Full download ebook at ebookgate.

com

Foundations of Cognitive Psychology 1st

Edition Fernand Gobet

https://ebookgate.com/product/foundations-of-
cognitive-psychology-1st-edition-fernand-gobet/

Download more ebook from https://ebookgate.com

More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Joint cognitive systems foundations of cognitive

systems engineering 1st Edition Erik Hollnagel

https://ebookgate.com/product/joint-cognitive-systems-
foundations-of-cognitive-systems-engineering-1st-edition-erik-
hollnagel/

Cognitive Foundations of Linguistic Usage Patterns

Empirical Studies Applications of Cognitive Linguistics
1st Edition Schmid

https://ebookgate.com/product/cognitive-foundations-of-
linguistic-usage-patterns-empirical-studies-applications-of-
cognitive-linguistics-1st-edition-schmid/

Cognitive Psychology First Edition Nick Braisby

https://ebookgate.com/product/cognitive-psychology-first-edition-
nick-braisby/

Foundations Of Physiological Psychology 6th Edition

Neil R. Carlson

https://ebookgate.com/product/foundations-of-physiological-
psychology-6th-edition-neil-r-carlson/
Foundations in Evolutionary Cognitive Neuroscience 1st
Edition Steven M. Platek

https://ebookgate.com/product/foundations-in-evolutionary-
cognitive-neuroscience-1st-edition-steven-m-platek/

A Cognitive Psychology of Mass Communication 6th

Edition Richard Jackson Harris

https://ebookgate.com/product/a-cognitive-psychology-of-mass-
communication-6th-edition-richard-jackson-harris/

Social Psychological Foundations of Clinical Psychology

1st Edition James E. Maddux Phd

https://ebookgate.com/product/social-psychological-foundations-
of-clinical-psychology-1st-edition-james-e-maddux-phd/

A Cognitive Psychology of Mass Communication 4th ed

Edition Richard Jackson Harris

https://ebookgate.com/product/a-cognitive-psychology-of-mass-
communication-4th-ed-edition-richard-jackson-harris/

The Tip of the Tongue State Essays in Cognitive

Psychology 1st Edition Alan S. Brown

https://ebookgate.com/product/the-tip-of-the-tongue-state-essays-
in-cognitive-psychology-1st-edition-alan-s-brown/
 Foundations
of Cognitive
Psychology

Fernand Gobet, Philippe Chassy

and Merim Bilalii

ffi
London Boston Burr Ridge, lL Dubuque, lA Madison, Wl New York San Francisco
5t. Louis Bangkok Bogotii Caracas Kuala Lumpur Lisbon Madrid Mexico City
Milan Montreal New Delhi Santiago Seoul Singapore Sydney Taipei Toronto
 Fountlations of Cognitive Ps.vchokrg-
Fcrnant'l Gobct, Philippe Chassy ancl l\{elini Bilalic
ISBN- l3 9-7{r-00-77 l l e08-e
ISBN-10 0-07-71 1908-8

Published by tr'lcGra*I{ill Educatiorr

Shoppenhangcrs Ro;rd
,\laidenheacl
Berkshire
sL6 2QL
TelePhone:44 (0) l618 502 500
Frrx: -l{ (0) 1628 77O 2.21
lVebsite: www.mheducation.co.uk

British Library Cataloguing in Publication Data

A cataluguc rccortl lor this book is available fi.onr the Bntrsh Libraw

Library of Congress Cataloguing in publication Data

The Librarv oiCongress data tbr this book has been appliecl tbr tl-om the Librarv of Congress

Acquisitions Editor: Natalie Jacobs

Developntent Editor: Jennifer Rotherhanr
N{arketing Nlanager: Kevin !\ratt
Ploduction Eclitor: Alistrn I)avi-;

Text design b1'Harcl Lines

Cover design by r\clam Renvoize
Printed and bound in Great Britain b5, Ashford Colour press Ltd.

Published by. NIcGrarv-FIill Edtrcarion (LrK) Limited an inrprinr o['Ihc i\,tcGrarv-Hill

companies, I'c., l2r.l Avenr-re of the Americas, Ne*. y.rk, Ny 10020. Copyright o 20 r l by
IlcGraw-Hill Education (UK) Limited. AJI rights reserved. No parr ol thii puf,licatiun .u1; b"
reproduced or distributed in any tbrm or bv any means! or stored in a database or retrieval
systenr, rvithout the prior rvritten consent of -fhe lllcGraw-Hill compauies,
Inc., inclucling, but
not linrited to, in any ne$vork or other electronic storage or lransniission, or broadcast lor
distance learning.

Fictitious narmes ofcontpanies, Products, peoplc, characters and/or rlata that may be usetl herein
(in case studies or in examples) are not intended to l'eprese
nt any rcal indiviLlual, compari)-,
product or event.

ISBN- I 3 97 8-OO-77 I t908-9

ISBN- I0 0-07-7 I 1908-8
o 201 t. Exclusive rights by The lVlcGlarv-Hil[ Companic's, Inc. for m4sulacture and export. This
book cann,t be from the country to ivhich it is s.ltl by lllcGrarv-Hill.
'e-exportetl
t

Ded ication
To Cliananda, FG
To r\nne, PC
To m)'parents Zir.a.linka and Niidhat, NIB

t--
 Preface xi
Guided tour xiii
Technology to enhance learning and teaching xiv
Acknowledgements xvii

PART 1
lntroduction 1

1 lntroduction to cognitive psychology 3

2 A brief history of cognitive psychology 13
3 A brief introduction to neuroscience 27

PART 2
Basic processes 45
4 Perception 47
5 Attention 71
5 Short-term memory 91
7 Long-term memory and learning 115

PART 3
Complex cognition 143
8 Concept formation and categorisation 145
9 Language 162
10 Reasoning and decision making 190
11 Problem solving 208
12 Expertise 22s
13 Creativity 241

PART 4
Advanced topics 253
14 Emotion and cognition 255
15 Computational modelling 267
15 The cognitive neuroscience of visual perception 279
17 Consciousness 293
18 Social cognition 301

PART 5
Putting it all together 315
19 Putting it all together 317

References 323
Glossary 342
lndex 355

vl
 Preface XI 2.2 lN FOCUS: The ditrerent meanings
Guided tour xiii ofthe term'theory' 17
Technology to enhance learning 2.3 CLASSIC EXPERIMENT Priming
and teaching xiv experiment 19
Custom publishing solutions xv 2.4 RESEARCH CLOSE.UP:
Make the grade! xvi Connectionist networks 22
Acknowledgements xvii 2.5 lN FOCUS: What is a good theory? 25
About the authors xviii
Abbreviations and acronyms xix 3 A brief introduction to
neuroscience 27
PART 1 lntroduction 1 lntroduction 27
I lntroduction to cognitive The neuron 28
psychology 3 Communication within and between
lntroduction 4 neurons 30
The experimental methods of Brain anatomy 32
cognitive psychology 4 Conclusions: from biology to
Developing theories that explain the psychology 41
data 7 Chapter summary 43
The subfields and applications of Further reading 43
cognitive psychology 10 Pedagogy boxes:
How to use this book 10 3.1 lN FOCUS: Reticularists vs
Chapter summary 11 neuronists 31
Further reading 12 3.2 RESEARCH CLOSE-UP: Brain.
Pedagogy boxes: imaging techniques 33
1.7 lN FOCUS: Cognitive psychology
and cognitive science 4 PART2 Basic processes 45
1.2 CUSSIC EXPERIMENT Sternberg's 4 Perception 47
experiment 5 lntroduction 47
1.3 lN FOCUS: Why do we need What is perception? 48
theories? 7 The psychophysics approach 53
Two-dimensional visual i nformation
2 A brief history of cognitive psychology 73 (Gestalt theory) 54
lntroduction 14 The third dimension: depth 57
The cognitive revolution 15 When the visual system is deceived:
Cognitive psychology comes of age 18 visual illusions 58
Connectionism strikes 21 Perception is cognition 52
The ascent ofcognitive neuroscience 23 Complex objects: faces 55
The search for unified theories of Chapter summary 69
cognition 24 Further reading 70
Chapter summary 26 Pedagogy boxes:
Further reading 25 4.1 RESEARCH CLOSE-UP: Oscillatory
Pedagogr boxes: processes 50
2.1 lN FOCUS: The influence of 4.2 RESEARCH CLOSE-UP: Stevens's law 55
engineering on cognitive 4.3 RESEARCH CLOSE.UP: ThE AdVANCC
psychology 16 ofscience 57

