sweatt, 2009).
Treatment
“The ‘holy grail’ of clinical utility is the positive effect of a change in the
diagnostic system on [treatment] outcome” (First et al., 2004, p. 951). As noted
previously, the central and fundamental importance of treatment planning for the
diagnostic manual was noted explicitly in the first paragraph of the introduction
to DSM-IV-TR: “Our highest priority has been to provide a helpful guide to
clinical practice” (American Psychiatric Association, 2000, p. xxiii).
In his argument against converting the diagnostic manual to a dimensional
system, First (2005) suggested that the existing diagnostic categories have clear
and compelling implications for treatment decisions. As he indicated more
recently, “given that most treatment research is keyed to the DSM diagnostic
categories, determination of a DSM-IV diagnosis for a particular patient
facilitates the selection of evidence-based treatments” (First, 2010, p. 471).
Elsewhere, however, it has been suggested that the issue of treatment planning
may be where the DSM-IV-TR categorical diagnoses are most problematic. As
suggested by the Chair and Vice Chair of DSM-5, in their evaluation of the
current success of the diagnostic manual, “With regard to treatment, lack of
treatment specificity is the rule rather than the exception” (Kupfer, First, &
Regier, 2002, p. xviii). It seems apparent that a diagnostic manual without clear
or specific treatment implications is fundamentally flawed with respect to its
purported highest priority.
Treatment utilizing the FFM. There are a number of texts based on clinical
experience and theoretical speculation that are helpful in providing suggestions
for the treatment of individual personality disorders (Millon, 2011). However,
there has been very little development of empirically supported therapies or
evidence-based treatment for the personality disorders. It has been well over 10
years since the American Psychiatric Association began publishing practice
guidelines for the diagnostic categories of the diagnostic manual and, as yet,
guidelines have been developed for only one of the 10 personality disorders (i.e.,
American Psychiatric Association, 2001). Similarly, the American Psychological
Association (2006) identifies modest to strong research support for the treatment
of only one personality disorder. Matusiewicz, Hopwood, Banducci, and Lejuez
(2010) identified 45 publications that evaluated the outcome of cognitive–
behavioral interventions for personality disorders. They suggested that only
borderline and avoidant personality disorders have treatments with empirical
support, whereas evidence for therapy for the others is limited to a small number
of open-label trials and case studies.
The absence of strong empirical backing for treatments for personality
disorders may be due to the problems inherent in the diagnostic categories
themselves. The diagnostic categories of DSM-IV-TR (now DSM-5) are not well
suited for specific and explicit treatment manuals as each disorder is a complex
combination of an array of maladaptive personality traits. People who meet the
diagnostic criteria for the same personality disorder may have few to no traits in
common, which is not at all conducive to developing a consistent, coherent, and
uniform treatment program.
The FFM provides a conceptually and empirically coherent structure of
homogeneous domains and traits that is far more suitable for specific treatment
implications than the syndromal constellations provided by the diagnostic
categories of DSM-IV-TR (Presnall, 2013). For instance, research has
documented that the five domains of the FFM have considerably more specific
implications for dysfunction and impairment than is provided by the DSM-IV-TR
categories (Hopwood et al., 2009; Mullins-Sweatt & Widiger, 2010).
Neuroticism is the domain of emotional and affective dysregulation,
agreeableness and extraversion are the domains of maladaptive interpersonal
relatedness, conscientiousness is the domain of work, career, and occupational
dysfunction, and openness is the domain of cognitive–perceptual dysregulation
(Mullins-Sweatt & Widiger, 2006). This specificity of dysfunction translates
well into comparably specific treatment implications (Presnall, 2013).
Extraversion and agreeableness, as the domains of interpersonal relatedness
(Costa & McCrae, 1992; Mullins-Sweatt & Widiger, 2006), will concern the
social and interpersonal relationships within and outside the therapy office.
Interpersonal models of therapy, marital–family therapy, and group therapy
might be particularly relevant to these two domains.
In contrast, neuroticism provides information with respect to mood, anxiety,
and emotional dyscontrol, often targets for pharmacologic interventions, as well
as cognitive, behavioral, and/or psychodynamic. There are very clear
pharmacologic implications for mood and anxiety dysregulation and emotional
instability (e.g., anxiolytics, antidepressants, and/or mood stabilizers), but little
to none for maladaptive antagonism or introversion, the interpersonal domains of
the FFM. This is not to suggest a limitation of the treatment implications for the
interpersonal domains. Quite the opposite, as it is an expression of the specificity
of treatment implications of an FFM of personality disorder. Barlow, Sauer-
Zavala, Carl, Bullis, and Ellard (2014) are indeed developing a specific and
manualized treatment approach for neuroticism. Maladaptively high openness
implies cognitive–perceptual aberrations, and so would likely have specific
pharmacologic implications (i.e., neuroleptics) that are quite different from those
for neuroticism. The domain of conscientiousness, in contrast to agreeableness
and extraversion, is the domain of most specific relevance to occupational
dysfunction, or impairments concerning work and career. Magidson, Roberts,
Collado, Rodriguez, and Lejuez (2014) discuss the substantial importance and
value of treating low conscientiousness, given the importance of this domain of
personality functioning for a wide variety of significant life outcomes (and
provide treatment recommendations for increasing a person’s level of
conscientiousness). Maladaptively high levels involve workaholism,
perfectionism, and compulsivity, whereas low levels involve laxness,
negligence, and irresponsibility. There might be specific pharmacologic
treatment implications for low conscientiousness (e.g., methylphenidates; Nigg
et al., 2002) although, as yet, none for maladaptively high conscientiousness.
This degree of specificity of treatment implications is nonexistent for the DSM-
IV-TR (now DSM-5) syndromes that overlap with one another and cut across the
FFM domains.
Impact of FFM on treatment decisions. The FFM can also provide a means
with which to consider the contribution of personality traits with treatment
planning in general (e.g., high conscientiousness that will suggest receptivity to
rigorous cognitive–behavioral treatments) in addition to where maladaptive traits
may lead to problems in treatment, such as introversion suggesting alienation
from social support (Mullins-Sweatt, 2013; Porter & Risler, 2013; Swickert,
Rosentreter, Hittner, & Mushrush, 2002; see also the chapter by Piedmont and
Rodgerson regarding marital–family therapy). There have indeed been several
clinical papers based on anecdotal experience addressing the FFM’s potential
ability to assist in treatment decisions (Chard & Widiger, 2005; Harkness &
McNulty, 2002; MacKenzie, 2002; Miller, 1991; Sanderson & Clarkin, 2002;
Stone, 2002; Widiger, 1997). For example, Stone (2002) suggested that he used
domains and facets of the FFM to guide treatment decisions for patients with
borderline personality disorder. For instance, “Neuroticism and agreeableness
scales picked up on the pathological aspects of the borderline patients (as did the
DSM items) but the extraversion, conscientiousness, and openness scales yielded
important information about … issues of perseverance at work, social abilities,
and openness to new ideas” (p. 412). As Stone (2002) suggested, “These
qualities, or their comparative deficiencies, play a vital role in determining
amenability to therapy” (p. 412).
A move toward a dimensional classification system could have implications
for treatment more generally and would be consistent with interest in the
development of transdiagnostic interventions (e.g., Barlow et al., 2014). As
individuals often present with complex clinical presentations (which often
includes personality pathology), clinicians may struggle with determining which
clinical problem is the primary concern or if multiple problems should be
addressed at the same time (Farchione et al., 2012). There has been little
research to guide clinicians in such treatment decisions. Thus, “Transdiagnostic
treatments may help eliminate the need for multiple diagnosis-specific treatment
manuals and simplify treatment planning, overall” (Farchione et al., 2012, p.
675). More specifically, Westen, Novotny, and Thomas-Brenner (2004) suggest
it would be useful to move beyond solely developing treatments for DSM-
defined disorders and to investigate treatments that explicitly target personality
processes. In a special issue of Psychological Assessment devoted to the
relationship between personality and psychopathology, Harkness and Lilienfeld
(1997) stated, “if treatment planning is to meet or surpass the standards
mandated by the field, then the fundamental rule of treatment planning applies:
The plan should be based on the best science available” (p. 349). Emphasized in
particular was that personality traits should be assessed when constructing and
implementing a treatment plan, given the considerable scientific support for the
reliability and validity of personality traits in predicting and accounting for a
wide variety of important life outcomes.
In the text by Lambert et al. (2004) on treatment research, Bergin and
Garfield’s Handbook of Psychotherapy and Behavior Change, Clarkin and Levy
(2004) likewise argued compellingly that “No two clients begin psychotherapy
in the same condition … many characteristics of the client may potentially
influence the therapeutic venture … the study of client variables may have much
to offer for our understanding of psychotherapy’s effectiveness. Identification of
premorbid clinical and personality characteristics predictive of outcome might
help clinician’s guide treatment choices and revise treatment methods based on
the needs of different types of clients” (pp. 194–195). Additionally, Lambert,
Bergin, and Garfield (2004) concluded that “Client characteristics make a
sizeable difference with respect to outcomes, and diagnosis per se may not be the
key variable in understanding treatment response” (p. 813). Therefore, a focus
on specific client characteristics, such as adaptive or maladaptive personality
traits of the FFM, should be a key consideration for treatment planning and
decision making.
Treatment implications can be further identified at the level of the lower-order
facets of the FFM. The literature suggests that treatment for personality disorders
focuses on specific traits rather than the global personality structure (Paris,
2006). Therefore, clinicians are more likely to treat the affective instability,
behavioral dyscontrol, or the identity disturbance of a person diagnosed with
borderline personality disorder. Those traits are specific facets of the FFM that
can be easily assessed in a variety of ways (Mullins-Sweatt et al., 2012; Widiger
& Mullins-Sweatt, 2009). Effective change occurs with respect to these specific
components, or traits, rather than the entire, global construct. Clinicians would
likely benefit from a classification system that concerns specifically and
explicitly their focus of clinical attention, such as cognitive–perceptual
aberrations (from the domain of openness), anxiousness, depressiveness, and
emotional dysregulation (from the domain of neuroticism), intense attachment
(from the domain of extraversion), meekness (from the domain of
agreeableness), or workaholism (from the domain of conscientiousness)
(Widiger & Mullins-Sweatt, 2009).
In addition to treatment outcomes, personality disorders and the traits of the
FFM may also be useful predictors of psychiatric treatment utilization and
satisfaction with care. Research suggests that those with borderline personality
disorder (for instance) utilize more services across all forms of psychiatric care
(Bender et al., 2001) but report lower treatment satisfaction (Kelstrup, Lund,
Lauritsen, & Bech, 1993). Recent research suggests that general personality
traits may also play a role in these aspects of psychiatric care. For example,
Miller, Pilkonis, and Mulvey (2006) suggested that the domain of agreeableness
was positively related to psychotropic medication usage and other psychiatric
services, extraversion was negatively related to medication use and other
treatment modalities, and openness to experience and conscientiousness were
related to the number of therapy sessions attended. In a similar study, Hopwood
et al. (2008) found that while the FFM demonstrated limited validity in
predicting prospective treatment utilization, extraversion and conscientiousness
tended to negatively relate to some forms of treatment and neuroticism was
related to utilization across treatment modalities (though this may have been due
to its association with psychiatric diagnoses).
Clinical Utility Research

“Changes in DSM-IV were made with the explicit goal of improving clinical
utility … [However,] no formal effort was made to empirically examine whether
these changes actually improved clinical utility. Instead, the field trials and data
reanalyses primarily evaluated proposed criteria sets in terms of reliability,
validity (using clinical diagnoses as the standard), and the extent to which the
proposed criteria set identified different individuals as having the disorder.
Purported improvements in clinical utility were simply assumed to be the case”
(First et al., 2004, p. 947). Authors of proposed revisions (or opponents to these
revisions) at times attempt to speak for the field, as if they know personally what
the predominant opinion of practicing clinicians is (Frances & Widiger, 2012).
More informative are studies that actually survey clinical opinion to determine
objectively whether a proposed revision is likely to be poorly or well received
and there indeed have been quite a few studies examining the potential clinical
utility of the FFM, and dimensional trait models more generally.
In an international survey of clinical psychologists and psychiatrists, Mullins-
Sweatt, Smit, Verheul, Oldham, and Widiger (2009) examined the potential
utility of traits found within various dimensional models of general personality.
Their results suggested that although clinicians considered abnormal personality
constructs to be relatively more useful for clinical treatment decisions than
normal personality constructs, a substantial number of normal personality
constructs were also identified that the psychologists and psychiatrists believed
should be included in a diagnostic manual. This finding is consistent with a
survey of the membership of the International Society for the Study of
Personality Disorders and the Association for Research on Personality Disorders,
which reported that 80% of the respondents believed that the DSM-IV-TR
personality disorders are best understood as maladaptive variants of normal
personality traits (Bernstein et al., 2007).
These findings though are inconsistent with a commentary within the
American Journal of Psychiatry (Shedler et al., 2010) that argued “clinicians
find dimensional approaches significantly less relevant and useful, and consider
them worse than the current DSM-IV system” (p. 1027). However, these authors
referenced only two (Rottman, Ahn, Sanislow, & Kim, 2009; Spitzer, First,
Shedler, Westen, & Skodol, 2008) of a number of studies that have examined
directly the opinions of clinicians with respect to the clinical utility of the FFM
(Mullins-Sweatt & Lengel, 2012).
The first study in which clinicians described patients in terms of the FFM was
provided by Blais (1997). Blais obtained FFM ratings by 100 clinicians
attending a workshop on the treatment of personality disorders. Each clinician
was asked to describe one of his or her patients who carried a primary diagnosis
of personality disorder. The clinicians rated the patient with respect to each of
the DSM-IV personality disorders as well as the FFM. As indicated by Blais
(1997), “it has been argued that the language of the FFM fails to capture
clinically important aspects of personality functioning and that clinicians will
have difficulty applying this model to their patients” (p. 388). However, he
found that the clinicians had little difficulty with the FFM and considered the
information to be clinically useful. Blais (1997) concluded that the “data suggest
that clinicians can meaningfully apply the FFM to their patients and that the
FFM of personality has utility for improving our understanding of the DSM
personality disorders” (p. 392).
In another study examining the application of the FFM to personality
disorders, Sprock (2002) sent licensed psychologists brief descriptions of
prototypic and nonprototypic cases of three personality disorders and asked them
to describe the case in terms of the 30 facets of the FFM. Their descriptions
converged significantly with FFM descriptions of the personality disorders
(Lynam & Widiger, 2001; Widiger, Trull, Clarkin, Sanderson, & Costa, 1994).
Sprock (2002) concluded that “practicing clinicians can directly apply the
dimensions of the FFM to cases of disordered personality with a moderate level
of reliability” (p. 417). Samuel and Widiger (2004) similarly found that
clinicians could apply the FFM descriptions of PDs. Importantly, interrater
reliability among diagnoses was good, suggesting that clinicians were able to
conceptualize PDs in a consistent manner using FFM traits.
Sprock (2003) was the first study to compare the potential clinical utility of
the FFM to the DSM. She asked one group of psychologists to rate brief,
fictitious case vignettes of prototypic and nonprototypic cases of three
personality disorders with respect to the FFM (as well as other dimensional
models of personality disorder), to indicate the confidence of their rating, and to
estimate the potential usefulness of the descriptions for professional
communication, case conceptualization, and treatment planning. Another group
of psychologists provided the same ratings for the DSM-IV personality disorder
diagnostic categories. Diagnostic confidence was higher for the DSM-IV
diagnostic categories, as were the ratings of utility for professional
communication, case conceptualization, and treatment planning. However,
Sprock acknowledged that much of her results could simply reflect the fact that
the clinicians had been trained with, and were much more familiar with, the
DSM-IV diagnostic categories in comparison to the FFM. She suggested that “it
may take a new cohort of clinicians, trained in a dimensional approach to
diagnosis, to obviate the need to translate back to the categories” (Sprock, 2003,
p. 1010).
Sprock’s (2003) findings were addressed in a subsequent follow-up study by
Samuel and Widiger (2006). They suggested that Sprock’s negative results were
due in large part to providing fictitious case vignettes that were written in terms
of the categorical diagnoses’ criterion sets, including even the nonprototypic
cases. The case vignettes used in Sprock were composed of sentences confined
largely to behavioral descriptions or illustrations of DSM diagnostic criteria. It is
perhaps not surprising for clinicians to indicate that the DSM-IV system is more
useful for conceptualizing, describing, and understanding persons who are
described explicitly in terms of the DSM-IV diagnostic criteria. Samuel and
Widiger (2006) provided lengthier descriptions of actual cases that were written
in a more neutral manner (i.e., using the same language to describe the cases as
was provided in the source materials). They reported that the clinicians rated the
FFM higher than the DSM-IV with respect to its ability to provide a global
description of the individual’s personality, to communicate information to
clients, to encompass all of the individual’s important personality difficulties,
and somewhat surprisingly, even to assist the clinician in formulating effective
treatment plans.
Spitzer et al. (2008) examined the clinical utility of five dimensional
diagnostic systems, including the FFM and the DSM-IV. A significant difference
from the studies of Sprock (2003) and Samuel and Widiger (2006) is that the
alternative models were applied to actual clients currently being seen in clinical
practice. Spitzer et al. (2008) reported higher utility ratings for a prototypal
matching model based on current DSM-IV diagnostic criteria and the prototypal
matching procedure of the Shedler and Westen Assessment Procedure (SWAP-
200; Shedler & Westen, 2004) than for the FFM profile description. However,
this study confounded the dimensional models with the method of assessment.
This was not by design as they used measures that were provided to them by
respective authors of these instruments. Nevertheless, this procedure
inadvertently resulted in using different methods for the assessment of each
model that would likely have a differential impact on ratings of clinical utility.
For example, the FFM assessment required the consideration and completion of
five to six pages of material, whereas Westen, Shedler, and Bradley (2006) have
suggested that with their prototypal matching “clinicians could make a complete
Axis II diagnosis in 1 or 2 minutes” (p. 855). This is perhaps an exaggeration of
how easy it is to conduct a prototypal matching, but it is evident that the
prototypal matching procedure required considerably less time and effort than
even completing an abbreviated measure such as the FFM rating form.
Lowe and Widiger (2009) also compared the clinical utility of the FFM, DSM-
IV, and SWAP-200 personality dimensions with rating forms for each model that
were comparable in length and time required for completion. They reported that
the SWAP-200 and FFM dimensions obtained higher clinical utility ratings than
the DSM-IV diagnostic constructs on five of six clinical utility questions, with no
difference in the clinical utility ratings between the FFM and the SWAP-200.
Samuel and Widiger (2011) asked clinicians to describe their clients in terms
of the DSM-IV-TR PDs as well as the FFM. The clinicians provided ratings of
both models’ clinical utility at the beginning of treatment and again after 6
months. At baseline, they rated the FFM as more clinically useful across all
domains, including ease of application, professional communication, client
communication, and treatment planning. The authors suggested the strength of
these results might have been due in part to actual clients being less prototypic
for the DSM-IV-TR categories. What is particularly interesting about this study is
that the clinicians also rated the FFM more clinically useful after 6 months of
treatment, providing direct evidence of the utility for treatment planning.
Other studies examining clinical utility from the perspective of practicing
clinicians have provided information on PDNOS. PDNOS is a diagnosis
provided by clinicians when they believe that a person has a personality disorder
but the person is not well described by one of the existing diagnostic categories
(e.g., antisocial or borderline). Surveys of clinicians, their clinical records, and
structured interview studies have suggested that PDNOS can be the most
common diagnosis in clinical practice and the most frequent diagnosis when it is
considered in empirical studies (Verheul & Widiger, 2004). Recent research has
provided evidence for the utility of the FFM in PDNOS cases. Mullins-Sweatt
and Widiger (2011) asked practicing psychologists to describe one or two clients
with personality pathology in terms of the FFM and DSM models. In some
instances, the client was someone who met the criteria for one of the 10 DSM-IV
personality disorders; in others, the client was someone who received a diagnosis
of personality disorder, not otherwise specified. Across both cases, the clinicians
rated the FFM as significantly more useful with respect to its ability to provide a
global description of the individual’s personality, to communicate information to
clients, to encompass all of the individual’s important personality difficulties,
and, perhaps surprisingly, to aid in treatment planning. Notably, within the
PDNOS case, clinicians also indicated that the FFM was significantly more
useful in terms of ease of application and professional communication (i.e.,
clinicians described the FFM as significantly more useful across all six clinical
utility questions).
Rottman et al. (2009) asked clinicians to produce DSM-IV personality disorder
diagnoses on the basis of either an FFM profile for a prototypic case (obtained
from Samuel & Widiger, 2004) or the presentation of the complete set of DSM-
IV diagnostic criteria for the respective personality disorder. They reported the
accuracy with which the DSM-IV diagnoses were obtained and the participants’
ratings of clinical utility for each method of obtaining a DSM-IV personality
disorder diagnosis. Although participants rated the FFM as more clinically
useful than the DSM-IV in terms of communicating with clients, the DSM-IV
criteria were rated as more clinically useful on three of six clinical utility
questions (i.e., prognoses, treatment planning, and professional communication).
The relevance of this study for an FFM diagnosis of personality disorder,
however, is not entirely clear. If the FFM was to replace the DSM-5 diagnostic
categories, the task of the clinicians would not be to reproduce the DSM-5
personality disorder diagnoses. Additionally, it is hardly surprising that
clinicians find it much easier to produce a DSM diagnosis when provided with
the respective diagnostic criteria than when provided with an FFM normal
personality trait profile. The authors stated that “the methods used in our studies
are not based on the assumption that the FFM, if adopted, would be used without
… diagnostic information” (Rottman et al., 2009, p. 432). However, this was in
fact precisely the methodology of the study as no FFM diagnostic information
was provided (whereas the full set of diagnostic criteria was provided for the
DSM-IV personality disorders). As noted above, the FFM of personality disorder
consists of four steps, the first being the obtainment of an FFM profile (which
the clinicians were provided) and the second being the identification of the
problems in living associated with each elevation (which the clinicians were not
provided). It would naturally be very difficult to speculate as to which
personality disorder is present in the absence of knowing the maladaptive
variants of the FFM trait elevations.
To address those concerns, Glover, Crego, and Widiger (2012) replicated the
methodology of Rottman et al. (2009) by asking clinicians to identify which
DSM-IV-TR PD is present when provided with the respective DSM-IV-TR
diagnostic criterion sets. However, rather than provide them with an FFM profile
of normal personality traits, they provided them with the maladaptive FFM
personality traits that are associated with each PD (comparable to the
maladaptive traits in DSM-5 Section III). In Rottman et al. (2009) clinicians
identified the correct PD diagnosis using the FFM only 47% of the time.
However, in the study by Glover et al. (2012) their accuracy improved to 89%,
which was comparable to the 91% accuracy using the DSM-IV-TR criterion sets.
The clinicians in Rottman et al. (2009) rated the DSM-IV-TR diagnostic
categories as more useful than the FFM with respect to making a prognosis,
devising treatment plans, and communicating with mental health professionals.
No significant difference was obtained between the DSM-IV-TR and FFM with
respect to describing all the important personality problems or describing the
individual’s global personality. However, when clinicians were provided with
the maladaptive variants of the respective FFM traits, the DSM-IV-TR and FFM
were rated equivalently with respect to communication with other professionals,
description of all problems, formulation of intervention strategy, or description
of global personality (Glover et al., 2012). The DSM-IV-TR was still considered
to be easier to use than the FFM, but this is to be expected given that the task
was to recover the DSM-IV-TR diagnostic categories. The FFM maladaptive
traits were considered to be better for communicating with clients.
In sum, when direct comparisons of the FFM and the DSM-IV-TR have been
empirically tested with respect to clinical utility, results have varied. What can
be noted is that the FFM has fared best when using comparable methods of
assessment (i.e., neutral case histories and/or presentation of diagnostic
information and measures equivalent in terms of length and time for
completion). In those studies, the results seem clear that the FFM has equivalent
or better clinical utility than the current DSM categorical model. It would be
useful for future research to go beyond clinical opinion studies to examine the
utility of the FFM of PD in implementing diagnostic evaluations and treatment
plans with actual clients and ease of usage in clinical settings (Mullins-Sweatt,
2013; Mullins-Sweatt & Lengel, 2012).
DSM-5
The previously discussed research has focused on DSM-IV-TR/DSM-5
categories. Given that DSM-5 appears to be shifting toward dimensional models,
and considerably closer to the FFM, research examining the clinical utility of
this model may be applicable to the clinical utility of broad dimensional
approaches while remaining relevant to the changing conceptualization of the
disorders. The alternative DSM-5 model attempts to address the criticisms and
limitations of the categorical approach, resulting in a different route of direct
clinical applications. The clinical application of the alternative model includes
four steps, all of which are anticipated to add important clinically significant
information (Skodol et al., 2011). The overall alternative model has been said to
be “an efficient assessment approach with considerable clinical utility” (Skodol
et al., 2013, p. 188). This model takes into account impairments in functioning
and pathological personality traits, both of which are anticipated to increase the
clinical utility of the model (American Psychiatric Association, 2013). After
determining whether the individual meets the general diagnostic criteria for
personality disorders, the first step of the four-step model is to assess the level of
personality functioning, which accounts for elements of impairment (i.e., self or
interpersonal) and severity. The next step assesses dimensional pathological
personality traits and the degree to which they are present. This step can
determine whether the individual fits the criteria for one of the six PD types,
selectively retained from DSM-IV-TR. If the individual does not fit within one of
the six categories, he or she can be described by other personality traits, also
known as personality disorder trait specified (American Psychiatric Association,
2013). These revised steps and new dimensional aspects allow for more precise
clinical utility including greater specification of problematic areas, targeted or
tailored treatments, and identification of the individual’s strengths that may
impact treatment (Skodol et al., 2011). Thus, in terms of ease of usage and
clinical applicability, it is important to consider how clinicians are collecting
information from clients regarding dysfunction, severity, and traits. Recent
research (e.g., Calabrese & Simms, 2014; Mullins-Sweatt & Widiger, 2010) has
suggested that functioning cannot necessarily be distinguished from pathological
traits. Calabrese and Simms (2014) examined this specifically and suggested that
general self-report trait measures given at baseline (e.g., SNAP-2) can assess the
type of personality pathology and the psychosocial functioning simultaneously;
however, other follow-up assessments may require separate, additional
measurements of functioning to best assess the types of impairment the client is
experiencing (Calabrese & Simms, 2014). Recent research has also suggested
that severity is related to neuroticism, conscientiousness, agreeableness, and
extraversion, and furthermore, severity is a strong predictor of prospective
dysfunction (Hopwood et al., 2011). However, the findings of Hopwood et al.
(2011) also demonstrated that personality traits are more related to severity and
may be less useful in identifying individual differences that we might see within
the features and impairment of a specific personality disorder. Ongoing research
should continue to determine the best way to assess personality disorders in
terms of both parsimony and clinical utility.
Again, one of the most crucial aspects of a new classification model for
personality disorders is clinical utility from the perspective of clinicians who
will be utilizing the model. Crego, Sleep, and Widiger (2016) surveyed
clinicians with respect to alternative lists of traits proposed for DSM-5 in
comparison to more extensive lists of traits derived from the FFM. They asked
the clinicians which sets of traits were more useful for treatment planning. A
consistent finding was that the clinicians preferred the more thorough,
comprehensive lists. For example, when describing patients with dependent
personality disorder, they preferred a more extensive list of traits (e.g.,
submissive, yielding, meek, anxious, gullible, helpless, self-effacing, and
insecure) rather than simply the three traits proposed for DSM-5 (i.e., separation
insecurity, submission, and anxiousness). These results are in direct contrast to
the presumption that clinicians would find a trait description to be too complex
and cumbersome to use (e.g., First, 2005; Shedler et al., 2010).
Morey, Skodol, and Oldham (2014) examined clinician’s perspectives on the
clinical utility of the current and alternative DSM-5 model. Psychiatrists and
psychologists provided an evaluation for each of the models, rating the current
model preferable for professional communication. Psychiatrists viewed both
classification systems as having higher utility than psychologists, which is
consistent with previous research showing that psychiatrists may find greater
utility in having diagnoses for a variety of reasons. The alternative model
personality disorder types (six) and severity rating (Level of Personality
Functioning scale) was preferred in regards to comprehensiveness, patient
communication, and treatment formulation from the point of view of
psychologists.
Psychiatrists viewed the current model as just as useful, or more useful than
the alternative model (Morey et al., 2014). Across psychologists and
psychiatrists, the dimensional trait ratings were preferred over the current model
in all aspects except for professional communication. For communication in
terms of the personality traits, the alternative model was rated similarly to the
current model. Overall, psychologists preferred the alternative model, which was
consistent with other studies examining clinical utility of dimensional
approaches (Glover et al., 2012; Lowe & Widiger, 2009; Mullins-Sweatt &
Lengel, 2012; Samuel & Widiger, 2006). Again, and consistent with previous
research, the preference for the current categorical model in regard to
communication with other professionals is not surprising as it is the system that
has been used for many years, and clinicians are expected to be more familiar
and comfortable with the current system than with the alternative model (Morey
et al., 2014). This study provides further evidence for a preference for a
dimensional model based on the FFM, in terms of comprehensiveness, patient
communication, and treatment formulation.
Conclusions
Valid concerns have been raised with respect to the potential clinical utility of
the FFM. The FFM is largely unknown to clinicians, it will include a substantial
number of traits, and there is considerably less written on the treatment of
maladaptive personality traits than on the treatment of the DSM-IV-TR (now
DSM-5) personality disorder syndromes. There have also been a few studies that
have reported negative results with respect to the FFM’s potential clinical utility,
at least in comparison to other approaches to personality disorder diagnosis.
However, anecdotal and empirical evidence for the clinical utility of the FFM
suggests that the FFM of PD is better than the existing categorical nomenclature
in addressing nonprototypic cases of personality pathology. The FFM may also
provide useful information for more prototypic cases, in terms of facilitating
communication and perhaps planning more specific and distinct treatment
decisions and interventions. Adopting a dimensional model, such as the FFM,
can address the need for the development of effective treatments for PDs, as they
can utilize and target personality traits rather than heterogeneous categories.
With the lack of empirically supported treatments for most of the existing PDs,
the potential for such treatments with a dimensional model would be consistent
with the shift toward transdiagnostic approaches to psychological disorders. An
FFM of personality disorder would also provide a clinician with a description of
abnormal personality functioning within the same model and language used to
describe general personality structure, thereby facilitating an integration of the
predominant personality models within psychiatry and psychology.
References
disorders (4th ed., rev. ed.). Washington, DC: Author.
American Psychiatric Association. (2001). Practice guidelines for the treatment of
patients with borderline personality disorder. Washington, DC: Author.
American Psychiatric Association. (2012). Rationale for the proposed changes to the
personality disorders classification in DSM-5. Retrieved from
http://www.dsm5.org/Documents/Personality%20Disorders/Rationale%20for%20the%20Proposed%2
5%205-1-12.pdf.
disorders (5th ed.). Arlington, VA: American Psychiatric Publishing.
American Psychological Association Presidential Task Force on Evidence-Based
Practice. (2006). Evidence-based practice in psychology. American Psychologist, 61,
271–285.
Aviram, R. B., Brodsky, B. S., & Stanley, B. (2006). Borderline personality disorder,
stigma, and treatment implications. Harvard Review of Psychiatry, 14, 249–256.
Bach, B., Markon, K., Simonsen, E., & Krueger, R. F. (2015). Clinical utility of the
DSM-5 alternative model of personality disorders: Six cases from practice. Journal of
Psychiatric Practice, 21, 3–25.
Barlow, D. H., Sauer-Zavala, S. E., Carl, J. R., Bullis, J. R., & Ellard, K. K. (2014). The
nature, diagnosis, and treatment of neuroticism: Back to the future. Clinical
Bender, D. S., Dolan, R. T., Skodol, A. E., Sanislow, C. A., Dyck, I. R., McGlashan, T.
H., … Gunderson, J. G. (2001). Treatment utilization by patients with personality
Bernstein, D. B., Iscan, C., Maser, J., & the Boards of the Directors of the Association for
Research in Personality Disorders and the International Society for the Study of
Personality Disorders. (2007). Opinions of personality disorder experts regarding the
DSM-IV personality disorders classification system. Journal of Personality Disorders,
21, 536–551.
DSM-IV personality disorders. Journal of Nervous and Mental Disease, 185, 388–393.
Calabrese, W. R., & Simms, L. J. (2014). Prediction of daily ratings of psychosocial
functioning: Can ratings of personality disorder traits and functioning be distinguished?
Chard, K. M., & Widiger, T. A. (2005). Abuse, coping, and treatment: A discussion of
Ms. S. Journal of Psychotherapy Integration, 15, 74–88.
Clarkin, J. F., & Levy, K. N. (2004). The influence of client variables on psychotherapy.
In M. J. Lambert (Ed.), Bergin and Garfield’s handbook of psychotherapy and
behavior change (5th ed., pp. 194–226). New York, NY: Wiley.
Costa, P. T., Jr., & McCrae, R. R. (1992). The NEO PI-R professional manual. Odessa,
Crego, C., Sleep, C. E., & Widiger, T. A. (2016). Clinicians’ judgments of the clinical
utility of personality disorder trait descriptions. Journal of Nervous and Mental
Disease, 204, 49–56.
Ellis, A., Abrams, M., & Abrams, L. D. (2009). Abnormal personality and personality
disorders. In A. Ellis, M. Abrams, & L. D. Abrams (Eds.), Personality theories:
Critical perspectives. Thousand Oaks, CA: Sage Publications.
Fanaian, M., Lewis, K. L., & Grenyer, B. F. S. (2013). Improving services for people
with personality disorders: Views of experienced clinicians. International Journal of
Mental Health Nursing, 22, 465–471.
Farchione, T. J., Fairholme, C. P., Ellard, K. K., Boisseau, C. L., Thompson-Hollands, J.,
Carl, J. R., … Barlow, D. H. (2012). Unified protocol for transdiagnostic treatment of
emotional disorders: A randomized controlled trial. Behavior Therapy, 43, 666–678.
First, M. B. (2005). Clinical utility: A prerequisite for the adoption of a dimensional
approach in DSM. Journal of Abnormal Psychology, 114, 560–564.
First, M. B. (2010). Clinical utility of the revision of the Diagnostic and Statistical
Manual of Mental Disorders (DSM). Professional Psychology: Research and Practice,
41, 465–473.
First, M. B., Pincus, H. A., Levine, J. B., Williams, J. B. W., Ustun, B., & Peele, R.
(2004). Clinical utility as a criterion for revising psychiatric diagnoses. American
Frances, A. J. (1980). The DSM-III personality disorders section: A commentary.
Frances, A. J. (1993). Dimensional diagnosis of personality—not whether, but when and
which. Psychological Inquiry, 4, 110–111.
Frances, A. J., First, M. B., & Pincus, H. A. (1995). DSM-IV guidebook. Washington,
DC: American Psychiatric Press.
Frances, A., First, M., Widiger, T., Miele, G., Tilly, S., Davis, W., & Pincus, H. (1991).
An A to Z guide to DSM-IV conundrums. Journal of Abnormal Psychology, 100, 407–
412.
Frances, A. J., & Widiger, T. A. (2012). Psychiatric diagnosis: Lessons from the DSM-IV
past and cautions for the DSM-5 future. Annual Review of Clinical Psychology, 8, 109–
130.
Garb, H. (2005). Clinical judgment and decision making. Annual Review of Clinical
Glover, N., Crego, C., & Widiger, T. A. (2012). The clinical utility of the five factor
model of personality disorder. Personality Disorders: Theory, Research, and
Harkness, A. R., & Lilienfeld, S. O. (1997). Individual differences science for treatment
planning: Personality traits. Psychological Assessment, 9(4), 349–360.
Harkness, A. R., & McNulty, J. L. (2002). Implications of personality individual
differences science for clinical work on personality disorders. In T. A. Widiger & P. T.
Costa, Jr. (Eds.), Personality disorders and the five factor model of personality (2nd
ed., pp. 391–403). Washington, D.C: American Psychiatric Association.
Helzer, J. E., Kraemer, H. C., Krueger, R. F., Wittchen, H-U., Sirovatka, P. J., & Regier,
D. A. (Eds.). (2008). Dimensional approaches in diagnostic classification.
Washington, DC: American Psychiatric Association.
Hinshaw, S. P., & Stier, A. (2008). Stigma as related to mental disorders. Annual Review
of Clinical Psychology, 4, 367–393.
Hopwood, C. J., Malone, J. C., Ansell, E. B., Sanislow, C. A., Grilo, C. M., McGlashan,
T. H., … Morey, L, C. (2011). Personality assessment in DSM-5: Empirical support for
rating severity, style, and traits. Journal of Personality Disorders, 25, 305–320.
Hopwood, C. J., Morey, L. C., Ansel, E. B., Grilo, C. M., Sanislow, C. A., McGlashan, T.
H., … Skodol, A. E. (2009). The convergent and discriminant validity of five factor
traits: Current and prospective social, work, and recreational dysfunction. Journal of
Personality Disorder, 23, 466–476.
Hopwood, C. J., Quigley, B. D., Grilo, C. M., Sanislow, C. A., McGlashan, T. H., Yen,
S., … Morey, L. C. (2008). Personality traits and mental health treatment utilization.
Kelstrup, A., Lund, K., Lauritsen, B., & Bech, P. (1993) Satisfaction with care reported
by psychiatric inpatients relationship to diagnosis and medical treatment. Acta
Psychiatrica Scandinavica, 87, 374–379.
Kendell, R. E. (1975). The role of diagnosis in psychiatry. London, England: Blackwell
Scientific.
Kendler, K., Kupfer, D., Narrow, W., Phillips, K., & Fawcett, J. (2009). Guidelines for
making changes to DSM-V. Retrieved from
http://www.dsm5.org/ProgressReports/Documents/Guidelines-for-Making-Changes-to-
DSM_1.pdf.
Kraemer, H. C., Noda, A., & O’Hara, R. (2004). Categorical versus dimensional
approaches to diagnosis: Methodological challenges. Journal of Psychiatric Research,
38, 17–25.
Krueger, R. F., & Eaton, N. R. (2010). Personality traits and the classification of mental
disorders: Toward a more complete integration in DSM-V and an empirical model of
psychopathology. Personality Disorders: Theory, Research, and Treatment, 1, 97–118.
Kupfer, D. J., First, M. B., & Regier, D. E. (2002). Introduction. In D. J. Kupfer, M. B.
First, & D. E. Regier (Eds.), A research agenda for DSM-V (pp. xv–xxiii). Washington,
Lambert, M. J., Bergin, A. E., & Garfield, S. L. (2004). Overview, trends, and future
issues. In M. J. Lambert (Ed.), Bergin and Garfield’s handbook of psychotherapy and
behavior change (5th ed., pp. 805–822). New York, NY: Wiley.
Livesley, W. J. (2001). Commentary on reconceptualizing personality disorder categories
using trait dimensions. Journal of Personality, 69, 277–286.
comparison of the DSM-IV-TR with dimensional models of personality disorder.
Lynam, D. R., & Vachon, D. D. (2012). Antisocial personality disorder in DSM-5:
Missteps and missed opportunities. Personality Disorders: Theory, Research, and
MacKenzie, K. R. (2002). Using personality measurements in clinical practice. In P. T.
Costa & T. A. Widiger (Eds.), Personality disorders and the five factor model of
personality (2nd ed., pp. 377–390). Washington, DC: American Psychological
Association.
Magidson, J. F., Roberts, B. W., Collado-Rodriguez, A., & Lejuez, C. W. (2014). Theory-
driven intervention for changing personality: Expectancy value theory, behavioral
activation, and conscientiousness. Developmental Psychology, 50, 1442–1450.
Maser, J. D., Kaelber, C., & Weise, R. D. (1991). International use and attitudes toward
DSM-III and DSM-III-R: Growing consensus in psychiatric classification. Journal of
Matusiewicz, A. K., Hopwood, C. J., Banducci, A. N., & Lejuez, C. W. (2010). The
effectiveness of cognitive behavioral therapy for personality disorders. Psychiatric
Clinics of North America, 33, 657–685.
McCrae, R. R., Costa, P. T., & Busch, C. M. (1986). Evaluating comprehensiveness in
personality systems: The California Q-Set and the five factor model. Journal of
Miller, J. D., Pilkonis, P. A., & Mulvey, E. P. (2006). Treatment utilization and
satisfaction: Examining the contributions of Axis II psychopathology and the Five
Factor Model of personality. Journal of Personality Disorders, 20, 369–387.
Miller, T. R. (1991). The psychotherapeutic utility of the five factor model of personality:
A clinician’s experience. Journal of Personality Assessment, 57, 415–433.
Millon, T. (2011). Disorders of personality: Introducing a DSM/ICD spectrum from
normal to abnormal (3rd ed.). Hoboken, NJ: Wiley.
Morey, L. C., Skodol, A. E., & Oldham, J. M. (2014). Clinician judgments of clinical
utility: A comparison of DSM-IV-TR personality disorders and the alternative model
for DSM-5 personality disorders. Journal of Abnormal Psychology, 123, 398–405.
Mullins-Sweatt, S. N. (2013). Conceptual and empirical support for the clinical utility of
five factor model personality disorder diagnoses. In T. A. Widiger & P. T. Costa, Jr.
(Eds.), Personality disorders and the Five Factor Model of personality (3rd ed., pp.
311–324). Washington, DC: American Psychological Association.
Mullins-Sweatt, S. N., Edmundson, M., Sauer-Zavala, S., Lynam, D. R., & Miller, J. D.
(2012). Five Factor measure of borderline personality traits. Journal of Personality
Mullins-Sweatt, S. N., Smit, V., Verheul, R., Oldham, J., & Widiger, T. A. (2009).
Dimensions of personality: Clinicians’ perspectives. Canadian Journal of Psychiatry,
54, 247–259.
Mullins-Sweatt, S. N., & Widiger, T. A. (2006). The five factor model of personality
disorder: A translation across science and practice. In R. F. Krueger & J. L. Tackett
(Eds.), Personality and psychopathology: Building bridges. (pp. 39–70). New York,
NY: Guilford Publications.
Mullins-Sweatt, S. N., & Widiger, T. A. (2007). Understanding the Shedler and Westen
Assessment Procedure from the perspective of general personality structure. Journal of
Mullins-Sweatt, S. N., & Widiger, T. A. (2009). Clinical utility and DSM-V.
Mullins-Sweatt, S. N., & Widiger, T. A. (2010). Personality-related problems in living:
An empirical approach. Personality Disorders: Theory, Research, and Treatment, 1,
230–238.
Mullins-Sweatt, S. N., & Widiger, T. A. (2011). Clinicians’ judgments of the utility of the
DSM-IV-TR and five factor models for personality disordered patients. Journal of
Nigg, J. T., John, O. P., Blaskey, L. J., Huang-Pollock, C. L., Willicut, E. G., Hinshaw, S.
P., & Pennington, B. (2002). Big five dimensions and ADHD symptoms: Links
between personality traits and clinical symptoms. Journal of Personality and Social
Paris, J. (2006, May). Personality disorders: Psychiatry’s stepchildren come of age.
Invited address at the 159th Annual Meeting of the American Psychiatric Association,
Toronto, Canada.
Pilkonis, P., Hallquist, M. N., Morse, J. Q., & Stepp, S. D. (2011). Striking the (im)proper
balance between scientific advances and clinical utility: Commentary on the DSM-5
proposal for personality disorders. Personality Disorders: Theory, Research, and
Porter, J. S., & Risler, E. (2013). The new alternative DSM-5 model for personality
disorders: Issues and controversies. Research on Social Work Practice, 24, 50–56.
Presnall, J. R. (2013). Disorders of personality: Clinical treatment from a five factor
model perspective. In. T. A. Widiger & P. T. Costa (Eds.), Personality disorders and
the five factor model of personality (3rd ed., pp. 395–408). Washington, DC: American
Regier, D. A., & Narrow, W. E. (2002). Defining clinically significant psychopathology
with epidemiologic data. In J. E. Helzer & J. J. Hudziak (Eds.), Defining
psychopathology in the 21st century. DSM-IV and beyond. Washington, DC: American
Psychiatric Publishing.
Rottman, B. M., Ahn, W. K., Sanislow, C. A., & Kim, N. S. (2009). Can clinicians
recognize DSM-IV personality disorders from five factor model descriptions of patient
cases? American Journal of Psychiatry, 196, 356–374.
Samuel, D. B., & Widiger, T. A. (2004). Clinicians’ personality descriptions of prototypic
comparison of the DSM-IV and five factor models. Journal of Abnormal Psychology,
115, 298–308.
Samuel, D. B., & Widiger, T. A. (2011). Clinicians’ use of personality disorder models
within a particular treatment setting: A longitudinal comparison of temporal
consistency and clinical utility. Personality and Mental Health, 5, 12–28.
Sanderson, C., & Clarkin, J. F. (2002). Further use of the NEO PI-R personality
dimensions in differential treatment planning. In P. T. Costa & T. A. Widiger (Eds.),
Personality disorders and the five factor model of personality (2nd ed., pp. 351–375).
New York, NY: American Psychological Association.
1085.
Schmoerus, G., Matschinger, H., & Angermeyer, M. C. (2013). Continuum beliefs and
stigmatizing attitudes towards persons with schizophrenia, depression and alcohol
dependence. Psychiatry Research, 209, 665–669.
Shedler, J., Beck, A., Fonagy, P., Gabbard, G. O., Gunderson, J., Kernberg, O., …
167, 1026–1028.
Shedler, J., & Westen, D. (2004). Dimensions of personality pathology. American
Skodol, A. (2012). Diagnosis and DSM-5: Work in progress. In T. A. Widiger (Ed.), The
Oxford handbook of personality disorders (pp. 35–57). New York, NY: Oxford
University Press.
Skodol, A. E., Bender, D. S., Oldham, J. M., Clark, L. A., Morey, L. C., Verheul, R., …
Siever, L. J. (2011). Proposed changes in personality and personality disorder
assessment and diagnosis for DSM-5 part II: Clinical Application. Personality
Skodol, A. E., Krueger, R. F., Bender, D. S., Morey, L. C., Clark, L. A., Bell, C. C., …
Oldham, J. M. (2013). Personality disorders in DSM-5 section III. Focus, 11, 187–188.
Skodol, A. E., Oldham, J. M., Bender, D. S., Dyck, I. R., Stout, R. L., Morey, L. C., …
Gunderson, J. G. (2005). Dimensional representations of DSM-IV-TR personality
disorders: Relationships to functional impairment. American Journal of Psychiatry,
162, 1919–1925.
Smith, G. T., & Combs, J. (2010). Issues of construct validity in psychological diagnoses.
In T. Millon, R. F. Krueger, & E. Simonsen (Eds.), Contemporary directions in
psychopathology: Scientific foundations of the DSM-V and ICD-11 (pp. 205–222).
Spitzer, R. L., First, M. B., Shedler, J., Westen, D., & Skodol, A. (2008). Clinical utility
of five dimensional systems for personality diagnosis. Journal of Nervous and Mental
Disease, 196, 356–374.
Spitzer, R. L., Williams, J. B. W., & Skodol, A. E. (1980). DSM-III: The major
achievements and an overview. American Journal of Psychiatry, 137, 151–164.
Disorders, 16, 402–423.
Sprock, J. (2003). Dimensional versus categorical classification of prototypic and
nonprototypic cases of personality disorder. Journal of Clinical Psychology, 59, 991–
1014.
Stone, M. H. (2002). Treatment of personality disorders from the perspective of the five
factor model. In P. T. Costa & T. A. Widiger (Eds.), Personality disorders and the five
factor model of personality (2nd ed., pp. 405–430). Washington, DC: American
Swickert, R. J., Rosentreter, C. J., Hittner, J. B., & Mushrush, J. E. (2002). Extraversion,
social support processes, and stress. Personality and Individual Differences, 32, 877–
891.
Trull, T. J., & Durrett, C. A. (2005). Categorical and dimensional models of personality
disorder. Annual Review of Clinical Psychology, 1, 355–380.
Verheul, R. (2005). Clinical utility of dimensional models for personality pathology.
Verheul, R., & Widiger, T. A. (2004). A meta-analysis of the prevalence and usage of the
personality disorder not otherwise specified (PDNOS) diagnosis. Journal of
Westen, D., Novotny, C. M., & Thomas-Brenner, H. (2004). The next generation of
psychotherapy research: Reply to Ablon and Marci (2004), Goldfried and Eubanks-
Carter (2004), and Haaga (2004). Psychological Bulletin, 130, 677–683.
Westen, D., Shedler, J., & Bradley, R. (2006). A prototype approach to personality
disorder diagnosis. American Journal of Psychiatry, 163, 846–856.
Widiger, T. A. (1997). Personality disorders as maladaptive variants of common
personality traits: Implications for treatment. Journal of Contemporary Psychotherapy,
27, 265–282.
Widiger, T. A. (2008). Classification and diagnosis: Historical development and
contemporary issues. In J. Maddux & B. Winstead (Eds.), Psychopathology:
Foundations for a contemporary understanding (2nd ed., pp. 83–101). Mahwah, NJ:
Lawrence Erlbaum Associates.
Widiger, T. A. (2011). A shaky future for personality disorders. Personality Disorders:
Widiger, T. A., & Boyd, S. (2009). Personality disorders assessment instruments. In J. N.
Butcher (Ed.), Oxford handbook of personality assessment (pp. 336–363). New York,
Widiger, T. A., & Mullins-Sweatt, S. N. (2009). Five Factor model of personality
Widiger, T. A., & Mullins-Sweatt, S. N. (2010). Clinical utility of a dimensional model of
personality disorder. Professional Psychology: Research and Practice, 41, 488–494.
Widiger, T. A., & Presnall, J. R. (2013). The clinical application of the five factor model.
Widiger, T. A., Trull, T. J., Clarkin, J. F., Sanderson, C., & Costa, P. T., Jr. (1994). A
description of the DSM-III-R and DSM-IV personality disorders with the five factor
model of personality. In P. T. Costa & T. A. Widiger (Eds.), Personality disorders and
the Five Factor Model of Personality (pp. 41–56). Washington, DC: American
Zimmerman, M. (2011). A critique of the proposed prototype rating system for
personality disorders in DSM-5. Journal of Personality Disorders, 25, 206–221.
Zimmerman, M., Rothschild, L., & Chelminski, I. (2005). The prevalence of DSM-IV
personality disorders in psychiatric outpatients. The American Journal of Psychiatry,
162, 1911–1918.
SECTION 4
Conclusions
A Five Factor Discussion
Thomas A. Widiger
Abstract
The purpose of this chapter is to pay homage to and provide a discussion of each
of the chapters included within this text. The first section of the book provided a
description of the Five Factor Model (FFM), followed by a chapter devoted to
each of the five domains. The second section concerned construct validity support
for the FFM. The third and final section considered various social and clinical
applications of the FFM, as well as issues and concerns with respect to these
applications. Each of the chapters included within each section is discussed in
turn.
Key Words: Five Factor Model, personality, trait, clinical applications,
extraversion, neuroticism, openness, agreeableness, conscientousness
I must begin by saying that I am happy with how this turned out. I am
certainly in a position of bias, but I do believe that this text serves a very useful
purpose, bringing together in one place diverse areas of investigation concerning
the Five Factor Model (FFM). Different versions of some of these chapters have
appeared in other texts, but never have they all appeared together within the
same text. This book may not only be useful in providing in one place much that
is known about the FFM, but it may also potentially inform one field of FFM
study about the methods and findings of another field of study, with
representation across different disciplines of psychology and input by both
young and senior investigators.
The authors of these chapters are not always in agreement with one another,
nor are they necessarily advocates of the FFM. Representing diversity of opinion
was intentional and indispensable. In any case, I consider the chapters that were
obtained to be wonderful. I will discuss each of them, but I really cannot do
justice to them. Their breadth of coverage and depth of coverage are just so
impressive. Any effort to summarize or to highlight these chapters runs the risk
of being superficial and neglectful of the most pithy, key points. In fact, at times
(actually much of the time) my comments reflect my own personal interests and
vantage point with respect to the FFM. My apologies to those who do not share
the same interests!
The first section of the book provides a description of the FFM, as it has been
articulated by Costa and McCrae (1995), or at least organized in a manner
consistent with their conceptualization. The second section concerns construct
validity support for the FFM. The third and final section concerns the various
social and clinical applications.
The Five Factor Model