YU

Ii
L
Viii DETAILED TABLE OF CONTENTS

4.4 lN FOCUS: Watching the ghost 59 Further reading 142

4.5 CLASSIC EXPERTMENT Hayworth Pedagogy boxes:
and Biederman's experi ment 57 7.1 RESEARCH CLOSE-|JP: Shopping
mall 125
5 Attention 71 7,2 RESEARCH CLOSE-UP: Ethicql
lntroduction 72 issues in psychology 125
Models of attention 74 7.3 CLASSIC EXPERIMENT
Automaticity 83 Remembering what never
Chapter summary 88 occurred 126
Further reading 89 7. 4 EV E RYDAY APP LI CATI O N :
Pedagogy boxes: Alzheimer's disease 141
5.7 RESEARCH CLOSE-|p: Attention
in perception 72 PART 3 Complex cognition 143
5. 2 EV E RYDAY APPL I CAT\ ON : Atte nti o n 8 Concept formation and
in magic 75 categorisation 145
5.3 CTASS/C EXPERTMENT: The Stroop lntroduction 145
effea 84 Classical theory of concepts 148
Prototype theories 152
6 Short-term memory 97 Exemplar-based theories 155
lntroduction 92 Explanation-based theo ries 1s8
Memory types 93 Role of strategies, attention and
The atoms of cognition: the chunking short-term memory 159
hypothesis 95 Chapter summary 160
Atkinson and Shiffrin's Further reading 161
multi-component model 97 Pedagogy boxes:
Sensory registers: stealth memory 8.1 lN FOCUS: Funesthe
stores 99 memorious 146
Short-term memory: verbal 8.2 RESEARCH CLOSE.IJP: A smal\
information 101 i ntrodu ctory experi ment 147
Short-term memory: visuospatial 8.3 RESEARCH CLOSE-|P: Exampte
information 103 of strategy use: conservative
Memory traces 10s focusing 1s0
Recent models of working memory 109 8. 4 H C LOSE-IJ p : prototy pes
R ESEARC
Chapter summary 114 in action: an example 153
Further reading 114 8.5 RESEARCH CLOSE-Up: Exemptars
Pedagogy boxes: in action: an example 156
6.1 RESEARCH CLOSE-|Jp: The free 8.6 RESEARCH CLOSE-IJP:
recall paradigm 93 Explanation-based theory i n action :
5.2 CLASSIC EXPERIMENT Chinese an example 158
memory 102
6.3 CLASSIC EXPERTMENT Finding 9 Language 152
your way on an island 104 lntroduction 153
Some definitions 153
7 Long-term memory and learning 11s The importance of development 155
lntroduction 116 Speech production 158
Long-term memory systems 117 Speech perception 158
Declarative memories 119 Theories of speech recognition 171
Non-declarative memories 131 Visual word recognition 174
Skill acquisition 131 Syntactic processing 177
Acquiring and changing habits 135 From printed word to sounds 180
Pathological forgetti ng 139 Neuropsychological synd romes of
Chapter summary 142 reading 182
 DETAILED TABLE OF CONTENTS ix

Theoretical explanations of acquired 12 Expertise 225

dyslexia 182 lntroduction 225
Text comprehension 185 Cognitive expertise 227
Chapter summary 188 Acquisition of (cognitive) skills 231
Further reading 189 Perceptual-motor expertise 234
Pedagogy boxes: Deliberate practice 235
9.1 EVERYDAY APPLICATION: FTom Cognitive and perceptual-motor skill -
Grice to Australia 164 two worlds apart? 238
9.2 lN FOCUS: Key questions in the Chapter summary 239
study of language 168 Further reading 240
9.3 CLASSIC EXPERIMENT The Pedagogy boxes:
McGurk effect 171 12.1 CLASSIC EXPERIMENT Chi's
9.4 RESEARCH CLOSE.UP: (7978) experiment on the recall of
Eye-movement studies of reading 175 chess positions and digits in adults
9.5 RESEARCH CLOSE-UP: and children 229
' Phrase-structure grammar 178 12.2 RESEARCH CLOSE-UP: Why do
good thoughts block better
10 Reasoning and decision making 190 ones? 233
lntroduction 191 1 2.3 RESEARCH CLOSE-UP: Ecological
The multiple forms of thinking 192 validity and expertise - are
Scientific reasoning 193 laboratory findings applicable to
Theories of reasoning 195 the real world? 23s
Decision making 200
(lr)Rationality? 204 13 Creativity 241
Chapter summary 206 lntroduction 241
Further reading 207 The creative product 243
Pedagogy boxes: The creative process 244
10.1 RESEARCH CLOSE.UP: ThC bASiS The creative person 250
I
of scientific thinking 193 Chapter summary 2s2
10.2 RESEARCH CLOSE-UP: Wason,s Further reading 252
245task 195 Pedagogy boxes:
10.3 CLASSIC EXPERIMENT Lord, 13.1 RESEARCH CLOSE-UP: ln search
Ross and Lepper's (1979) of the creative potential 245
experiment on the death penalty 195 1 3,2 RESEARCH CLOSE-UP:
1 0.4 EVERYDAY APPLICATION: P ro b I e m -space ex p I a n ati o n of
Paranormal beliefs 203 incubation 245
13,3 RESEARCH CLOSE-UP: NaturAI
11 Problem solving 208 and artifi ci al creativity 248
lntroduction 208
The Gestalt approach 210 PART 4 Advanced topics 253
What is restructuring? The theory 14 Emotion and cognition 255
behind the insight phenomenon 215 lntroduction 255
The i nformation-processing approach 217 What emotions are and what they are
Making analogies 220 not 255
Chapter summary 223 What are emotions for? 255
Further reading 224 The structure of emotions 257
Pedagogy boxes: Basic emotions and ontogenetic
11.1 RESEARCH CLOSE-UP: The emotions 259
feathered ape in your garden 211 The interaction between emotion and
11.2 CLASSIC EXPERIMENT Luchins' cognition 259
(1942) Einstellung (mental set) Positive emotions 250
effect 213 Theories of emotions, or better said:

L
DETAILED TABLE OF CONTENTS

attempts to theorise emotionsl 263 15.2 lN FOCUS: Fqilures of the visual

Chapter summary 265 system 288
Further reading 265 15,3 CLASSIC EXPERIMENT Motion
Pedagogy boxes: coherence 289
14.1 CLASSIC EXPERIMENT Bechara -
emotions and decision making 251 17 Consciousness 293
1 4.2 EVERYDAY APPLICATION: lntroduction 293
Therapy - memory structures, H istorical perspective 294
emotions and cognitive therapy 261 Modern materialism and theories of
14.3 lN FOCUS: Music - perceptual consciousness 295
structures, emotions and music 252 An illusion? 297
14.4|N FOCUS: Love 253 Conclusion 299
Chapter summary 300
15 Computational modelling 267 Further reading 300
lntroduction 267 Pedagogy boxes:
A taxonomy of formal modelling 258 17.1 IN FOCUS: Neuropsychology and
Strengths of computational modelling 220 consciousness 295
lssues 274 17.2 CLASSIC EXPERTMENT Libet's
Scope of models 277 clock and the question of free will 299
Chapter summary 277
Further reading 278 18 Social cognition 301
Pedagogy boxes: lntroduction 302
15.1 lN FOCUS: Cognitive modelling vs What is social cognition? 302
-
artificial intelligence the case of Social perception 303
Deeper BIue 269 Social memory 305
7 5.2 EVERYDAY EXAMPLE: Mobile
Social inferences 306
phones and driving 271 Chapter summary 312
15.3 RESEARCH CLOSE-UP: Can Further reading 313
models predid new phenomena? 272 Pedagogy boxes:
15.4 RESEARCH CLOSE-UP: Fitting and 1 8.1 EVERYDAY APPLICATION:
Self-fu lfi ll i ng prophecy, Pygmal ion
overfitting data 275
and the halo effect revisited 307
18.2 CI/,SSIC EXPERIMENT
15 The cognitive neuroscience of Obedience to authority: Milgram's
visual perception 279 experiments 309
lntroduction 280
Point of contact 280 PART 5 Putting it all together 315
The visual cortex 284 19 Putting it all together 317
The ventral pathway 286 Preparing for an exam il7
The dorsal pathway 288 A simple theory of human cognition 319
Object recognition 290 Some direct consequences ofthe
The binding problem 291 theory 321
Chapter summary 292 A final chunk 322
Further reading 292
Pedagoglr boxes: References 323
16.1 RESEARCH CLOSE-UP: Receptive Glossary 342
field - receptive fields in Vl 28s lndex 355
Cognitive psychology is an exciting but also complex topic. The fascinating questions it addresses
about consciousness, problem solving and memory, to take just a few examples, require the use
of sophisticated methods for collecting data and developing theories. Moreover, cognitive psy-
chology has become increasingly interdisciplinary. In particular, the last two decades have seen
a close collaboration between cognitive psychology and neuroscience in what is often called
co gn it iv e n eu r o s cien c e.
Cognitive psychology is a very active area, with intensive research being carried out to
address both basic and applied questions. Given the rapid developments taking place, there is
an abundance of ideas, concepts and techniques, The number and variety of methods used in
cognitive psychology (e.g, behavioural experiments, brain imaging, cognitive neuropsychology,
computer modelling) can be overwhelming and confusing for the student.
Rather than providing an encyclopaedic treatment, we have written this textbook as a first
introduction to coglitive psychology, To achieve this goal, we obviously had to be selective with
respect to the material covered. In general, we have aimed to strike a balance between discussing
classic studies and presenting cutting-edge research. In doing so, we have tried to keep the sense
of excitement and discovery that is palpable in the field, but also to provide material that is suit-
able at the early stages of learning. Thus, we offer a fairly simple introduction that is technically
correct and that provides a solid foundation for further study.
The intended book audience is primarily first-year and second-year psychology undergradu-
ates taking a first course in cognitive psychology. The audience also includes students majoring in
other fields taking cognitive psychology as an elective. The book can also be used for self-study.
While the structure of the book is standard (starting from low-level mechanisms such as
neural transmission and perception, and moving to higher-level functions such as creativiry and
expertise), its originality is in using a unique theoretical framework that organises and inte-
grates the material in all chapters. More advanced topics (e.g. consciousness or computer mod-
elling) are not ignored, but they are dealt with in detail only after the basics of cognitive
psychology have been fully introduced.
The book is organised around four main sections:

r Part I introduces the topic, offers some historical pointers and provides a brief introduc-
tion to neuroscience.
r Part 2 covers basic cognitive processes, such as perception and learning; these processes
are to some extent shared by other animals.
r Part 3 discusses higher-level cognitive functions, such as language and thinking; these
functions tend to be uniquely human.
r Part 4 deals lvith advanced topics, such as the neuroscience of visual perception, and com-
putational modelling. The goal is not to provide an extensive coverage of these topics,
which is beyond the scope of this book, but rather to show how they link with the topics
discussed in the first three sections ofthe book.

lg
Xii PREFACE

The framework used in this textbook is loosely based on theories and computational models
developed to explain, amongst other phenomena, expert perception, Iearning, memory and
problem solving (Gobet et a1.,2001' Gobet and Simon, 2000; Richman et al., 1995). Following
the principle that the building blocks of knor.r'ledge should be taught first, rve will not present
our framervork in detail here or even in the nert chapter, but rather we lvill build it gradually in
the first three sections of the book. The final chapter, 'Putting it ali together', rvill bind the
t'arious strands together and shorv horv all aspects of cognition - most notably perception,
memory, learning, decision making and problem solving - hang together closely. At this stage,
it is sufficient to point out that this framework can summarise a large amount of knowledge by
a series of propositions that are accepted by most cognitive psychologists, and that together can
explain a surprisingly large number of phenomena. As previously mentioned, the statements
lvill be introduced incrementally as the empirical evidence is presented in the respective chap-
ters. A few examples lvill suffice at this stage:

r The brain and the mind are organised hierarchically.

r Perception is strongly parallel.
r Attention is limited and is essentially serial.
r Short-term memory is limited in capacity.
r Rationality is limited.

While such statements can be found in most textbooks, the originality of our book is that
they will be used systematically to build connections between chapters, and thus to facilitate
learning. Using this framework provides a coherent and integrative structure that will help
counteract the fragmented impression often given by cognitive psychology, Anchoring informa-
tion in a single framework does not mean that we will ignore the main competing theories. On
the contrary, a fair amount of text will be devoted to such theories.
 re_
C cognni.r p{.holog! irth€ (i.ntiiir
rhc inrcrdj*iplinary itud' ofrhc nind.

C csitne p.y.hologins rnr+cr thcor.iol

d.pcnd.nt !nriab!6 rrc th< ptr*rrageof
Chapter Preview
eftoc, rnt !.rbJ prorGoL
As an introductory overview, learning
d 8.sit.'h..lrhrdulthrman3,.ogniti!.
oaticnt: rufrcrin3 ioar braia Jrmagc. objectives help you quickly identify the
d Tl6ri!:!nabi.si.ntinsrosurlnari*
ComFutE mod.lr mrk.;t p6sibl. :o essentials 1'ou will learn in each chapter
'5at an rddrcs complcxphcnomcna.
d Cognti\'. Srt.holS' consists
.ppLi.d in6tions.

C Rccrnh hr shom rhar bcing a.ti!.

Glossary
and rctcntion oftbc nratcrirl, and

rFp.ori nnbli.rpprorb :

lhe d.nlo!:.Fn,!.
Key Terms and Glossary it.n(i.g snih..,itiJ.F
rhil c.en'.n tu.!i<i cf'h.
nrnigdrdoi.it,rkL ud
Key terms are highlighted in the chapters and rith !6 S..oqd.r

defined in the glossary at the end ofthe book. rdub.nr: rn turd rod:rr
Di&:6..ry, S. rrriiqnfr
An ideal tool for last-minute revision or to srq!lodri{!tni) ofSr
1*onl. i.r.ttrt,)n.n
check definitions as you read. nd-t.:t@q: r rrlq.F
.nn'i'!irr.lhumn tuJc
rtdrion:ts.i,.€rirn. tr{.$
drEd utoi-.tu' :,n,
rhrt
pdkuLrr!rrd!!J1 ilr:

Figures and Tables

Each chapter provides a number offigures,
illustrations and photos to help you to
visualise the key theories and studies of
fi&r.1.2 Th. @mpurcr rdog for th. i.forman
ra: tdtu al@nd. did.6416 hn: i. fi cognitive psychology.
rF*n SSr,k @d@ftd*i^rb.q
,@v&-ft cru :d'l F-6tha,.nl dr#r r.

nln!!. In iddition, r control gour;u* rc*cJ

that.orld bc conrpar.d to the Erous uho haJ
cl.ar fen.rn.Th. 161in8
tsk. rmon6rhee sroups,
uJ rhe i{or* bt th. 9oIP
rhat it, ifrrorJs

Stop and Think inr.ricr.nres Jhor rh. oriinal m.mo,1

Slopond fhiDk iii,:irllr.< tle!' -;.ial

These questions have been designed to : .:r r'-:..,1_:,

challenge you to think criticaliy in evaluating

themes in cognitive psychology and its Recent models of $/orking
relevance to your own life. Shon-t..m m.mon h:r b..n r let
t.Y.rrl
!r3
on th: infornation'pro:es'i.A rp?ir ot
trar nrzrorl ir norr increxingiy.nlcd

Fll

L
 xiY GUIDED TOUR

Box $:i EVERYD/IV APPLICATIoN:

Everyday Applications
ne sdotk @ofthirdrpts j5
Drarving on exarnples from every.day life, including film and
literature, these examples are used to illustrate horv cognitive
lnawn tu tri(*ptEd d1! dsn. drq.
of
ps,,'chologD'can be aprplied in the lrorld around us,
infoftEd a@ tu ntuG of tlE erFdffit E,
sp6.d dE 016. drs. AJl
.im's ,AE qhir ddn *dkr. Yd. Ss tu
anb,rho hJ intriJlyfirld ro ss rft
don.. Surprsiql enryh, frl! oNua^6
Bor14.4lilFOCUS:t {c
tr lilhq, Sc rirht ffir) - in tufid dJ

ln Focus 5urpri9in8ly, Iove has not ben (he focus of

kd negleded: fo. the thouends of 5tudi6
ttc sd rdst ffc Utl rd d in the find only ahandful rbout love. This ts !tunt*
th nffil cdfrion, rh qician d dt These boxes encourage you to hhaviou.. Even the {aw icknowled86 the vr
fi6dm - h ldd dtlrc N rhn ft e6 itr sideration Sen fninS $e tarif ia mu(der i
15 suFdlt 6 s. rlr. wtrcn th" t+ 16 take a more detailed look at ndenovgh ofenemotion or ttrat lt Is & aon
bf pi.idi3m.
some oi cognitive psychology's Ac<ording to Sternba(t (1986r, love cn E
an be .omJntic or compan;onate. To accoul
key theories and challenge vou p(opored that lo€ is the rsultofthe intsacj
ri6 ir artimacy, lhe sd p&sion (rcftrr
to think critically about the d**ion f,alidg 'lntimaq'dexribB alt lhr
person and b gensate the wilm @rity tl
issues presented. dod not indude phyrical a&adion, whi.l
f 5ion @mponenten@piulatq all the ton
the decision{aking aompoaent rell*ts tht
loved one.
aoi al nEgARcH o-osE Uftnl.. -l*R:

l@ ru|tr. Th aypi.J €xrrilMr h5 6.d

intuhdon- fiR h.e i. S. nrtue of tu ru
rtuc i5 tk re of il*trtdm. wkrkr an I
tud.. rcing.n. snd pha 15tto d€r Research Close Up and Classic Experiments
rcntdb of th it 6 and ftr rfralt. tu
pa(icipd6dca[M ro rch.e i^form& These features ofier you a detailed understanding of hor,v
hd bqby6md.tnr mturtdr(h 6ro
ch@n fttridt nffi&r.ft M,mFd by
.apuc b r&e. ne td phe B cJLwh,
cognitive psychologists use various research methods to
md KU.q. (Rqe rid qtrts.yb tu p
rim inEryJ adtu bi6ningof th tsdciFd explore fascinating questions such as'horr. do lve remember
ft blhi4ffirlc illEdgrE 6.eptu
Yd 4 Fd dtlr . lk of moqlbic * things?
'ilu, ufl,GN,wM, rcg.h.tu a$
Fidly, FU bE b 6d k la h bq s- t

Chapter summary
Thi3;trlt.r hs brrclr J:rrtch.l rh. slrar;.
..hoi!a tl:e .ililurion: rlCh!tr.r I I or l3Ll
Chapter Summary As r .c:rs.qrkflr. oiiri{ linritlrjor:t ir rt.IIi,
!is,:'. are !r!r..lr cc::rrainarl in cur cos,ririr
ri,!. \\ 1.., nrri;nr:o.iJ juti5crr:rrs rnri dcci;
The detaiied chapter summaries r::tt.r;ti!..ess trd iroiiur.o!s1r g...rii:r<
.ne,1l. \\'i Jso :iricliir our ii\ * L:r judqicg I
help you to remember key .iir.') :.C b) rsira rch:ma ,thtn :ralirg so,
rhir r13 r3i!, oRen lork,.rnd do not a.c.nirfi
.\enrs. The rcpft nnr3ii en.$ h.uris!i..orresi, concepts and issues. They also hi:;os:s:tieTirak:dl5tou..ster.oirF.i,r
or ):lr r a.,r irlllj..s rhrrrcrcririr oi r rhole
insrret:cha.ioLrrofot|err,,nJ ro i:ll preyoi
\\'hil. (rhr..ian :iJ T\crsla rrinr I n.,\.r serve as an excellent study and ;rr:onirs:rirlen:r :ni r.ile:rinE.:.l.n.i I
riJ ot..rs 'ikc r r::ore foiiti!. n:n;e. H.rrirr
Tl.r. bir- 6iqln hr!€ b€r;An .oils.qu.n..r
,:r.i : I \' l,e ,\ :!i ! or rh. r.:.g.ir on ;:.r:i, revision guide. l.r!!r.r oilo Ce.Ll. r{herh.r to irvitc ;omebo.
,.-l b! us:rr .r::Jir rre in'.rn,r ".tft:r,. ,':
!.Ji;:r rrh:n r.{ ji::t.hildren r.d!ririrn .
b::r.r i! !hiikiir i.l d.c;!ion nrr\in; .r; r.jl (ii:.:nrrleLr t,.lniQlt.r.i.6 iotiii.at trlje
'er.. E o, c.l Di'iri rf. 15 ,l litrr: ,r:r.;
Ii.: tur r1,o, or .\ er rnairlr, rc::use ofthe r:
i:st:r:es girl;our arrors do nor o,icr

Further readtng
\lrrl'rrn trn.i ri.rrr.r I ltal ) proridr Lrref!
higl:.r.le!.l.ogiirior. Th. p.;.,r r" X,
r\errk. :.1 I'rn::rI,,:1,:hi:,lJ;:.
"1.,,t.t
':-
iii. .jii.ririz.r;.rit.rtrinin3 book R:.ir,:'i:a
irl,iri'r (ir :ir:,."r,'r.r,il'.ionUri i i),re:..
Further Reading
{h. r"o r,pulrr !o,!"r iurhc:lrn.l':1jii,.
reler,rnr every.lrr eranpl.r oi th. rh.ar;* F:?'. At the end of each chapter you'11 find a ftirther reading section
r.,j rl. i' i: i .).rr\i\l: r.rr ro r c..orr.rrerJ .hcr
'r to help you research and read around the subject in more
depth.

I
 I I

Visit www.rnheducation.co.uUtextbooks/gobet'l today

Online Learning Centre (OLC)

After completing each chapter, log on to the supporting Learning,Centte
Online Learning Centre website. Take advantage of the
study tools offered to reinforce the material you have
read in the text, and to develop your knowledge of
cognitive psychology in a fun and effective way'

Resources for students include:

t Self-testmuhiple-choice &a*ffi
.;-o;aJJn -,i i i..
,11 .: j:'
dbffid-5ffi .., .l ii i !,'i
questions
-hlkffi

Also available for lecturers: d4&Q-hts

t hlE .d.a 6 t6 Fd t t5 G h
t dE
!1,6rikJ4
t PowerPointpresentations '*@ffir4|@c,G'',c

t Seminar material
6*,GWtddbil-''
*t{it-d
b-hdt6D@dry'

t Test bank

rffi*.ddff

Visit www.mheducation.€o.uk/textbooks/gobetl today

xv'
j

L
 ACT-R Adaptive Control of Thought - Rational
AD Alzheimer's disease
AI artificial intelligence
ANS autonomous nervous system
CHREST Chunk Hierarchy and REtrieval STructures
cm centimetre
DNA deoxyribonucleic acid
EEG electroencephalography
fMRI functional Magnetic Resonance Imaging
GABA gamma-aminobutyric acid
IT inferior temporal cortex
km kilometre
LTM long-term memory
LTP long-term potentiation
m metre
MEG magnetoencephalography
M cells magnocellular cells
MRI magnetic resonance imaging
ms millisecond
MT medial-temporal cortex
NP noun-phrase
PET Positron Emission Tomography
P cells parvocellular cells
S second
S sentence
SAS supervisory attentional system
SQ3R survey, question, read, recite and review
STM short-term memory
UTC unified theory of cognition
VP verb-phrase
V1 visual area I
v2 visual area 2
v4 visual area 4
VSSP visuospatial sketchpad (also known as the visuospatial scratchpad)

iE, )o(
 Part contents
1 lntroduction to cognitive psycholog;y 3

2 A brief hlstory of cognitive psychology 13

3 A brief introduction to neuroscience 27

L
i
 lntroduction to
cognitive psychology
d
d Cognitive psycholog, is the scientific study of mental processes, Cognitive science is
the interdisciplinary study of the mind.
d Cognitive psychologists answer theoretical questions by running experiments. Tlpical
dependent variables are the percentage of correct anslvers, response times, rlpes of
errors, and verbal protocols. Brain-imaging data are increasingly used.
d Besides healthy adult humans, cognitive psychology also studies children, animals and
patients suffering from brain damage.

d Theories enable scientists to summarise and explain data, and to make predictions.
Computer models make it possible to develop theories that are highlv specified and
that can address complex phenomena.
d Cognitive psychology consists of numerous subfields that address both basic and
applied questions.
d Research has shorvn that being active during learning leads to better understanding
and retention of the material, and cognitive psychoiogy has developed methods for
studl.ing efficiently.

L
 4 CHAPTER I INTRoDUCTION To CoGNITIVE PsYcHoLoGY

lntroduction
Horv does the mind r,vork? \Vhat is the link betrveen brain and ntind? Why do rve tbrget? Does
strong emotion help or hinder mernory? What is the nature of consciousness? Hou,do ne think?
These questions - some of the nrost difficult ihat humans have ever. tackled are the subject
-
matter of cognitive psychology. As you can see, you are about to engage in an exciting voyage of
discoveryl
Cognitive ps,vchology is the scientific study olmental processes that is, processes that allorv
-
us to perceire our environment, memorise intbrmation, use language to communicate, and
make decisions' It also covers topics such as attention and consciousness. Nlost ofthese proc-
esses occur automatically, without us even noticing they have happened, In spite of this, eyery
rvaking minute is 6lled rvith events that are studied by cognitive psvchology.
The difficulty of the questions cognitive psychology aims to ansrver, as rgell as the inter-
disciplinary nature of its research, has a hvo-fold consequence. On the one hand, it is a very
exciting field, in which new discoveries are made almost daily. On the other hand, it is a fairly
complex field: ideas from different sciences (cognitive psychologl'has often teamed up with
other fields, such as neuroscience and computer science; see Box 1.1) must be assimilated and
combined; many topics are covered; students must have at least elemenlary knorvledge of dif-
ferent tlpes of experimental methods; and, perhaps more than in any other field of psychology,
the sometimes complex interplay between data collection and theory development must be
understood.

Ue4t1_$ fut sj.G ,p$trgibo1t,end,co.gitft,i,$$*

Cognitive psychologists study a staggering number of topics: they are interested in the way
people perceive, learn, speak, read, make decisions, do mathematics, play chess, and even
develop new scientific theoriesl Given this range of interest, it is not surprising that they
often carry out interdisciplinary research. For instance, an understanding of language com-
prehension may reguire collaboration with linguists, and the study of consciousness may put
together a group of cognitive psychologists, computer scientists and philosophers. The
boundaries between disciplines sometimes become fuzzy, and the interdisciplinary study of
how the mind works has become a scientific field in itself: cognitive science. cogniiive
science can be seen as the intersection ofseveral fields including psychology, anthropology,
artificial intelligence, computer science, neuroscience, linguistics and philosophy.