The first chapter of this section is by Costa and McCrae (this volume) who,
appropriately so, provide an historical overview of their work on the FFM
(which was clearly quite extensive) along with an articulation of their particular
Five Factor Theory for the existence and nature of the domains. The amount of
research Costa and McCrae have conducted in support of the validity of the FFM
is really quite remarkable! This was facilitated in part by their highly productive
longitudinal study at the Gerontology Research Center within the National
Institute on Aging, but I also want to note their tremendous skill as collaborators,
marshalling many large-scale cross-national studies. Whereas the authors of a
particular measure might devote one entire article to the replication of gender
differences or to the replication of the factor structure within a new language or
culture, McCrae and Costa would include 50 such replications within one
publication (e.g., McCrae, Terracciano, & 78 Members of the Personality
Profiles of Cultures Project, 2005).
One of the substantive issues discussed by Costa and McCrae (this volume) is
that the FFM domains are not truly orthogonal and that some facets occupy to a
degree interstitial space. I have always enjoyed using galaxies of stars as a
metaphor for the clustering of trait terms within the language; that is, there are
essentially five galaxies of trait terms, with the largest and most clearly defined
being extraversion and the smallest and least clearly defined being openness.
The metaphor is useful. The galaxies are vast, and they are real. Plus, it is
naturally difficult to identify one word that summarizes or represents well all of
the members of a trait galaxy. But the metaphor is also frankly quite misleading
because in between galaxies of stars there is a vast empty space, whereas the
galaxies of trait terms shade into one another.
There was a period of time when there was a concern that the NEO
Personality Inventory-Revised (NEO PI-R; Costa & McCrae, 1992c) was failing
to obtain adequate results via confirmatory factor analysis. McCrae and
colleagues defended the NEO PI-R in part by arguing that confirmatory factor
analyses (CFA) included unrealistic assumptions of simple structure (e.g.,
McCrae, Zonderman, Costa, Bond, & Paunonen, 1996). This position has been
largely vindicated (Hopwood & Donnellan, 2010; Marsh, Morin, Parker, &
Kaur, 2014; see also the chapter by Wright). It is now questionable whether CFA
is the optimal model for testing the validity of any model of personality
structure. CFA presumes a simple structure that would not exist in nature, as if
the Big Five had a perfect simple structure within the lexicon. Stars reside in one
and only one galaxy, with empty space between them. The universe of trait
terms does appear to be well organized by five axes, but the interstitial space is
well packed with terms. This should not be at all surprising. Humans create
words to describe themselves and other persons, and there will naturally be
plenty of terms to describe combinations of the fundamental axes, as well as the
axes themselves. With respect to the NEO PI-R, it is to be expected that an
important construct such as angry hostility will correlate with both neuroticism
(negative affectivity) and antagonism; that impulsivity will correlate with
neuroticism as well as low conscientiousness; and that warmth will correlate
with both extraversion and agreeableness (Costa & McCrae, 1992c). We could
confine FFM facets to constructs that produce the best simple structure, but these
are unlikely to be the best constructs for clinical, social, and/or business
applications (Costa & McCrae, 1995). The interpersonal circumplex structure for
FFM agreeableness and extraversion clearly embraces this interstitial space,
representing each of the octants in between each domain (McCrae & Costa,
1989b; Wiggins & Pincus, 1989).
Costa and McCrae (this volume) are forthright though in their
acknowledgment that their original selection of FFM facets (Costa & McCrae,
1995) may no longer represent the optimal choices. As they expressed
elsewhere, “It is clearly not the case that these 30 [NEO PI-R] scales exhaust the
full range of traits related to each of the factors; punctuality is a good example of
a marker of conscientiousness that is not included” (McCrae & Costa, 2013, p.
23). Alternative facet choices for neuroticism are presented in the chapter by
Tackett and Lahey (this volume), for extraversion by Wilt and Revelle (this
volume), for openness by Sutin (this volume), for agreeableness by Graziano and
Tobin (this volume), and for conscientiousness by Jackson and Roberts (this
volume). Allen and DeYoung (this volume) offer their own proposals for all five
domains in their chapter on personality and neuroscience. Personally, I find the
inclusion of the trait of volatility within the Big Five Aspects Scale (DeYoung,
Quilty, & Peterson, 2007) to be a major strength of this measure (Mullins-Sweatt
et al., 2012). And, of course, there are alternative facet choices for the FFM as
applied to children and adolescents (De Fruyt, De Clercq, and De Bolle, this
volume; De Pauw, this volume). Nevertheless, the NEO PI-R facets identified by
Costa and McCrae have clearly held up reasonably well over the years.
Neuroticism
Neuroticism “reflects individual differences in tendencies toward negative
affect (including sadness, anxiety, & anger)” (Tackett and Lahey, this volume).
Neuroticism is an enduring tendency or disposition to experience negative
emotional states. It is otherwise known as negative affectivity or negative
emotionality (Watson & Clark, 1994), or simply as emotional instability
(Goldberg, 1993). Persons high in neuroticism respond poorly to environmental
stress, interpret ordinary situations as threatening, and can experience minor
frustrations as hopelessly overwhelming (Widiger, 2009). They have difficulty
controlling negative emotional states, including anxiousness, anger, and
depression.
It is sorely tempting to propose that neuroticism is the most important domain
of the FFM. However, to do so would be to suggest that one or more of the other
domains is not as important, and a case can be made for the singular importance
of each domain. There are also compelling reasons to argue that neuroticism is
not among the most important domains of the FFM.
Neuroticism is typically the fourth domain to be extracted from a language
(De Raad et al., 2010), suggesting that within the general population neuroticism
is of lesser importance than extraversion, agreeableness, and conscientiousness.
In fact, De Raad et al. (2010) suggest that this domain may not in fact be truly
universal (see also the chapter by De Raad and Mlačić). A person’s degree of
emotional stability though would seem to be fundamentally important to all
cultures. Tackett and Lahey (this volume) indeed make a very good case for the
special significance and relevance of neuroticism. Neuroticism is certainly the
predominant personality dimension with clinical populations, pervading many
measures of psychopathology (Widiger, 2009). Neuroticism is arguably the first
domain of personality assessed within psychology, via Woodworth’s (1919,
1920) Personal Data Sheet (Butcher, 2010). Imagine if the description of
personality structure was confined to just the three domains that have
unambiguous, undebatable lexical universality (De Raad et al., 2010). We would
just have a three-dimensional trait model of extraversion, agreeableness, and
conscientiousness. Neuroticism would not be included. The loss of the
personality domain of neuroticism to clinical psychology and psychiatry would
be considerable, if not devastating.
Neuroticism has enormous public health care implications. It provides a
dispositional vulnerability to a wide array of different forms of psychopathology.
Neuroticism is comparably associated with a wide array of physical maladies,
both subjective and objective, including cardiac problems, disrupted immune
functioning, asthma, atopic eczema, irritable bowel syndrome, and mortality
(Tackett and Lahey, this volume). It is also associated with a variety of
contributors to quality of life, including subjective well-being, occupational
success, emotional exhaustion, and marital dissatisfaction. Given the
contribution of neuroticism to so many negative life outcomes, Lahey (2009) and
Widiger and Trull (2007) have recommended that the general population be
screened for high levels of neuroticism, perhaps on the Internet or during routine
medical visits. “Not only would such efforts help identify those at potentially
high risk for a range of adverse outcomes, information about levels of
neuroticism could be incorporated into more personalized &, ideally, more
effective treatments” (Tackett and Lahey, this volume). It is routine to screen for
blood pressure and cholesterol levels, and would seem only natural and sensible
to do the same for a trait with so many public health care implications. “Even if
the indirect reduction in the prevalence of each individual adverse outcome were
modest, it is possible that such a strategy could be cost-effective because of the
sheer number of adverse outcomes associated with neuroticism” (Lahey, 2009,
p. 14).
Lahey (2009) had originally suggested that “to date, no interventions for
reducing neuroticism have been identified” (p. 14). Screening in the absence of
available treatment would perhaps be problematic. This shortcoming though is
perhaps no longer the case, as Barlow and colleagues (2011) have developed an
empirically validated Unified Protocol (UP) for the transdiagnostic treatment of
emotional disorders, which is essentially a treatment for neuroticism. Barlow et
al. suggested that current psychological treatments, which have been driven
largely by the fragmented categorical approach embodied in the American
Psychiatric Association’s (APA) Diagnostic and Statistical Manual of Mental
Disorders (DSM-5; APA, 2013), have become overly specialized, focusing on
disorder-specific symptoms (Barlow, Sauer-Zavala, Carl, Bullis, & Ellard,
2014). The UP protocol was initially designed to be transdiagnostic with respect
to mood and anxiety disorders, but it has become evident that it is indeed “a
cognitive-behavioral intervention designed to address core temperamental
processes in emotional disorders” (Barlow et al., 2014, p. 357). The UP targets
identification and modification of the strong negative reactions to emotions that
lead to problematic, avoidant coping across emotional disorders. “Amelioration
of negative reactions to emotions in turn changes the frequency and intensity of
future emotional experiences and thereby affects temperamental constructs”
(Barlow et al., 2014, p. 357). “The public-health implications of directly treating
and even preventing the development of neuroticism would be substantial”
(Barlow et al., 2014, p. 344).
Extraversion
As indicated in the chapter by Wilt and Revelle (this volume), the personality
domain of extraversion subsumes a wide array of related traits. Extraversion
concerns the disposition to be energetic, outgoing, assertive, adventurous,
gregarious, friendly, warm, poised, self-disclosing, talkative, sociable,
enthusiastic, active, lively, expressive, excitement seeking, and bold. It is the
predominant trait within most languages, the first to emerge in factor analytic
explorations of the structure of language (De Raad et al., 2010). Extraversion
versus introversion is then arguably the most important trait in personality
description. Extraversion (versus introversion) is also among the very first traits
to be assessed by psychologists, and included within the measures developed by
Bernreuter (1931) and Jung (Jung & Baynes, 1921). Its polar opposite,
introversion, was even included within the original Minnesota Multiphasic
Personality Inventory (Hathaway & McKinley, 1943).
Wilt and Revelle (this volume) document well that much is owed to Eysenck
for further developing a strong scientific base for understanding the nature and
mechanisms of extraversion. “Perhaps nobody has done as much for
extraversion as Eysenck” (Wilt and Revelle, this volume). He demonstrated the
importance of extraversion as a fundamental dimension of personality in a series
of both experimental and individual differences studies (Eysenck, 1952). There
is at times a large divide between the individual differences and experimental
laboratory approaches, with the former (or latter) even failing to correlate to any
meaningful degree with the latter (or former). Many researchers confine their
careers to just one of the two methods. Hans Eysenck was a very productive and
innovative exception. With respect to the individual differences approach, he
was also, of course, the principal author of a number of very influential and
widely used personality inventories, including (for instance), the Eysenck
Personality Inventory (Eysenck & Eysenck, 1964), the Eysenck Personality
Questionnaire (Eysenck & Eysenck, 1975), and the Eysenck Personality Profiler
(Eysenck & Wilson, 1991).
Clark and Watson (2008) have long proposed that positive emotionality is the
temperament, or driving force, of extraversion (positive emotionality is one of
the six facets of extraversion within the FFM of Costa & McCrae, 1995). I have
always embraced this notion, and not only because of its theoretical and
empirical support (Clark & Watson, 2008). Increased positive mood is readily
understood as the engine that drives increased levels of talkativeness,
gregariousness, adventurousness, boldness, and excitement-seeking.
Furthermore, in a complementary fashion, severe deficits in positive
emotionality equate well with the construct of anhedonia (Widiger, Trull,
Clarkin, Sanderson, & Costa, 1994). Anhedonia, the inability (or at least severe
deficit in the ability) to experience pleasure, has long been considered the central
pathology of schizoid personality disorder (Hopwood & Thomas, 2012; Kalus,
Bernstein, & Siever, 1993). Anhedonia is essentially having a dead, or severely
depleted, furnace of the energy-driving positive emotionality of extraversion
and, indeed, schizoid personality disorder is largely isomorphic to FFM
introversion (Hopwood & Thomas, 2012; Samuel & Widiger, 2008).
Wilt and Revelle (this volume) also document well the benefits of
extraversion (as well as the costs of introversion). There are numerous benefits
for getting along and getting ahead (Hogan, 1982). Extraversion is associated
with subjective well-being (Ozer & Benet-Martinez, 2006), having more sexual
partners (Nettle, 2006), greater satisfaction with a marriage (Watson et al.,
2004), satisfaction with a job (Thoresen, Kaplan, Barsky, Warren, & de
Chermont, 2003), and higher levels of job performance (Sackett & Walmsley,
2014), However, Wilt and Revelle (this volume) also point out that there are
potential costs associated with high levels of extraversion: “Extremely high
extraversion poses risks for personality pathology as well, as people falling at
this end of the continuum are more likely to be sexually promiscuous,
emotionally intrusive, and engage in excessive self-disclosure and thrill-seeking
behaviors.”
Millon (1981) referred to DSM-III (APA, 1980) histrionic personality disorder
as the “gregarious pattern” (p. 131). Histrionic personality disorder largely
represents a maladaptive variant of extraversion (Lynam & Widiger, 2001).
Many of the traits (or symptoms) of histrionic personality disorder are clearly
variants of extraversion. Persons diagnosed with histrionic personality disorder
tend to be high-spirited, buoyant, and dramatic (positive emotions), to be flashy
and seek strong stimulation (high excitement-seeking), to actively seek social
contact, to be talkative and the center of attention (high gregariousness), to be
outgoing and affectionate (high warmth), to be assertive and forceful (high
assertiveness), and to be energetic, fast-paced, and vigorous (high activity).
There is indeed compelling empirical support for the relationship between
histrionic personality disorder and FFM extraversion (Furnham, 2014; Gore,
Tomiatti, & Widiger, 2011; Samuel & Widiger, 2008).
However, if the DSM-5 Personality and Personality Disorders Work Group
had their way, there would no longer have been a histrionic personality disorder,
and in the next edition of the diagnostic manual of the American Psychiatric
Association it will likely be gone. The DSM-5 Personality and Personality
Disorders Work Group initially proposed to delete the narcissistic, dependent,
schizoid, paranoid, and histrionic personality disorders. Many came to the aid of
the narcissistic personality disorder (e.g., Miller, Widiger, & Campbell, 2010;
Ronningstam, 2011), and some came to the aid of the dependent (e.g., Bornstein,
2011, 2012; Gore & Pincus, 2013). Nobody spoke on behalf of the histrionic.
It is admittedly difficult to defend histrionic personality disorder (see
Blashfield, Reynolds, & Stennett, 2012, for an excellent argument for its
demise). There is some research explicitly concerned with histrionic personality
disorder, but considerably less than for the borderline, antisocial, schizotypal,
narcissistic, or dependent disorders. In addition, the diagnosis has a long
problematic history of gender bias concerns (Blashfield et al., 2012).
Nevertheless, if histrionic personality disorder was eliminated from the
diagnostic manual, we would lose the only personality disorder that represents
well maladaptive variants of extraversion (Lynam & Widiger, 2001) and,
concurrently, maladaptive variants of the extraverted octant of the interpersonal
circumplex. There are well-established maladaptive variants of all eight octants
of the interpersonal circumplex (Pincus & Hopwood, 2012), including the
location occupied by extraversion. There are even well-established measures of
the maladaptive variants for every octant of the circumplex, such as the
Inventory of Interpersonal Problems (Horowitz, Alden, Wiggins, & Pincus,
2000). If we removed histrionic personality disorder from the diagnostic manual,
along with the dependent, we would lose representation of literally half of the
interpersonal circumplex (Widiger, 2010a).
Openness
Sutin (this volume) suggests that FFM openness is perhaps the most resilient
domain within the FFM. In contrast to domains such as neuroticism or
extraversion, which are well represented in most models of personality, openness
has at times struggled to be recognized. It is the smallest and most troubled of
the FFM domains, as it is typically the last to emerge within factor analytic
studies of alternative languages (De Raad et al., 2010). There is less consensus
over how it is best described or even labeled, alternatively being referred to as a
domain of intellect, culture, imagination, and unconventionality, as well as
openness (Sutin, this volume). However, it could similarly be argued that
openness is the most interesting and stimulating of the FFM domains because it
is the most problematic, and perhaps even controversial. A special section of the
Journal of Personality Assessment was devoted to its conceptualization and
assessment (Connelly, Ones, & Chernyshenko, 2014).
Indeed, as Sutin (this volume) suggests, openness has survived and perhaps
even thrived. Sutin documents well that the consequences of openness have
stretched far beyond intellectual and artistic pursuits. One of the current
questions concerning openness is whether high openness is associated with any
potential costs or maladaptive variants. It is difficult to imagine that there would
be no maladaptive variants associated with any pole of an FFM domain. Nettle
(2006) suggests that high openness is associated with paranormal beliefs and
schizotypal thinking, but this is precisely the controversy within the personality
disorder field (Watson, Clark, & Chmielewski, 2008). It is noteworthy that Costa
and McCrae (1980) did not conceptualize openness as having any maladaptive
variants, considering it instead to concern ideal personality traits such as self-
actualization, an open mind, and self-realization (Coan, 1974; Rogers, 1961;
Rokeach, 1960). Costa and McCrae (1980) began with just a three-factor model,
assessed by the NEO Inventory (e.g., McCrae & Costa, 1983). Shortly after the
development of the NEO Inventory, they became aware of the Big Five and they
extended their instrument to include agreeableness and conscientiousness.
However, they did not revise their scales assessing neuroticism, extraversion, or
openness, and they subsequently acknowledged that FFM openness does not
align as well with the Big Five (McCrae, 1990).
The assessment of this domain by DeYoung, Quilty, and Peterson (2007)
distinguishes between openness and intellect. DeYoung, Grazioplene, and
Peterson (2012) suggest that their version of openness is associated with
schizotypal thinking but intellect is not. The assessment of openness by Lee and
Ashton (2004) includes a facet scale titled Unconventionality that assesses the
disposition to be eccentric, weird, peculiar, odd, and strange, clearly aligning
with schizotypal thinking. Tellegen (1993) in fact identifies the domain as
“unconventionality,” the scale for which contains items that assess normal
openness (e.g., curious, inquisitive, imaginative, and creative) as well as items
that concern attributes such as having ideas or beliefs that have little basis within
reality, dwelling upon fantasies, or often engaging in activities that are bizarre,
deviant, or aberrant (Tellegen & Waller, 1987). Tellegen (1993) indeed
suggested years ago with respect to his conceptualization of openness that
“markers of this type are clinically suggestive, particularly of schizotypal
personality disorder” (p. 126).
Agreeableness
Graziano and Tobin (this volume) acknowledge that FFM agreeableness has
not received the recognition given to other domains of the FFM, such as
neuroticism, extraversion, or conscientiousness, or at least this attention has been
slower to come. They conducted a PsycINFO search of the keyword
“agreeableness” and identified 2,872 peer-reviewed journal articles written in
English from 1860 to February 7, 2015. More than 97% of these articles were
written after 1992 (the year of the publication of the NEO PI-R; Costa &
McCrae, 1992c). In a related search, they found 135 articles with
“agreeableness” in the title between 1900 and 2014. For purposes of comparison,
the corresponding numbers for titles during this same time frame were 1,600 for
neuroticism, 1,398 for extraversion, 363 for conscientiousness, and 175 for
openness. It might have been interesting to repeat this search using the word
“antagonism,” the opposite pole of the same domain. Perhaps there is a much
larger body of research concerning antagonism, which would nevertheless
concern research addressing the same domain of personality.
Nevertheless, their point is well taken. As they indicated, De Raad (2000)
described agreeableness as the Big Five dimension having “the shortest history”
(p. 91). Neither agreeableness nor antagonism was included within the
influential temperament model of Clark and Watson (2008), nor even within the
initial, short-lived three-factor model of McCrae and Costa (1983), which, as
noted earlier, was confined to neuroticism, extraversion, and openness. Yet
agreeableness is the second domain to emerge from factor analytic studies of the
language and is among the three domains that everyone would agree are
lexically universal (De Raad et al., 2010). Agreeableness is second only to
extraversion with respect to its lexical importance and when paired with
extraversion, defines the highly influential interpersonal circumplex (Leary,
1957; Wiggins, 1991).
The discovery of the circumplex was an epiphany for Leary (1957). Finding
that all manner of interpersonal relatedness falls neatly within a circumplex
arrangement is indeed striking. The discovery is perhaps tempered somewhat by
the possibility that a circumplex can be created by combining any two of the
FFM domains (DeGeest & Schmidt, 2015; Goldberg, 1999; Hofstee, De Raad, &
Goldberg, 1992). Costa and McCrae (this volume) suggest that not all
combinations are equally compelling, empirically or conceptually, but certainly
the most compelling illustration of an additional example is provided by
Tellegen, Watson, and Clark (1999). As extraversion is neatly combined with
agreeableness to create an interpersonal circumplex, Tellegen et al. (1999)
likewise created another circumplex by combining extraversion (reframed as
positive affectivity) with neuroticism (reframed as negative affectivity) to form
an affectivity circumplex.
Graziano and Tobin (this volume) document well various costs to being
antagonistic and benefits to being agreeable. Areas of further investigation are
the potential costs of being too agreeable and the potential benefits of being
antagonistic. There is indeed a body of research supporting, to a degree, the
cliché that nice “guys” finish last and antagonistic “guys” can finish first (e.g.,
Ahmetoglu & Swami, 2012; Doerrenberg, Duncan, Fuest, & Peichl, 2014;
Judge, Livingston, & Hurst, 2012; Lin-Healy & Small, 2013). This is consistent
with the sociobiological view of Nettle (2006) that there are potential costs and
benefits for all traits. Leary (1957) in fact referred to the agreeable octant of the
interpersonal circumplex as the “docile-dependent” octant. In its less severe
form it was said to involve a “poignant or trustful conformity,” in its more
severe form a “helpless dependency” (p. 292). As reviewed in the chapter by
Widiger and colleagues (this volume), and summarized as well by Gore and
Pincus (2013) and Samuel and Gore (2012), there is a considerable body of
research to suggest that maladaptive variants of agreeableness involve dependent
personality traits (e.g., Bagby et al., 2001; Blais, 1997; Haigler & Widiger, 2001;
Lowe et al., 2009; Lynam & Widiger, 2001; Mullins-Sweatt & Widiger, 2007;
Samuel & Widiger, 2004; Sprock, 2002; Trobst, Ayearst, & Salekin, 2004;
Wiggins & Pincus, 1989). Oakley (2012) in fact edited an entire text devoted to
the illustration of variants of pathological altruism.
Conscientiousness
Jackson and Roberts (this volume) indicate that conscientiousness includes “a
broad swath of constructs that reflect the propensity to be self-controlled,
responsible to others, hardworking, orderly, and rule abiding.” The importance
of conscientiousness as a domain of personality is unassailable.
Conscientiousness is one of the three domains that is unambiguously universal
across all languages (De Raad et al., 2010). Traits of conscientiousness are
evident in most every personality trait model and measure (John et al., 2008;
O’Connor, 2002, 2005). Jackson and Roberts (this volume) emphasize in
particular the importance of conscientiousness in predicting a wide array of
significant life outcomes, including occupational success (e.g., Dudley, Orvis,
Lebiecki, & Cortina, 2006; Seibert & Kraimer, 2001), marital stability (e.g.,
Roberts, Harms, Caspi, & Moffitt, 2007), academic achievement (e.g., Noftle &
Robins, 2007; Ozer & Benet-Martinez, 2006), and even health and longevity
(e.g., Friedman & Kern, 2014; Moffitt et al., 2001). As they indicate,
conscientiousness is a well-studied personality trait across other scientific
disciplines (beyond psychology), including economics, political science, and
public policy.
Comparable to the recommendation of Lahey (2009) and Widiger and Trull
(2007) for mass screening for levels of neuroticism, given its considerable public
health care implications, Magidson, Roberts, Collado-Rodriguez, and Lejuez
(2014) have made a similar recommendation for improving persons’ levels of
conscientiousness. “Efforts to change conscientiousness may hold great public
health significance in enabling changes across key outcomes related to health,
functioning, and quality of life” (Magidson et al., 2014, p. 1443). Given the
impact of the level of conscientiousness across such a wide array of important
life outcomes, it would seem that even minimal to mild improvements would
have substantial social, occupational, and personal benefits (Chapman,
Hampson, & Clarkin, 2014; Reiss, Eccles, & Nielsen, 2014). And, just as
Barlow et al. (2014) have developed a pandiagnostic treatment program for
reducing the global level of neuroticism, Magidson et al. (2014) have been
developing a treatment program for increasing persons’ global level of
conscientiousness, using a relevant behavioral intervention considered within the
motivational framework of expectancy value theory (Hopko, Lejuez, Ruggiero,
& Eifert, 2003).
The time is perhaps right for the development of a treatment manual for all of
the domains of the FFM (Widiger & Presnall, 2013). There are studies
supporting the responsivity of FFM traits to therapeutic intervention, including
conscientiousness (e.g., Jorm, 1989; Krasner et al., 2009; Piedmont, 2001; Reiss
et al., 2014), and there are suggestions for treatment from the perspective of the
FFM (e.g., Stepp, Whalen, & Smith, 2013; Stone, 2013), but there is, as yet, no
effort to develop a comprehensive treatment manual, let alone empirically
validated therapies, for developing changes to all five domains of personality
functioning.
However, it is worth noting that no such manual has also been developed for
the DSM-IV (now DSM-5) personality disorders. It has been over 20 years since
the APA has been developing practice guidelines for each of the mental
disorders included within DSM-5 and to date guidelines have been published for
only one personality disorder: borderline (APA, 2001). The absence of
manualized treatment programs does not appear to be due to an assumption that
personality disorders are untreatable, as there is empirical support for change in
personality secondary to pharmacologic and psychotherapeutic intervention
(Leichsenring, & Leibing, 2003; Perry & Bond, 2000).
One good reason for the absence of empirically based manualized treatment
plans is the complex heterogeneity of the DSM-IV-TR personality syndromes
(Widiger & Mullins-Sweatt, 2009; Widiger & Presnall, 2013; Zapolski, Guller,
& Smith, 2013). Each DSM-IV-TR personality disorder is a compound
assortment of maladaptive personality traits (Clark, 2007; Lynam & Widiger,
2001; Widiger & Trull, 2007). Two patients, each meeting the diagnostic criteria
for the same DSM-IV-TR personality disorder, can have few traits in common
(Trull & Durrett, 2005). Given the substantial variability of the defining features
within each diagnostic category, it would be understandably difficult to develop
a uniform treatment program for persons sharing the same personality disorder
diagnosis (Verheul, 2005). A uniform treatment plan could hardly be developed
when the patients sharing the same personality disorder diagnosis vary
tremendously in their personality trait profiles.
Presnall (2013) outlined what might be considered an initial draft of an FFM
treatment manual, indicating potential psychotherapeutic and pharmacologic
treatment recommendations for all 10 poles of all five domains. She suggested
that the five domains of the FFM are much better suited for treatment planning
than the DSM-IV-TR personality disorder diagnoses because they are
considerably more distinct and homogeneous. Extraversion and agreeableness
are the domains of interpersonal relatedness, neuroticism is a domain of
emotional instability and dysregulation, conscientiousness is a domain of work-
related behavior, constraint, and responsibility, and openness is a domain of
cognitive intellect, curiosity, unconventionality, and creativity (Mullins-Sweatt
& Widiger, 2006). Extraversion and agreeableness are confined specifically to
social, interpersonal dysfunction, an area of functioning that is relevant to
relationship quality both outside and within the therapy office. Presnall (2013)
suggested that interpersonal models of therapy, marital–family therapy, and
group therapy might be particularly suitable for these two domains. There is
currently no pharmacotherapy for this interpersonal dysfunction, which is not
actually a criticism; it is simply a further indication of the specificity of the
treatment implications. In contrast to agreeableness and extraversion,
neuroticism concerns depressive, anxiety, and emotional dyscontrol, often
targets for effective pharmacologic interventions, as well as cognitive,
behavioral, and/or psychodynamic. Maladaptively high openness implies
cognitive–perceptual aberrations, and so would likely have pharmacologic
implications (i.e., neuroleptics) that are quite different from the anxiolytics,
antidepressants, and mood stabilizers that would be used for neuroticism. The
domain of conscientiousness is the domain of most specific relevance to
occupational dysfunction, or impairments concerning work and career.
Maladaptively high levels involve workaholism, perfectionism, and
compulsivity, and low levels involve laxness, negligence, and irresponsibility.
There might be specific pharmacologic treatment implications for low
conscientiousness (e.g., methylphenidates; Nigg et al., 2002), although, as yet,
none for maladaptively high conscientiousness. Perhaps there never will be a
pharmacotherapy for high conscientiousness, but the point is that the structure of
the FFM is considerably more commensurate with specific treatment
implications than the heterogeneous and overlapping DSM-IV-TR personality
syndromes.
Construct Validity
Construct validity is the validation of the theoretical model for a particular
construct (Cronbach & Meehl, 1955; Strauss & Smith, 2009). There are many
components to FFM theory. Included within this text was a consideration of its
robustness, its universality, the lexical foundation, factor analytic support,
childhood variants and antecedents, the application to animal personality,
behavior and molecular genetics, and personality neuroscience. Each chapter
will be discussed briefly in turn.
Robustness
One of the major strengths of the FFM is that it appears to capture or subsume
the primary traits and dimensions that exist in other personality inventories.
McCrae and Costa’s (2003) initial FFM research was heavily devoted to this
effort, indicating how one particular measure or model after another could be
well understood in terms of the domains and facets of the FFM, including (but
certainly not limited to) the constructs of the interpersonal circumplex (McCrae
& Costa, 1989b), Henry Murray’s 20 need dispositions (Costa & McCrae, 1988),
the California Psychological Inventory (McCrae, Costa, & Piedmont, 1993), the
Myers–Briggs Type Indicator (McCrae & Costa, 1989a), the Minnesota
Multiphasic Personality Inventory (Costa, Zonderman, McCrae, & Williams,
1985) and many others (see also the chapter by Costa and McCrae). This was not
always welcomed by the proponents of these alternative models. At times it did
perhaps feel as if the Borg was assimilating all manner of individual differences
(Widiger, 2010b). Resistance though was perhaps indeed futile.
O’Connor (2002, 2005) has arguably conducted the most informative studies
demonstrating the robustness of the FFM, addressing numerous inventories and
models within integrative factor analyses. In this chapter, O’Connor (this
volume) extends these findings to address more nuanced questions concerning
robustness, including robustness with respect to the dimensional structure. One
of the intriguing aspects of the chapter is the suggestion for somehow reaching a
consensus as to the optimal location of the axes or anchors, thereby providing
even more clarity as to where other measures of personality are located within
the five dimensional map. As O’Connor (this volume) expresses, “It seems clear
that the field now has a powerful, robust dimensional model, that the
disagreements are over specifics, and that even more progress could be made if
the field agreed on the measurement anchoring points for the dimensions.” There
has long been disagreement with respect to the FFM and the interpersonal
circumplex as to the optimal location of the axes that define the two
interpersonal domains (e.g., McCrae & Costa, 1989b; Wiggins & Pincus, 1989).
O’Connor (this volume) also addresses the suggestion of Ashton and Lee
(2010) that honesty–humility is a major, previously neglected, sixth dimension
of personality (see also the chapter by de Raad and Mlačić). I would add the
simple point that it is readily apparent that honesty and humility are closely
aligned with the straightforwardness and modesty facets of the Costa and
McCrae (1992c) version of the FFM. We can separate them from agreeableness
if we include enough indicators of their presence, but the benefits in doing so are
unclear and there is perhaps much that might be lost. For example, there is
considerable elegance and coherence in the circumplex structure of interpersonal
traits (Benjamin, 1996; Horowitz, 2004; Pincus & Hopwood, 2012; Wiggins,
1991) that would perhaps be sorely maimed, if not dismantled, if honesty and
humility were removed from the interpersonal circumplex.
Universality
There are two fundamental methods of cross-cultural validation, the etic and
the emic (Cheung, van de Vijver, & Leong, 2011; Pike, 1954). The etic approach
applies a particular model, variable, or measure across the universe of different
cultures, testing whether it maintains consistent construct validity. The emic
approach identifies and explores the constructs indigenous to a particular culture,
seeking to determine whether the traits inherent or innate to one culture are also
evident in all other cultures. The chapter by Allik and Realo (this volume) is
largely an etic approach to the question of universality, and they clearly provide
a very sophisticated and cutting edge understanding and conceptualization of this
research.
As Allik and Realo (this volume) indicate, perhaps the largest cross-cultural
study to date has been conducted by Schmitt et al. (2003, 2007, 2008) as part of
the International Sexuality Description project, which included 100 scientists
from 56 countries. In this project they administered the Big Five Inventory
(Benet-Martinez & John, 1998), translated into 29 languages and administered to
17,837 participants from 56 different countries. The results indicated that the
five-dimensional structure was highly robust across major regions of the world,
including North America, South America, Western Europe, Eastern Europe,
Southern Europe, the Middle East, Africa, Oceania, South-Southeast Asia, and
East Asia (Schmitt et al., 2007).
The emic approach though can at times be considered the more compelling
(Church & Katigbak, 1988). It is indeed quite striking to know that extraversion
is the predominant individual differences domain across all languages studied to
date (De Raad et al., 2010). In addition, the etic approach can at times be
perceived as perhaps imposing from outside a particular construct onto a local
perspective that should be given credible respect. On the other hand, a very
compelling case can also be made for the fundamental importance of the etic
approach.
The etic–emic comparison applies as well to other models of classification
and description. Consider, for example, the classification of psychopathology,
and even personality disorders in particular. An emic universal diagnostic
system (i.e., one in which the members of all cultures and countries agree on the
existence of the syndrome) will have an international social utility and consensus
validity (Kessler, 1999), but even if a syndrome is not recognized within one or
more particular cultures that does not necessarily mean that it should not be
recognized. Cultures, like any belief systems, will vary in their veridicality.
Recognition of and appreciation for alternative belief systems are important for
adequate functioning within an international community, but respect for
alternative belief systems does not necessarily imply that all belief systems are
equally valid (Widiger, in press). Perhaps a particular cultural perspective is
relatively inaccurate in its failure to recognize the presence of a particular
syndrome, such as schizophrenia or alcohol abuse. The FFM can be difficult to
replicate within preliterate or less educated populations (Church, 2016; Gurven,
von Rueden, Massenkoff, Kaplan, & Lero Vie, 2013), but the implications of
this for the validity of the FFM, even within that group, are not really clear.
Yamagata et al. (2006) examined whether the etic universality of the FFM is
genetically supported. They conducted factor analyses of matrices of phenotypic,
genetic, and environmental correlations estimated in a sample of 1,209
monozygotic and 701 dizygotic twin pairs from Canada, Germany, and Japan.
Five genetic and environmental factors were obtained for each sample. High
congruence coefficients were observed when phenotypic, genetic, and
environmental factors were compared in each sample as well as when each
factor was compared across samples. If the genetic structure appears to exist
within a particular culture, would it matter that one of the domains was not well
recognized within that culture?
Lexical
De Raad and colleagues are arguably conducting the most authoritative lexical
research (e.g., De Raad et al., 2010). The lexical paradigm stands on a very
compelling foundation. Many alternative models of personality structure have
relied on the insight and brilliance of a compelling theorist, such as Millon
(1981, 2011) or Cloninger (2000). However, the lexical paradigm is more purely
empirical. It is guided by the hypothesis that what is of most importance,
interest, or meaning to persons will be encoded within the language. Language
can be understood as a sedimentary deposit of people’s observations over
thousands of years during the language’s development and transformation. The
most important domains of personality functioning will be those with the
greatest number of terms to describe and differentiate their various
manifestations and nuances, and the structure of personality will be evident by
the empirical relationship among these trait terms (Goldberg, 1993). Any
particular personality structure that is guided by such findings will be consistent
with how persons naturally think about personality structure, and will thereby be
relatively comfortable and straightforward to implement and understand.
The lexical paradigm has a long and eminent history. Galton (1884) is
arguably the first to propose a lexical approach to developing a model of
personality structure. Allport and Odbert (1936) though are appropriately given
due credit for initiating a much more thorough effort, apparently culling 4,500
trait descriptive terms from the second edition of Webster’s dictionary
(Goldberg, 1993). They considered current slang, such as “booster, rooter,
knocker, hoodlum, climber, yes-man, four-flusher, crabber, cake-eater, chiseler,
gigolo, flapper, racketeer, [and] Babbitt” (Allport & Odbert, 1936, p. 3), but
most of these terms did not make the final list of “clearly ‘real’ traits of
personality” (p. 26), albeit “knocker” did. Included within the final list, however,
were auld-farrant, bibliomaniac, bloadshedder, boswellian, chesterfieldian,
creepmouse, deipnosophistic, devil-dodger, soothfast, maungy, swashbuckling,
supersubtle (there were 19 variations of super-), nothingarian, loppy, giggish,
giant-rude, eleutheromaniac, hugger-mugger, hyppish, schoolmasterish, and
jiggish (which mean, respectfully, wise beyond one’s years, book-hoarder,
murderer, companion and observer, elegant or urbane, shy, adept
conversationalist, clergyman, truthful, whiny, heroic, very subtle, holds no
beliefs, actively performing poorly, trifling, very rude, passionate for freedom,
confused, rejecting conventional values, resembling a schoolmaster, and playful
or frisky).
Allport and Odbert (1936) found the list frankly overwhelming, and could not
do much with it. Thurstone (1934), an early pioneer of factor analysis,
administered a substantially reduced list of 60 adjectives to 1,300 raters, and
suggested that “the scientific description of personality may not be so hopelessly
complex as it is sometimes thought to be” (p. 14). Well, maybe Thurstone (1934)
was being overly optimistic. Anyone who reads Allport and Odbert (1936) will
appreciate the hugger-bugger complexity of the task. Anyone reading the chapter
by de Raad and Mlačić (this volume) will appreciate that perhaps little has
changed with respect to this fact.
De Raad and Mlačić (this volume) suggest that only three domains of
personality have true universal, cross-language lexical support: Extraversion,
agreeableness, and conscientiousness (De Raad et al., 2010). This excludes not
only honesty-humility of Lee and Ashton (2008), but also FFM openness and
even neuroticism. Ashton and Lee (2010), however, responded in part by
suggesting that De Raad et al. (2010) were imposing an arbitrarily high threshold
for replication. It is only natural that support for replication will decrease with
each new factor that is extracted. It wasn’t that there was virtually no support for
replication beyond the first three factors. Ashton and Lee (2010) suggest that
there was just relatively less support for an apparent replication of honesty–
humility. “Small differences in the congruence coefficients of factors were
interpreted by De Raad et al. (2010) as indicating large differences in factor
replicability” (Ashton & Lee, 2010, p. 436). To the extent that the arguments of
Ashton and Lee (2010) are correct, this would also support the lexical validity of
the FFM domains of neuroticism and openness. Perhaps the only question would
then be whether it is valid or useful to extract an additional sixth factor, or leave
honesty–humility within agreeableness.
Factor Analysis
Factor analysis is a highly useful analytic device (Clark & Watson, 1995). It is
particularly useful for identifying the common variance among a set of
measures, reducing them to a more manageable set of underlying factors, or,
more theoretically, to identify a set of latent variables that provides a structural
model that defines a set of manifest scales. As indicated by Wright (this
volume), factor analysis has been a central tool in the development and
validation of the FFM (e.g., see the chapters by Costa and McCrae, by
O’Connor, and by De Raad and Mlačić). Nevertheless, factor analysis is not
without its limitations (read Lykken, 1971). As acknowledged by Clark and
Watson (1995), “put simply, factor analytic results provide information, not
answers or solutions” (p. 314). Any single factor analysis can be sorely
misleading and far from conclusive (Schmitt, 2011). It is not at all unusual, for
instance, to find a series of studies in which various laboratories have offered
different factor analytic solutions for the same measure (Osborne & Fitzpatrick,
2012).
As Wright (this volume) indicates, two common problems of factor analysis
complement one another: one is the insufficient representation for a potential
factor and the other is providing an excessive representation for what is actually
just a component of an existing factor. With respect to the first problem, if “one
includes too few measures of a construct (e.g., a single openness scale), it is
unlikely to emerge as a stand-alone factor, and instead these indicators may join
another factor or be orphaned with low loadings on all factors” (Wright, this
volume). The FFM will not emerge in a factor analysis if one of the domains
lacks adequate representation relative to the other domains. Likewise, there can
be an artifactual appearance of a sixth factor if a component of one of the FFM
domains is represented excessively. This latter problem was first identified by
Cattell and Tsujioka (1964), who coined this a “bloated specific factor.”
A bloated specific factor occurs when a trait that would normally be just a
component of a larger higher-order domain is overrepresented by an excessive
number of scales relative to other traits within that same domain. These scales
will correlate more highly with one another than they will with the other scales
from that same domain, and will then naturally bind together to form their own
separate factor. As expressed by DeYoung, Grazioplene, and Peterson (2012),
“if multiple measures of a single lower-level trait are present among the
variables to be factor analyzed, their intercorrelations may be strong enough to
cause them to form a separate factor, even when the other factors recovered are
at a higher level of the trait hierarchy and one of them should subsume the
lower-level trait in question” (p. 65). Wright (this volume) and DeYoung (2011)
even suggest that it might be possible to separate a well-established facet, such
as anxiousness, from the domain of neuroticism. There is little doubt that
anxiousness is well understood as a facet of neuroticism (Costa & McCrae,
1992c, 1995; Goldberg, 1993; John et al., 2008; Mervielde, De Clercq, De Fruyt,
& Van Leeuwen, 2005; O’Connor, 2002). Perhaps if multiple scales of
anxiousness were included within a factor analysis of FFM neuroticism scales,
the anxiousness scales would tightly bind together to form a separate factor, as if
anxiousness is a higher order construct distinct from neuroticism and comparable
to extraversion and agreeableness.
Consider what would happen if one included 10 scales measuring different types of
anxiety in a factor analysis with the 30 facets of the Big Five measured by the NEO
PI-R. One would be likely to find a sixth factor for anxiety, in addition to the usual
Neuroticism factor encompassing traits like depression, vulnerability, and self-
consciousness. This would be considered a bloated specific factor because the
location of anxiety as a lower-level trait within Neuroticism is well established (John
et al., 2008, Markon et al., 2005).
(DeYoung, 2011, p. 718)
Berghuis, Kamphuis, and Verheul (2012) administered the 19 scales of the