The experimental methods of cognitive psychology

Cognitive psychology is first of all an experimental science. A theoretical question is addressed
by rulning an experirnent in which the independent yariable of interest is manipulated, ancl
the effEct of the manipulation is assessed by measuring dependentvariables (see Box r.2). The
dependent variables most often used are response times, the percentage of responses that
are
correct, and the t)?e of errors made. Eye-movement recorclings are sometimes used, as the
duration and location of fixations provid.e useful information about where the focus of

ll
 THE EXPERIMENTAL fuIETHODS OF COGNITIVE PSYCHOLOGY 5

attelttion is directed. \Vhen stucll.ing hieh-level functions such as problem solving antl decision
rlaking, cognitive Lrs),chologists olten ask participar.rts to think irloud. The verbal protocols thlt
are obtained crul then be analysecl at cliiterent levels of granulariq': for example, to look at the
speed ivith rvhich particip.rnts carn' out mental processes, or, at e highel level of anll;'sis, to
identity the strategie's rhat ale being Lrsed.

Stop and Thlnk Sternberg

_ j .,: -
j :. . . : , , , - ,. ,'
. '

r,ta ,,-: -a: '

For many years cognitive ps1'chologists have been interested in horv cognition is imple-
mented in the brain. Horvever, breakthloughs in brain imaging technology in the past 20 years,
combined ivith the broad ar.aiiabiliry of high-speed computers, have brought about a revolutioll
in the rvay cognitive psychologists collect data. As rve rviil see in Chapter 3, it is nolv possible to
observe brain activation in real time rvhen participants carry out experimental tasks.

Box 1.2 CLASSIC EXPERIMENT: Sternbergis experiment

Saul Sternberg (1966) carried out one of the most famous experiments in cognitive psychol-
ogy, and, although we will cover short-term memory in detail in Chapter 6, it is instructive to
describe this experiment here. Sternberg was interested in how we access the contents of our
short-term memory. ls it done serially, one item after the other, or can we access all the items
simultaneously? For example, if you memorise a list of numbers, say 7 3 9 2, and are asked
to say whether 9 was a member of the list, do you have to mentally go through 7 and 3 to
finally reach 9, or can you access 9 as rapidly as any of the other numbers?
The task we have just described actually closely follows Sternberg's procedure. He pre-
sented a list of digits (which he called the 'memory set') and then a probe digit, and the task
of the participants was to press either a YES button or a NO button as rapidly as possible to
indicate whether the probe was, or was not, part of the memory set. The independent vari-
ables that Sternberg manipulated were the size of the memory set, from 1 digit to 5 digits,
and whether the probe was, or was not, part ofthe memory set. The dependent variable was
the time taken to press the YES or No button. sternberg wanted to answertwo questions. we
have already mentioned the first one: is the access to short-term memory serial or parallel? lf
short-term memory isSccessed serially, then it should take longer to respond with a large
memory set, say five lems, than with a small memory set, say two items. lf short'term
memory is accessed in parallel, there should not be any difference. With his second question,
Sternberg wanted to kriow whether search is exhaustive or not. That is, after having found the
target, does the search stop, as one would expea intuitively, or does it continue to the end of
the list? To return to the example above, does search stop after reaching 9, or is 2 considered
as well? Note that, with NO answers, the search is expected to be exhaustive: the entire list
has to be scanned if one wants to be sure that the probe is absent. For each memory size, half
the required answers were YES and half were NO. With the YES trials, the position of the
target in the list was changed with equal probability.
 6 CHAPTER 1 INTRoDUCTIoN To CoGNITIVE PSYcHoLoGY

Participants were very accurate, with only't.3 per cent of their responses being incorrect.
As shown in Figure 1 .1 , response time increased as a function of the size of the memory set,
and there was little difference between the YES and the NO answers. The results are very well
fitted by a straight line, and the slope of the line indicates that every additional digit increases
scanning time by 37.9 milliseconds (ms). Sternberg considered that his results gave a clear
answerto his two questions. First, search is serlal, as larger memory sets incur longer scan-
ning times. Second, scanning does not stop when a target is found and thus search is exhaus-
tive: this is indicated by the fact that the yES answers take as long as the No answers.

o Positive responses
600 I Negative responses

E
F
e.
at
E 500
o
u
6 RT=397.2+(37.9xs)
E
F 400
o

300
3 6
Memory set size (s)

Figure 1.1 Results of Sternberg's experiment

The graph shows the mean response latency as a function ofthe number ofsymbols in memory, as well as the
best-fi tting linear regression line.
Source: Adapted from Sternbert (196G)

Cognitive psychologists.rr.t.*0"t,-.nts with diflerent rypes of population. The default is to

use healthy adult participants - Wpically psychology students - who are naive about the ques-
tion being studied. In the field of expertise, the performance of experts is contrasted with that of
intermediates or ngvicqs;In neuropsychology, the comparison is between patients suffering
from a brain lesion and healthy participlnts, and the question of interest is the extent to which
brain damage affects co$nition. Developmental aspects are studied with children and even
infants' Some questions can be addressed by studlng animals, typically non-human primates,
In all cases, experiments are designed that both address specific theoretical questions and are
ethically appropriate to the population under study.

ll
.) ii.
 DEVELOPING THEORIES THAT EXPLAII\ THE DATA 7

.r:
Developing theories that explain the data
i; Once data have been collected, cognitive psychologists develop theories to explain thenr.
\\iithout good theolies, er,en the best data ale ollittle use (see Box 1,3). in addition, theot'ies are
also a porverful *,a).to generate nelv predictions and thus nelv ideas tbr future experinlents,
rvhich enables a better understanding of the phenomena under study.
in general, the more precise a theory, the better it is - that is, lr,e should prefer theories that
are explicit about the structures and the mechanisms involved to lheories that leave manl'details
unspecified. One rvay to do this is to use mathernatical models. For example, the rate of forget-
ting can be expressed mathematicaliy (see Chapter 7), However, these models make assump-
tions that are often not met b1.the type of data collected in cognitive psychology, ancl they also
force more simplifications than is desirable. As rve rvill see in the next chapter, on the historv ot
cognitive psychology, one of the great contributions of the 'cognitive revolution' is the iclea that
.olp,rt"., can be used to iinplement scientific theories in psychology, This idea really has trvo
meanings, rvhich you should ciearly distinguish.
The first meaning is that, at some level of abstraction, human cognition is roughly similar
to a digital computer (see Figure 1.2); in particular, both include input and output
mechanisms, programs, different types of memory, serial processing (i.e. only one operation is
carried out at a time), and so on. This meaning, rvhich is the one used by the information-
processing approach, is best illustrated by Newell and Simon's (1972) work on ho',v
humans solve problems in domains such as logic and chess. This rvork lvill play a central role in
Chapter 11.

p,g!.!$,{N FffiuSi'Hhy do *e tre{ tlie.origq?, I ., , ,1 '',,,. ",

A large number of theorles have been developed in ctgnitive psychology, and this book
will discuss a fair number of them. But $/hy bother with theories, in particular if one is
more interested in applications (e.g.'in applying cognitive therapies to hetp depressed
people, or in improving financial .decision making in a business)? Why not simply cotlect
data?
Well, thousands of experiments are carried outt every year on different aspects of cogni-
tion, and it would be impractiial, if not impos'sible, to survey even a subset of them when
developing a specific application.'The initial relevance of theories is thus to summarise data.
In some cases, this can.be done to the point where scientific laws can be derived (see the
power law of learning ih Chapter 7). But.we can go one step further. Theories can explain
data: that is, they can highlight the causal mechanisms that underlie a phenomenon' Finally,
theories make it possible to predict future behaviours.
It is essential to realise that theories are important for applied science. lf you want to
construct a plane, you'd better know about theoretical physics. lf you want to develop medi-
cines, you'd better know about theoretical biochemistry. The same applies with (cognitive)
psychology. lf you want to develop better educational techniques, improve on the current
cognitive therapies, or provide better environments for the elderly, you need to use and
understand the current theories in your specific field of interest.
8 CHAPTER 1 INTRODUCTION TO COGNITIVE PSYCHOLOGY

lnput
devices
CPU

Control Arithmetic
unit logic unit

Output
devices Memory

External
storage

Effectors
E nviron ment Processor Memory
Receptors

Figure 1.2 The computer anatogy for the information-procesling approach

Top: simplified diagram ofa digital computer. Bottom: the standard view ofcognition
in the information-processing
approach. Note that the input and output devices in the computer correspond to the receptors
and effectors,
respectively. The cPu (central processing unft) corresponds to the processor.
The final component, memory, is present
in both cases.

Stop and Think The computer analogy

Tne ailalcSy feiweer c!n-rDLrter anc :ogi.:iiicn has hac a lcnsiJerable
nfluence ln :cgr iilYs
ani i; s:itl c1-r:r.r ,;sed icCay. For- e,(ample researchers ir ccgirritve teuroscielcS l.t:e corcepts Psy'chcl;g;,
slch as
npLrt, ouipul: cr prcces;ing lf ;ricrrnaticn jce.tiay Ine aspecis of lnis analogy tirar a.e tat ci_r arly
Sifong. anc l.ilse lqai are Weak.

The second meaning is that the computer can be used as a modelling tool to simulate
cogni-
tive behaviour' without necessarily accepting the first meaning. Just like other sciences
such as
physics or chemistry, cognitive psychology can use computer modelling to
understand complex
systems that evolve as a function of time (Figure 1.3 illustrates the ideaJ. Weather
forecasting is
possible because scientists have a sufficiently detailed understanding of the physical "are
laws that
implicated. (A scientific law is a verbal or mathematical statement that describes a relation
between rwo or more empirical variables,) These laws can be expressed as mathematical
equa-
tions' which in turn can be implemented as computer programs. If you want to predict the
weather in London for the next day, you can run such computer programs togethei
with data
about the previous days'conditions (e.g. temperature and air pressure) at numerous spots
in and
around London. Given the amount of data to process and the complexity of the computations
to
carry out, serious predictions cannot be made without using powerful computers.
Human
 DEVELOPING THEORIES THAT EXPLAIN THE DATA 9

cognition is more conrplex than the rveather in London - rr hich is rather predictable and ttsuall,v
not very enjoyable - and computer modeliing is a crucial tool tbr understanding psychological
phenomena at different levels of abstraction. Figure 1.3 illustrates this rvith an example of social
cognition. A group of business people meet to make an important decision. We can measure the
knorvledge of these individuals, their goals and their emotional state, Togethel rvith our under-
standing of cognition and emotion, tvhich can be stated as laws and used to develop a computer
program, we can make predictions about the like\' outcome of the meeting - for exampie, that
the chief executive,n'ill be put in a defensive position and react angrily in order to convince the
other members of the board. This example illustrates the use o[ computational modelling rvith
high-level cognition. Computer models can also be developed to explain lorv-level aspects of
cognition (e,g. the perceptual mechanisms used to discriminate behveen two different letters),
and they can also be used to simuiate phenomena at the brain level (e.g. how the connection
between two neurons can be strengthened by the simultaneous presentation of rwo stimuli).

l;::i':Il',*'.,t,:T!:1,.J';3,:"i'.i,ili'.ai-3r, i.:c : --!iai- i.r r-j, :; ,-:; j.;i-re:! : t):::,t

3rpCr- i,r-ii l,:: ts:x 2r?

-
lnitial conditions: Laws of physics Weather forecast
. temperature computer simulations
. air pressure, etc.

+ +

lnitial conditions: Laws of psychology Prediction of behaviour

'. emotional state computer simulations
environment, etc.
>.;

+ +

Figure 1.3 Computer modetling

Computer modelling in weather forecasting and in psychology. ln both cases, initial conditions combined with laws
make it possible to derive predictions.
 IO CHAPTER 1 INTRODUCTION TO COGNITIVE PSYCHOLOGY

The subfields and applications of cognitive psychology

Cognitive psychology is not a homogeneous field, but consists of a number of subfields. This
organisation is apparent in the Qpe of scientific journals lvhere research is published or the Eype of
conferences that are organised to disseminate results. The subfields can be identified either using
the type ofbehaviour they study (e.g. psychology ofperception, psychology ofreading) or rhe rype
of experimental technique they used (e.g. eye movements, brain imaging). These fields can also be
grouped as doing either basic research (e.g. statistical properties of reaction times) or applied
research (e.g. cognitive design, psychology ofdriving, educational psychology).
As we shall see in the next chapter, one of the reasons behind the birth of cognitive psychol-
ogy was the need to answer very practical questions, such as hotv to keep the vigilance level of
radar operators high. The field has maintained this interest for practical issues, and the scientific
principles it has uncovered have generated countless real-life applications (e.g. Pezdek et al.,
2006). We will discuss many of these applications in this book, btl? here is a small sample.
Research into attention has clearly shown the dangers of using cell phones when driving. Work
on long-term memory and learning has led to new methods for instructing new drivers, training
technicians in industrial settings, and teaching mathematics to children; this last application
includes the development of intelligent computer-tutoring systems. This work has also pro-
vided guidelines on how to improve the reliabiliry of eyewitnesses in court. The study of lan-
guage has led to advice concerning how to write texts th3t are easier to read (Hayes, 1989;
Kintsch, 1998) - you may want to apply these principles in your own writing. In a field called
human-computer interaction, principles from perception and short-term memory research have
been used to improve the design of computers and computer applications. Research into exper-
tise has led to novel and literally life-saving applications: methods for training physiCians and
land mine detection experts, and for improving security in nuclear plants. On a more cheerful
note, it has generated guidance about how to train your dog in a more efficient way and even
holv to protect yourselfagainst burglars.

How to use this book

Learning has been extensively researched in cognitive psychology and in other fields ofpsychology.
Several laws have been identified (see Chapter 7) and you can take advantage ofthis knowledge to
make your own learning more efficient. This textbook has been structured to help you do so.
The overarching principle is that practice makes perfect, and this is valid too for understand-
ing and memorising information from a textbook. In general, practice increases the strength of
memory traces and increases the number of different ways through which information can be
retrieved. Together, stronger memory traces and numerous access paths make forgetting less
likely. However, practice requires time. Give yourself enough time to study this textbook and
any additional material that your instructor recommends.
Research has also shorvn that people remember information better when this information
has been subjected to encoding that elaborates the material, organises it and links it to previous
knowledge. A particularly powerful way to make sure you memorise and understand material is
to force yourself to recall or othenvise generate (e.g, by answering questions) the information
you have learned. Among other things, generating material creates additional retrieval cues and

,il

-ii i-

,: .,:- ,, r. '' l,-; i ' :l -

..,:.ta:: t..::r.,,..,, ,:
 CHAPTER SUMMARY I1

when you learn' do not

leads to the creation of nerv links in your knowledge base. In a nutshell:
read passively, but be activel
Tle validiry of these study principles has been tested exPerimentally, and it has been repeat-
edly found that students tvho follolv them learn better than those who
do not' (If yor"r want to
1990.) In addition, a variefy of tech-
find more about this research, you can check Anderson,
learning habits' The SQ3R method
niques have been developed to help students develop effrcient
of which has been confirmed by
ltadlock, t978) is a particularly efficient one' the efficary
and Review' lVhen s$rdying a
research. The acronym stands for Survey, Question, Read, Recall
of the chapter
chapter, first surtey its contents, by reading the preview section at the beginning
you lvould like the chapter
and skimming its contents. Then, think about the types of cluestion
go back to these questions, and also try to link
to answer. When you renclthechapter, regularly
topic. A1so' read critically, and do not hesitate
the material to rvhat you already know about the
to question the material. At the end of each section of a chapter, force yourself to recall the
essential points of what you have learned, Finally, after finishing a
chapter, rwinv il regularly'
and try to link its contents to the preceding and following chapters.
This rhethod is certainly
consuming than reading the chapters once whilst daydreaming, but it will
also ensure
more time
a much better understanding and memory of the material'
indeed to
This textbook has been written to encourage you to be active when reading, and
Learning objectives at the beginning enable
make use of the sQ3R method. In each chapter, lhe
Summary althe end of each chapter for
you to survey the chapter (you can also uselhe Chapter
and connections with the other
this purpose). In the body of each chapter, Thought Questions
the pres-
.nupi.., will force you to think deeply about the issues, Reviewing is made easier by
material that lvill
.n.e of a Chapter Summary. Further Readingwtlldirect you to more advanced
allow you an even deeper processing of the information'

Stop and lhlnk SQ3R

iSCn'i'C;3;' iCr ii-'-:"OV rlE
V\jian:iLrCyiirg In iir: CaSi. ta\./a ylU .jSeJ:eS.rniqUe;, rrCllCilirg Seii-de\€lC-taa
riy11 r13.'r3 fev3 ir-ieJ
y3r, aO.,ain,-liln aili iear'iringi .'rci,v dC:neSe:eChniCljej Clffei ir3i-i' SQ3R?
tr..rCi-, IeCrniilrreS, Siirdy Cne c.,ire, Cirnis bCC<ih3 lvayyCrl ncrl-taliy jttrJy, a'i anCtne'Chacler rsrtrq
SQ3i.. Car ycrr oDse rve a drl+ei'e;rce?