General Assessment of Personality Disorders (GAPD; Livesley, 2006), the 16
scales from the Severity Indices for Personality Problems (SIPP-118; Verheul et
al., 2008), and the 30 facet scales of the NEO PI-R (Costa & McCrae, 1992c) to
a joint factor analysis. The GAPD includes 15 scales that assess for different
forms of self-pathology and four scales that assess for different forms of
interpersonal dysfunction. The SIPP-118 includes four additional scales of self-
pathology. Berghuis et al. (2012) submitted the correlations among these scales
to a factor analysis that yielded seven factors. The NEO PI-R scales loaded
substantially on and helped to define six of the seven factors. Only the first
factor was not well defined in large part by multiple NEO PI-R scales. This first
factor was confined to 19 scales of self-pathology (15 from the GAPD and four
from the SIPP-118). Berghuis et al. (2012) concluded that the core components
of personality disorder (i.e., self-pathology) and the FFM formed “clearly
distinct components of personality” (Berghuis et al., 2012, p. 704).
Oltmanns and Widiger (in press), however, suggested that the factor analysis
of Berghuis et al. (2012) demonstrated an instance of a bloated specific factor.
The 15 GAPD scales assess alternative forms of self-pathology, such as lack of
self-clarity, self-state disjunctions, fragmentary self–other representations,
defective sense of self, and poorly differentiated images of others, constructs
(and scales) that are highly correlated with one another. However, self-pathology
can be understood as a component of neuroticism (i.e., maladaptive view of the
self or self-image). Indeed, self-pathology scales correlate substantially with
neuroticism (DeShong et al., 2015; Mullins-Sweatt et al., 2012). The NEO PI-R
includes only six scales per domain, and these scales (e.g., Angry Hostility, Self-
Consciousness, and Vulnerability) would not correlate with one another to the
same extent as the self-pathology scales. Across two independent samples and
two different measures of the FFM, Oltmanns and Widiger (in press) reported
that deficits in the sense of self separated from neuroticism when all 15 GAPD
scales of self-pathology were included. However, self-pathology loaded with
FFM neuroticism when only a subset of the GAPD self-pathology scales was
included. They further demonstrated the bloated specific factor phenomenon by
even separating social withdrawal scales from introversion and, as suggested by
DeYoung (2011), anxiousness from neuroticism, through an excessive
representation of these facets of introversion and neuroticism, respectively.
Wright (this volume) suggests that the joint factor analysis of FFM scales
along with measures of schizotypy and dissociation by Watson, Clark, and
Chmielewski (2008) represented another instance of a bloated specific factor.
Watson et al. (2008) reported a separation of schizotypy and dissociation from
FFM openness, concluding that schizotypal traits (e.g., cognitive–perceptual
aberrations) lie outside of the FFM. However, as stated by Wright (this volume),
“In each of their analyses it could be argued that they over-saturated their
models with scales related to schizotypy/oddity, which may have served to
virtually guarantee that a separate factor would emerge for schizotypy/oddity.”
For example, Watson et al. (2008) administered 10 scales assessing the domains
of the FFM (two for each domain), four alternative measures of dissociation, and
nine scales assessing different features of schizotypal personality disorder. In
another factor analysis they included five alternative measures of dissociation. In
both cases the schizotypal and dissociation scales formed their own distinct
factor, separate from openness (and other domains of the FFM). It is perhaps not
surprising that multiple measures of a maladaptive variant of a specific facet of
openness would bind together to form a factor separate from the normal variants
of openness.
Childhood and Adolescence