Chapter summary
has provided an
Cognitive psychology is the scientific study of mental Processes. The chapter
the kind of methods
overview of the research questions addressed by cognitive psychology, and
psychology is first of all an experimental science'
that are used to answer ihem. Cognitive
computer plays an important
although the development of theories, often realised as Programs'
for the devel-
role as well. A central message of this chapter is that theories are essential not only
for instruc-
opment of fundamental knlrvledge, buialso for devising applications - example,
tional methods,
also expands on
The next chapter, which provides a brief history of cognitive psychology,
requires some
some of the topics we have lurt couered. A good understanding of cognition
a full chapter
knowledge of the biophysical medium in which it is implemented - the brain - and
(Chapter 3) will be devoted to the basics of neuroscience'
 a

12 cHAprER 1 rNTRoDUcroN To coGNrlvE psycHolocy

Further reading
The four-volume collection edited by Pashler (2002) provides extensive coverage of the experi-
mental methods used in cognitive psychology. Newell (1990) contains an interesting discuision
about the link betrveen data and theory, and how this affects the developrnent of theories.
Among the many journals that publish general research on cognitive psychology and cognitive
science, the most important are Cognitive Psychology, Cogtitive Science, Cognition, the Ettlropeott
Journal of Cognitive Psychology, lournal of Experimental Psychology: Learning, N[emory, arul
Cognition, Memory d- Cogtrition, Ps1'chological Review and Trencls in Cognitive Science. Apptietl
cognitive Psychology contains papers on how cognition relates to everyday life,
 A brief history of
cognitive psychology
d
d the German lVilhelm Wundt opened the first laboratory of psycholog,v in 1879, and
other scientists started studyrng psychology experimentally around the same time. In
1890 the American William james published a textbook of psychology that anticipated
many of the questions that rvere later addressed by cognitive psychologists'

d lVtrite psychologists rvere actively researching cognition in Europe during the first half
of the twentieth century, in the USA behaviourism proscribed the study of mental
processes and directed attention to learning behaviours in animals'

d the Second lVorld War put an end to the Gestalt and lVr,irzburg schools in Germany,
rvhich rvere both interested in cognition, but also alerted psychologists to the
importance of attention and vigilance.in human behaviour. The rvar effort speeded up
the development of the digital computer, which .,vas essential in the birth of modern
cognitive psychology.
d the cognitive revolution happened over several years, but 1956 is often seen as its
foundation date. Three key factors led to this revolution: the computer analogy
developed by Broadbent in the UK and Newell and Simon in USA; the development of
new linguistic theories by Noam Chomsky; and the use of intbrmation theorT to
explain psychological phenomena.
d Cognitive psychology has split into numerous subfields and has increasingly influenced
other scientific fields.

l3 H
fr
14 CHAPTER 2 A BRIEF HISToRY oF CoGNITIVE PSYcHoLoGY

d In the 1980s cognitive psychologists started using models inspired by the biology of the
brain (connectionist models). Critical developments in brain-imaging techniques in
the early 1990s led to the growth of cognitive neuroscience, which uses knowledge
from biology to explain the way the mind works.
d Unified theories of cognition, which are embodied as computer programs, aim to
provide a single theory explaining every aspect of cognition.

lntroduction
Humans have been interested in their mental processes at least since the Ancient Greeks. For
example, Aristotle proposed that our memory is like a clay tablet on which information can be
written, and Socrates thought that knowledge is innate: one can recollect it, but not acquire it.
However, we had to wait until the end of the nineteenth century to see philosophical specula-
tions replaced by truly scientific investigations. In 1879 Wilhelm Wundt opened a psychology
laboratory at the University of Leipzig, in Germany. At about the same time, a number of scien-
tists were carrying out experiments on cognitive processes. The French academic Alfred Binet
was investigating the strategies used by mental calculators, using measures that are now stand-
ard in the field (such as reaction times and errors), and the German Hermann Ebbinghaus
designed nonsense verbal stimuli (such as BOF, GIF or ZID) Io study memory and learning.
Wundt ( 1874) himself used a rich repertoire of methods, of which introspection is of special
interest. This method consisted in training participants to report and analyse their own percep-
tions and sensations. This method was anchored in structuralism, a school of thought that
assumed that the contents of consciousness can be broken down in elementary parts, just like a
chemical compounds can be broken down into atoms. Unfortunately, this method was riddled
with problems. For example, as we shall see in this book, most mental processes occur uncon-
sciously and thus are simply not reportable. The problems plaguing introspection quickly
became apparent as participants in different laboratories were reporting different elementary
Processes. It is now clear that the participants' introspections were influenced by the theoretical
preconceptions of the experimenters.
In 1890 the American William fames wrote a textbook, Principles of Psychology, that became
particularly influential. While Wundt was interested in the structure of consciousness, |ames
was intereste din its funaion: that is, he wanted to know how the mind enables adaptation to the
environment. His approach was non-experimental and rather speculative, but he anticipated
some of the questions that would later occupy cognitive psychologists for decades. For example,
as we shall see in Chapter 6, current research is still busy studying |ames's distinction between
primarymemory (the content of which is immediately available to consciousness) and second-
ary memorF (which is much larger and can be seen as the repository of previous experiences).
The study of cognitive processes in the USA was stalled almost at its inception by the advent
of behaviourism, which had started just before the turn of the century. The weaknesses of
introspection as a method and structuralism as a framework offered an easy target at which
experimental psychologists lashed out. A particularly clear statement of the aims of behaviour-
ism was given in John Watson's (1913) article, 'Psychology as the behaviorist views it', which is
 THE COGNITIVE REVOLUTION 15

often seen as the behaviourist manifesto. \\ratson argued forcefully that, being a branch of the
natural sciences, psychology should strive to gather objective and experimental data. Scientific
lalvs in psychology should quantif,/ the relation berrveen observable behaviour and observable
stimuli in the environment, the so-called 'stimulus-reaction (S.R) bond'. These laws are the
same tbr animals and humans. Thus, there is no place for ill-defined concepts such as mental
processes, consciousness and mind,
For more than half a centurT (roughly from 1900 to 1960), most experimental research in
America focused on animal learning, as this made possible the close control of both stimuli and
responses. Little research was carried out by psychologists on issues such as decision making,
thinking and memory, u'hich all had to presuppose more complex mechanisms than stimulus-
response bonds, It shouid be pointed out that, during this American 'cognitive winter', the
study of mental processes was active in Europe. For example, in the Soviet Union Alexander
Luria was investigating phenomena ranging from aphasia (impairment of language production
and comprehension due to brain damage) to the processes involved in lying, and in Sr,vitzerland
fean Piaget was working on a general theory of cognitive development, Research was particu-
larly active in Germany. Gestalt psychologists were studying perception and problem solving
(see Chapters 4 and 1 I ), and the members of the Wtirzburg School examined in great detail the
processes underpinning high-level thinking such as arguing philosophical questions.

The cognitive revolution

The Second lVorld War had two opposite effects on the development of cognitive psychology.
On the one hand, it slowed down the vigorous research that was carried out in Germany. Several
of the researchers of the Gestalt and lVtrzburg schools were persecuted under the Nazis and
had to escape Germany. On the other hand, the war effort in the USA and the UK encouraged
psychologists to study factors such as attention and fatigue. For example, radar operators had to
stay vigilant for hours in conditions where nothing happened for most of the time, but vvhere
one second of inattention could mean missing an enemy plane and could have disastrous con-
sequences. This was clearly a topic for which cognitive psychology could provide important
insight. As we shall see in Chapter 5, a substantial amount has been learned about attention
from this research.
By accelerating the technological advances in engineering that led to the construction of
computers, the Second World War provided another critical thrust for the development of cog-
nitive psychology, In England, just after the war, Donald Broadbent proposed the computer-
mind analogy: both have receptors dealing with inputs and effectors dealing with outputs. In
addition, both are characterised by a fairly small short-term memory and channels of limited
capacity, which has the consequence that a fair amount of processing is carried out serially (i.e.
only one thing is done at a time).
One the other side of the Atlantic, Newell and Simon went a step further. They noted that
both computers and humans manipulate syrnbols (i.e. physical patterns standing for something
else). If both carry out the same tJpe of information processing - at some level of abstraction -
then one should be able to develop computer programs that simulate aspects of human cogni-
tion, or even that could think, This insight was not only the motivation behind an important
approach in cognitive psychology - the information-processing approach, also called rhe
 16 CHAPTER 2 A BRIEF HISTORY OF COGNITIVE PSYCI-IOLOGY