There was a period of time when an assessment of personality in childhood
and adolescence was discouraged. Personality is generally understood to be the
result of the interaction of innate temperament with environment, and the
presumption was that the result of this interaction would not be fully settled until
perhaps young adulthood.
However, as demonstrated well by De Pauw (this volume), those days are now
largely gone. Personality traits are readily evident in childhood. The
classification of personality disorder by the APA (2013), and even
psychopathology more generally, has not been so well-informed. The APA
diagnostic manual has long included a section devoted to the diagnosis of
childhood disorders, distinct from the disorders seen within adults. This
distinction grew out of the historical context of clinical specialization, wherein
some clinicians devote their practice and research to disorders of childhood and
an entirely separate group of clinicians and researchers will focus on disorders of
adulthood. There has typically been little communication or perhaps even much
awareness across these disciplines (Frances, First, & Pincus, 1995). This
specialization has contributed to an arbitrary bifurcation in the conceptualization
and classification of mental disorders across developmental stages, further
hampering the study of the longitudinal course of psychopathology (Pine et al.,
2002). One of the odd aspects of this bifurcation is that once a child became an
adult, she or he might no longer have a mental disorder simply because it was
not actually recognized in adulthood, such as attention deficit hyperactivity
disorder. Some disorders of childhood are indeed primarily a concern,
disruption, or trouble for the adults interacting with that child. Once the person is
able to determine whether she or he considers it to be a significant problem, she
or he does not seek treatment. Oppositional defiant might be such an example (it
used to be speculated that oppositional defiant disorder in some cases grew up to
become a passive-aggressive personality disorder; APA, 1980).
In other cases, the childhood diagnosis might switch to a different diagnosis of
adulthood, as if the disorder somehow, suddenly changed at the age of 18 years,
such as avoidant disorder of childhood becoming avoidant personality disorder
in adulthood, identity disturbance of adolescence becoming borderline
personality disorder, or conduct disorder becoming antisocial personality
disorder (APA, 1980). The authors of DSM-IV (APA, 1994) worked toward a
more unified classification, in some instances breaking down the arbitrary
boundary between childhood and adulthood to achieve a more developmentally
informed, lifespan perspective (Frances et al., 1995). One example included the
decision to subsume the DSM-III-R (APA, 1987) “childhood” diagnosis of
overanxious disorder within the more general diagnoses of generalized anxiety
disorder (Klein, Tancer, & Werry, 1997). Rather than suggest that overanxious
disorder of childhood changes to a new disorder upon reaching the age of 18
years, DSM-IV (APA, 1994) indicated that the same disorder can be present
across the lifespan but varies in its expression due to differences in age and
developmental context (Shaffer, Widiger, & Pincus, 1998). Regrettably,
however, avoidant disorder of childhood did not become a childhood variant of
avoidant personality disorder (as conduct disorder is a childhood variant of
antisocial); it was annexed instead by the anxiety disorders work group as a
childhood variant of generalized social phobia.
The stark independence of the child and adult research fields has contributed
to a lack of knowledge of the childhood antecedents of some disorders,
particularly the personality disorders (Cohen & Crawford, 2005; De Fruyt & De
Clercq, 2013; Widiger, De Clercq, & De Fruyt, 2009). The one exception was
the empirical support for conduct disorder as the childhood variant of antisocial
personality disorder (Robins, 1978). In addition, advances in one domain went
essentially unknown to the other. One of the more productive and influential
shifts in the classification of adult psychopathology has been the recognition of
the cross-cutting domains of internalization and externalization (Krueger &
Markon, 2006) that had been identified by Achenbach (1966) many years ago
within childhood psychopathology.
De Pauw (this volume) though provides “a plea for the study of child
personality traits in its own right, rather than assuming that child personality
taxonomies are identical to those established in adults.” This is perhaps
somewhat analogous to the interplay and/or tension between etic and emic cross-
cultural research (Cheung et al., 2011; Pike, 1954). It should not be at all
surprising to find that the Big Five does not immediately or suddenly appear in
infancy. A more developmental perspective for its emergence is necessary, both
with respect to the number and nature of the primary domains. In particular De
Pauw (this volume) cites the study by Soto and John (2014). Soto and John
suggested that children’s personality might be better described in terms of six
dimensions, with activity split off as an independent domain, with the Big Five
structure becoming more clearly evident from ages 15 to 18–20 years. Similarly,
Caspi, Roberts, and Shiner (2005) suggest that openness to experience may
begin to develop early in life, but due to ongoing cognitive development, its
conceptual core may shift from exploration of the physical world to an
intellectual curiosity and imagination by middle childhood and early
adolescence. “Hence, openness in young children would not only be defined by
cognitive characteristics (e.g., imagination, creativity), but also by aspects of
motor activity and physical exploration” (De Pauw, this volume).
Animal Personality
Very few individual differences might be imagined among fruit flies. Every
one I have known has seemed to be pretty much the same one. In fact, it would
be difficult for me to even classify the species as a whole in terms of the FFM,
let alone notice any meaningful differences among them, other than finding them
to be tremendously annoying flying around my line of vision when I type.
Gosling (2008), however, readily distinguished two fruit flies in particular.
Frank was an aggressive fruit fly, pushing, punching, and kicking other fruit
flies, whereas Fred, in stark contrast, was clearly quite timid.
Consider, as well, sloths. When placing these adorable, seemingly docile
animals within the FFM dimensions, we might think of them as being low in
neuroticism (not fearful, emotionally reactive, or excitable), low in extraversion
(i.e., they live alone and appear to lack much surgency, sociability, energy,
vivacity), low in openness (not especially exploratory), high in agreeableness
(affability, affection, social closeness), and low in conscientiousness (lack of
attention and goal directedness, erratic, unpredictable, disorganized). However,
these impressions are based largely on a cross-species perspective, that is,
relative to us. Although there is little research on the traits of sloths, let alone
FFM personality profiles for sloths, there is evidence to suggest that, as
hypothesized, sloths are indeed solitary animals, traveling through trees either
alone or with just an infant family member, as well as perhaps being low in
openness as they are unlikely to go out exploring (although this can be blamed
on the fact that their maximum speed is 6 feet per minute). However,
inconsistent with expectations, there is also evidence to suggest that sloths are
aggressive and willing to fight predators when danger is present and perhaps
pose higher levels of conscientiousness, which can be seen in their attention to
detail and goal directedness surrounding their ritualistic bathroom habits. Most
importantly, when considered in their own right, from a more emic view, we will
in fact observe a considerable amount of individual differences among the sloths
with respect to (for instance) their neuroticism, agreeableness, and introversion.
The study of animal personality is indeed a rapidly growing line of
investigation. There was at one time some resistance to any such research
(Gosling, 2008). Weiss and Gartner (this volume) also refer to a period of
decline that was shared with human personality research. However, it is now
becoming readily apparent that there is much to be gained from studying
personality within other species. As Weiss and Gartner (this volume) suggest,
“studies of animal personality and outcomes related to welfare,
psychopathology, and psychological needs could be used to better understand
these associations in humans.” There are clearly many potential advantages of
addressing questions regarding the impact and importance of personality
functioning in animals, including (for instance) greater experimental control,
better access to physiological parameters, increased opportunity for naturalistic
observations, and accelerated lifespan allowing for productive longitudinal
research (Gosling, 2008).
For the construct validity of the FFM, animal personality research provides
fairly strong arguments for its universality. Weiss and Gartner (this volume) note
how the etic and emic principles and issues that pervade the discussion of cross-
cultural research apply as well to cross-species research. If cross-cultural
research is thought to be complicated, imagine the effort to research the same
traits and structure across species. It can indeed be quite difficult to translate the
expression or meaning of a behavior from one species to another. Equally
important, although the pole of one domain may predominate within a particular
species, we will also observe considerable individual differences that are quite
meaningful within that species but are not readily apparent to the human who
sees only the one forest rather than the individually different trees.
Genetics
“The study of behavior and molecular genetics can help us examine where
personality traits come from, how they develop, and how they change over time”
(Jarnecke and South, this volume). Where they came from is clear: genetics.
Why they came is not so clear. Yet, if one considers the five domains of the
FFM, perhaps it is understandable (Nettle, 2006). The five domains of the FFM
do appear to represent the primary domains of personality functioning, covering
exhaustively the most important traits for describing oneself and others
(Goldberg, 1993).
The five broad domains in their order of typical extraction and size from
English and other languages are extraversion, agreeableness, conscientiousness,
emotional instability, and openness (or unconventionality). In other words, the
first two relatively largest domains concern a person’s manner of interpersonal
relatedness. It is to be expected that the domains of personality functioning
considered to be relatively most important to persons across cultures and
languages when describing themselves and other persons would concern how
persons relate to one another. All manner of interpersonal relatedness is
contained within the FFM domains of agreeableness and extraversion, and how
the members of a society, country, culture, or tribe relate to one another would
naturally be of primary importance to most everyone.
The third domain of personality extracted from all languages is
conscientiousness (otherwise known as constraint). This domain concerns the
control and regulation of behavior, and contrasts being disciplined, compulsive,
dutiful, conscientious, deliberate, workaholic, and achievement-oriented, with
being carefree, irresponsible, lax, impulsive, spontaneous, disinhibited,
negligent, and hedonistic (Jackson and Roberts, this volume). It is again perhaps
self-evident that all cultures would consider it to be important to describe the
extent to which its members are responsible, conscientious, competent, and
diligent as a mate, parent, friend, employee, soldier, or colleague (versus being
negligent, lax, disinhibited, and impulsive).
The fourth domain, emotional instability, albeit not as important or as large as
the first three, and perhaps not as reliably or consistently defined (De Raad et al.,
2010), is certainly of considerable importance (as noted earlier) in the fields of
clinical psychology and psychiatry, saturating most measures of personality
disorder and psychopathology more generally (Lahey, 2009; Widiger, 2009; see
also the chapter by Tackett and Lahey). It would again not be surprising that
most, and perhaps all, cultures consider the emotional stability (anxiousness,
depressiveness, irritability, volatility, anger, and vulnerability) of its partners,
children, friends, workers, laborers, and employees to be of considerable
importance.
The fifth domain, openness, intellect, or unconventionality, reflects a culture
or society’s interest in creativity, intellect, and imagination, contrasting being
open-minded, unusual, odd, weird, creative, peculiar, and unconventional with
being closed-minded, practical, conventional, and rigid. Cognitive, intellectual,
and creative growth, the advancement and expansion of knowledge, and the
drive for curiosity, investigation, and inquisitiveness are all dependent upon this
domain (see also the chapter by Sutin). It is very difficult to imagine a society,
culture, or tribe progressing without this domain, nor a model of personality
structure.
Additionally, as suggested by O’Connor (this volume), the “consistent power
and effectiveness of the FFM in capturing the dimensions that exist in other
measures, as reviewed [within his chapter], makes it difficult to imagine that
additional dimensions have been missed all of this time.” It is frankly difficult to
find an important trait that lies outside of these five domains. The five domains
do appear to be reasonably comprehensive in their coverage (Funder, 2001;
Goldberg, 1993; John et al., 2008; McCrae & Costa, 2003; Ozer & Reise, 1994).
Ozer and Reise (1994) and Goldberg (1993) even likened the domains of the
FFM to the coordinates of latitude and longitude that cartographers use to map
the world, suggesting that the FFM might be similarly useful in comparing and
contrasting all personality measures and all of the important personality trait
constructs with respect to their relative saturation of the five fundamental
personality traits.
There is, of course, tremendous variability across persons within each domain,
considerable variations among the traits within the same domain, and substantial
interstitial presence in between the domains. The genetics of this hierarchical
dimensional trait model is unlikely to be simple, as indicated well by Jarnecke
and South (this volume). Nevertheless, this structure is clearly more coherent
genetically than comparable or competing models. As noted earlier, Yamagata et
al. (2006) demonstrated that “the phenotypic structure of the FFM as assessed by
the NEO-PI-R is reflective of an underlying genetic structure and whether the
genetic structure is universal across populations from Canada, Germany, and
Japan” (p. 994), “suggesting that the FFM may represent the common heritage of
the human species” (p. 996). Their final conclusion might be a bit strong (see
also the chapter by Jarnecke and South), but no such comparable support would
likely be obtained for the international classification of maladaptive personality
structure provided by the World Health Organization’s (WHO, 1992)
International Classification of Diseases (ICD-10), or the DSM-IV (now DSM-5;
APA, 2013).
Distel et al. (2009) examined the phenotypic and genetic association between
borderline personality and FFM personality traits in 4,403 monozygotic twins,
4,425 dizygotic twins, and 1,661 siblings from 6,140 Dutch, Belgian, and
Australian families. Multivariate genetic analyses indicated that the genetic
factors that influenced individual differences in neuroticism, agreeableness,
conscientiousness, and extraversion accounted for all of the genetic liability for
borderline personality (albeit unique environmental effects were not completely
shared with the FFM traits). This is a fairly compelling argument that DSM-IV
(now DSM-5) borderline personality disorder can be understood as maladaptive
variants of the FFM domains. It would be of interest to conduct comparable twin
studies concerning the antisocial, narcissistic, obsessive–compulsive, and
avoidant personality disorders. I would predict much less success though for the
schizotypal personality disorder, given how openness is conceptualized and
assessed by the NEO PI-R (Gore & Widiger, 2013) and the genetic association
of schizotypal personality disorder with schizophrenia (Kwapil & Barrantes-
Vidal, 2012).
Neuroscience
The goal of personality neuroscience is to identify the neurobiological systems
and substrates of respective personality traits. Allen and DeYoung (this volume)
review more specifically the neuroscience research concerning FFM traits,
including the available methods, such as neuroimaging techniques,
electrophysiological techniques, molecular genetics, psychopharmacological
manipulation, and assays of endogenous psychoactive substances. Allen and
DeYoung highlight pitfalls and best practices in personality neuroscience,
emphasizing the importance of theoretically informed neuroscience by framing
results in light of a theory of the psychological functions underlying each of the
Big Five. Emphasis was given in particular to the theoretical model of the
authors, which is certainly quite appropriate given that they are arguably the
leading investigators of personality neuroscience (DeYoung, 2010, 2015;
DeYoung & Gray, 2009).
The chapter by Allen and DeYoung is also quite timely, to say the least. The
National Institute of Mental Health (NIMH) has made it clear that the institute is
not very interested in funding studies concerned with the DSM-5 diagnoses. “It
is critical to realize that we cannot succeed if we use DSM categories” (Insel,
2013). This would, of course, include the DSM-5 personality disorders. NIMH
has developed its own diagnostic system, called the Research Domain Criteria
(RDoC; Sanislow et al., 2010). The RDoC consists of five broad domains:
negative valence systems, positive valence systems, cognitive systems, systems
for social processes, and arousal/modulatory systems. More importantly,
perhaps, NIMH would clearly prefer that any such studies embrace a cognitive
neuroscience perspective. “Mental disorders can be addressed as disorders of
brain circuits” (Insel et al., 2010, p. 749). “The primary focus of RDoC is on
neural circuitry, with levels of analysis progressing in one of two directions:
upwards from measures of circuitry function to clinically relevant variation, or
downwards to the genetic and molecular/cellular factors that ultimately influence
such function” (Insel et al., 2010, p. 749). “The first step is to inventory the
fundamental, primary behavioral functions that the brain has evolved to carry
out, and to specify the neural systems that are primarily responsible for
implementing these functions” (Cuthbert & Insel, 2013, p. 4).
There is arguably a strong alignment of the domains of the FFM with the
RDoC (Widiger, 2012). RDoC negative valence (i.e., anxiety, fear, threat)
clearly aligns well with FFM neuroticism. Positive valence (reward, approach)
aligns well with FFM extraversion, as positive affectivity is the driving
temperament for extraversion (Clark & Watson, 2008). Social processes align
with FFM agreeableness (versus antagonism) and introversion (versus
extraversion), as these are the two fundamental domains of all manner of
interpersonal relatedness. The RDoC domain of cognitive systems would include
the psychoticism of DSM-5, which aligns with FFM openness (otherwise known
as intellect). RDoC arousal regulatory systems align with FFM
conscientiousness (DSM-5 low disinhibition), as this domain concerns regulatory
constraint (Clark & Watson, 2008).
The alignment of RDoC with the FFM is perhaps a stretch in one or two cases.
The RDoC domain of cognitive systems concerns attention, perception, working
memory, declarative memory, and language behavior. The cognitive and
perceptual processes and mechanisms included within this domain go beyond the
individual differences in intellect (cognition and perception) covered within the
domain of openness and, of course, does not even necessarily concern
personality traits. Likewise, the RDoC domain of arousal, involving biological
rhythms and the sleep–wake cycle, may have nothing to do with constraint or
conscientiousness. Being entirely and precisely aligned with RDoC would
actually be a rather surprising coincidence. The authors of the RDoC clearly had
no interest or intention of aligning their classification system with a model of
personality. Nevertheless, the alignment that does exist is rather striking. The
FFM would certainly appear to be more commensurate with the NIMH RDoC
perspective than the DSM-IV (now DSM-5) personality disorders, and the
chapter by Allen and DeYoung provides a conceptualization that NIMH should
find very encouraging and intriguing.
Application
There is a wide range of potential and actual applications of the FFM. The
FFM is not just a scholarly academic model of personality structure. FFM
personality traits have been associated with a wide array of consequential life
outcomes and real world applications. Included within this section was a
consideration of its assessment, its application to business and industry, health
psychology, marital and family therapy, the conceptualization of personality
disorders, adult psychopathology, child psychopathology, and clinical utility.
Each of these chapters will be discussed briefly in turn.
Assessment
A testament to the significance of the FFM is simply the sheer number of
alternative measures that have been developed for its assessment. Widiger and
Trull, back in 1997, devoted a review paper within a special section of the
Journal of Personality Assessment precisely to this topic. They indicated that
“the FFM has become so compelling that a variety of instruments [for its
assessment] have been developed, and many existing instruments have been
modified to assess the FFM” (Widiger & Trull, 1997, p. 230). They noted that
they could not cover all of the relevant measures as space limitations would be
prohibitive and therefore they confined their “review (perhaps appropriately) [to]
five alternative instruments: the Goldberg (1992) Big Five Markers, the
Interpersonal Adjective Scales-Big Five (IASR-B5; Trapnell & Wiggins, 1990),
the NEO Personality Inventory-Revised (NEO PI-R; Costa & McCrae, 1992c),
the Personality Psychopathology-Five (PSY-5; Harkness & McNulty, 1994), and
the Hogan Personality Inventory (HPI; Hogan, 1986; Hogan & Hogan, 1992)”
(Widiger & Trull, 1997, p. 230). Subsequently, De Raad and Perugini (2002)
devoted an entire book to this topic, including many additional measures, such as
the Hierarchical Personality Inventory for Children (Mervielde & De Fruyt,
2002), the Structured Interview for the Five Factor Model (Trull & Widiger,
2002), the Global Personality Inventory (Schmit, Kihm, & Robie, 2002), the Big
Five Marker Scales (Perugini & Di Blas, 2002), and the Zuckerman–Kuhlman
Personality Questionnaire (Zuckerman, 2002). There have been quite a few more
instruments since then (Samuel, 2013), covered well in the chapter by Simms,
Williams, and Simms (this volume).
Simms et al. (this volume) also consider the development of measures of
maladaptive variants of the FFM. They identify a number of such measures,
including the Five Factor Form (FFF; Rojas & Widiger, 2014). They consider in
particular the Personality Inventory for DSM-5 (PID-5; Krueger et al., 2012) and
the Computerized Adaptive Test-Personality Disorder (CAT-PD; Simms et al.,
2011). Worth recognizing as well, perhaps, are eight Five Factor Model
Personality Disorder (FFMPD) scales (Lynam, 2012; Widiger, Lynam, Miller, &
Oltmanns, 2012). Many of the FFMPD scales align closely with scales from the
CAT-PD (Simms et al., 2011) and the PID-5 (Krueger et al., 2012) as well as the
Dimensional Assessment of Personality Pathology-Basic Questionnaire
(Livesley & Jackson, 2009) and the Schedule for Nonadaptive and Adaptive
Personality-2 (Clark, Simms, Wu, & Casillas, in press). Crego and Widiger (in
press) reported the results of joint factor analyses of the CAT-PD, PID-5, and
selected FFMPD scales. As they indicated, “these measures were constructed
with different rationales and methods, yet the end result was highly congruent.”
Indeed, the FFMPD scales were explicitly developed to be measures of
maladaptive variants of the FFM (Lynam, 2012). The scales within these
FFMPD inventories are aligned not only with particular domains of the FFM,
but as well with the more specific facets, such as Workaholism aligning with the
facet of achievement-striving from FFM conscientiousness (Samuel, Riddell,
Lynam, Miller, & Widiger, 2012), Rashness aligning with low deliberation from
FFM conscientiousness (Mullins-Sweatt et al., 2012), Gullibility aligning with
the facet of trust from FFM agreeableness (Gore, Presnall, Lynam, Miller, &
Widiger, 2012), Exploitative aligning with the facet of low altruism from FFM
agreeableness (Glover, Miller, Lynam, Crego, & Widiger, 2012), Attention-
Seeking with the facet of gregariousness from FFM extraversion (Tomiatti,
Gore, Lynam, Miller, & Widiger, 2012), and Social Withdrawal with the facet of
low gregariousness from the domain of FFM extraversion (Edmundson, Lynam,
Miller, Gore, & Widiger, 2011). Over 100 such scales have been developed and
validated (see also the chapter by Widiger, Gore, Crego, Rojas, and Oltmanns,
for further description). All of the scales from all eight of the FFMPD
instruments have been validated as measures of their respective FFM domains,
as well as the underlying facets (e.g., Edmundson et al., 2011; Glover et al.,
2012; Gore et al., 2012; Lynam et al., 2011; Mullins-Sweatt et al., 2012; Samuel
et al., 2012; Tomiatti et al., 2012).
The FFMPD inventories have a number of scales for maladaptive variants of
extraversion, conscientiousness, and agreeableness (as well as for high and low
openness, and even for low neuroticism; Widiger et al., 2012). It is important to
note that the FFMPD scales are not alone in the representation of maladaptive
variants of both poles of the FFM. For instance, the CAT-PD (Simms et al.,
2011) has three scales for maladaptive conscientiousness (i.e., Perfectionism,
Rigidity, and Workaholism), and one scale, Exhibitionism, for maladaptive
extraversion (Wright & Simms, 2014).
Business and Industry

Seibert and De Geest (this volume) document well the productive application
of the FFM within business and industry. As they indicate, the initial
consideration of personality traits in industrial–organizational psychology,
human resource management, and organizational behavior was overly optimistic,
and the pendulum eventually swung back in the other direction to excessive
pessimism when the initial high expectations were not met. However, “through
the decline and the resurgence [of the interest in personality traits], the Five
Factor Model has predominated as an integrative personality structure for
conceptualizing and researching the potential importance of personality traits for
optimal and/or problematic performance” (Seibert and De Geest, this volume).
Seibert and De Geest focus on three particular issues: the validity of personality
variables as predictors of important attitudes, behaviors, and outcomes;
generalization across situations; and bandwidth—fidelity.
Another potential line of investigation is the assessment of maladaptive
variants of the FFM that would be of particular relevance to business and
industry. As indicated in the chapter by Simms, Williams, and Simms (this
volume), considerable attention is currently being given to the development of
scales to assess maladaptive variants of the FFM. As noted earlier, these
instruments include (but are not limited to) the PID-5 (Krueger et al., 2012), the
CAT-PD (Simms et al., 2011), and the FFMPD scales (Lynam, 2012; Widiger et
al., 2012). The selection of these scales for these instruments was governed
largely by their potential application within clinical psychology. It is quite
possible that these selected scales would have a relevance comparable to the
assessment of maladaptive personality functioning within business and industry,
but also possibly not.
One of the more commonly used FFM measures in business and industry is
the HPI (Hogan & Hogan, 1992, 2002). The HPI is a well validated measure that
converges with other measures of the FFM, although with some important
differences, due in part to Hogan’s socioanalytic model of personality but also
out of an interest to be slanted toward traits of particular relevance to business,
including primary scales such as Ambition, Prudence, Inquisitive, and Learning
Approach, and subscales such as Competitive, Leadership, No Social Anxiety,
Entertaining, Easy to Live With, Likes People, No Hostility, Science Ability,
and Generates Ideas. The same rationale could also be made for a selection of
scales to assess maladaptive variants for optimal use in business and industry.
There is also though some reluctance to assess for maladaptive traits within
business employees due in part to the stigma and pejorative stereotyping that
such labeling could entail. Hogan and Hogan (1997) in fact developed the Hogan
Developmental Survey (HDS) as a means to assess DSM-IV-TR personality
disorders within employees in a manner that appeared to be more complimentary
than critical (Furnham, Milner, Akhtar, & De Fruyt 2014). The instrument is
explicitly intended to assess “dark side” personality traits, while at the same time
suggesting that they emerge only when under stress. Each personality disorder is
also retitled in a manner that conveyed a positive personality expression:
borderline is renamed excitable, paranoid becomes skeptical, schizoid is just
reserved, histrionic is actually colorful, the narcissistic is bold, and schizotypal
becomes imaginative. Even antisocial personality disorder is conveyed in a
positive manner, with its scale titled Mischievous. The items within the
respective scales assess for these adaptive, positive variants of each personality
disorder. It would then, in theory, be possible to assess for the DSM-IV-TR
personality disorders in a nonpejorative manner. It is perhaps reasonable to
question whether the personality disorders can be validly assessed when they
involve socially desirable, adaptive traits, but research has provided support for
the validity of the HPS (e.g., De Fruyt et al., 2009; Furnham & Crump, 2005;
Furnham, Trickey, & Hyde, 2012; Rolland & de Fruyt, 2003).
In addition, it is reasonable to question whether the DSM-IV-TR personality
disorders would provide the optimal constructs for maladaptive personality
functioning within business and industry. There is certainly a good deal of
interest in the recognition and assessment of psychopathy in business (Babiak &
Hare, 2006), but it is perhaps questionable whether there would be a comparable
interest in assessing, for instance, the schizotypal personality disorder. The DSM
personality disorder nomenclature would likely be much different if it was
authored by experts identifying the most problematic personality syndromes
within business and industry.
The HPI does assess for some maladaptive personality traits (such as the
troublemaker, or having narrow interests). However, for a measure developed to
provide an assessment at the facet level of maladaptive traits at the workplace,
there should probably be a particularly heavy representation of traits from high
and low conscientiousness. The DSM-IV-TR personality syndromes are weighted
heavily in favor of traits from neuroticism (Lynam & Widiger, 2001), as
neuroticism pervades clinical practice (Widiger, 2009). In contrast, as indicated
by Seibert and De Geest (this volume), conscientiousness is the domain of
personality that is of most relevance to success and failure at the workplace.
Therefore, it would seem natural that a measure of maladaptive personality traits
for the workplace (comparable to the PID-5, SNAP-2, and/or FFMPD scales)
would include a particular emphasis on scales assessing maladaptively high and
low conscientiousness.
Health Psychology
Kern and Friedman (this volume) provide an outstanding review of the
research on the presence, importance, and mechanisms of the relationship of
personality to physical health. As they demonstrated well, “the five factor
approach has shown excellent utility for understanding health, including physical
and mental health, longevity, cognitive function, social competence, and
productivity” (Kern and Friedman, this volume). They emphasize in particular
the importance of obtaining longitudinal data across persons’ lives. “Much of
our existing knowledge of life course personality-health associations stem from
longitudinal data—extensive data gathered prospectively across time
documenting people’s lives” (Kern and Friedman, this volume). This research
should begin early in a person’s life, childhood and/or adolescence, and then
followed well into adulthood and even older age. Of course, no researchers could
ever conduct such a study, unless they began when they were children, or
managed to live past much of the lives of their oldest participants.
One study that Kern and Friedman referred to in particular was the results of
Friedman et al. (1993). This study is worth describing in some detail, as it was so
creative, interesting, and informative. Friedman and colleagues (1993) obtained
raw data from the Terman Life-Cycle Study of children, beginning in 1921–
1922. Terman began with 1,500 bright male and female children (age 11 years)
and continued to assess them every 5 to 10 years. The sample was fairly
homogeneous (i.e., bright, mostly white and mostly middle-class). The sample
was at times referred to as a longitudinal study of budding geniuses, but this did
not turn out to be the case. The sample though was certainly bright.
Approximately 70% became college educated, and many were business persons,
physicians, lawyers, teachers, scientists, and writers. Despite the apparent
homogeneity, the restriction in range was not so severe as to prohibit meaningful
results. By the time Friedman and colleagues (1993) considered these data, the
longitudinal information was quite wealthy with respect to the prediction of
longevity. Indeed, the fact that meaningfully significant results were obtained
despite the restriction in range is perhaps itself a testament to the strength of the
results.
In 1922, a participant’s parent provided personality ratings of his or her child
(age 11 years) on 25 trait dimensions. There were also a few additional questions
that were scored for personality. “We endeavored to select personality variables
that basic theory and research have shown are reliable and theoretically
meaningful-dimensions that appear in the Big Five factors of personality”
(Friedman et al., 1993, p. 177). More specifically, they developed proxy scales
for extraversion, conscientiousness, and neuroticism. Sadly, there was
insufficient information to score for agreeableness or openness. In any case, the
findings demonstrated that even when controlling for other obvious predictors,
“conscientiousness in childhood was clearly related to survival in middle to old
age” (Friedman et al., 1993, p. 176), the potential mechanisms for which are
discussed in the chapter by Kern and Friedman (see also the chapter by Jackson
and Roberts).
The success of their study inspired researchers to seek other treasure troves of
long-forgotten personality data. Imagine obtaining access to original Woodworth
(1919, 1920) Personal Data Sheet results, although that would perhaps be
confined largely to neuroticism. Even better would be Bernreuter (1931)
personality data, or perhaps original MMPI (Hathaway & McKinley, 1943)
findings, although the latter would not likely have much representation of
conscientiousness (Costa, Busch, Zonderman, & McCrae, 1986; Costa et al.,
1985; Han, Weed, & McNeal, 1996). Nevertheless, a clear strength of the FFM
is its robustness across alternative personality measures and models, with most
any instrument readily rescored with respect to the heavily researched FFM
domains (O’Connor, 2002, 2005).
An additional focus of discussion within this chapter was the construct of the
Type A personality. The Type A personality was first described as a potential
risk factor for heart disease in the 1950s by the cardiologists Meyer Friedman
and Ray Rosenman. Type A persons were described as having personality traits
such as excessive competitiveness, time urgency (impatience), and angry
hostility. As indicated by Kern and Friedman (this volume), “Although early
health psychology research focused on Type A behavior and related domains of
hostility, over the past two decades the Five Factor Model (FFM) has become
the dominant organizing framework for integrating studies of personality and
health (Smith & Williams, 1992).”
Although the original syndrome has not held up well, with perhaps angry
hostility being the driving force behind the health risk, it is perhaps remarkable
that never in the history of the APA diagnostic manual, from DSM-III through
DSM-III-R, DSM-IV, and up to DSM-5, has there ever been even a proposal to
consider including Type A within the personality disorders section. This is
surprising and perhaps quite regrettable. Here was a personality syndrome with
strong empirical support for quite important life health implications, yet it has
largely been ignored within the personality disorder field. The inclusion of such
a syndrome within the personality disorders section would have provided a
degree of credibility across the entire domain of medicine for the importance of
considering personality disorders. This was an opportunity that was perhaps
sorely missed.
However, with the shift toward a dimensional trait model, the potential for the
diagnostic manual to have an impact across medicine might become truly
realized. The FFM (Costa & McCrae, 1992c) and the DSM-5 (Krueger et al.,
2012) dimensional trait models both include a scale for angry hostility, as well as
for other traits of the Type A personality syndrome. This has long been one of
the advantages of a dimensional trait model, the FFM in particular, being able to
construct new, clinically useful syndromes out of the building blocks of the
reasonably comprehensive FFM lexicon (Widiger & Costa, 2012; Widiger et al.,
1994). There are indeed many other maladaptive traits that could be described,
such as alexithymia (Taylor & Bagby, 2013), racism (Flynn, 2005), and the
depressive personality (Bagby, Watson, & Ryder, 2013).
Couple’s Therapy and Crossover Analysis

The chapter by Piedmont and Rodgerson (this volume) describes the
application of FFM personality description for couple’s therapy, more
specifically, for crossover analysis. Crossover analysis concerns a comparison of
each person’s self-description with the description provided by the spouse. The
essence of a crossover analysis is identifying those points of discrepancy across
self and spouse. The areas of disagreement identify potential points of conflict,
or at least identify and clarify how the couple may have an inaccurate view of
themselves and their spouse within the relationship. The FFM thereby offers an
interesting basis for a couple’s therapeutic analysis and intervention.
I think one of the more interesting illustrations of crossover analysis was
provided by Costa and Piedmont (2003) in their interpretation of the NEO PI-R
profile of Madeline G., a case study presented in the engaging text of Wiggins
(2003). Wiggins wrote a text on what he considered to be the five basic
paradigms of personality assessment (i.e., psychodynamic, interpersonal,
personological, multivariate, and empirical). Anyone familiar with personality
assessment must be familiar with the seminal and highly influential contributions
of Wiggins (e.g., Wiggins, 1973) and his 2003 text is also a delight to read. To
illustrate each paradigm Wiggins invited experts for each paradigm to provide an
assessment of the same person, Madeline G. Madeline was a very engaging,
inspirational, colorful, and intriguing woman. At the time of the assessments, she
was a civil rights lawyer newly employed in a prestigious law firm. She was
currently living with her common-law husband, in what she felt was a very
successful and satisfying relationship for the both of them, or so she thought.
Madeline completed the NEO PI-R, as did her husband, describing her. What
was immediately evident was the sharp discrepancy in their FFM profiles of
Madeline. Madeline described herself in a very positive light, whereas her
husband, not so much. Whereas she described herself as being high in trust and
altruism, her husband described her as being very low (they both agreed that she
was very low in straightforwardness, compliance, and modesty). Whereas she
described herself as being high in warmth, he described her as very low (they
both agreed she was high to very high in gregariousness, assertiveness, and
excitement-seeking). Whereas she described herself as being high to very high in
competence, dutifulness, self-discipline, and deliberation, he described her as
being low to very low on each of these facets of conscientiousness (they agreed
that she was high to very high in achievement-striving). These sharp
discrepancies did not bode well for a successful relationship, and if her husband
was correct, did not bode well for Madeline.
Wiggins’ text was delayed in its publication, at least 3 years since the
assessments were conducted. In that time, a lot changed in Madeline’s life. At
the time of the assessment she was a rising star in the law profession. However,
subsequent to the assessment “Madeline and her boss apparently agreed that
Madeline wasn’t capable of being an employee” (Trobst & Wiggins, 2003, p.
317). Approximately 1.5 years after the assessment, her husband left her, much
to her surprise and substantial dismay. It is questionable whether any of the
experts did or could have predicted such dramatic changes in Madeline’s life,
but it is certainly the case that her crossover FFM profile comparison may have
been quite prescient, at least with respect to the relationship. Madeline did not
appear to have a clue as to her husband’s critical view of her, nor would she be
aware that a good deal of his view was perhaps quite accurate.
Personality Disorders
As indicated by Widiger, Gore, Crego, Rojas, and Oltmanns (this volume),
there is a substantial body of research supporting the conceptualization of the
DSM-IV-TR (now DSM-5; APA, 2013) personality disorders as maladaptive
variants of both poles of all five domains of the FFM. Wiggins and Pincus
(1989) provided the first published study concerned explicitly with the empirical
relationship of the FFM to personality disorder symptomatology, conducting
joint factor analyses of measures of the FFM, the interpersonal circumplex, and
DSM personality disorders. They concluded that “conceptions of personality
disorders were strongly and clearly related to dimensions of normal personality
traits” (Wiggins & Pincus, 1989, p. 305), including (but not limited to) a close
relationship of schizotypal symptoms with openness; dependent with
agreeableness; antisocial, paranoid, and narcissistic with antagonism; borderline
with neuroticism; histrionic and narcissistic with extraversion; schizoid with
introversion; and compulsive with conscientiousness. Although the interpersonal
circumplex was able to provide a meaningful and informative understanding of a
subset of the personality disorders, Wiggins and Pincus (1989) reached the
conclusion that “the full 5-factor model was required to capture and clarify the
entire range of personality disorders” (p. 305).
Trull (1992) provided the first study to include the administration of measures
of the FFM and personality disorder symptomatology within a clinical sample.
He administered the NEO-PI (Costa & McCrae, 1992c) and three independent
measures of the DSM-III-R personality disorders. He concluded, “the FFM
appears to be useful in conceptualizing and differentiating among the DSM-III-R
personality disorders” (Trull, 1992, p. 557), with some findings replicating
“across all three personality disorder assessment instruments” (p. 557).
Nevertheless, there is considerable opposition to understanding the DSM-IV-
TR personality disorders as maladaptive variants of the FFM (e.g., Gunderson,
2010; Shedler et al., 2010). This opposition is in some respects difficult to
understand, as the FFM would in fact improve considerably the construct
validity of the DSM-IV syndromes. As demonstrated across the chapters of this
text, the empirical support for the FFM is extensive, to say the least. As
acknowledged by the Chair of the DSM-5 Personality and Personality Disorders
Work Group, “similar construct validity has been more elusive to attain with the
current DSM-IV personality disorder categories” (Skodol et al., 2005, p. 1923).
The FFM conceptualization of the DSM-IV-TR personality disorders helps to
address and resolve many of the problems with the DSM-IV-TR syndromes
(Zapolski, Guller, & Smith, 2013). For example, one fundamental problem for
the personality disorders has been their extensive co-occurrence, which in fact
was a primary reason for the proposal by the DSM-5 Personality and Personality
Disorders Work Group to delete half of the diagnoses from the manual (Skodol,
2010, 2012). Lynam and Widiger (2001) and O’Connor (2005) indicated how
the FFM can in fact account for the personality disorder diagnostic co-
occurrence. Lynam and Widiger (2001) had personality disorder researchers
describe prototype cases of each DSM-IV-TR personality disorder in terms of the
30 facets of the FFM. They then demonstrated empirically that the extent to
which these disorders shared FFM traits explained the co-occurrence among
them. The “overlap among FFM profiles reproduced well the covariation
obtained for the schizoid, schizotypal, antisocial, borderline, histrionic,
narcissistic, avoidant, and compulsive personality disorders aggregated across
several sets of studies” (Lynam & Widiger, 2001, p. 410).
Another longstanding problem has been the differential gender prevalence
rates, a source of considerable controversy (Oltmanns & Powers, 2012), which
some have suggested reflects a bias in a respective disorder’s conceptualization,
diagnosis, and/or assessment. The differential gender prevalence rates obtained
for the personality disorders are difficult to justify in the absence of any
theoretical basis for knowing what differential prevalence should be obtained
(Widiger & Spitzer, 1991). In contrast, the FFM has proved useful in helping to
explain and understand gender differences in personality (Feingold, 1994).
Lynam and Widiger (2007) demonstrated empirically that the differential gender
prevalence rates obtained for the personality disorders through a meta-analytic
aggregation of prior studies were consistent with the gender differences that
would be predicted if the personality disorders are understood to be maladaptive
variants of the FFM.
Axis I Disorders
This was probably one of the more difficult chapters to provide. I am not
saying that any one of the others was easy to provide. I am just saying that the
literature on the relationship of FFM domains (and facets) with Axis I disorders
is so incredibly vast and diverse. Even for just one domain, such as neuroticism,
there are meta-analyses upon meta-analyses (e.g., Cassin & von-Ranson, 2005;
Connor-Smith & Flachsbart, 2007; Deneve & Cooper, 1998; Hoyle, Fejfar, &
Miller, 2000; Jorm, 1989; Kotov, Gamez, Schmidt, & Watson, 2010; Malouff,
Thorsteinsson, Rooke, & Schutte, 2007; Malouff, Thorsteinsson, & Schutte,
2005, 2006; Munafo, Zetteler, & Clark, 2007; Ruiz, Pincus, & Schinka, 2008;
Steel, Schmidt, & Schultz, 2008).
Bagby, Uliaszek, Gralnick, and Al-Dajani (this volume) address this literature
with respect to the common forms of possible relationship: causal, pathoplastic,
and spectrum (Andersen & Bienvenu, 2011; Widiger & Smith, 2008). The
greatest interest tends to be with respect to the causal, etiological relationships,
but many such studies are perhaps instead focusing on the pathoplistic and/or the
spectrum. Disentangling these different forms of relationship in any one
particular study can be exceedingly difficult (De Bolle, Beyers, De Clercq, & De
Fruyt, 2012; De Fruyt & De Clercq, 2013, 2014; De Fruyt, Van Leeuwen,
Bagby, Rolland, & Rouillon, 2006).
The influence of neuroticism and psychopathology on the presentation,
appearance, or expression of one another is typically referred to as a pathoplastic
relationship (Bagby et al., this volume; Widiger, 2009). This pathoplastic
relationship can be bidirectional, as a psychopathology can be altered in its
appearance or course due to a person’s premorbid level of neuroticism, and the
appearance of neuroticism can be similarly affected by the presence of a current
or even recent psychopathology. With respect to the former, Duberstein and
Heisel (2007), for instance, reported that neuroticism is associated with a higher
level of overreporting of affective symptoms in persons with clinical depression
(overreporting was suggested by higher levels of self-report relative to a
clinician-based assessment of depression). A well-established finding is that
persons high in neuroticism complain of more medical symptoms (ten Have,
Oldehinkel, Vollebergh, & Ormel, 2005; see also the chapters by Tackett and
Lahey and by Kern and Friedman). Objectively, they may be no more medically
ill than the person low in neuroticism, but they are more likely to report the
presence of symptoms and to seek treatment for them (Chapman, Duberstein,
Sorensen, Lyness, & Emery, 2006). The same phenomenon will occur for the
experience and reporting of symptoms of psychopathology.
Equally important, of course, is the pathoplastic effect of psychopathology on
the self-report or perception of neuroticism (Farmer, 2000; Widiger & Samuel,
2005). Researchers will at times assess the trait of neuroticism while patients are
clinically depressed, but persons who are very depressed will routinely fail to
provide accurate descriptions of their personality traits (Widiger, 2009).
Distortion in self-image is one of the direct effects of a mood disorder (APA,
2013), and it should not be surprising to find that persons who are depressed
provide an inaccurate description of their level of FFM depressiveness, self-
consciousness, or vulnerability that was present prior to or independent of their
current depressed mood. Once the mood disorder is successfully treated, the
level of self-described neuroticism decreases, not because of a change in
personality but simply because of the remission of the mood disorder.
Jorm (1989) summarized the results of 63 therapy outcome studies that
included measures of trait anxiety or neuroticism. The results indicated a
significant reduction in neuroticism over the course of treatment (particularly by
rational-emotive therapies). It is difficult not to be concerned that this change in
self-reported levels of neuroticism was artifactual, resulting simply from changes
in levels of Axis I psychopathology for which the patients were actually seeking
and receiving treatment, rather than reflecting real changes in premorbid
personality functioning. Piedmont (2001) similarly reported changes in FFM
self-report personality assessments for 132 persons in a 6-week outpatient drug
rehabilitation program. Significant changes in levels of neuroticism,
agreeableness, and conscientiousness were maintained on follow-up
approximately 15 months after the termination of treatment. He concluded that
“personality change may be possible in the context of treatment” (p. 500).
However, it is again worth noting that the change in neuroticism scores was also
associated with changes in Axis I psychopathology, suggesting once more that
the original assessment of neuroticism might have been an artifact of the
psychopathology.
Costa, Bagby, Herbst, and McCrae (2005), however, offered a different view
of these changes to mood and levels of neuroticism. To the extent that
neuroticism is a disposition to experience and express negative affects, increases
(and decreases) in the expression of these moods could be understood as
fluctuating expressions of (and changes to) the personality trait. Costa et al.
(2005) argued, “rather than regard these depression-caused changes in assessed
personality trait levels as a distortion, we interpret them as accurate reflections
of the current condition of the individual” (p. 45). They suggested that the
elevated neuroticism scores at the beginning of treatment should not be
understood simply as an artifact of a depressed mood state but can and should be
understood instead as a true fluctuation in the level of neuroticism secondary to
environmental events (e.g., stress). Our levels of neuroticism will not simply
remain flat and stable no matter what is happening within our lives. Fluctuations
in levels of agreeableness and extraversion, and other domains of the FFM, will
also occur in response to situational changes. Costa et al. (2005) indicated that
support for the validity of the NEO PI-R neuroticism scores was their sustained
correlation with variables unrelated to depression and their incremental validity
in the prediction of personality-relevant criteria above and beyond the effects of
severity of depression. They also indicated that there were no significant changes
in the shape of the NEO PI-R profiles over the course of treatment (although
there was substantially more change in the depression and vulnerability facets of
neuroticism than for any other NEO PI-R facet scores). In sum, “psychometric
analyses demonstrate that the baseline NEO PI-R provides a reliable and valid
assessment of personality at the time it was administered” (Costa et al., 2005, p.
52).
If Costa et al. (2005) are correct, then perhaps Widiger and Samuel (2005)
were being too dismissive of personality change scores resulting from brief,
pharmacotherapies of a mood disorder. Personality can indeed change in
response to pharmacotherapy (Widiger, 2009). Knutson et al. (1998) “examined
the effects of a serotonergic reuptake blockade on personality and social
behavior in a double-blind protocol by randomly assigning 51 medically and
psychiatrically healthy volunteers to treatment with a selective serotonin
reuptake inhibitor (SSRI), paroxetine … (N = 25), or placebo (N = 26)” (p. 374).
None of the participants currently met, or apparently met throughout their
lifetime, diagnostic criteria for any mental disorder, as assessed with a
semistructured interview. None of the participants had ever received a
psychotropic medication, had ever abused drugs, or had ever been in any form of
psychiatric treatment. In other words, they were in many respects above normal
in psychological functioning. Certainly any subsequent changes in their
personality traits could not be attributed to the effect of treating a co-occurring
mood disorder.
The paroxetine (and placebo) treatment continued for 4 weeks. Knutson et al.
(1998) reported that the SSRI administration (relative to placebo) significantly
reduced scores on a self-report measure of neuroticism. The magnitude of
change even correlated well with plasma levels of SSRI within the SSRI
treatment group. As concluded by Knutson et al. (1998), this was a clear
“empirical demonstration that chronic administration of a selective serotonin
reuptake blockade can have significant personality and behavioral effects in
normal humans in the absence of baseline depression or other psychopathology”
(p. 378). In sum, normal personality can be altered through pharmacotherapy.
It is generally accepted that medical conditions, such as Alzheimer’s disease,
Parkinson’s disease, and traumatic brain injury, can produce actual changes in
personality functioning (Costa et al., 2005). The APA (2013) diagnostic manual
recognizes the concept of personality change due to a general medical condition.
Why not then change due to a mental disorder? The WHO (1992) indeed
recognizes within ICD-10 personality change secondary to severe mental illness,
which would include, for instance, changes in levels of neuroticism secondary to
a mood disorder.
However, to the extent that self-report descriptions of neuroticism secondary
to a mood disorder are considered to represent fluctuations in personality
functioning or actual changes in personality functioning, it becomes difficult to
conduct research on the etiological contribution of neuroticism to the respective
mood disorder. They are no longer distinct, independent constructs. They would
then represent the same underlying disposition, with the mood disorder being a
momentary, transient elevation of trait neuroticism.
There is indeed a considerable body of research to indicate that neuroticism
may provide the foundation for a wide variety of Axis I psychopathologies.
There is a great deal of comorbidity among Axis I disorders that may be due to
the presence of a common underlying disposition. “Comorbidity may be trying
to show us that many current treatments are not so much treatments for transient
‘state’ mental disorders of affect and anxiety as they are treatments for core
processes, such as negative affectivity, that span normal and abnormal variation
as well as undergird multiple mental disorders” (Krueger, 2002, p. 44).
Krueger and his colleagues have replicated well across a variety of
populations the two dimensions of internalization and externalization that cut
across a number of different Axis I disorders, and the broad domain of
internalization maps well onto the personality temperament of neuroticism
(Clark, 2005; Krueger & Markon, 2006; Watson, Gamez, & Simms, 2005).
Kendler, Prescott, Myers, and Neale (2003) applied multivariate genetic analyses
to 10 mental disorders assessed in more than 5,600 members of male–male and
female–female twin pairs from a population-based registry. They concluded that
“the patterns of comorbidity of these disorders (internalizing vs. externalizing,
and within internalizing, anxious misery vs. fear) [are] driven largely by
[common] genetic factors” (p. 936). Kahn et al. (2005) reported large effect sizes
for the association of neuroticism with depression, generalized anxiety disorder,
and panic disorder in a sample of 7,588 twin pairs. Neuroticism explained 20%
to 45% of the comorbidity among the depression and anxiety disorders. Hettema,
Neale, Myers, Prescott, and Kendler (2006) reported similarly that one-third to
two-thirds of the genetic variance in mood (depressive) and anxiety disorders
was shared with neuroticism.
There are some Axis I mental disorders that are even difficult to distinguish
from the personality trait or temperament of neuroticism, such as social anxiety
disorder and persistent depressive disorder (APA, 2000, 2013). DSM-5 social
anxiety disorder was DSM-IV generalized social phobia, an Axis I mental
disorder that was diagnosed when the social phobia included most every social
situation. Generalized social phobia was said to have an early onset and chronic
course, “emerging out of a childhood history of social inhibition or shyness”
(APA, 2000, p. 453). It would seem difficult to distinguish this Axis I mental
disorder from elevations on the anxiousness, self-consciousness, and
vulnerability facets of NEO PI-R Neuroticism, along with facets of introversion
(Widiger, 2009). Likewise, persistent depressive disorder was previously early
onset dysthymia whose initial literature review relied heavily on the research
concerning charactological depression (i.e., Keller, 1989) and is in turn
essentially equivalent to FFM depressiveness. In sum, neuroticism is not
currently conceptualized as a mental disorder, but it may not be long before a
temperament of neuroticism is in fact explicitly identified as a mental disorder,
consistent with RDoC negative valence (Widiger, 2012).
Childhood and Adolescent Dysfunction
Tyrer (2015), a leading international personality disorder researcher, and
Chair of the WHO ICD-11 Personality Disorders Work Group, called for an
examination of “a new hypothesis explaining the relationship between
personality and mental state disorders” (p. 1). He suggested that this would
require overcoming the reluctance to assess for personality dysfunction in
childhood and adolescence. “There is a general reluctance among all those who
treat people with developmental disorders to think in terms of personality
dysfunction when describing people whom others in adult psychiatry would have
no difficulty in identifying as having personality problems” (Tyrer, 2015, p. 5).
Tyrer (2015) ended his paper by stating, “the time has now come for all those
involved with improving understanding and awareness of personality disorders
to extend their scope across the boundary between adolescence and adult life”
(p. 5).
Tyrer (2015) would probably find the chapter by De Fruyt, De Clercq, and De
Bolle (this volume) to be very encouraging, if not satisfying. All that Tyrer was
calling for has in fact been realized within the work and research of De Fruyt, De
Clercq, and others with respect to the study and application of maladaptive
variants of FFM personality traits in childhood and adolescence (Widiger, 2015).
Whereas there is considerable reluctance to identify DSM-IV-TR and/or ICD-10
personality disorders in childhood, contributing to the limited understanding of
the childhood antecedents for their development (Widiger, De Clercq, & De
Fruyt, 2009), there is a substantial body of research on personality and
personality dysfunction in childhood and adolescence (De Fruyt & De Clercq,
2014). De Clercq, De Fruyt, Van Leeuwen, and Mervielde (2006) developed a
measure of maladaptive personality functioning, the Dimensional Personality
Symptom Item Pool (DIPSI), that is aligned explicitly with the FFM, as assessed
by the Hierarchical Personality Inventory for Children (HiPIC; Mervielde & De
Fruyt, 1999, 2002). As indicated by De Fruyt and De Clercq (2014), “an
integrative model of personality pathology precipitants for childhood and
adolescence is available now” (p. 469).
One of the reasons there is tremendous reluctance to identify personality
dysfunction in children and adolescents is the stigma that can be associated with
a personality disorder diagnosis. Personality disorders are among the more
stigmatizing labels within the diagnostic manual. They are relatively unique in
concerning “ego-syntonic” aspects of the self, or one’s characteristic manner of
thinking, feeling, behaving and relating to others throughout one’s life. Whereas
an Axis I disorder is something that momentarily happens to the person, a
personality disorder is who that person is and perhaps always will be (Millon,
2011). A personality disorder diagnosis therefore suggests that who you are is
itself a disorder. The FFM of personality disorder, in contrast, provides a more
complete description of each person’s self that recognizes and appreciates that
the person is more than just a diagnosis and that other aspects of the self can be
adaptive, even commendable, despite the presence of some maladaptive
personality traits (Widiger, Samuel, Mullins-Sweat, Gore, & Crego, 2012). It
would be the very rare person indeed who lacked any exemplary, admirable
traits. In addition, from the perspective of the FFM, a personality disorder is no
longer something that is qualitatively distinct from normal personality
functioning. Instead, a personality disorder simply represents the presence of
maladaptive variants of the same personality traits that are evident within all
persons.
The WHO (2014) recently launched a call for a stronger focus on adolescent
health, consistent with the plea by Tyrer (2015). As indicated by De Fruyt, De
Clercq, and De Bolle (this volume), the top three causes of adolescent deaths
across 109 countries were road traffic injuries, human immunodeficiency virus,
acquired immune deficiency syndrome, and suicide. The predominant cause of
illness and disability among adolescents aged 10 to 19 years was depression
(WHO, 2014). Personality traits in childhood and adolescence impact these
conditions. The traits of impulsiveness, self-control, and sensation-seeking
contribute to unsafe sex and reckless drunk driving. Neuroticism is a strong
vulnerability factor for depression and suicide (see also the chapter by Tackett
and Lahey). As suggested by De Fruyt, De Clercq, and De Bolle (this volume)
“given that traits can be reliably assessed in childhood and adolescence, and that
half of all people developing a kind of mental disorder show an onset of
symptoms in late childhood, it is clear that professionals and policy makers need
to pay more attention to the developing personalities of individuals.”
In sum, Tyrer (2015) appropriately bemoaned the fact that the importance of
personality dysfunction to mental and physical health in childhood or
adolescence has not been sufficiently well recognized or appreciated. He
suggested that the possibility of such a relationship in childhood is “a testable
hypothesis” (Tyrer, 2015, p. 1) that would first require though the field of
personality disorder to overcome its reluctance to assess for the presence of
personality disorder in childhood and adolescence. However, no such reluctance
is evident in the work of De Fruyt and De Clercq (2013) or their colleagues (e.g.,
De Bolle, Beyers, De Clercq, & De Fruyt, 2012; De Pauw & Mervielde, 2010;
Shiner & Tackett, 2013; Tackett, Kushner, Mervielde, & De Fruyt, 2013). To the
extent that personality disorder and dysfunction are understood as maladaptive
variants of the FFM, the relationship of personality disorder to consequential life
outcomes in childhood and adolescence is not just a testable hypothesis, it is a
well-established empirical finding and a vibrant line of investigation.
Clinical Utility
Quite some time ago, Costa and McCrae (1992a) suggested that the “five
factor model is a theoretical advance that has important implications for many
applied areas, including clinical practice” (p. 5). They argued that “most clinical
populations are not dramatically different from normal volunteer samples with
regard to the structure of personality” (p. 9). They offered a number of different
ways in which the FFM can be useful to a clinician, including (but not limited
to) the provision of a reasonably complete profile of the patient’s personality,
which would include both strengths and weaknesses with respect to the patient’s
characteristic manner of thinking, feeling, and relating to others. They indicated
how this profile could help the clinician to better understand a patient’s problems
in a manner that would be readily communicable to the patient, as well as
anticipating the course of therapy.
The suggestions of Costa and McCrae (1992a) were accompanied by a critical
review by Ben-Porath and Waller (1992b), followed by a rejoinder by Costa and
McCrae (1992b), and then followed by a further response by Ben-Porath and
Waller (1992a). This was a debate that was also presaged by commentaries
published 5 years earlier in the American Psychologist (i.e., McCrae & Costa,
1986; Waller & Ben-Porath, 1987). The debate was at times in reference to the
FFM, but much of it actually focused on the relative strengths or weaknesses of
the NEO PI-R (Costa & McCrae, 1992c) as compared to the MMPI (Hathaway
& McKinley, 1943). It is difficult not to perceive this debate as reflecting some
degree of a competitive rivalry between two leading measures of individual
differences, with both sides taking perhaps rather strong positions. Both
measures and models would appear to have their unique strengths and could
perhaps in fact nicely complement one another.
Many of the recommendations by Costa and McCrae (1992a) were
subsequently mirrored years later by Widiger and Mullins-Sweatt (2009) and
Krueger and Eaton (2010) in their extolling of the potential value for a clinician
of a dimensional trait model. Widiger and Mullins-Sweatt (2009) discussed the
benefits of a personality descriptive system that recognizes the strengths of the
patient, along with the maladaptive traits that were being diagnosed as a
personality disorder. “Some of these strengths may also be quite relevant to
treatment, such as openness to experience indicating an interest in exploratory
psychotherapy, agreeableness indicating an engagement in group therapy, and
conscientiousness indicating a willingness and ability to adhere to the demands
and rigor of dialectical behavior therapy (Sanderson & Clarkin, 2002)” (Widiger
& Mullins-Sweatt, 2009, p. 203). Krueger and Eaton (2010), similarly, noted the
virtues of having a truly integrative model of normal and abnormal personality.
They described a person with borderline personality disorder whose high
openness and extraversion had important treatment implications. “The high
openness might also suggest that this person would be open to a therapeutic
approach where depth and underling motives for behavior are explored”
(Krueger & Eaton, 2010, p. 102).
As indicated by Mullins-Sweatt, Samuel, and Helle (this volume), concerns
with respect to clinical utility have been a major objection with respect to a shift
to a dimensional trait model of classification (e.g., Clarkin & Huprich, 2011;
First, 2005; Shedler et al., 2010). This concern is somewhat ironic given that the
heterogeneity of diagnostic membership, the lack of precision in description, the
excessive diagnostic co-occurrence, the reliance on the “not otherwise specified”
wastebasket diagnosis, and the unstable and arbitrary diagnostic boundaries of
the DSM-IV-TR (now DSM-5) diagnostic categories are sources of considerable
frustration for clinicians. Verheul (2005) systematically reviewed various
components of clinical utility for both categorical and dimensional models and
concluded, “overall, the categorical system has the least evidence for clinical
utility, especially with respect to coverage, reliability, subtlety, and clinical
decision-making” (p. 295). The FFM provision of more individualized and
precise personality profiles, the increased homogeneity of trait constructs, and
the inclusion of normal personality traits would considerably improve the
clinical utility of personality disorder assessments (Widiger & Mullins-Sweatt,
2009).
This opinion is also supported by empirical research, well documented by
Mullins-Sweatt and colleagues. The primary components of clinical utility are
ease of usage, the facilitation of communication, and treatment planning
(Mullins-Sweatt, Lengel, & DeShong, in press), and there have been multiple
studies surveying the experiences, perspectives, and opinions of clinicians that
have directly compared the FFM and the DSM-IV-TR diagnostic categories with
respect to these components. Clinicians have agreed with Costa and McCrae
(1992a) that the FFM has a great deal of clinical utility with respect to ease of
usage, communication, and treatment planning (e.g., Glover, Crego, & Widiger,
2012; Lowe & Widiger, 2009; Miller, 1991; Mullins-Sweatt & Lengel, 2012;
Mullins-Sweatt & Widiger, 2011; Presnall, 2013; Samuel & Widiger, 2006;
Stone, 2013).
Conclusions
The FFM is the predominant dimensional model of general personality
structure, the reasons for which should be apparent from this text. The FFM is
not without criticism, dispute, and concerns. These should also be apparent from
this text. Nevertheless, the FFM is a powerful and robust theoretical model of
personality structure that has considerable construct validity and social, clinical
utility. Persons applying the FFM to their research program or to their social,
clinical practice typically discover rather quickly its power, richness, and
potency. Still, there remains room for growth, development, revision, and
expansion. There are different versions of the FFM, different interpretations, and
different measures. As more and more investigators become involved in FFM
research, it will continue to advance and evolve. Hopefully this text will serve as
a useful stimulant for such future research. I hope you enjoyed it!
References
Achenbach, T. M. (1966). The classification of children’s psychiatric symptoms: A factor
analytic study. Psychological Monographs, 80 (No. 615).
Ahmetoglu, G., & Swami, V. (2012). Do women prefer “nice guys”? The effect of male
dominance behavior on women’s ratings of sexual attractiveness. Social Behavior and
Personality: An International Journal, 40, 667–672.
Psychological Monographs, 47 (1, Whole No. 211).
disorders (3rd ed.). Washington, DC: Author.
disorders (3rd ed., rev. ed.). Washington, DC: Author.
disorders (4th ed., text rev.). Washington, DC: Author.
American Psychiatric Association. (2001). Practice guidelines for the treatment of
patients with borderline personality disorder. Washington, DC, American Psychiatric
Association.
Ashton, M. C., & Lee, K. (2010). On the cross-language replicability of personality
factors. Journal of Research in Personality, 44(4), 436–441.
Babiak, P., & Hare, R. D. (2006). Snakes in suits: When psychopaths go to work. New
York, NY: Regan Books.
Bagby, R. M., Gilchrist, E. J., Rector, N. A., Dickens, S. E., Joffe, R. T., Levitt, A., …
Kennedy, S. H. (2001). The stability and validity of the sociotropy and autonomy
personality dimensions as measured by the Revised Personal Style Inventory. Cognitive
Therapy and Research, 25, 765–779.
Bagby, R. M., Watson, C., & Ryder, A. G. (2013). Depressive personality disorder and
the five factor model. In T. A. Widiger & P. T. Costa (Eds.), Personality disorders and
the five factor model of personality (3rd. ed., pp. 179–192). Washington, DC:
Barlow, D. H., Ellard, K. K., Fairholme, C., Farchione, T. J., Boisseau, C., Allen, L., &
Ehrenreich-May, J. (2011). Unified protocol for the transdiagnostic treatment of
emotional disorders. New York, NY: Oxford University Press.
Barlow, D. H., Sauer-Zavala, S., Carl, J. R., Bullis, J. R., & Ellard, K. K. (2014). The
nature, diagnosis, and treatment of neuroticism back to the future. Clinical
Benet-Martinez, V., & John, O. P. (1998). Los Cinco Grandes across cultures and ethnic
groups: Multitrait multimethod analyses of the Big Five in Spanish and English.
Benjamin, L. S. (1996). Interpersonal diagnosis and treatment of personality disorders
(2nd ed.). New York, NY: Guilford Press.
Ben-Porath, Y. S., & Waller, N. G. (1992a). Five big issues in clinical personality
assessment: A rejoinder to Costa and McCrae. Psychological Assessment, 4, 23–25.
Ben-Porath, Y. S., & Waller, N. G. (1992b). “Normal” personality inventories in clinical
assessment: General requirements and the potential for using the NEO Personality
Inventory. Psychological Assessment, 4(1), 14–19.
Berghuis, H., Kamphuis, J. H., & Verheul, R. (2012). Core features of personality
disorder: Differentiating general personality dysfunctioning from personality traits.
Bernreuter, R. G. (1931). The Personality Inventory. Palo Alto, CA: Consulting
DSM-IV personality disorders. The Journal of Nervous and Mental Disease, 185, 388–
394.
Blashfield, R. K., Reynolds, S. M., & Stennett, B. (2012). The death of histrionic
personality disorder. In T. A. Widiger (Ed.), The Oxford handbook of personality
Bornstein, R. F. (2011). Reconceptualizing personality pathology in DSM-5: Limitations
in evidence for eliminating dependent personality disorder and other DSM-IV
syndromes. Journal of Personality Disorders, 25, 235–247.
Bornstein, R. F. (2012). Illuminating a neglected clinical issue: Societal costs of
interpersonal dependency and dependent personality disorder. Journal of Clinical
Butcher, J. N. (2010). Personality assessment from the nineteenth to the early twenty-first
century: Past achievements and contemporary challenges. Annual Review of Clinical
Cassin, S. E., & von-Ranson, K. M. (2005). Personality and eating disorders: A decade in
review. Clinical Psychology Review, 25, 895–916.
Cattell, R. B., & Tsujioka, B. (1964). The importance of factor-trueness and validity,
versus homogeneity and orthogonality, in test scales. Educational and Psychological
Measurement, 1, 3–30.
Chapman, B. P., Duberstein, P. R., Sorensen, S., Lyness, J. M., & Emery, L. (2006).
Personality and perceived health in older adults: The five factor model in primary care.
Journals of Gerontology: Series B: Psychological Sciences and Social Sciences, 61,
P362–P365.
Chapman, B. P., Hampson, S., & Clarkin, J. (2014). Personality-informed interventions
for healthy aging: Conclusions from a National Institute on Aging work group.
Cheung, F. M., van de Vijver, F. J., & Leong, F. T. (2011). Toward a new approach to the
study of personality in culture. American Psychologist, 66, 593–603.
Church, A. T. (2016). Personality traits across cultures. Current Opinion in Psychology,
8, 22–30.
Church, A. T., & Katigbak, M. S. (1988). The emic strategy in the identification and
assessment of personality dimensions in a non-western culture rationale, steps, and a
Philippine illustration. Journal of Cross-Cultural Psychology, 19, 140–163.
Clark, L. A. (2005). Temperament as a unifying basis for personality and
psychopathology. Journal of Abnormal Psychology, 114, 505–521.
Clark, L. A., Simms, L. J., Wu, K. D., & Casillas, A. (in press). Manual for the Schedule
for Nonadaptive and Adaptive Personality (SNAP-2). Minneapolis, MN: University of
Minnesota Press.
Clark, L. A., & Watson, D. (1995). Constructing validity: Basic issues in objective scale
development. Psychological Assessment, 7, 309–319.
Clark, L. A., & Watson, D. (2008). Temperament: An organizing paradigm for trait
psychology. In O. P. John, R. W. Robins, & L. A. Pervin (Eds.), Handbook of
personality. Theory and research (3rd ed., pp. 265–286). New York, NY: Guilford
Press.
Cloninger, C. R. (2000). A practical way to diagnose personality disorder: A proposal.
Press.
Cohen, P., & Crawford, T. N. (2005). Developmental issues. In D. S Bender, J. M.
Oldham, & A. E. Skodol (Eds.), The American Psychiatric Publishing textbook of
personality disorders (pp. 171–185). Washington, DC: American Psychiatric
Publishing.
nomological net. Journal of Personality Assessment, 96, 1–16.
Connor-Smith, J. K., & Flachsbart, C. (2007). Relations between personality and coping.
Costa, P. T., Bagby, R. M., Herbst, J. F., & McCrae, R. R. (2005). Personality self-reports
are concurrently reliable and valid during acute depressive episodes. Journal of
Costa, P. T., Busch, C. M., Zonderman, A. B., & McCrae, R. R. (1986). Correlations of
MMPI factor scales with measures of the Five Factor Model of personality. Journal of
Costa, P. T., & McCrae, R. R. (1980). Still stable after all these years: Personality as a
key to some issues in adulthood and old age. In P. B. Baltes & O. G. Brim (Eds.), Life
span development and behavior (Vol. 3). New York, NY: Academic Press.
Costa, P. T., & McCrae, R. R. (1988). From catalogue to classification: Murray’s needs
and the Five Factor Model. Journal of Personality and Social Psychology, 55, 258–
265.
Costa, P. T., & McCrae, R. R. (1992a). Normal personality assessment in clinical
practice: The NEO Personality Inventory. Psychological Assessment, 4, 5–13.
Costa, P. T., & McCrae, R. R. (1992b). “‘Normal’ personality inventories in clinical
assessment: General requirements and the potential for using the NEO Personality
Inventory”: Reply. Psychological Assessment, 4, 20–22.
Costa, P. T., & McCrae, R. R. (1992c). Revised NEO Personality Inventory (NEO PI-R)
and NEO Five Factor Inventory (NEO-FFI) professional manual. Odessa, FL:
Costa, P. T., & McCrae, R. R. (1995). Domains and facets: Hierarchical personality
Costa, P. T., & Piedmont, R. L. (2003). Multivariate assessment: NEO PI-R profiles of
Madeline G. In J. S. Wiggins (Ed.), Paradigms of personality assessment (pp. 262–
Costa, P. T., Zonderman, A. B., McCrae, R. R., & Williams, R. B. (1985). Content and
comprehensiveness in the MMPI: An item factor analysis in a normal adult sample.
Crego, C., & Widiger, T. A. (in press). Convergent and discriminant validity of
alternative measures of maladaptive personality traits. Psychological Assessment.
Cronbach, L. J., & Meehl, P. E. (1955). Construct validity in psychological tests.
Cuthbert, B. N., & Insel, T. R. (2013). Toward the future of psychiatric diagnosis: The
seven pillars of RDoC. BMC Medicine, 11, 1–8.
psychopathology in referred and nonreferred children and adolescents: An investigation
121, 958–970.
perspective for DSM-V. Journal of Abnormal Psychology, 115, 639–657.
De Fruyt, F., & De Clercq, B. (2013). Childhood antecedents of personality disorder: A
five factor perspective. In T. A. Widiger & P. T. Costa (Eds.), Personality disorders
and the five factor model of personality (3rd. ed., pp. 43–60). Washington, DC:
De Fruyt, F., De Clercq, B. J., Miller, J., Rolland, J. P., Jung, S. C., Taris, R., … Van
Hiel, A. (2009). Assessing personality at risk in personnel selection and development.
De Fruyt, F., Van Leeuwen, K., Bagby, R. M., Rolland, J. P., & Rouillon, F. (2006).
Assessing and interpreting personality change and continuity in patients treated for
major depression. Psychological Assessment, 18(1), 71–80.
De Pauw, S. S., & Mervielde, I. (2010). Temperament, personality & developmental
traits. Child Psychiatry & Human Development, 41, 313–329.
De Raad, B., Barelds, D. P. H., Levert, E., Ostendorf, F., Mlacic, B., Blais, L. D., …
Hogrefe & Huber.
DeGeest, D. S., & Schmidt, F. (2015). A rigorous test of the fit of the circumplex model
to Big Five personality data: Theoretical and methodological issues and two large
sample empirical tests. Multivariate Behavioral Research, 50, 350–364.
DeNeve, K., & Cooper, H. (1998). The happy personality: A meta-analysis of 137
personality traits and subjective well-being. Psychological Bulletin, 124, 197–229.
DeShong, H. L., Lengel, G. J., Sauer-Zavala, S. E., O’Meara, M., & Mullins-Sweatt, S.
N. (2015). Construct validity of the Five Factor Borderline Inventory. Assessment,
22(3), 319–331.
DeYoung, C. G. (2010). Personality neuroscience and the biology of traits. Social and
DeYoung, C. G. (2011). Intelligence and personality. In R. J. Sternberg & S. B. Kaufman
(Eds.), The Cambridge handbook of intelligence (pp. 711–737). New York, NY:
DeYoung, C. G. (2015). Cybernetic Big Five theory. Journal of Research in Personality,
36, 33–58.
Cambridge handbook of personality (pp. 323–346). New York, NY: Cambridge
University Press.
aspects of the big five. Journal of Personality and Social Psychology, 93(5), 880–896.
Distel, M. A., Trull, T. J., Willemsen, G., Vink, J. M., Derom, C. A, Lynskey, M. M., …
Boomsma, D. I. (2009). The five factor model of personality and borderline personality
disorder: A genetic analysis of comorbidity. Biological Psychiatry, 66, 1131–1138.
Doerrenberg, P., Duncan, D., Fuest, C., & Peichl, A. (2014). Nice guys finish last: Do
honest taxpayers face higher tax rates? Kyklos, 67(1), 29–53.
Duberstein, P. R., & Heisel, M. J. (2007). Personality traits and the reporting of affective
disorder symptoms in depressed patients. Journal of Affective Disorders, 103, 165–171.
Dudley, N. M., Orvis, K. A., Lebiecki, J. E., & Cortina, J. M. (2006). A meta-analytic
investigation of conscientiousness in the prediction of job performance: Examining the
intercorrelations and the incremental validity of narrow traits. Journal of Applied
Edmundson, M., Lynam, D. R., Miller, J. D., Gore, W. L., & Widiger, T. A. (2011). A
five factor measure of schizotypal personality traits. Assessment, 18, 321–334.
Eysenck, H. J. (1952). The scientific study of personality. London, England: Routledge &
K. Paul.
Eysenck, H. J., & Eysenck, S. B. (1964). Eysenck Personality Inventory. San Diego, CA:
Educational and Industrial Testing Service.
Eysenck, S. B., & Eysenck, H. J. (1975). Manual of the Eysenck Personality
Questionnaire. London, England: Houer & Stoughton.
Eysenck, H. J., & Wilson, G. D. (1991). The Eysenck Personality Profiler (version 6).
Worthing, England: Psi-Press.
Farmer, R. F. (2000). Issues in the assessment and conceptualization of personality
disorders. Clinical Psychology Review, 20, 823–851.
Bulletin, 116, 429–456.
First, M. B. (2005). Clinical utility: A prerequisite for the adoption of a dimensional
approach in DSM. Journal of Abnormal Psychology, 114, 560–564.
Flynn, F. J. (2005). Having an open mind: The impact of openness to experience on
interracial attitudes and impression formation. Journal of Personality and Social
Frances, A. J., First, M. B., & Pincus, H. A. (1995). DSM-IV guidebook. Washington,
DC: American Psychiatric Press.
Friedman, H. S., Tucker, J., Tomlinson-Keasey, C., Schwartz, J., Wingard, D., & Criqui,
M. H. (1993). Does childhood personality predict longevity? Journal of Personality
Furnham, A. (2014). A bright side, facet analysis of histrionic personality disorder: The
relationship between the HDS colourful factor and the NEO-PI-R facets in a large adult
sample. The Journal of Social Psychology, 154, 527–536.
Furnham, A., & Crump, J. (2005). Personality traits, types, and disorders: An examination
of the relationship between three self-report measures. European Journal of
Furnham, A., Milner, R., Akhtar, R., & De Fruyt, F. (2014). A review of the measures
designed to assess DSM-5 personality disorders. Psychology, 5, 1646–1686.
Furnham, A., Trickey, G., & Hyde, G. (2012). Bright aspects to dark side traits: Dark side
traits associated with work success. Personality and Individual Differences, 52, 908–
913.
Galton, F. (1884). Measurement of character. Fortnightly Review, 36, 179–185.
Glover, N., Crego, C., & Widiger, T. A. (2012). The clinical utility of the five factor
model of personality disorder. Personality Disorders: Theory, Research, and
Glover, N., Miller, J. D., Lynam, D. R., Crego, C., & Widiger, T. A. (2012). The Five
Factor Narcissism Inventory: A five factor measure of narcissistic personality traits.
measuring the lower-level facets of several five factor models. In I. Mervielde, I.
Gore, W. L., & Pincus, A. L. (2013). Dependency and the five factor model. In T. A.
Widiger & P. T. Costa (Eds.), Personality disorders and the five factor model (3rd ed.,
Gore, W. L., Presnall, J., Lynam, D. R., Miller, J. D., & Widiger, T. A. (2012). A five
factor measure of dependent personality traits. Journal of Personality Assessment, 94,
488–499.
Gore, W. L., Tomiatti, M., & Widiger, T. A. (2011). The home for histrionism.
Gore, W. L., & Widiger, T. A. (2013). The DSM-5 dimensional trait model and five
Gosling, S. D. (2008). Personality in non-human animals. Social and Personality
Psychology Compass, 2, 985–1001.
Gunderson, J. G. (2010). Commentary on “Personality traits and the classification of
mental disorders: Toward a more complete integration in DSM-5 and an empirical
model of psychopathology.” Personality Disorders: Theory, Research, and Treatment,
1, 119–122.
universal is the Big Five? Testing the five factor model of personality variation among
forager–farmers in the Bolivian Amazon. Journal of Personality and Social
Han, K., Weed, N. C., & McNeal, T. P. (1996). Searching for conscientiousness on the
MMPI-2. Journal of Personality Assessment, 67, 354–363.
Harkness, A. R., & McNulty, J. L. (1994). The Personality Psychopathology Five (PSY–
York, NY: Springer.
disorders. American Journal of Psychiatry, 163, 857–864.
Hofstee, W. K., De Raad, B., & Goldberg, L. R. (1992). Integration of the big five and
63, 146–163.
Hogan, R. (1982). A socioanalytic theory of personality, vol. 30 of Nebraska Symposium
on Motivation. Lincoln, NE: University of Nebraska Press.
Hogan, R. (1986). Hogan Personality Inventory manual. Minneapolis, MN: National
Computer Systems.
Hogan, R., & Hogan, J. (1992). Hogan Personality Inventory manual. Tulsa, OK: Hogan
Assessment Systems.
Hogan, R., & Hogan, J. (1997). Hogan Developmental Survey manual. Tulsa, OK: Hogan
Assessment Systems.
Hogan, R., & Hogan, J. (2002). The Hogan Personality Inventory. In B. de Raad & M.
Perugini (Eds.), Big five assessment (pp. 329–351). Bern, Switzerland: Hogrefe &
Huber.
Hopko, D. R., Lejuez, C. W., Ruggiero, K. J., & Eifert, G. H. (2003). Contemporary
behavioral activation treatments for depression: Procedures, principles, and progress.
Clinical Psychology Review, 23(5), 699–717.
Hopwood, C. J., & Donnellan, M. B. (2010). How should the internal structure of
personality inventories be evaluated? Personality and Social Psychology Review, 14,
332–346.
Hopwood, C. J., & Thomas, K. M. (2012). Paranoid and schizoid personality disorders. In
T. A. Widiger (Ed.), The Oxford handbook of personality disorders (pp. 582–602).
New York, NY: Oxford University Press.
Horowitz, L. M. (2004). Interpersonal foundations of psychopathology. Washington, DC:
Horowitz, L. M., Alden, L. E., Wiggins, J. S., & Pincus, A. L. (2000). IIP-64/IIP-32
professional manual. San Antonio, TX: The Psychological Corporation.
Hoyle, R. H., Fejfar, M. C., & Miller, J. D. (2000). Personality and sexual risk-taking: A
quantitative review. Journal of Personality, 68, 1203–1231.
Insel, T. R. (2013). Director’s blog: Transforming diagnosis. Retrieved from
http://www.nimh.nih.gov/about/director/2013/transforming-diagnosis.shtml
Insel, T. R., Cuthbert, B., Garvey, M., Heinssen, R., Pine, D. S., Quinn, K., … Wang, P.
(2010). Research domain criteria (RDoC): Toward a new classification framework for
research on mental disorders. American Journal of Psychiatry, 167, 748–751.
Jorm, A. F. (1989). Modifiability of trait anxiety and neuroticism: A meta-analysis of the
literature. Australian and New Zealand Journal of Psychiatry, 23, 21–29.
Judge, T. A., Livingston, B. A., & Hurst, C. (2012). Do nice guys—and gals—really
finish last? The joint effects of sex and agreeableness on income. Journal of
Jung, C. G., & Baynes, H. G. (1921). Psychological types. London, England: Kegan Paul
Trench Trubner.
Kahn, A. A., Jacobson, K. C., Gardner, C. O., Prescott, C. A., & Kendler, K. S. (2005).
Personality and comorbidity of common psychiatric disorders. British Journal of
Kalus, O., Bernstein, D. P., & Siever, L. J. (1993). Schizoid personality disorder: A
review of current status and implications for DSM-IV. Journal of Personality
Disorders, 7, 43–52.
Keller, M. B. (1989). Current concepts in affective disorders. Journal of Clinical
Kendler, K. S., Prescott, C. A., Myers, J., & Neale, M. C. (2003). The structure of genetic
and environmental risk factors for common psychiatric and substance use disorders in
men and women. Archives of General Psychiatry, 60, 929–937.
Kessler, R. C. (1999). The World Health Organization International Consortium in
Psychiatric Epidemiology: Initial work and future directions—the NAPE lecture. Acta
Psychiatrica Scandinavica, 99, 2–9.
Klein, R. G., Tancer, N. K., & Werry, J. S. (1997). Anxiety disorders of childhood or
adolescence. In T. A. Widiger, A. J. Frances, H. A. Pincus, R. Ross, M. B. First, & W.
Davis (Eds.), DSM-IV sourcebook (Vol. 3, pp. 221–239). Washington, DC: American
Psychiatric Association.
Knutson, B., Wolkowitz, O. M., Cole, S. W., Chan, T., Moore, E. A, Johnson, R. C., …
Reus, V. I. (1998). Selective alteration of personality and social behavior by
serotonergic intervention. American Journal of Psychiatry, 155, 373–379.
Bulletin, 136(5), 768.
Krasner, M. S., Epstein, R. M., Beckman, H., Suchman, A. L., Chapman, B., Mooney, C.
J., & Quill, T. E. (2009). Association of an educational program in mindful
communication with burnout, empathy, and attitudes among primary care physicians.
Journal of the American Medical Association, 302(12), 1284–1293.
Krueger, R. F. (2002). Psychometric perspectives on comorbidity. In J. E. Helzer & J. J.
Hudziak (Eds.), Defining psychopathology in the 21st century. DSM V and beyond (pp.
41–54). Washington, DC: American Psychiatric Publishing.
Krueger, R. F., Derringer, J., Markon, K. F., Watson, D., & Skodol, A. E. (2012). Initial
Krueger, R. F., & Eaton, N. R. (2010). Personality traits and the classification of mental
disorders: Toward a complete integration in DSM-V and an empirical model of
psychopathology. Personality Disorders: Theory, Research, and Treatment, 1, 97–118.
Krueger, R. F., & Markon, K. E. (2006). Reinterpreting comorbidity: A model-based
approach to understanding and classifying psychopathology. Annual Review of Clinical
Kwapil, T. R., & Barrantes-Vidal, N. (2012). Schizotypal personality disorder: An
integrative review. In T. A. Widiger (Ed.), The Oxford handbook of personality
64, 241–256.
personality lexicons of English and 11 other languages. Journal of Personality, 76,
1001–1053.
Leichsenring, F., & Leibing, F. (2003). The effectiveness of psychodynamic therapy and
cognitive behavior therapy in the treatment of personality disorders: A meta-analysis.
Lin-Healy, F., & Small, D. A. (2013). Nice guys finish last and guys in last are nice. The
clash between doing well and doing good. Social Psychological and Personality
Science, 4, 692–694.
Livesley, W. J. (2006). General Assessment of Personality Disorder (GAPD).
Unpublished manuscript. Department of Psychiatry, University of British Columbia.
Livesley, W. J., & Jackson, D. (2009). Manual for the Dimensional Assessment of
Personality Pathology-Basic Questionnaire. Port Huron, MI: Sigma.
Lowe, J. R., Edmundson, M., & Widiger, T. A. (2009). Assessment of dependency,
agreeableness, and their relationship. Psychological Assessment, 21, 543–553.
comparison of the DSM-IV-TR with dimensional models of personality disorder.
Lykken, D. T. (1971). Multiple factor analysis and personality research. Journal of
Experimental Research in Personality, 5, 161–170.
Lynam, D. R. (2012). Assessment of maladaptive variants of five factor model traits.
Lynam, D. R., Gaughan, E. T., Miller, J. D., Miller, D. J., Mullins-Sweatt, S., & Widiger,
T. A. (2011). Assessing the basic traits associated with psychopathy: Development and
validation of the Elemental Psychopathy Assessment. Psychological Assessment, 23,
108–124.
Lynam, D. R., & Widiger, T. A. (2007). Using a general model of personality to
understand sex differences in the personality disorders. Journal of Personality
Disorders, 21, 583–602.
Magidson, J. F., Roberts, B. W., Collado-Rodriguez, A., & Lejuez, C. W. (2014). Theory-
driven intervention for changing personality: Expectancy value theory, behavioral
activation, and conscientiousness. Developmental Psychology, 50, 1442–1450.
Malouff, J. M., Thorsteinsson, E. B., Rooke, S. E., & Schutte, N. S. (2007). Alcohol
involvement and the five factor model of personality: A meta-analysis. Journal of Drug
Education, 37, 277–294.
Malouff, J. M., Thorsteinsson, E. B., & Schutte, N. S. (2005). The relationship between
the five factor model of personality and symptoms of clinical disorders: A meta-
analysis. Journal of Psychopathology and Behavioral Assessment, 27, 101–114.
Malouff, J. M., Thorsteinsson, E. B., & Schutte, N. S. (2006). The five factor, model of
personality and smoking: A meta-analysis. Journal of Drug Education, 36, 47–58.
Marsh, H. W., Morin, A. J., Parker, P. D., & Kaur, G. (2014). Exploratory structural
equation modeling: An integration of the best features of exploratory and confirmatory
factor analysis. Annual Review of Clinical Psychology, 10, 85–110.
McCrae, R. R., & Costa, P. T. (1983). Joint factors in self-reports and ratings:
Neuroticism, Extraversion, and Openness to Experience. Personality and Individual
McCrae, R. R., & Costa, P. T. (1986). Clinical assessment can benefit from recent
advances in personality psychology. American Psychologist, 41, 1001–1003.
McCrae, R. R., & Costa, P. T. (1989a). Reinterpreting the Myers-Briggs Type Indicator
from the perspective of the Five Factor Model of personality. Journal of Personality,
57, 17–40.
McCrae, R. R., & Costa, P. T. (1989b). The structure of interpersonal traits: Wiggins’
circumplex and the Five Factor Model. Journal of Personality and Social Psychology,
56, 586–595.
McCrae, R. R., & Costa, P. T. (2003). Personality in adulthood. A five factor theory
McCrae, R. R., & Costa, P. T. (2013). Introduction to the empirical and theoretical status
of the Five Factor Model of personality traits. In T. A. Widiger & P. T. Costa (Eds.),
Personality disorders and the five factor model of personality (pp. 15–27).
McCrae, R. R., Costa, P. T., & Piedmont, R. L. (1993). Folk concepts, natural language,
and psychological constructs: The California Psychological Inventory and the Five
Factor Model. Journal of Personality, 61, 1–26.
McCrae, R. R., Zonderman, A. B., Costa, P. T., Bond, M. H., & Paunonen, S. V. (1996).
Evaluating replicability of factors in the Revised NEO Personality Inventory:
personality and developmental psychopathology as childhood antecedents of
Inventory for Children (HiPIC). In I. Mervielde, I. Deary, F. De Fruyt, & F. Ostendorf
(Eds.), Personality Psychology in Europe, Proceedings of the Eighth European
Conference on Personality Psychology (pp. 107–127). Tilburg, The Netherlands:
Mervielde, I., & De Fruyt, F. (2002). Assessing children’s traits with the Hierarchical
Personality I Inventory for Children. In B. de Raad & M. Perugini (Eds.), Big five
assessment (pp. 129–146). Bern, Switzerland: Hogrefe & Huber.
Miller, J. D., Widiger, T. A., & Campbell, W. K. (2010). Narcissistic personality disorder
and the DSM-V. Journal of Abnormal Psychology, 119, 640–649.
Miller, T. R. (1991). The psychotherapeutic utility of the five factor model of personality:
A clinician’s experience. Journal of Personality Assessment, 57, 415–433.
Millon, T. (1981). Disorders of personality: DSM-III: Axis II. New York, NY: Wiley.
Millon, T. (2011). Disorders of personality. Introducing a DSM/ICD spectrum from
normal to abnormal (3rd ed.). New York, NY: Wiley.
Moffitt, T. E., Arseneault, L., Belsky, D., Dickson, N., Hancox, R. J., Harrington, H. L.,
… Caspi, A. (2001). A gradient of childhood self-control predicts health, wealth, and
public safety. Proceedings of the National Academy of Sciences, 108, 2693–2698.
Mullins-Sweatt, S. N., Edmundson, M., Sauer-Zavala, S., Lynam, D. R., Miller, J. D., &
Widiger, T. A. (2012). five factor measure of borderline personality traits. Journal of
Mullins-Sweatt, S. N., Lengel, G. J., & DeShong, H. L. (in press). The importance of
considering clinical utility in the construction of a diagnostic manual. Annual Review of
Clinical Psychology.
Mullins-Sweatt, S. N., & Widiger, T. A. (2006). The five factor model of personality
disorder: A translation across science and practice. In R. F. Krueger & J. L. Tackett
(Eds.), Personality and psychopathology (pp. 39–70). New York, NY: Guilford Press.
Mullins-Sweatt, S. N., & Widiger, T. A. (2007). The Shedler-Westen Assessment
Procedure from the perspective of general personality structure. Journal of Abnormal
Mullins-Sweatt, S. N., & Widiger, T. A. (2011). Clinicians’ judgments of the utility of the
DSM-IV-TR and five factor models for personality disordered patients. Journal of
Munafo, M. R., Zetteler, J. I., & Clark, T. G. (2007). Personality and smoking status: A
meta-analysis. Nicotine and Tobacco Research, 9, 405–413.
Nigg, J. T., John, O. P., Blaskey, L. J., Huang-Pollock, C. L., Willicut, E. G., Hinshaw, S.
P., & Pennington, B. (2002). Big five dimensions and ADHD symptoms: Links
between personality traits and clinical symptoms. Journal of Personality and Social
93, 116–130.
Oakley, B. (Ed.). (2012). Pathological altruism. New York, NY: Oxford University
Press.
O’Connor, B. P. (2002). A quantitative review of the comprehensiveness of the five
Oltmanns, J., & Widiger, T. A. (in press). Self-pathology, the five factor model, and
bloated specific factors: A cautionary tale. Journal of Abnormal Psychology.
Oltmanns, T. F., & Powers, A. D. (2012). Gender and personality disorders. In T. A.
Widiger (Ed.), The Oxford handbook of personality disorders (pp. 206–218). New
York, NY: Oxford University Press.
Osborne, J. S., & Fitzpatrick, D. C. (2012). Replication analysis in exploratory factor
analysis: What it is and why it makes your analysis better. Practical Assessment,
Research, and Evaluation, 17, Available online: http://pareonline.net/getvn.asp?
v=17&n=15
Ozer, D. J., & Reise, S. P. (1994). Personality assessment. Annual Review of Psychology,
45, 357–388.
Perry, J. C., & Bond, M. (2000). Empirical studies of psychotherapy for personality
disorders. In J. Gunderson & G. Gabbard (Eds.), Psychotherapy for personality
disorders (pp. 1–31). Washington, DC: American Psychiatric Press.
Perugini, M., & Di Blas, L. (2002). The Big Five Marker Scales (BGMS) and the Italian
AB5C taxonomy–analyses from an etic-emic perspective. In B. de Raad & M. Perugini
(Eds.), Big five assessment (pp. 281–304). Bern, Switzerland: Hogrefe & Huber.
Piedmont, R. L. (2001). Cracking the plaster cast: Big Five personality change during
intensive outpatient counseling. Journal of Research in Personality, 35, 500–520.
Pike, K. L. (1954). Language in relation to a unified theory of the structure of human
behavior. The Hague, Netherlands: Mouton.
Pincus, A. L., & Hopwood, C. J. (2012). A contemporary interpersonal model of
personality pathology and personality disorder. In T. A. Widiger (Ed.), Oxford
Press.
Pine, D. S., Alegria, M., Cook, E. H., Costello, E. J., Dahl, R. E., Koretz, D., … Vitiello,
B. (2002). Advances in developmental science and DSM-V. In D. J. Kupfer, M. B.
First, & D. E. Regier (Eds.), A research agenda for DSM-V (pp. 85–122). Washington,
Presnall, J. R. (2013). Disorders of personality: Clinical treatment from a five factor
model perspective. In. T. A. Widiger & P. T. Costa (Eds.), Personality disorders and
the five factor model of personality (3rd ed., pp. 395–408). Washington, DC: American
Reiss, D., Eccles, J. S., & Nielsen, L. (2014). Conscientiousness and public health:
Synthesizing current research to promote healthy aging. Developmental Psychology,
50, 1303–1314.
Roberts, B. W., Harms, P. D., Caspi, A., & Moffitt, T. E. (2007). Predicting the
counterproductive employee in a child-to-adult prospective study. The Journal of
Robins, L. N. (1978). Sturdy childhood predictors of adult outcomes: Replication from
longitudinal studies. Psychological Medicine, 8, 611–622.
Rogers, C. R. (1961). On becoming a person: A therapist’s view of psychotherapy.
Boston, MA: Houghton-Mifflin.
Rojas, S. L., & Widiger, T. A. (2014). The convergent and discriminant validity of the
Five Factor Form. Assessment, 21, 143–157.
Rokeach, M. (1960). The open and closed mind. New York, NY: Basic Books.
Rolland, J-P., & de Fruyt, F. (2003). The validity of FFM personality dimensions and
maladaptive traits to predict negative affect at work. European Journal of Personality,
17, 101–121.
Ronningstam, E. (2011). Narcissistic personality disorder in DSM-V. In support of
retaining a significant diagnosis. Journal of Personality Disorders, 25, 248–259.
Sackett, P. R., & Walmsley, P. T. (2014). Which personality attributes are most important
in the workplace? Perspectives on Psychological Science, 9, 538–551.
Samuel, D. B. (2013). Assessing the five factor model of personality disorder. In T. A.
Widiger & P. T. Costa (Eds.), Personality disorders and the five factor model of
personality (3rd ed., pp. 221–232). Washington, DC: American Psychological
Association.
Samuel, D. B., & Gore, W. L. (2012). Maladaptive variants of conscientiousness and
agreeableness. Journal of Personality, 80, 1669–1696.
Samuel, D. B., Riddell, A. D. B., Lynam, D. R., Miller, J. D., & Widiger, T. A. (2012). A
five factor measure of obsessive-compulsive personality traits. Journal of Personality
Samuel, D. B., & Widiger, T. A. (2004). Clinicians’ descriptions of prototypic personality
disorders. Journal of Personality Disorders, 18, 286–308.
comparison of the DSM-IV and five factor models. Journal of Abnormal Psychology,
115, 298–308.
Sanderson, C., & Clarkin, J. F. (2002). Further use of the NEO PI-R personality
dimensions in differential treatment planning. In P. T. Costa & T. A. Widiger (Eds.),
Personality disorders and the five factor model of personality (2nd ed., pp. 351–375).
New York, NY: American Psychological Association.
Sanislow, C. A., Pine, D. S., Quinn, K. J., Kozak, M. J., Garvey, M. A., Heinssen, R. K.,
… Cuthbert, B. N. (2010). Developing constructs for psychopathology research:
Research domain criteria. Journal of Abnormal Psychology, 119, 631–638.
Schmit, M. J., Kihm, J. A., & Robie, C. (2002). The Global Personality Inventory (GPI).
In B. de Raad & M. Perugini (Eds.), Big five assessment (pp. 195–236). Bern,
Switzerland: Hogrefe & Huber.
Schmitt, D. P., Alcalay, L., Allik, J., Ault, L., Austers, I., Bennett, K. L., … Zupaneie, A.
(2003). Universal sex differences in the desire for sexual variety: Tests from 52
nations, 6 continents, and 13 islands. Journal of Personality and Social Psychology, 85,
85–104.
Schmitt, D. P., Allik, J., McCrae, R. R., Benet-Martinez, V., Alcalay, L., Ault, L., …
Zupancic, A. (2007). The geographic distribution of Big Five personality traits:
Patterns and profiles of human self-description across 56 nations. Journal of Cross-
Cultural Psychology, 38, 173–212.
Schmitt, D. P., Realo, A., Voracek, M., & Allik, J. (2008). Why can’t a man be more like
a woman? Sex differences in Big Five personality traits across 55 cultures. Journal of
Schmitt, T. A. (2011). Current methodological considerations in exploratory and
confirmatory factor analysis. Journal of Psychoeducational Assessment, 29, 304–321.
Seibert, S. E., & Kraimer, M. L. (2001). The five factor model of personality and career
success. Journal of Vocational Behavior, 58, 1–21.
Shaffer, D., Widiger, T. A., & Pincus, H. A. (1998). DSM-IV child disorders, part II:
Final overview. In T. A. Widiger, A. J. Frances, H. A. Pincus, R. Ross, M. B. First, W.
Davis, & M. Kline (Eds.), DSM-IV sourcebook (Vol. 4, pp. 963–977). Washington,
Shedler, J., Beck, A., Fonagy, P., Gabbard, G. O., Gunderson, J. G., Kernberg, O., …
167, 1027–1028.
Shiner, R., & Tackett, J. L. (2013). Personality & personality disorders. In E. Mash & R.
Barkley (Eds.), Child psychopathology (3rd ed.). New York, NY: Guilford Press.
CAT-PD project. Journal of Personality Assessment, 93, 380–389.
Skodol, A. E. (2010, February 10). Rationale for proposing five specific personality
types. Retrieved from
http://www.dsm5.org/ProposedRevisions/Pages/RationaleforProposingFiveSpecificPersonalityDisord
Skodol, A. E. (2012). Personality disorders in DSM-5. Annual Review of Clinical
Skodol, A. E., Gunderson, J. G., Shea, M. T., McGlashan, T. H., Morey, L. C., Sanislow,
C. A., … Stout, R. L. (2005). The Collaborative Longitudinal Personality Disorders
Study (CLPS): Overview and implications. Journal of Personality Disorders, 19, 487–
504.
Smith, T. W., & Williams, P. G. (1992). Personality and health: Advantages and
limitations of the five factor model. Journal of Personality, 60, 395–423.
personality traits from early childhood to early adulthood. Journal of Personality, 82,
182–199.
Disorders, 16, 402–423.
Steel, P., Schmidt, J., & Schultz, J. (2008). Refining the relationship between personality
Stepp, S. D., Whalen, D. J., & Smith, T. D. (2013). Dialectical behavior therapy from the
perspective of the five factor model of personality. In T. A. Widiger & P. T. Costa
(Eds.), Personality disorders and the five factor model of personality (3rd ed., pp. 395–
Stone, M. H. (2013). Treatment of personality disorders from the perspective of the five
Strauss, M. E., & Smith, G. T. (2009). Construct validity: Advances in theory and
methodology. Annual Review of Clinical Psychology, 5, 1–25.
Tackett, J. L., Kushner, S. C., Mervielde, I., & De Fruyt, F. (2013). Delineating
personality traits in childhood & adolescence: Associations across measures,
temperament, & behavioral problems. Assessment, 20, 738–751.
Taylor, G. J., & Bagby, R. M. (2013). Alexithymia and the five factor model of
factor model of personality (pp. 193–208). Washington, DC: American Psychological
Association.
Tellegen, A. (1993). Folk concepts and psychological concepts of personality and
personality disorder. Psychological Inquiry, 4, 122–130.
Development of the Multidimensional Personality Questionnaire. Unpublished
manuscript, Minneapolis, Minnesota.
Tellegen, A., Watson, D., & Clark, L. A. (1999). On the dimensional and hierarchical
structure of affect. Psychological Science, 10, 297–303.
ten Have, M., Oldehinkel, A., Vollebergh, W., & Ormel, J. (2005). Does neuroticism
explain variations in care service use for mental health problems in the general
population? Results from the Netherlands Mental Health Survey and Incidence Study
(NEMESIS). Social Psychiatry and Psychiatric Epidemiology, 40, 425–431.
Thoresen, C. J., Kaplan, S. A., Barsky, A. P., Warren, C. R., & de Chermont, K. (2003).
The affective underpinnings of job perceptions and attitudes: A meta-analytic review
and integration. Psychological Bulletin, 129, 914–945.
Thurstone, L. L. (1934). The vectors of mind. Psychological Review, 41, 1–32.
Tomiatti, M., Gore, W. L., Lynam, D. R., Miller, J. D., & Widiger, T. A. (2012). A five
factor measure of histrionic personality traits. In A. M. Columbus (Ed.), Advances in
psychology research (Vol. 87, pp. 113–138). Hauppauge, NY: Nova Science
Publishers.
to include the big five of personality. Journal of Personality and Social Psychology,
59(4), 781–790.
Trobst, K. K., Ayearst, L. E., & Salekin, R. T. (2004). Where is the personality in
personality disorder assessment? A comparison across four sets of personality disorder
scales. Multivariate Behavioral Research, 39, 231–271.
Trobst, K. K., & Wiggins, J. S. (2003). Constructive alternativism in personality
assessment. Madeline G. from multiple perspectives. In J. S. Wiggins (Ed.), Paradigms
of personality assessment (pp. 296–324). New York, NY: Guilford Press.
Trull, T. J. (1992). DSM-III-R personality disorders and the five factor model of
Trull, T. J., & Durrett, C. A. (2005). Categorical and dimensional models of personality
disorder. Annual Review of Clinical Psychology, 1, 355–380.
Trull, T. J., & Widiger, T. A. (2002). The Structured Interview for the Five Factor Model
of Personality (SIFFM). In B. de Raad & M. Perugini (Eds.), Big five assessment (pp.
147–170). Bern, Switzerland: Hogrefe & Huber.
Tyrer, P. (2015). Personality dysfunction is the cause of recurrent non-cognitive mental
disorder: A testable hypothesis. Personality and Mental Health, 9, 1–7.
Verheul, R. (2005). Clinical utility for dimensional models of personality pathology.
Verheul, R., Andrea, H., Berghout, C. C., Dolan, C., Busschbach, J. J., van der Kroft, P.
J., … Fonagy, P. (2008). Severity Indices of Personality Problems (SIPP-118):
Development, factor structure, reliability, and validity. Psychological Assessment, 20,
23–34.
Waller, N. G., & Ben-Porath, Y. S. (1987). Is it time for clinical psychology to embrace
the five factor model of personality? American Psychologist, 42, 887–889.
Watson, D., & Clark, L. A. (1994). Emotions, moods, traits, and temperaments:
Conceptual distinctions and empirical findings. In P. Ekman & R. J. Davidson (Eds.),
The nature of emotion: Fundamental questions (pp. 89–93). New York, NY: Oxford
University Press.
Watson, D., Gamez, W., & Simms, L. J. (2005). Basic dimensions of temperament and
their relation to anxiety and depression: A symptom-based perspective. Journal of
Watson, D., Klohnen, E. C., Casillas, A., Nus Simms, E., Haig, J., & Berry, D. S. (2004).
Widiger, T. A. (2009). Neuroticism. In M. R. Leary & R. H. Hoyle (Eds.), Handbook of
individual differences in social behavior (pp. 129–146). New York, NY: Guilford
Press.
Widiger, T. A. (2010a). Personality, interpersonal circumplex, and DSM-5: A
commentary on five studies. Journal of Personality Assessment, 92, 528–532.
Widiger, T. A. (May, 2010b). Scientific construction of a diagnostic manual. DSM-I
through DSM-V. Society for the Science of Clinical Psychology Distinguished
Scientist Award Address, 22nd Annual Meeting of the American Psychological
Society, Boston, MA.
Widiger, T. A. (2012). Historical developments and current issues. In T. A. Widiger
(Ed.), The Oxford handbook of personality disorders (pp. 13–34). New York, NY:
Widiger, T. A. (2015). The consequential outcomes of personality dysfunction.
Widiger, T. A. (in press). Classification and diagnosis: Historical development and
contemporary issues. In J. Maddux & B. Winstead (Eds.), Psychopathology:
Foundations for a contemporary understanding (4th ed.). Mahwah, NJ: Lawrence
Erlbaum Associates.
Widiger, T. A., & Costa, P. T. (2012). Integrating normal and abnormal personality
structure: The five factor model. Journal of Personality, 80, 1471–1506.
Widiger, T. A., De Clercq, B., & De Fruyt, F. (2009). Childhood antecedents of
personality disorder: An alternative perspective. Development and Psychopathology,
21, 771–791.
maladaptive variants of the five factor model. Journal of Personality Assessment, 94,
450–455.
Widiger, T. A., & Mullins-Sweatt, S. N. (2009). Five Factor model of personality
Widiger, T. A., & Presnall, J. R. (2013). Clinical application of the five factor model.
Widiger, T. A., & Samuel, D. B. (2005). Evidence based assessment of personality
disorders. Psychological Assessment, 17, 278–287.
Integrating normal and abnormal personality structure: The five factor model. In T. A.
Widiger, T. A., & Smith, G. T. (2008). Personality and psychopathology. Handbook of
Press.
Widiger, T. A., & Spitzer, R. L. (1991). Sex bias in the diagnosis of personality disorders.
Clinical Psychology Review, 11, 1–22.
Widiger, T. A., & Trull, T. J. (1997). Assessment of the five factor model of personality.
Widiger, T. A., Trull, T. S., Clarkin, J. F., Sanderson, C., & Costa, P. T. (1994). A
description of the DSM-III-R and DSM-IV personality disorders with the five factor
model of personality. In P. T. Costa & T. A. Widiger (Eds.), Personality disorders and
the five factor model of personality (pp. 41–56). Washington, DC: American
Wiggins, J. S. (1973). Personality and prediction: Principles of personality assessment.
Reading, MA: Addison-Wesley.
Wiggins, J. S. (1991). Agency and communion as conceptual coordinates for the
understanding of and measurement of interpersonal behavior. In W. M. Grove & D.
Ciccetti (Eds.), Thinking clearly about psychology: Vol. 2. Personality and
Wiggins, J. S. (2003). Paradigms of personality assessment. New York, NY: Guilford
Press.
Wiggins, J. S., & Pincus, H. A. (1989). Conceptions of personality disorder and
dimensions of personality. Psychological Assessment, 1, 305–316.
Woodworth, R. S. (1919). Examination of emotional fitness for war. Psychological
Bulletin, 15, 59–60.
Woodworth, R. S. (1920). Personal data sheet. Chicago, IL: Stoelting.
World Health Organization. (1992). The ICD-10 classification of mental and behavioural
disorders. Clinical descriptions and diagnostic guidelines. Geneva, Switzerland:
Author.
World Health Organization. (2014). WHO calls for stronger focus on adolescent health.
From http://www.who.int/mediacentre/news/releases/2014/focus-adolescent-health/en/
Conjoint analysis of the CAT-PD, PID-5, and NEO PI-3 trait models. Personality
Yamagata, S., Suzuki, A., Ando, J., One, Y., Kijima, N., Yoshimura, K., … Jang, K. L.
Zapolski, T. C. B., Guller, L., & Smith, G. T. (2013). On the valid description of
personality dysfunction. In T. A. Widiger & P. T. Costa (Eds.), Personality disorders
and the five factor model of personality (3rd ed., pp. 29–42). Washington, DC:
Zuckerman, M. (2002). Zuckerman-Kuhlman Personality Questionnaire (ZKPQ): An
alternative five-factorial model. In B. de Raad & M. Perugini (Eds.), Big five
assessment (pp. 377–396). Bern, Switzerland: Hogrefe & Huber.
INDEX
Page numbers followed by f and t indicate figures and tables.
AB5C. See Abridged Big Five Circumplex

ABC. See affective, behavioral, and cognitive states
ABCDs. See affect, behavior, cognition, and desire
abnormal personality structure, 229–33
Abridged Big Five Circumplex (AB5C), 17–18, 59–61, 119, 137, 194, 366
absolute stability, of Neuroticism, 41
abstract trait words, 193
acculturation, 184
activity facet, 244, 254, 259–60, 301
adaptability, of Extraversion, 64
adjective-based measures, 111, 196, 229, 365
AB5C, 366
IPIP, 366
IPIP-AB5C, 366–67
adjective lists, 138, 196, 229, 245, 251, 266
adolescence, 176, 550–51. See also consequential outcomes
conscientiousness during, 140, 265
extraversion development in, 67–68
FFM personality antecedents, 245
neuroticism conceptualization in, 42–43
openness in, 88–89
openness to experience throughout, 265
personality and temperament, 275t
Rothbart’s taxonomies for, 271
adolescence personality measurements, 245, 246t, 247t, 248t, 249t, 250t, 253–55, 265
adult questionnaires for, 251–52
BFI, 252
CCQ, 254, 259–60
FFPI, 252
international classification study, 256
NEO-PI-R, 252, 256–57, 263
self-report and peer reports, 251, 255, 262, 266
Adolescent Personality Style Inventory, 252
adoption studies, 311
adulthood
agreeableness/Benevolence in, 260–61, 265, 273
conscientiousness during, 139–40, 262
extraversion development and, 67–68, 260
neuroticism, emotional stability in, 47, 261–62
openness in, 88, 89–90, 92–93, 262–64, 273
adults, 302. See also top-down research approach
personality traits and, 28
psychopathology, 2, 4, 28
questionnaires, in childhood and adolescent measures, 121, 251–52
self-report for, 266
Adult Temperament Questionnaire, 121
affect, 61
extraversion and, 68–69, 71
affect, behavior, cognition, and desire (ABCDs), of personality, 57
in extraversion, 61, 62t, 63, 68–73
affective, behavioral, and cognitive (ABC) states, 60
age
in animal personality facets, 289
childhood and adolescence consequential outcomes on, 509
higher-order invariant differences in, 180–81
agoraphobia, 487
agreeableness, 2, 127–28, 544–45
animal personality and, 285
BIS and BAS relation to, 120
CB5T on, 339–40
characteristic adaptations, 123, 124–26
conscientiousness and, 120–21, 429, 431t
construct specification and elaboration strategies, 110–11
cybernetic function of, 326t
emotions, motives, social behavior and, 105
empathy and, 120
empirical concept emergence, 107–10
FFT and, 122–23
fMRI on, 339
implications for, 124–25
leadership and, 390
maladaptive measures on, 462–63
motivational and cognitive processes, 123–24
neuroticism contrast with conscientiousness and, 16
opponent process to, 119–20, 124–26, 125f
personality neuroscience and, 339–40
self-reports on, 118
social accommodation and, 121–22
temperament and, 120–21
theoretical explanations for, 119–23
trait word descriptions, 111–12
warmth and, 106
agreeableness/benevolence, adult and child differences, 260–61, 265, 273
agreeableness facets, 106, 127
altruism, 108, 109–10, 111, 113, 115, 261, 339
compliance, 108, 109, 110, 113, 115, 261
Dark Triad related to, 115
destructive behaviors and, 115
extraction problem, 111–17, 119
FFM empirical research on, 117–19
HEXACO, 116–17
modesty, 111, 113
multiple, 115–16
NEO PI-R on, 110–11, 118
PscyINFO searches on, 112–13, 112t
psychopathology and, 115
straightforwardness, 111, 113–14, 115
surface similarity and, 118
tender-mindedness, 109, 111, 114, 115
trust, 111, 114–15
TTRF and IASR-B5 on, 111
Allport-Odbert personality list, 108
alphabetical psychology, German, 195
alpha factors, 227, 228
altruism, 108, 109–10, 111, 113, 115, 261, 339
amygdala, 329, 330, 332–34, 335, 340
anhedonic introversion, in EFA studies, 234
animal personality, 2, 4, 296, 551–52
beyond behavior, 289–90
of chimpanzees, 281–83, 285, 289, 292, 294–95, 337, 416
cladogram of, 292, 294f
cultures research, 294
dog studies, 284, 291
ethology and, 282, 283, 285
FFM dimensions comparison to, 284
future directions, 295
genetics in, 285, 289, 294–95
health psychology studies and, 416
history of, 281–85
immune function and, 289–90
learning and social relationships, 291–92
morbidity and mortality, 290–91
personality development and, 294–95
personality traits understanding and interpretation, 287–88
questionnaire approach, 285
research contribution, 285–95, 294f
rhesus macaques monkey studies, 283–84, 287, 291
species-typical structures identification, 285–87
stress and, 290
traits understanding for, 287–88
antagonism, 235, 260
antisocial behavior, 115
antisocial personality disorder (ASPD), 453–54
anxiety disorders, 487–90
ARAS. See ascending reticular activating system
ASA. See attraction-selection-attrition model
ascending reticular activating system (ARAS), 64–65
ASPD. See antisocial personality disorder
aspects, of trait hierarchy, 17, 326–27
assertiveness, 326t, 328
compassion, 326t, 339
cybernetic functions of, 326t
enthusiasm, 326t
industriousness, 326t
intellect, 326t
openness to experience, 326t
orderliness, 326t
politeness, 326t
volatility, 326t
withdrawal, 326t, 331, 334
assays of endogenous psychoactive substances, 321
assertiveness, 59, 60–61, 65, 68, 70, 72, 301, 326t, 328
assessment, of FFM, 373–75, 554–55. See also clinical utility
alternative FFM facet models and measures, 372–74
lexical tradition, 356t, 357t, 358t, 365–67
maladaptive measures, 358t, 359t, 368–69
observer and interview measures, 364–65
PSY-5 measures, 369–71
questionnaire tradition, 353–54, 355t, 356t, 360, 364–65
related lexical models and measures, 367–68
short Big Five scales, 367
trait and scale polarity, 371–72
unresolved issues in, 371
assessments. See also personality assessment
AB5C, 17–18, 59
couples, in cross-over analysis, 424–26
literature on NEO Inventories, 354, 360
Little Five personality dimensions methods, 245
in top-down approach, 252–55, 257–58
Association for Research on Personality Disorders, 528
assortative mating, univariate twin studies and, 304
atheoretical research, in personality neuroscience, 320
attraction-selection-attrition (ASA) model, 395
avoidant personality disorder, 234
Axis I psychopathology, 479, 499–500, 499t, 559–62
BD, 92, 232, 486–87
common cause model, 480, 480f, 482–83, 486–87, 489, 492
complication and scar models, 480, 480f, 485, 490, 493, 494, 497
depression disorders, 480–86
eating disorder, 496
OCD, 490–91
pathoplasty model, 480, 480f, 483–84, 487, 489–90, 492–98
schizophrenia and psychotic disorders, 493–96
spectrum model, 480, 480f, 486, 490, 491, 495–96
substance use disorders, 497–98
trauma, 491–93
vulnerability model, 43–44, 480, 480f, 481–82, 488–89, 491, 492, 494, 496, 497
B5/FFM, Big Five/Five–Factor Model

Balanced Emotional Empathy Scale, 339
Baltimore Longitudinal Study of Aging (BLSA)
on FFM comprehensiveness, 23
NEO Inventories Form S administered to, 13, 364–65
bandwidth-fidelity dilemma, in leadership, 389
Barratt Impulsivity Scale, 338
BAS. See Behavioral Approach System
Basic Tendency, in Five-Factory Theory of Personality, 24–25, 28, 58–63, 122
of agreeableness, 123, 124–26
basic traits, in personality, 272
bass-ackwards method, 259, 267
BD. See bipolar disorder
BDNF. See brain-derived neurotropic factor
behavior
antisocial, 115
definition of, 61
employee, 386–88
extraversion and, 70–71
FFM and animal personality beyond, 289–90
molecular genetics and, 4, 301–15, 321
personality neuroscience and, 320
reward-pursuit, 66
social, 105, 292
Behavioral Approach System (BAS), of RST, 65, 120, 328
behavioral data, in health psychology, 414
Behavioral Facilitation System (BFS), 65–66
Behavioral Inhibition System (BIS), of RST, 65, 120, 331
behavior genetics, 313–15
adoption studies and extended family designs, 311
biometric moderation models, 310–11
genes and environment influence on personality trait, 301
GWAS and personality, 86–87
mate selection and Openness, 96
multivariate and longitudinal twin studies, 306–10
SNPs and, 86–87
univariate twin studies, 302–6, 311
beta factors, 227, 228
BFAS. See Big Five Aspects Scales
BFI. See Big Five Inventory
BFQ. See Big Five Questionnaire
BFQ-C. See Big Five Questionnaire-Children version
BFS. See Behavioral Facilitation System
Big Two model, 206–7, 209
Big Three model, 207, 209
Big Five Aspects Scales (BFAS), 59, 228, 327, 373, 456
Big Five/Five-Factor Model (B5/FFM)
domains structuring, 209–10
HEXACO compared to, 116
international taxonomy of traits, 198–201, 211
psycholexical approach to, 191, 192, 197–99, 206–10
Big Five Inventory (BFI), 138, 175–77, 252, 367
Big Five Inventory for Children, 252, 508
Big Five Mini-Markers, 366
Big Five Modular Markers, 366, 372
Big Five personality traits perspective, 1, 2, 30, 58, 197
AB5C assessment of, 17–18, 59
Big Five Questionnaire (BFQ), 373, 374, 498–99
Big Five Questionnaire-Children version (BFQ-C), 508
Big Six model, 200, 201, 207–8
Big Seven model, 203, 208, 367
biological basis
for openness, 86–88
for personality structure, 28, 32n4
biomarkers, 88
biometric moderation models, 310–11
bipolar disorder (BD), 232, 486–87
openness and, 92
bipolarity, 232, 371–72, 457–60, 459t
BIS. See Behavioral Inhibition System
blood-oxygen-level-dependent (BOLD) signal, 320
BLSA. See Baltimore Longitudinal Study of Aging
BOLD. See blood-oxygen-level-dependent signal
borderline personality disorder, 3, 424, 467, 524, 527–28, 550, 553
bottom-up research approach, 251, 354
CCQ, on adolescent personality structure, 254, 259–60
conscientiousness as childhood personality factor, 262
from free description research, 255–57, 261
free-response techniques, 256
on infants, 264–65
jingle-jangle fallacy, 255
on Openness or Intellect, 263
brain-derived neurotropic factor (BDNF), 88
brain function. See also neuroscience
dopamine and, 65–66, 325–26, 328–29, 335
extraversion and, 329–30
serotonin and, 325–26, 331–32, 336–37, 340
brain systems, underlying Extraversion
ARAS arousal, 64–65
EEG and fMRI on Extraversion and cortical arousal, 65
reward processing and dopaminergic functioning, 65–66
broad trait domain, of Conscientiousness, 135–36
bullying, 511–12
business and industry, 4, 381, 555–56
debates on FFM and, 396–97
employee job attitudes, motivation and behavior, 386–88
entrepreneurship, 394–96
leadership, 388–91
personnel selection/job performance, 382–86
research on personality in, 383
teams, 391–94
Buss and Plomin model, of childhood temperament, 269, 271
C. See common environment

CAC. See Conscientiousness Adjective Checklist
California Child Q-set (CCQ), 261, 508
Activity domain, 254
bottom-up perspective of adolescent personality structure, 254, 259–60
Common Language adaptation of, 253–54
dependency domain, 254
ego-control and ego-resiliency dimensions, 254
on FFM personality dimensions of infants and toddlers, 264
on openness to experience/intellect, 262
self-ratings and peer-ratings, 255, 266
Sociability domain, 254
strengths of, 257–58
California Psychological Inventory (CPI)
on agreeableness, 116
on traits, 24
California Q-Set, on personality structure, 15–16, 113, 117–18, 450, 524
candidate gene analysis, 311–12
CAPs. See Cognitive-Affective Processing system
caregiver or teacher, as informant, 246t, 247t, 248t, 251, 266, 270
Care system, 125
CAT. See computerized adaptive testing
catechol-O-methyltransferase (COMT) gene, 87
CAT-PD. See Computerized Adaptive Test of Personality Disorder
cat personality studies, 284
Cattell’s trait sphere, 195–96, 222–23
CB5T. See Cybernetic Big Five Theory
CBCL. See Child Behavior Checklist
CCICL. See Couples Critical Incidents Check List
CCQ. See California Child Q-set
CCS. See Chernyshenko Conscientiousness Scales
CFA. See confirmatory factor analysis
CFAs. See common factor analyses
characteristic adaptations, 24–25, 28, 272
of agreeableness, 123, 124–26
of extraversion, 68–73
in FFT of Personality, 122
Chernyshenko Conscientiousness Scales (CCS), 137–38
Child Behavior Checklist (CBCL), 508
childhood, 261, 550–51. See also consequential outcomes
conscientiousness during, 140
FFM personality antecedents, 2, 245
neuroticism antecedents, 2
neuroticism conceptualization in, 42–43
openness in, 88
personality universals and, 176
childhood personality, 243, 274–75, 275t
activity dimension, 244
childhood temperament compared to, 244–45, 258
discovery and measurement of, 244–58
language of, 274
research challenges, 264–68
temperament linkages research, 271–73
childhood personality measurements, 245, 246t, 247t, 248t, 249t, 250t, 252–57. See also
bottom-up research approach; top-down research approach
Activity factor, 259
child versus adult personality taxonomies, 258–64
FFM traits above and beyond, 258–59
questionnaire approach, 267–68
theory-driven construction methods, 251
childhood temperament, 4, 275t
childhood personality compared to, 244–45, 258
Extraversion development and, 66–67
language of, 274
NE and, 42
neuroticism and difficult, 42
childhood temperament research
Buss and Plomin model, 269, 271
child personality and temperament linkages, 271–73
on complementary languages of trait differences, 273–74
Rothbart model, 269–70, 271
taxonomy of childhood traits, 268
on temperament and personality concepts, 268
on temperament structure, 271
Thomas and Chess model, 268–69, 271
child psychopathology, 4
Chimpanzee Personality Questionnaire, 284
chimpanzees, 281–83, 416
Extraversion and Conscientiousness facets, 285, 289, 337
personality development, 294–95
social behavior, 292
chi-square test, in CFA, 221
Cholesky decomposition model, 306, 307f, 308, 309
circadian rhythms, Neuroticism and physical health, 46–47
circumplex structure, 22, 210. See also interpersonal circumplex
cladogram, of animal and human personality, 292, 294f
clinical applications, 416–17, 531
clinical utility, of FFM, 4, 374–75, 563–64
communication, 524–25
DSM-IV-TR, DSM-5 and, 522–26, 530–32
ease of usage, 522–24
research, 528–31
treatment, 525–28
Cloninger’s Harm Avoidance, 331
CMI. See Cornell Medical Index
COA. See cross-over analysis
cognition, 339
defined, 335
description of, 61
extraversion and, 71
function, health and, 407
openness and, 84, 86–88, 92–93
personality neuroscience use of, 320
Cognitive-Affective Processing system (CAPs), 134
cognitive process, of Agreeableness, 123–24
cohort effects challenges, in universality, 185
Collaborative Longitudinal Personality Disorders Study, 114
common cause model, 480, 480f
anxiety disorders, 489
bipolar disorder, 486–87
trauma, 492
common environment (C), 303–4, 306
common factor analyses (CFAs), 160
Common Language adaptation, of CCQ, 253–54
common pathways (CP) model, 307, 308f
common variance, 156, 157, 159–60
communication, 70, 524–25
compassion, 326t, 339
compatibility
in COA case study, 436–44, 437f, 439f, 441f, 443f
COA identification of, 430–32, 432t
compliance
as agreeableness facet, 108, 109, 113, 115, 261
California Q-set and NEO PI-R on, 113, 117–18
in HiPIC, 261
complication and scar model, 480, 480f
anxiety disorders, 490
depression disorders, 483
substance use disorders, 497
trauma, 493
comprehensiveness
NEO Inventories, BLSA, MMPI and, 22–23
robustness compared to, 151
computational social science, 414–15
computerized adaptive testing (CAT), 370, 413–14
Computerized Adaptive Test of Personality Disorder (CAT-PD), 231, 370–71, 373, 374,
457, 465, 467, 468t, 469t
COMT. See catechol-O-methyltransferase gene
confirmatory factor analysis (CFA), 217, 220f, 237n3
chi-square test, 221
covariance in, 220
FFM factors correlations with, 226
for FFM structure, 224–25
goodness-of-fit indices, 220–21
hypothesis testing in, 219–20
on mental disorders, 234
personality structure restrictive assumptions, 224
conflict areas
in COA case study, 436–44, 437f, 439f, 441f, 443f
COA identification of, 430
confounding variables, 409–10
conscientiousness, 2, 326t, 545–46
during adolescence, 140, 265
adult and child taxonomy differences, 139–40, 262
agreeableness and, 120–21, 429, 431t
CB5T on, 337–39
chimpanzee personality and, 285, 289, 337
constraint, 138
demographic differences in, 182
difficulties with trait of, 133–34
effortful control, 139
entrepreneurship and, 395
fusiform gyrus association with, 339
gorilla personality and, 292
gratification delay, 139
grit correlation with, 139
health and health care behaviors, 141, 142, 288, 405
hierarchical structure of, 135–37
intelligence and, 141
job performance and, 385
leadership and, 390
lifespan development of, 139–41
maladaptive measures on, 459–62
mortality and, 406, 408
Neuroticism contrast with Agreeableness and, 16
as personality trait, 134–35, 143
PFC and, 337–38
predictive utility of, 141–42
proactive and inhibitory component of, 135–36
romantic relationships and, 141, 142
self-regulation, 139
teams and, 393
workplace and, 142
Conscientiousness Adjective Checklist (CAC), 138
conscientiousness facets
formality, 136
industriousness, 136, 138, 326t
orderliness, 136, 137, 138, 140, 326t
punctuality, 136
responsibility, 136
self-control, 136, 138, 139
traditionality, 136
conscientiousness measures
AB5C, 137
BFI, 138
CAC, 138
CCS, 137–38
FFMRF, 138
HEXACO-PI, 138
IPIP, 138
NEO-FFI, 137
NEO PI-R, 136, 137
TIPI, 138
consequential outcomes, in childhood and adolescence, 407, 562–63
age differences, 509
bullying, 511–12
FFM measures for, 508–9
gender differences, 509–10
learning achievement and, 512–14
learning styles, 514
long-term outcomes and, 516–17
mid-level constructs, 510t
pediatric diseases, 516
peer interaction and relationships, 251, 255, 262, 266, 511
psychopathology, 514–16
Considerate, in ICID, 261
consideration, in leadership, 390
constraint, MPQ on, 138
construct specification and elaboration strategies, Agreeableness and, 110–11
construct validity, 546
of FFM, 3, 11, 384
of NEO Inventories, 14
contextualized nature, of personality traits, 135
conventionality, 140
convergent validity, 11
of NEO-PI-R, 59
Cornell Medical Index (CMI), 13–14
cortical arousal, Extraversion and, 65, 329
cortical thickness, Neuroticism and, 333, 334
corticotropin-releasing hormone (CRH), 332
cortisol, Neuroticism link, 332
couples assessments, 424–26
Couples Critical Incidents Check List (CCICL), 433, 436, 438. 440, 442, 447
couple therapy
crossover analysis and, 4, 423–47
IBCT, 424
NEO PI-3 use, 424, 446
CP. See common pathways model
CPI. See California Psychological Inventory
creativity, openness and, 83, 88–89, 90, 94
CRH. See corticotropin-releasing hormone
cross-observer agreement, for traits, 12–13
cross-over analysis (COA), 426, 558
CCICL use, 433, 436, 438. 440, 442, 447
compatibility identification, 430–32, 432t
conflict areas identification, 430
self-reports comparison, 427–30, 431, 431t
cross-over analysis case study
background, 432–34
conflict and compatibility areas, 436–44, 437f, 439f, 441f, 443f
findings, 444–45
self-reports comparison, 434–36, 435f
therapeutic epilogue, 445–46
cultural relativism, on universality, 174
culture-level analysis, of FFM, 179, 181–82
cultures
animal personality research and, 294
openness and, 98–99
personality and, 28–29
personality universals and, 177–78
universality exceptions, 176
validation across, 267
Cybernetic Big Five Theory (CB5T)
advantage of, 324–25
on agreeableness, 339–40
on conscientiousness, 337–39
on extraversion, 327–31
on metatraits, 325
MIMIC approach, 324
on neuroticism, 331–35
traits functioning, 326t, 327
DAPP. See Dimensional Assessment of Personality Pathology

Dark Triad, Agreeableness facets related to, 115
data sets and analytic methods, 152
common, specific, and error variances, 156, 157, 159–60
congruences between factor solutions, 160
interbattery factor analyses and varimax rotation, 160
methods and illustrative data overview, 161
number of factors, 160
principal axis factor analyses and varimax rotation, 160
results, 161, 163
statistical significance tests, 160–61
targeted rotations, 160
decontextualized nature, of personality traits, 135
default network, 321
defense, self-descriptions and, 13
dementia, Openness and, 92–93
dependent personality disorder (DPD), 462–63
desire, 61, 71–72, 74n2
destructive behaviors, Agreeableness facet and, 115
development
animal personality and personality, 294–95
conscientiousness and lifespan, 139–41
extraversion and, 66–68, 260
generic description lists of, 252
multivariate twin studies and, 309–10
neuroticism and, 41–42
openness trajectory, 88–90
personality universals and, 176
Diagnostic and Statistical Manual of Mental Disorders (DSM)
clinical syndromes associated with FFM, 233–34
on Openness, 84–85, 455–57
on personality disorders, 84–85, 229–33, 368, 370, 528–31
Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR), 3
clinical utility, 522–26, 530–32
on PDs, 452–53, 464, 466, 521
Diagnostic and Statistical Manual of Mental Disorders (DSM-5)
clinical utility, 522–26, 530–32
PD model, 231, 233, 235, 451, 454–55, 457–58, 530
Personality and Personality Disorder Work Group, 231, 237n7, 521
PID-5, 231, 374, 463, 464–65, 467, 468t, 469t
psychoticism, Openness and, 455–57
Section III, 454–55, 523–24, 530
differences in perspective, in FFM higher order variants, 181
differential stability
of extraversion, 67
of neuroticism, 41
diffusion tensor imaging (DTI), 334
on empathy, 340
on Openness/Intellect and IQ, 336–37
Dimensional Assessment of Personality Pathology (DAPP), 230–31, 451, 457
dimensional classification system, 527
Dimensional Personality Symptom Item Pool (DIPSI), 464
dimensional structure level research, 167–68
dimensional trait models, 3
dimensions, 151, 157–59
ego-control and ego-resiliency, 254
emic, 178, 208, 211, 294, 296n4
etic, 208, 211, 296n4
factor analysis, 156, 158f, 163
Little Five personality, 245
Little Six personality, 259–60, 264
DIPSI. See Dimensional Personality Symptom Item Pool
discriminant validity, 11, 59, 360
disease progression, Openness and, 90
disease prone personality, 404
dizygotic (DZ) twins, 302
DNA analysis, 321
dog
learning and, 291
personality studies, 284
domains
conscientiousness broad trait, 135–36
discriminant validity within, 360
Hogan Likeability, 111
Dominance trait
in animal personality, 285
Fearless, in Psychopathic Personality Inventory, 285
in HiPIC, 260
dopamine
brain function, 325–26
extraversion influenced by, 65–66, 328–29
openness/Intellect and, 335
Plasticity influenced by, 325
dopamine D3 receptor gene (DRD3), 312
dopamine D4 receptor gene (DRD4), 87, 311, 312, 323–24
DPD. See dependent personality disorder
DRD3 gene. See dopamine D3 receptor gene
DRD4 gene. See dopamine D4 receptor gene
DSM. See Diagnostic and Statistical Manual of Mental Disorders
DSM-5. See Diagnostic and Statistical Manual of Mental Disorders
DSM-IV-TR. See Diagnostic and Statistical Manual of Mental Disorders
DTI. See diffusion tensor imaging
duality principle, 30, 31
personality content and, 22–25
traits and characteristic adaptations, 24–25
Dutch taxonomic project, 197, 198
DZ. See dizygotic twins
EAR. See Electronically Activated Recorder

East Baltimore Epidemiologic Catchment Area study, 113
education, 508
Edwards Personal Preference Schedule (EPPS), 161
EEA. See equal environments assumption
EEG. See electroencephalography
effortful control, 139
Egocentrism, in HiPIC, 260
ego-control and ego-resiliency dimension, of CCQ, 254
electroencephalography (EEG), 321, 330, 331
on extraversion and cortical arousal, 65
neuroticism research, 334
research on FRN and Extraversion, 329
Electronically Activated Recorder (EAR), 414
electrophysiological techniques, 321
Elemental Psychopathy Assessment (EPA), 454
emic dimensions, of personality, 178, 208, 211, 294, 296n4
Emotional Reactivity, in animal personality, 290
emotional stability
adolescence and, 265
neuroticism and, 47, 261–62
emotion regulation, 333–34, 340
emotions
Agreeableness and, 105
positive, 301
Emotions Profile Index, 287
empathy, 120, 339–40
Empathy Quotient, 339
empirical concept emergence, in Agreeableness, 107–10
employee job attitudes, motivation and behavior, 386–88
employment, Openness and, 93–94
ENRICH/PREPARE scale, 425
enthusiasm, 326t
entrepreneurship, 394–96
environmental influences
on childhood temperament and personality, 271–72
on neuroticism, 41, 44
openness and, 86
on personality, 31, 32
on personality traits, 301
EPA. See Elemental Psychopathy Assessment; exploratory factor analysis
EPI. See Eysenck Personality Inventory
epithets, personality traits in, 195
EPP. See Eysenck Personality Profiler
EPPS. See Edwards Personal Preference Schedule
EPQ. See Eysenck Personality Questionnaire
equal environments assumption (EEA), 304, 311
ERN. See error-related negativity
error-related negativity (ERN), 494
error variance, 156, 159–60
ESEM. See exploratory structural equation modeling
ethology, 282, 283, 285
etic dimension, of personality, 208, 211, 296n4
evaluation, in personality questionnaires, 233
evolution, 63, 281, 284–86
Experience Inventory, 197
explanations division, of traits, 63
exploratory factor analysis (EPA), 217, 220f
factor rotation in, 218–19, 225
Kaiser-Guttman rule, 218
pathological or anhedonic introversion in, 234
personality structure robustness and, 227
quantitatively rigorous criteria, 218
RMSEA and, 218
variable selection, 219
exploratory structural equation modeling (ESEM), 218, 220f, 231
DSM-PD criteria, 235
early applications of, 225–26
estimation methods, 222
flexibility of, 226
oblique rotation, 226
extended family designs, 311
external influences, on personality, 31, 182
externalizing spectrum, 234
extraction problem, for Agreeableness facets, 111–17, 119
extraversion, 2, 57, 74, 138, 542–43
ABCD approach, 61, 62t, 63, 68–73
ABC states, 60
adaptability of, 64
adult psychopathology research, 2
as Basic Tendency, 58–63
brain function and structure, 329–30
brain systems underlying, 64–66
CB5T on, 327–31
characteristic adaptations, 68–73
cortical arousal and, 65, 329
development, 66–68
dopamine influence on, 65–66, 328–29
EEG and fMRI on cortical arousal and, 65, 329
EEG research on FRN and, 329
evolution, genetics, and biology, 63–64
evolutionary perspectives, 63
facets of, 59–60, 289, 337
fitness contribution from, 63–64
fMRI and MRI studies on, 329, 330–31
genetic basis of, 63
habitual and specific responses, 60
Honesty-Humility trait in children, 260
job performance and, 383–84
lower-order structure, 59–61
neurobiological basis of, 64–65
openness link with, 16
PEM and, 60, 328, 329–30
positive affect association, 68–69, 71
resting EEG hemispheric asymmetry, 330
sociability, 58
teams and, 393
warmth and, 106
extraversion/sociability, adult and child differences, 260
extreme groups, for small samples, 322
Eysenck Personality Inventory (EPI), 58, 60, 197, 287–88
Eysenck Personality Profiler (EPP), 18, 19t, 20t, 58
Eysenck Personality Questionnaire (EPQ), 58, 60, 176, 311
Facebook users, 95–96, 414, 415f

Faceted Inventory of the Five Factor Model (FI-FFM), 373–74
facets, 17, 19t, 20t, 21t, 22, 211. See also agreeableness facets; conscientiousness facets
Activity, 244, 254, 259–60, 301
in animal personality research, 288–89
assertiveness, 59, 60–61, 65, 68, 72, 301, 326t, 328
defined, 326
EPP, 18
excitement-seeking, 301
of extraversion, 59–60, 289, 337
gregariousness, 301
HiPIC and, 260, 261
intermediate measures, 372–73
NEO Inventories, 18, 19t, 20t, 21t, 22t
of neuroticism, 39–40
normal range and maladaptive measures, 361t, 362t
of Openness, 84, 85
positive emotions, 301
scale level research, 164
substructure, 106
symmetry limitations, 360
warmth, 106, 224–25, 289, 301
factor analytic support, for FFM, 3–4, 237, 353, 548–50
alpha and beta, 227, 228
conditional independence assumption, 236
confirmatory, 219–21
ESEM, 221–22
exploratory, 218–19
first CFA study for testing, 224
future challenges in research, 227–36
methodological underpinnings, 217–22
network modeling, 235–36
of personality attributes, 222–27
family, Openness and, 96–98
feedback-related negativity (FRN)
EEG on Neuroticism and, 331
EEG research on, 329
feline immunodeficiency virus (FIV), 290–91
FFAvA. See Five Factor Avoidant Assessment
FFBI. See Five Factor Borderline Inventory
FFDI. See Five Factor Dependency Inventory
FFFS. See fight-flight-freeze system
FFM. See Five-Factor Model
FFMPD. See Five-Factor Model Personality Disorder
FFMRF. See Five-Factor Model Rating Form
FFMSS. See Five-Factor Model Score Sheet
FFNI. See Five Factor Narcissism Inventory
FFOCI. See Five-Factor Obsessive-Compulsive Inventory
FFPI. See Five-Factor Personality Inventory
FFS. See Fight-Flight System
FFT. See Five-Factor Theory
FI-FFM. See Faceted Inventory of the Five Factor Model
fight-flight-freeze system (FFFS), 331
Fight-Flight System (FFS), of RST, 65, 74n3, 125
FIPI. See Five-Item Personality Inventory
fitness, Extraversion contribution to, 63–64
FIV. See feline immunodeficiency virus
5DPT. See Five-Dimensional Personality Test
5-HTTLPR polymorphism, 43, 44, 47, 312
Five-Dimensional Personality Test (5DPT), 456
five-dimensional space, of FFC, 164–67
Five Factor Avoidant Assessment (FFAvA), 466
Five Factor Borderline Inventory (FFBI), 466, 467
Five Factor Dependency Inventory (FFDI), 466
Five-Factor Model (FFM), 1
description, 539–40
discussion, 539–64
five-dimensional space locations, 164–65
history, 223, 236n1, 360
life history strategies link, 292–93
MMPI dimensions factor analysis and, 156, 158f
strengths of, 2–3
system representation, 29f
theories of psychological function in, 324–27, 326t
variance in measures of personality inventories, 152, 153t, 154t, 155t, 156
Five-Factor Model Personality Disorder (FFMPD) scales, 466–67, 468t, 469t
Five-Factor Model Rating Form (FFMRF), 111, 138, 369, 372, 465–66
Five-Factor Model Score Sheet (FFMSS), 369
Five Factor Narcissism Inventory (FFNI), 466, 467
Five-Factor Obsessive-Compulsive Inventory (FFOCI), 461, 466
Five-Factor Personality Inventory (FFPI), 208, 252
Five-Factor Personality Inventory for Children, 252
Five-Factor Theory (FFT) of Personality
Agreeableness and, 122–23
Basic Tendency in, 24–25, 28, 58–63, 122–26
characteristic adaptation in, 122
NEO-PI to, 197
personality universals, 183–84
Self-Concept in, 122
Five-Item Personality Inventory (FIPI), 367
fMRI. See functional magnetic resonance imaging
formality, 136
fraternal twins. See dizygotic twins
free description research, 255–57, 261, 262, 263
free response techniques, 256
friendliness, 59, 70
FRN. See feedback-related negativity
frontoperietal or cognitive control network, 339
functional connectivity patterns, 320–21
functional magnetic resonance imaging (fMRI)
on agreeableness, 339
BOLD signal and, 320
default network discovery, 321
on extraversion and cortical arousal, 65, 329
functional connectivity patterns, 320–21
on neuroticism, 332
on openness/intellect, 335–36
fusiform gyrus, 339
GAD. See generalized anxiety disorder

GCTA. See genome-wide complex trait analysis
gender
in animal personality facets, 289
differences, in consequential outcomes, 509–10
FFM higher-order invariant differences in, 180
universals and, 174
general factor of personality (GFP), 17, 228
generalized anxiety disorder (GAD), 487–90
genes, 87, 301, 311, 312, 323–24
genetics, 552–53. See also behavior genetics; molecular genetics
in animal personality, 285, 289, 294–95
extraversion influenced by, 63
neuroticism influenced by, 40–41, 46
gene x environment interactions, 312
gene x gene interactions, 312
genome-wide association studies (GWAS), 86–87, 312–13, 321, 324
genome-wide complex trait analysis (GCTA), 313–14
geographic distribution, 182
geographic Openness, 98–99
German alphabetical psychology, 195
German Socio-Economic Panel Study, 410
German taxonomic project, 197, 198, 199
GFP. See general factor of personality
Goldberg’s taxonomy of descriptive terms, 196, 222, 223, 365–66
goodness-of-fit indices, in CFA, 220–21
gorilla personality study, 284, 292
Gough’s Adjective Check List, 196
grammatical categories, 201, 206
adjectives, nouns, verbs, adverbs combined, 195, 203–4
nouns, 202–3
gratification delay, Conscientiousness and, 139
Gray’s Reinforcement Sensitivity Theory, 328
Great Man theory of leadership, 389, 394
gregariousness, 59, 60, 61, 65, 70
grit, 139
Guilford-Zimmerman Temperament Survey (GZTS), 58, 161, 162t
GWAS. See genome-wide association studies
GZTS. See Guilford-Zimmerman Temperament Survey
habitual responses, of Extraversion, 60

health and healthcare behaviors, 408
Conscientiousness and, 141–42, 288, 405
health outcomes
cognitive function, 407
longevity, 406
mental health and well-being, 406–7
physical (biological or medical) health, 404–5
self-rated health, 405
social competence and productive function, 407–8
health psychology, 4, 403, 556–58
animal studies, 416
application and intervention, 416–17
behavioral data, 414
computational social science, 414–15
health behaviors, 408
historical perspectives, 404
personality as moderator, 410
personality-health research, 410–12
personality neuroscience, 415
physiological pathways, 409
social pathways, 408–9
third or confounding variables, 409–10
heritability
adoption studies and, 311
univariate twin studies and, 303, 304–5
heterogeneous traits, of Openness, 84
heterotypic continuity, for Neuroticism and psychopathology, 44
HEXACO. See Honesty-Humility, Emotionality, Extraversion, Agreeableness,
Conscientiousness, and Openness to Experience
HEXACO Personality Inventory (HEXACO-PI), 138, 456, 463
Hierarchical Personality Inventory for Children (HiPIC), 256–57, 464, 508–9
adolescence and, 265
Altruism facet in, 261
compliance facet in, 261
Dominance, Egocentrism, Irritability facets in, 260
on intellect and imaginative tendencies, 263
principal component analysis of, 273
raters and instruments validation, 267
strength of, 258
weakness of, 258
hierarchical structure, 259, 416
aspects in, 17, 326–27, 326–29, 326t, 331, 334
in Big Five metatraits, 326t
of Conscientiousness, 135–37
higher-order invariants, of FFM, 180–81, 227
HiPIC. See Hierarchical Personality Inventory for Children
history of personality traits
Cattell’s trait sphere, 195–96, 222–23
German alphabetical psychology, 195
Goldberg’s taxonomy of descriptive terms, 196, 222, 223, 365–66
NEO-PI to FFT, 197
Norman’s 2,800 stable traits, 196
past archives, 194
text characters, 194–95
histrionic personality disorder (HPD), 235, 453
Hogan Likeability domain, 111
Hogan Personality Inventory (HPI), 111, 116
Honesty-Humility, Emotionality, Extraversion, Agreeableness, Conscientiousness, and
Openness to Experience (HEXACO), 207, 368
Agreeableness facets and, 116–17
B5/FFM compared to, 116
personality model, 59
Honesty-Humility trait in children, 260
Hopkins Epidemiology of Personality Disorders Study, 114
HPA. See hypothalamic-pituitary-adrenal axis
HPD. See histrionic personality disorder
HPI. See Hogan Personality Inventory
human resource (HR) management, 382, 383
hypothalamic-pituitary-adrenal (HPA) axis, 332
hypothesis testing, in CFA, 219–20
IALSA. See Integrative Analysis of Longitudinal Studies on Aging

IASR-B5. See Interpersonal Adjective Scales, Revised-Big Five
IBCT couple therapy, 424
ICD-10. See International Classification of Diseases
ICID. See Inventory for Childhood Individual Differences
ICID(-S). See Inventory for Childhood Individual Differences, Short Version
identical twins. See monozygotic twins
identity
Extraversion and, 72–73
narrative, 73
self-concept and, 72–73
imagination
Openness and, 83, 84, 85, 88–89, 97
Openness/Intellect and, 336
immune function, animal personality and, 289–90
impulsivity, 453
self-control relationship, 138
UPPS model on, 337
independent pathways (IP) model, 307, 307f, 309, 310
indicator variables, 236n2
individual differences
extrapsychological, 25
taxonomy of, 25, 26f
industrial/organizational (I/O) psychology, 381
industriousness, 136, 138, 326t
infants
personality in, 264–65
temperament models, 264
informants
in Little Five personality dimensions, 245, 246t, 247t, 248t, 249t, 250t
questionnaires approach from, 320
research issues on, 266
initiation of structure, in leadership, 390
instruments validation, 266–67
Integrative Analysis of Longitudinal Studies on Aging (IALSA), 413
integrative data analysis, 413
integrative trait model, of FFM, 1
Intellect
Openness and, 83, 84, 86, 88–89, 90, 92–95
Openness to Experience or, 205
working memory association with, 336
intelligence
conscientiousness and, 141
EPA and, 218
modesty and, 113
16PF measure of, 23
intelligence quotient (IQ), 336–37
interbattery factor analysis technique, 160
intermediate facet measures, 372–73
internalizing spectrum, 234
International Classification of Diseases (ICD-10), 173
International Personality Item Pool (IPIP), 61, 252, 364, 366, 370
AB5C in, 59
on Agreeableness facet, NEO PI-R compared to, 110–11
on conscientiousness, 138
International Society for the Study of Personality Disorders, 528
Interpersonal Adjective Scales, Revised-Big Five (IASR-B5), 23, 111, 229
interpersonal circumplex (IPC), 3, 70, 458, 462
interpersonal perspective of personality, 411
Interpersonal Reactivity Index (IRI), 339, 340
interpersonal relations, Agreeableness and, 105
interpersonal theoretical perspective, 2
interview measures
observer measures, of FFM and, 364–65
SIFFM, 365, 368–69, 464
intraindividual level analysis, of FFM, 179
introversion
characteristics of, 57
health protective factor, 64
Inventory for Childhood Individual Differences (ICID), 256, 257, 508–9
Considerate facet in, 261
Obedience in, 261
Openness domain, 263
strength of, 258
Strong-Willed and Antagonism facets in, 260
weakness of, 258
Inventory for Childhood Individual Differences, Short Version [ICID(-S)], 267
Inventory of Personal Characteristics-7 (IPC-7), 367–68
I/O. See industrial/organizational
IP. See independent pathways model
IPC. See interpersonal circumplex
IPC-7. See Inventory of Personal Characteristics-7
IPIP. See International Personality Item Pool
IPIP-AB5C measure, 366–67
IPIP-NEO scales, 364
IQ. See intelligence quotient
IRI. See Interpersonal Reactivity Index
Irritability, in ICID, 260
item response theory (IRT), 313, 451–52, 455
Jackson Personality Inventory (JPI), 161

jingle-jangle fallacy, 255
job performance, 382–84
change over time, 386
conscientiousness and, 385
job satisfaction, personality relationship with, 387
joint factor analyses, 157–59, 158f
Journal of Abnormal Psychology, 522
Journal of Personality, 114
JPI. See Jackson Personality Inventory
Kaiser-Guttman rule, 218

knowability, 12
languages
of childhood personality and temperament, 274
everyday, 193–94
in international taxonomy of traits, 198–201, 211
trait differences complementary, 273–74
universality of, 174, 184–85
LCP. See Life Course of Personality model
leadership, 59, 389–90, 394
learning
achievement, 512–14
animal personality and, 291–92
styles, 514
lexical foundation, of FFM, 3, 23, 191, 211–12, 353–54, 548
abstract trait words, 193
competition and validation, 206–10
dictionary, 211
everyday language, 193–94
history of, 194–97
introduction to, 192–94
lingual means for personality characterization, 201–6
person-talk, 192, 193, 210
roles and effects, 205–6
trait taxonomies, international, 197–201
lexical tradition measures, 356t, 357t, 358t
adjective-based measures, 111, 196, 229, 365–67
BFI, 367
Big Seven model, 203, 208, 367
HEXACO, 368
IPC-7, 367–68
lexicon, 192–93, 197
Life Course of Personality (LCP) model, 411
life history, FFM links to, 292, 294
lifespan pathways, 403
lingual means grammatical categories, for personality characterization, 201–6
linkage analysis, in molecular genetics, 312
Little Five personality dimensions
activity trait addition, 259–60
assessment methods, 245
for infants and toddlers, 264
informants, 245
strategies, 245
Little Six personality dimensions
Activity addition, 259–60
for infants and toddlers, 264
Locke-Wallace Marital Satisfaction Scale, 425
longevity, 406
longitudinal research, 412
health ongoing and new, 413–14
twin studies, 306–10
lower-order scales, 228, 230
low statistical power, in molecular genetics, 323–24
low statistical power, in neuroimaging techniques, 321
methods to increase, 322
Type I errors, 322–23
Type II error rates from, 322
M5-PS-Inventories, 252
Madingley Questionnaire, 283, 284, 287, 296n2
magnetic resonance imaging (MRI), 320–21
on Agreeableness, 339
on Conscientiousness, 338
on Extraversion, 329, 330–31
major depressive disorder (MDD), 92, 481–86
maladaptive FFM scales, PDs, 464–67, 468t, 469t
maladaptive measures, 358t, 359t, 464–67
on agreeableness, 462–63
FFMRF, 369
FFMSS, 369
of personality, 228–29, 231–32
SIFFM, 365, 368–69, 464
maladaptive traits, 370, 458–64, 459t
marijuana use, 91, 498
marital therapy, 423
marriage, 408
family therapy and, 4
Openness and, 96–98
mate selection, Openness and, 96
MATLAB, 166
Maudsley Personality Questionnaire (MPQ), 58
MBTI. See Myers-Briggs Type Indicator
MCMI-II. See Million Clinical Multiaxial Inventory
MCMI-III. See Millon Clinical Multiaxial Inventory
MDD. See major depressive disorder
mental health
ICD-10 on disorders, 173
mental health disorders
CFA applied to, 234
literature on quantitative modeling of, 233–34
neuroticism comorbidity with, 43, 49
metastructural model, of personality and psychopathology, 233–35
metatraits, in Big Five trait hierarchy
Plasticity, 325
Stability, 325
Midlife in the United States Study (MIDUS), 302, 309, 408, 410, 412
Million Clinical Multiaxial Inventory-II (MCMI-II), 161
Millon Clinical Multiaxial Inventory-III (MCMI-III), 230, 452–53
MIMIC. See multiple indicators, multiple causes
Mini-Modular Markers, 366
Minnesota Multiphasic Personality Inventory (MMPI), 13, 14, 229, 369, 374–75
dimensions factor analysis, 156, 158f, 163
on FFM comprehensiveness, 23
five factors added to, 197
Minnesota Multiphasic Personality Inventory-2 (MMPI-2), on PDs, 453, 455
Minnesota Personality Questionnaire (MPQ), 40, 59
Minnesota Twin Registry, 304
ML7. See Multi-Language 7
MMPI. See Minnesota Multiphasic Personality Inventory
MMPI-2. See Minnesota Multiphasic Personality Inventory-2
moderator variables, 165–67
modesty, 111, 113
molecular genetics
behavior and, 4, 301–15, 321
candidate gene analysis, 311–12
DNA analysis, 321
linkage analysis, 312
low power and inconsistent findings, 323–24
molecular personality scales (MPSs), 313
monozygotic (MZ) twins, 302
moral character traits, 194
morbidity, animal personality and, 290–91
mortality, 404
animal personality and, 290–91
neuroticism and physical health, 46, 406
openness and physical health, 90
motivational process, of Agreeableness, 123–24, 125f
motives, Agreeableness and, 104, 105
MPQ. See Maudsley Personality Questionnaire; Minnesota Personality Questionnaire;
Multidimensional Personality Questionnaire
MPSs. See molecular personality scales
MRI. See magnetic resonance imaging
Multidimensional Personality Questionnaire (MPQ), 255
on constraint, 138
on higher-order personality traits, 309
Minnesota Twin Registry and, 304
on PEM, 329–30
Multi-Language 7 (ML7), 200, 208
multiple indicators, multiple causes (MIMIC) approach, in CB5T, 324
multivariate twin studies
Cholesky decomposition model, 306, 307f, 308, 309
CP model, 307, 308f
development and stability, 309–10
findings from environmental correlates, 307–8
IP model, 307, 307f
personality structure and, 308–9
psychopathology and well-being relationships, 310
twin methodology, 306–7
Myers-Briggs Type Indicator (MBTI), 16, 24
five factors added to, 197
Openness and, 86
MZ. See monozygotic twins
NA. See negative affectivity

narcissistic personality disorder (NPD), 235, 453
narrative identity, 73
NAS. See Normative Aging Study
natural selection, 63
nature and nurture personality contributions, 304
NE. See negative emotionality
near-universals, 174
negative affectivity (NA), 39, 463
MMPI Dependent, 290
in Rothbart model, 270
negative emotionality (NE), 39
childhood temperament and, 42
MPQ facets of, 40
Negative Valance systems, 166, 199–201, 203–4, 207–8, 367–68, 554
NEO-FFI. See NEO Five-Factor Inventory
NEO Five-Factor Inventory (NEO-FFI)
on conscientiousness, 137
ESEM and, 225–26
on language, 184
on openness, 85
research base, 364
on twin studies, 305
NEO Inventories, 11, 32, 252
Agreeableness in, 106
assessment literature most used, 354, 360
on comprehensiveness, 22–23
duality principle, 22–25
facets, 18, 19t, 20t, 21t, 22t
FFM theoretical implications, 25–31
forced symmetry and internal consistency, 364
Form R, 364–65
Form S administered to BLSA, 13, 364–65
metatheoretical premises underlying, 12–14
orthogonal factor scores, 16–17
personality structure, 14–22
proactivity and traits, 12–13
rationality and self-reports, 13–14
NEO Inventory (NEO-I) history, 360
NEO Personality Inventory (NEO-PI), 13, 223, 360
Agreeableness in, 106
Big Five as measured with, 118–19
personality traits history, 197
NEO Personality Inventory, Revised (NEO-PI-R), 16, 17, 22, 24, 176, 223, 360
as adolescence personality measurement, 252, 256–57, 263
on Agreeableness facet, 106, 110–11
on altruism, 110
on compliance, 113, 117–18
as conscientiousness measure, 136, 137
convergent and discriminant validity of, 59
facets of, 326–27
FFM intercorrelations with, 152
lower-order scales, 228
on maladaptivity, 451–52, 464–65
measurement by design, 227
on MZ and DZ twins, 302
network perspective, 60
Openness and, 84, 85, 205–6, 263
replicability, 208
scales, facets in, 225
on twin studies, 305, 309
warmth of, 224–25
NEO Personality Inventory-3 (NEO PI-3), 364
Agreeableness, Conscientiousness, 429, 431t
couples analysis, 425
couples therapy use, 424, 446
cross-observer validity of, 426
Neuroticism, Extraversion, Openness, 428, 431t
PAC, 430
value of, 427
NEO-PI. See NEO Personality Inventory
NEO-PI-3 First Half (NEO-PI-3FH), 364
NEO-PI-R. See NEO Personality Inventory, Revised
network modeling approach
criticism of, 235–36
graphic representations in, 236
personality and psychopathology causal associations, 236
network perspective, of NEO-PI-R, 60
neurobiological theoretical perspective, 2
assays of endogenous psychoactive substances, 321
electrophysiological techniques, 321
molecular genetics, 321
neuroimaging techniques, 320–21
psychopharmacological manipulation, 321
neurobiology, 291, 331, 332
neuroimaging techniques
BOLD signal from fMRI, 320
DTI, 334, 336–37, 340
EEG, 65, 111, 321, 329, 330, 331, 334
fMRI, 65, 320–21, 335–36, 339–40
low statistical power in, 321–23
MRI, 320–21, 330–31, 338, 339
PET, 321, 332
quality of, 323
TBM, 323
variability in methods, 323
VBM, 323
neuroscience, personality, 4, 319, 553–54
agreeableness, 339–40
atheoretical research in, 320
behavioral and cognitive tasks use, 320
conscientiousness, 337–39
Extraversion and, 327–31
future directions in, 340–41
health psychology and, 415
informants and self-reports, 320
methodological issues in, 320–24
neuroticism, 331–35
openness/intellect, 335–37
Type II error rates, 322
neuroticism, 138, 179, 540–42
absolute and differential stability, 41
agreeableness and conscientiousness contrast with, 16
BIS sensitivity, 331
CB5T on, 331–35
childhood antecedents to, 2
conceptualization at younger ages, 41–43
cortisol link, 332
description of, 39–42
developmental considerations, 41–42
difficult childhood temperament and, 42
emotional stability and, 47, 261–62
environmental influences on, 41, 44
facets of, 39–40
genetic influences on, 40–41, 46
hemispheric asymmetry, 333–34
informant disagreement, 42
mental health disorders comorbidity with, 43, 49
NEO PI-3 on, 428, 431t
neuroscience and, 331–35
personality variability, 41
physical health and, 45–48, 405, 406
proxy scale for child, 253
psychopathology, 42–45, 47, 331
psychopharmacological treatment, 48
quality of life and, 48–49
serotonin, noradrenaline and, 331–32
SSRIs reduction of, 331
treatment protocol, 48
universality of, 2
word clouds, 415f
nomological network
Agreeableness and, 110
Openness and, 86
nonshared or unique environment, 304
noradrenaline, Neuroticism and, 331–32
normal personality structure, 229–33
Norman 2,800 stable traits, 196, 199, 223
Norman self-report and other-report scales, 224
Normative Aging Study (NAS), 12, 13–14
NPD. See narcissistic personality disorder
OB. See organizational behavior

Obedience, in ICID, 261
obesity, Openness and, 91
objective biography, Extraversion and, 72
oblique rotation factors, 219, 226
observable traits, personality perception and, 42
observer ratings (R)
interview measures, of FFM and, 364–65
in NEO Inventories, 13
in personnel selection, 385
obsessive-compulsive personality disorder (OCPD), 3, 452–53, 460–61, 490–91
OMNI Personality Inventory-IV (OMNI-IV), 453
openness, 2, 285, 326t
in adulthood, 88, 89–90, 92–93, 262–64, 273
BD, MDD and, 92
biological basis, 86–88
cognition and, 84, 86–88, 92–93
creativity and, 83, 88–89, 90, 94
definition, 84–85
dementia and, 92–93
developmental trajectory, 88–90
disease progression and, 90
DRD4 and COMT genes, 87
DSM on, 84–85, 455–57
employment and, 93–94
extraversion link with, 16
Facebook users and, 95–96
facets of, 84, 85
genetics, 86–87
geographic, 98–99
GWAS and SNPs on, 86–87
heterogeneous traits, 84
inflammation and, 90
marijuana use and diet, 91
measurement considerations, 85
mental health and, 92–93
mortality risk, 90
NEO-PI-R and, 84, 85, 205–6, 263
neuroscience, 87–88
nomological net, 86
person presentation and perception, 94–96
physical health and, 90–91
positive schizotypy, 92
relationships, marriage, family and, 96–98
resilience of, 83
SWB, 92
openness/intellect, 23, 205–6, 543–44
dopamine and, 335
DTI on IQ and, 336–37
fMRI on, 335–36
imagination and, 336
serotonin and, 336–37
openness to experience, throughout adolescence, 265
openness to experience/imagination, 273, 326t
openness to experience/intellect, 205, 262–64, 273
opponent process, to agreeableness, 119–20, 124–26, 125f
orderliness, 136, 137, 138, 140, 326t
organizational behavior (OB), 382, 383
organizational commitment, traits relationship with, 387
Orthogonal-40 Markers, 366
orthogonal rotation factors, 219, 233, 366
Overcontrolled type, 15
PAC. See profile agreement coefficient

PACL. See Personality Adjective Check List
PAFs. See principal axis factor analyses
panic disorder, 487
pathological or anhedonic introversion, in EFA studies, 234
pathoplasty model, 480, 480f
BD, 487
trauma, 492–93
PDNOS. See personality disorder not otherwise specified
PDQ-4. See Personality Diagnostic Questionnaire-4
PDs. See personality disorders
Pearson correlations, 161
pediatric diseases, 516
peer interaction and relationships, 251, 255, 262, 266, 511
PEM. See Positive Emotionality
personality
ABCDs of, 57, 61, 62t, 63, 68–73
adaptive and maladaptive measures, 228–29, 231–32
Allport-Odbert list of, 108
animal personality and development of, 294–95
basic traits in, 272
content, duality principle and, 22–25
cultures and, 28–29
emic dimensions of, 178
GWAS on, 86–87
HEXACO model of, 59
interpersonal perspective of, 411
metastructural model of psychopathology and, 233–35
personal narratives, 272
Personality Adjective Check List (PACL), 229
personality assessment, 4
of conscientiousness, 133–34, 137, 138
in early life, 42
extraversion and, 2
for neuroticism, 48
Personality Diagnostic Questionnaire-4 (PDQ-4), 230
personality disorder not otherwise specified (PDNOS), 523, 530
personality disorders (PDs), 3, 4, 114, 449–52, 470, 528–31, 558–59
antagonism domain, 235
ASPDl, 453–54
avoidant, 234
bipolarity of maladaptive personality structure, 457–60, 459t
borderline, 3, 424, 467, 524, 527–28, 550, 553
CAT-PD, 231, 370–71, 373, 374, 457, 465, 467, 468t, 469t
DPD, 234, 462
DSM-5 psychoticism and Openness, 455–57
DSM-5 Section III, 454–55
DSM-IV-TR, 452–53, 464, 466, 521
DSM on, 84–85, 229–33, 368, 370, 528–31
EFA of scales, 230
FFMPD scales, 466–67, 468t, 469t
HPD, 235, 453
maladaptive agreeableness, 462–63
maladaptive conscientiousness, 460–62
maladaptive FFM scales, 464–67, 468t, 469t
maladaptive traits underlying, 370
MMPI-2, 453, 455
NEO, IASR-B5, MMPI, and PACL on, 229–30
NEO domain-level and facet-level profiles, 368
NPD, 235, 453
OCPD, 460–61
personality traits of, 173
scales, 452–54
social phobia, 234
STPD, 3, 92, 232, 336–37, 456
traits and, 16
personality inventories
dimensions in, 151, 157–59
FFM intercorrelations with, 152
scales in, 151, 152, 153t, 154t, 155t, 156, 164
Personality Inventory for the DSM-5 (PID-5), 231, 374, 463, 464–65, 467, 468t, 469t
personality perceptions, observable traits and, 42
personality psychology
from animal research, 292
data, nature of dimensions in, 156–57
lexicon for, 192–93, 197
pathology decline with age, 181
WASP and WEIRD research, 175
personality questionnaires, evaluation in, 233
Personality Research Form (PRF), 24
personality states, 68
personality structure, 2, 4, 175–78
biological basis for, 28, 32n4
bipolarity of maladaptive, 457–60, 459t
California Q-Set on, 15–16, 113, 117–18, 450, 524
CFA restrictive assumptions, 224
circumplex, 22, 210
hierarchy shape, 22
historical foundations, 222–24
MBTI on, 16
model fit in modern era, 224–27
multivariate twin studies and, 308–9
normal and abnormal convergence, 229–33
parents and teachers as informants, 266
person-centered approach to, 15–16
primary domains mapping, 231–32
research on, 226, 265–66
robustness, EFA and, 227
structural validity and, 14
trait hierarchy levels, 16–22
traits and types, 15–16
traits hierarchy, 227–29
variable-centered approach, 15–16
personality tests, 151, 157, 164–65
personality theory, 14
personality traits
attitudes, 204–5
broad trait domain, 135
conscientiousness as, 134–35, 143
contextualized and/or decontextualized nature of, 135
genes and environment influence on, 301
hierarchical structure of, 227–29
hierarchy, 325f
of PDs, 173
physical appearance, 204–5
research, 2, 178
Sociogenomic model of, 134–35
personality variability, 41
personal narratives, 272
person-centered approach, to personality, 15–16
personnel selection, 382–86
person presentation and perception, Openness, 94–96
person-talk, 192, 193, 210
PET. See positron emission tomography
PFC. See prefrontal cortex
phenotypes
C influence on, 303–4, 306
variation, 302–4, 303f
phylogeny, 292, 294, 295
physical appearance, 204
physical health
conscientiousness and, 141–42, 288, 405
neuroticism and, 45–47, 49
physical (biological or medical) health outcomes, 404–5
physiological pathways, health and, 409
PID-5. See Personality Inventory for the DSM-5
Plasticity
CB5T on, 325
dopamine influence on, 325
polarity, of traits and scales, 371–72
politeness, 326t, 339
Positive Emotionality (PEM), 60, 328, 329–30
positive schizotypy, Openness and, 92
Positive Valance, 17, 199–201, 203, 368, 458, 511, 554
positron emission tomography (PET), 321, 332
posttraumatic stress disorder (PTSD), 491–93
PPI. See Psychopathic Personality Inventory
PPI-R. See Psychopathic Personality Inventory, Revised
predictive utility, of neuroticism, 43, 45–46
predictive validity, 383, 384, 385, 388–89, 396
prefrontal cortex (PFC)
conscientiousness and, 337–38
intellect and, 336
neuroticism and, 333
PRF. See Personality Research Form
principal axis factor analyses (PAFs), 160
principle components analysis, 237n6
proactivity
cross-observer agreement, 12–13
stability in self descriptions, 12
profile agreement coefficient (PAC), 430
Project Competence longitudinal study, 253
PSY-5 measures, 369, 455
psychodynamic theoretical perspective, 2
psycholexical approach, to B5/FFM, 191, 210
cross-cultural findings, 208–9
departing models, 206–7
dictionary use, 192
German study, 197, 198, 199
Psychological Assessment Resources, 375
psychological theory, NEO Inventories as instruments of, 11–32
Psychopathic Personality Inventory (PPI), 285
Psychopathic Personality Inventory, Revised (PPI-R), 454
psychopathology, 236. See also Axis I psychopathology
of adults, 2, 4, 28
agreeableness facet and, 115
childhood and adolescence consequential outcomes, 514–16
metastructure model of personality and, 233–35
neuroticism and, 42–45, 47, 331
studies on structure of, 234
trust and, 114
well-being relationships, in twins, 310
psychopharmacological treatment, 48, 321
PsycINFO, 112–13, 112t, 354
PTSD. See posttraumatic stress disorder
punctuality, 136
quality of life, 47
questionnaire tradition, 191, 353–54, 355t, 356t, 360, 364–65
adult questionnaires, 251–52
in childhood personality research, 267–68
through informant reports, 320
R. See observer
rank-order stability. See differential stability
raters, validation of, 266–67
rationality, self-reports interpretation, 13–14
RDoC. See Research Domain Criteria
Reinforcement Sensitivity Theory (RST), BAS, BIS, and FFS of, 65, 120, 328
relational functioning, in Neuroticism and psychopathology, 44–45
relationships
animal personality and social, 291–92
conscientiousness and romantic, 141, 142
impulsivity and self-control, 138
of job satisfaction with personality, 387
organizational commitment with traits, 387
peer interaction and, 251, 255, 262, 266, 511
twins well-being, 310
research, 3
adult personality and traits, 2, 28
animal personality contribution, 285–95, 294f
childhood personality challenges, 264–68
clinical utility, 528–31
on culture, 28–29
EEG, on FRN, 329
established facts, 26–28
factor analytic support challenges, 227–36
free description, 255–57, 261, 262, 263
neuroscience atheoretical, 320
new tests of, 29–31
on OB, 383
on personality biological basis, 28
personality-health, 410–12
on personality structure, 226, 265–66
personality trait, 2, 178
robustness and, 151, 164–68
theoretical implications of, 25–31
on X-chromosome, 28
Research Domain Criteria (RDoC), 43
resilience, 15, 83
responsibility, 136
reward process, Extraversion and, 65–66
rhesus macaques monkey studies, 283–84, 287, 291
RMSEA. See root mean square error of approximation
robustness, 168, 227, 546–47
comprehensiveness compared to, 151
data sets and analytic methods, 152, 159–61
dimensional level research, 167–68
at dimensional structure level, 151–52, 157–59
on personality inventories, 151
personality psychology data dimensions, 156–57
on personality research literatures, 151
at scale level, 151–52, 164–67
of thought disorder/psychosis spectrum, 234
variance in measures, 152, 155
root mean square error of approximation (RMSEA), 218
rotation factors, in EPA, 218–19, 225
Rothbart model of childhood temperament, 269–70, 271
RST. See Reinforcement Sensitivity Theory
SAD. See social anxiety disorder

SAGE. See Study of Addiction: Genetics and Environment
scales
Balanced Emotional Empathy Scale, 339
Barratt Impulsivity Scale, 338
BFAS, 59, 228, 327, 373, 456
CCS, 137–38
ENRICH/PREPARE, 425
facets and, 164, 225
FFMPD, 466–67, 468t, 469t
IPIP-NEO, 364
Locke-Wallace Marital Satisfaction Scale, 425
lower-order, 228, 230
maladaptive FFM, for PDs, 464–67, 468t, 469t
measurement invariance of, 226
MPSs, 313
Norman self-report and other-report, 224
in personality inventories, 151, 152, 153t, 154t, 155t, 156, 164
proxy for child, Neuroticism and, 253
robustness at level of, 151–52, 164–67
trait polarity and, 371–72
Schedule for Non-Adaptive and Adaptive Personality (SNAP), 230–31, 233, 451, 452–
53, 455, 457, 463
schizotypal personality disorder (STPD), 3, 92, 232, 336–37, 456
School Behavior Checklist, 255
SDS. See Self-Directed Search
selective serotonin reuptake inhibitors (SSRIs), 331
Self-Concept, 72–73, 122
self-control, 136, 138, 139
self descriptions, 12
defense and blind spots in, 13
Self-Directed Search (SDS), 163
self-healing, 405
self-regulation, 139
self-reported health, 404, 405
Self-Report Psychopathy-Version III (SRP-III), 454
self-reports, 11
in adolescence personality measurements, 251, 255, 262, 266
for adults, 266
on Agreeableness, 118
children puppet interviews for, 266
in COA case study, 434–36, 435f
COA comparison and, 427–30, 431, 431t
in couples assessment, 425–26
in NEO Inventories, 13–14
in personality neuroscience, 320
rationality interpretation of, 13–14
SEM. See structural equation modeling
serotonin
agreeableness and, 340
brain function, 325–26
neuroticism and, 331–32
openness/intellect and, 336–37
stability influenced by, 325
SES. See socioeconomic status
Shedler-Westen Assessment Procedure-200 (SWAP-200), 450–51
SIFFM. See Structured Interview for the Five-Factor Model of Personality
simian immunodeficiency virus (SIV), 289–90
single nucleotide polymorphisms (SNPs), 86–87, 313
situations division, of traits, 63
SIV. See simian immunodeficiency virus
Sixteen Personality Factor Questionnaire (16PF), 18, 19t, 20t, 23, 58, 196–97
Slavic languages trait taxonomy, 198–99
small samples, extreme groups use for, 322
SNAP. See Schedule for Non-Adaptive and Adaptive Personality
SNPs. See single nucleotide polymorphisms
sociability, 57–61, 63, 67, 70, 73, 254
social accommodation, 121–22
social activity hypothesis, 69
social anxiety disorder (SAD), 487–90
social behavior, 105, 292
social competence, 407–8
social effects, 205–6
Social-Investment Theory, 181, 294
social pathways, health and, 408–9
social phobia, 234
social relationships, animal personality and, 291–92
social supports, Extraversion and, 72
societies, Extraversion development and, 67–68
Socioanalytic Theory, 59
socioeconomic status (SES), 303–4, 405
Sociogonomic model, of personality traits, 134–35, 139, 140
specific responses, of Extraversion, 60
specific variance, 156, 159–60
spectrum model, 43–44, 480, 480f, 486, 490, 491, 495–96
SRP-III. See Self-Report Psychopathy-Version III
SSRIs. See selective serotonin reuptake inhibitors
stability, 228
CB5T on, 325
differential or rank-order, 41, 67
emotional, 47, 261–62, 265
in multivariate twin studies, 309–10
serotonin influence on, 325
state-level, of Sociogenomic model, 134–35
statistical significance tests, 160–61
STPD. See schizotypal personality disorder
straightforwardness, as Agreeableness facet, 111, 113–14, 115
strategies, in Little Five personality dimensions, 245
stress response, 409
neuroticism and physical health, 46
Strong Vocational Interest Blank (SVIB), 14
Strong-Willed, in ICID, 260
Stroop Test, 121
structural equation modeling (SEM), 411, 412
structural validity, 14
structure analyses, of FFM, 178–79
Structured Interview for the Five-Factor Model of Personality (SIFFM), 365, 464
Study of Addiction: Genetics and Environment (SAGE), 313
subjective well-being (SWB), 47, 92, 291, 406–7
surface similarity, Agreeableness facets and, 118
surgency, 58–59
SVIB. See Strong Vocational Interest Blank
SWAP-200. See Shedler-Westen Assessment Procedure-200
SWB. See subjective well-being
targeted rotations, 160, 226

taxonomy
of childhood and adolescence temperament and personality, 275t
of childhood traits, 268
Goldberg’s, of descriptive terms, 196, 222, 223, 365–66
of individual differences, 25, 26f
taxonomy, adult and child differences, 244, 258–59, 267–68
agreeableness/Benevolence, 260–61, 265, 273
conscientiousness, 262
extraversion/sociability, 260
neuroticism/emotional stability, 47, 261–62
openness to experience/intellect, 205, 262–64, 273
taxonomy of traits, international
B5/FFM in Germanic and Slavic languages, 198–99
B5/FFM in Indo-European languages, 199, 211
B5/FFM in non-Indo-European languages, 200–201, 211
Dutch method, 197, 198
German psycholexical study, 197, 198, 199
selection procedures, 197–98
TBM. See tensor-based morphometry
TCI. See Temperament and Character Inventory
teacher. See caregiver or teacher
teams, 391–94, 393
Temperament and Character Inventory (TCI), 309, 455
tender-mindedness, 109, 111, 114, 115
Ten-Item Personality Inventory (TIPI), 138, 367
tensor-based morphometry (TBM), 323
Terman Life Cycle Study, 410, 412
theoretical explanations
for Agreeableness, 119–23
for animal personality, 288
theories of psychological function, CB5T and, 324–40, 326t
theory-driven construction methods, 251
of Rothbart model, on childhood temperament, 269–70, 271
theory of purposeful work behavior (TPWB), 387–88
third or confounding variables, 409–10
Thomas and Chess model, of childhood temperament, 268–69, 271
thought disorder/psychosis spectrum, 234
TIPI. See Ten-Item Personality Inventory
top-down research approach, 259, 354
adjective lists, for children, 251, 266
adult questionnaires, 251–52
advantage of, 257
CCQ and Common Language adaptation, 253–54
childhood personality measurements, 251–52
conscientiousness as childhood personality factor, 262
from existing assessment measures, 252–55, 257–58
FFM model for childhood personality, 253
FFM modification, 25
generic description lists of development, 252
on infants, 264–65
on openness or intellect, 262–63
problems with, 255
TPQ. See Tridimensional Personality Questionnaire
TPWB. See theory of purposeful work behavior
traditionality, 136
trait hierarchy, 16, 210, 211
aspects of, 17
facets, 17–18, 19t, 20t, 21t, 22
GFP in, 17
NEO-PI-R on, 17
traits, 15, 68
adult personality and, 28
Agreeableness word descriptions, 111–12
as Basic Tendencies, 24–25, 28
challenges of unique, 184
CPI, PRF, MBTI and, 24
cross-observer agreement, 12–13
default network for, 321
descriptive words, 195
duality principle and, 24–25
of entrepreneurs, 394–95
explanations division of, 63
factor analytic work on, 274
international taxonomy of, 197–201
moral character, 194
nature of, 23–24
Openness and, 86
past archives, 194
PDs and, 16
Principal Components of, 210
proactivity and, 12–13
scale polarity and, 371–72
self-reports of, 11
situations division of, 63
taxonomy of childhood, 268
variance, 157
Transparent Trait Rating Form (TTRF), 111
trauma, 491–93
treatment
clinical utility and, 526–28
FFM impact on decisions for, 526–27
for neuroticism, 48
utilizing FFM, 525–26
Tridimensional Personality Questionnaire (TPQ), 311–12, 452
trust
as Agreeableness facet, 111, 114–15
in animal personality, 289
Collaborative Longitudinal Personality Disorders Study and, 114
Hopkins Epidemiology of Personality Disorders Study, 114
in workplace, 114
TTRF. See Transparent Trait Rating Form
Tucker-Burt-Wrigley-Neuhaus congruence coefficient, 160
twins
biometric modeling of data of, 303
fraternal or DZ, 302
identical or MZ, 302
methodology of, 302–4, 306–7
MIDUS sample of adult, 302
multivariate studies, 306–10, 307f
NEO-FFI on, 305
NEO-PI-R on, 302, 305, 309
raised apart, 304
univariate studies, 302–6, 303f, 311
Type I errors, from low statistical power, 322–23
Type II error rates, in personality neuroscience, 322
unconventionality, 83, 85, 200, 230, 375, 456

Undercontrolled type, 15
unipolarity, bipolarity compared to, 232
univariate twin studies
assortative mating and, 304
EEA reliance, 304
findings from, 304–6
heritability and, 303, 304–5
limitations and assumptions of, 304
nature and nurture personality contributions, 304
nonadditive influences, 305, 311
nonshared environment, 304
phenotype variation, 302–4, 303f
shared compared to nonshared environment difficulties, 304
twin methodology, 302–4
on twins raised apart, 304
universality, 547
acculturation, 184
challenges to, 184–85
cohort effects challenges, 185
cultural relativism on, 174
culture exceptions, 176
of FFM, 3, 294
of languages, 174, 184–85
of neuroticism, 2
unique traits, 184
universals, personality
background of, 183–84
cultures and, 177–78
development and, 176
FFM alternatives, 178
FFT and, 183–84
Internet study on, 177
large-scale cross-cultural studies, 176–78
mean levels and, 181–83
structure of, 175–78
UPPS model, on impulsivity, 337
validation
across cultures, 267
in HiPIC and ICID(-S) personality measures, 267
lexical foundation and, 206–10
of raters and instruments, 266–67
VA Normative Aging Study, 412
variability, 12
informant disagreement, 42
personality, 41
variable-centered approach, to personality, 15–16
variables
confounding, 409–10
EPA selection of, 219
indicator, 236n2
moderator, 165–67
variance
common, 156, 157, 159–60
error, 156, 159–60
robustness and, 152, 155
specific, 156, 159–60
trait, 157
VBM. See voxel-based morphometry
ventral attention or salience network, 338
ventromedial prefrontal cortex (VMPFC), Extraversion and, 329–30
volatility, 326t, 331, 334
voxel-based morphometry (VBM), 323
vulnerability model, 480, 480f
for neuroticism and psychopathology, 43–44
OCD, 491
schizophrenia and psychotic disorders, 494, 496
substance use disorders, 497
trauma, 492
warmth, 106, 224–25, 289, 301

WASP. See Western Academic Scientific Psychology
Webster’s New International Dictionary, 365
Webster’s Unabridged Dictionary, 195
WEIRD. See Western, Educated, Industrialized, Rich, and Democratic
well-being, 291, 310. See also subjective well-being
Western, Educated, Industrialized, Rich, and Democratic (WEIRD) people, 175, 294
Western Academic Scientific Psychology (WASP), 175
Wisconsin Card Sorting Task, 121
withdrawal, 326t, 331, 334
Wonderlic Personnel Test, 113
word clouds, Neuroticism and, 415f
workplace, 114, 142
work-related motivation, 387
X-chromosome, 28
Yale Laboratories of Primate Biology, 281–82

Oxford Handbook of The Five Factors Model

Uploaded by

Copyright:

Available Formats

Oxford Handbook of The Five Factors Model

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Oxford Handbook of The Five Factors Model

Uploaded by

Copyright:

Available Formats

The Oxford Handbook of the Five Factor Model

OXFORD LIBRARY OF PSYCHOLOGY

Methods and Measurement

Personality and Social Psychology

Published in the United States of America by Oxford University Press

© Oxford University Press 2017

CIP data is on file at the Library of Congress

Oxford Library of Psychology

About the Editor

The Oxford Library of Psychology, a landmark series of handbooks, is published

Section 1 • The Five Factor Model

Section 2 • Construct Validity

The Five Factor Model

Paul T. Costa Jr. and Robert R. McCrae

Reading Loevinger’s 1957 monograph, “Objective tests as instruments of

Metatheoretical Premises Underlying the NEO Inventories

Proactivity: The Reality of Traits

Rationality: The Interpretation of Self-Reports

The Structure of Personality

Traits and Types

Table 2.1. Correspondence of Facet-Level Scales

The Shape of the Hierarchy

The Content of Personality and the Duality Principle

The Nature of Traits

The Duality Principle

A TAXONOMY OF INDIVIDUAL DIFFERENCES

Theoretical Implications of FFM Research

New Tests of FFT

Jennifer L. Tackett and Benjamin B. Lahey

The purpose of this chapter is to provide a broad and comprehensive

Genetic Influences on Neuroticism

Environmental Influences on Neuroticism

Neuroticism and Physical Health

Neuroticism and Quality of Life

Conclusions and Future Directions

Joshua Wilt and William Revelle

Personality trait dimensions are abstractions used to describe and explain

Extraversion as a Basic Tendency

Defining Extraversion’s Lower Order Structure

The ABCD Approach as an Organizing Framework

Table 4.1. Extraversion Items and ABCD Content

Evolution, Genetics, and Biology

Brain Systems Underlying Variation in Extraversion

Eysenck and Gray

Reward Processing and Dopaminergic Functioning

Are Societies as a Whole Becoming More Extraverted?

Extraversion Characteristic Adaptations

Self-Concept and Identity

Openness: From Biology to Society

PHYSICAL, MENTAL, AND COGNITIVE HEALTH

Person Presentation and Perception

Relationships, Marriage, and Family

William G. Graziano and Renée M. Tobin

Agreeableness: Emergence of an Empirical Concept

Strategies for Construct Specification and Elaboration

Table 6.1. Results of PsycINFO Searches for Agreeableness Facet Keywords

Such nuances notwithstanding, empirical research on personality would be

Theoretical Explanations for Agreeableness

Agreeableness and Empathy