.Box 2.1 lN FOCUS: The influence of engineering on cognitiYe psychology

Several fields of engineering had a direct impact on the birth of modern cognitive psychology,
and their influence is still detectable in some of the terms cognitive psychologists use rou-
tinely today. tnformation theory (a branch of engineering interested in measuring informa-
tion: for example, the amount of information that is exchanged during a phone contrersation)
has given us several terms. These include input, output, channel capacity (the largest amount
of information that can reliably be transmitted on a channel), redundancy (the idea that the
components of a system or a message are duplicated in order to augment reliability), code (a
specific type of representation) and encoding (the process by which information in one type
of representation is converted in a different type of representation). From computer science
we have terms such as buffer (a memory store that keeps information for a limited duration
only), production (a rule consisting of a condition and an action) and algorithm (a set of
instructions). Computer science has also provided cognitlve psychology with diagrams called
flowcharts (charts that show the path ofdata through an information-processing system and
the operations performed on the data; see Figure 1.2 in Chapter 1, and almost any chapter of
this book), and the programming techniques and languages used for developing computa-
tional models. These techniques have turned out to be highly useful in enabling cognitive
psychologists to state theories unambiguously, which enables them to simulate complex phe
nomena and make quantitative predictions (e.9. the number of errors a child will make when
learning multi-column subtraction). €omputers have also been tremendously important for
collecting and analysing data - think oftheir use nowadays for processing the huge data sets
produced by brain-imaging techniques such as event-related potentials and fMRl.

syrnbolic approach but also led to the creation of the field of artificial inteliigence (AI).
-
Whereas cognitive psychologists use computer modelling to understand cognitive processes, AI
researchers are interested in producing computer programs or robots that behave intelligently,
even if this entails mechanisms that are unlike those characterising animal and human cogni-
tion (see Box 2.1).
Another critical development consisted in a ... book reviewl As noted above, behaviourism
was rampant, and Skinner, one of its leaders, published a book called Verbal Behaviour (1957),
where he argued that the simple mechanisms identified by the research on animal classical con-
ditioning could explain language, a capacity that is uniquely human, Two years later, a young
linguist named Noam Chomsky destroyed Skinner's argument in an article that must count as
the most influential book review ever written in science (Chomsky, 1959). Written in a surpris-
ingly aggressive tone by today's standards, Chomslcy essentially accused Skinner of scientific
incompetence at best and dishonesry at worst, and argued that the mechanisms proposed by
Skinner simply could not explain even simple aspects of human language. Chomsky based his
virulent attack not on empirical data, but on an abstract analysis of the properties of human lan-
guage. As we lvill see in Chapter 9, this preference for abstract considerations over empirical data
has remained a characteristic feature of linguistics. It is fair to say that Chomsky's article dealt a
serious blow to Skinner's enterprise, and to almost any attempts to explain language using simple
associative mechanisms such as S-R bonds. The field would have to tvait nearly three decades and
for the advent of connectionism to finally rvitness a serious attempt to do so (see belotv),

it
 TI-IE COGNITIVE REVOLUTION 17

of both (modern) cognitive psycholog)'- and

The 1,ear 1956 is often taken as the starting point
took place thtlt year' \vhich were altencled b)'
artihcial i1telligence. Two epoch-making colferences research-
The Dartmouth conference brought together
man),of the researchers \ve have mentioned' held at the
the symposium of Intbrmation Theory that rvas
er-s lvliting the first Ai programs, and
and others on com-
talks by lrlervell, Simo', Chomsky
Nlirssachusetts Instirute ofiechnology sarv inf-lu-
year 1956 also saw the publication olhighly
r-lullication, language and problem soh'ing. The
(1956) p.,bllrh.d his'magical number seven' paper in rvhich he
ential papers and books. Nliller cognition' rvas
as lvell as other aspects of human
argued that the capacity of short-term memory'
important notion of chunking: that grouping
limited to seven items. Nliller aiso introduced the is dif-
of shoi-term memory' For example' rvhile it
items together as units heifs obvixe the limits
ficulttomemorisethesequenceofletters.THISSENTENCEISEASYTot\4Erv.loRIS
in meaningful are grouped
E,, it is much easier to memorise the foilowing sequence where letters

Box?.2 iNFocus: rrre dtff€rent meanings of uldtem.theorlr

referring to a single 'thing'. However, the

we have so far used the term theory as if it were
on the context, and it is useful to be explicit
term is used with different meanings depending
in many scien^ces, including cognitive
about these meanings. Three meaning, ,r. "plrrunt
psychology (Newell, Tggo)' rnuo'ies referring tq l:*"
t:ll:1:T:-:l^*t*pts and mechan-
ismsareknownasframeworks.Forexample,information-processingpsychologyassumes
abstraction, to the digital computer: both
that human cognition is similar, at some levei of
canbecharacterisedbythepresenceofinputsandoutputs,acentralprocessorandmemG valid-
enable any clear{ut predictions, and their
ries. Frameworks are not specified enough to generate'
(in the strict sense) they
ity is judged in terms ofthe success ofthe theories
between concepts and mechanisms are speci-
With theories (in the strict sense), the relations
macrb. However, various parameters are
left unspecified,
fied, and qualitative ;;il;;;;;fe framework of informa-
example, within the
which makes quantitative prediction difficult. For
tion.processingpsycholoey,atheorycouldspecirythatmemoryisdividedintotwostores(short-
termmemoryandlong.termmemory),thecapacityofshort.termmemoryislimited,andthe
would predict that
is fairly lonB. such a.theory
time to encode information in long-term memory
of 2O digits, but could not make
people are not parri;;larly good it memorisinj a sequence
be memorised and the type and number
precise predictions aUout inJ number of digi6 t[at can
for this, which are provided by models' with
of errors made. Additional details are necessary
models, detailed r"in"nis6 are proposed and
all parameters are specified, which makes it
possibleforquantitativepredictionstobemade.lnourexampleofmemoryfordigits,sucha
the prob-
the capacity of short-term memory,
model would n"u. to ip"iiry, among other things,
information from short-term to long-term
ability of making an .rror, ,nO tt't. rite of transferring
memory.Anumberofsuchmodelshavebeendeveloped,amongstwhichthemostinfluential
hasbeenthemodelproposedbyAtkinsonandShiffrin(1968)(seeChapterS).
18 CHAPTER 2 A BRIEF HISTORY OF COGNITIVE PSYCHOLOGY

units (chunks): 'THIS SENTENCE IS EASY TO lvlElvlORISE'. Another imporlant publication n'as
the book published by Bruner et ul. (1956), which systematically investigated horv people forn.r
concepts, and highlighted the role of strategies in thinking'

Stop and fhlnk Cognitive revolution -'''

a're:vet-:: il(,''g I-1,3:a :i i?15 1i: f,i-.' .ad'ar' 15 r-1i13i1r-; i:a .j-i
i:.j\v:\/: 1;1"': il\,e a j.J sa:;l 1;:3. -,'r1;i' .:saa'ar :rl?i.l -'3ii- J,:)i'c a)r i-"3'-'
t.t'ii,e. ,S :-3 rJal q;i.i ;;;qi, -;t,c i'ai,,ll'-ll::rl" j-::: iry;-r; iiie ::a lV A ;"'ir:
a::;r: ,;i 3' 13'r2i;i.-;.,;r;i-lt -r' .l:. :l^':-' -g g= - le',r'll:iti?

Cognitive psychology comes of age

The developments lve have just described clearly jolted psychologists. But one had to wait until
nearly ten y.urs afte. the seminal Dartmouth and MIT conferences to see the new ideas firmly
crystallised, with the publication of the first textbook on cognitive psychology by Ulrich Neisser
(tSSl) . Reflecting the interests of psychology at the time, Neisser's book focused on perception
and attention, and devoted relatively little space to memory, language and thinking' It is also
interesting to note that Neisser, although using an information-framework approach, was fairly
critical about the possibility of simulating mental Processes with computer programs, as he
argued that mental processes were too complex to be explained by simple mechanisms' This
an-itude is still widespread and many cognitive psychologists, while Payrng lip service to the
information-processing framework, are more reluctant when it comes to using computer Pro-
grams as models of specific phenomena (Ericsson and Kintsch,2000). Indeed, it is fair to say
ihat the majority of cognitive psychologists would define their field as the use of elegant and
sophisticated experimental paradigms to answer specific questions about the mind and/or brain,
rather than the development of theories implemented as computer programs.
Since the early 1960s cognitive psychology has flourished. Numerous experimental paradigms
have been created, and numerous theories have been developed to explain the large amount of
empirical data collected. The field has rapidly divided into several subfields (see Chapter l), to the
extent that today a cognitive psychologist interested in language hardly speaks to a colleague inter-
ested in, say, visual perception, Cognitive psychology has also increasingly influenced other fields
of psychology. For example, the concept of a schema, which was developed to explain phenomena
in memory and language, started being used to explain depression in clinical psychology (Beck,
1975), stereotypes in social psychology (Brewer and Hewston e,2004;see also Chapter 18), and the
way schoolchildren learn in educational psychology (Siegler, 1986)'
Cognitive neuropsychology has become an important subfield of cognitive psychology.
Here, fatients with biain lesions are studied with the hope that the pattern of cognitive deficits
they display will shed light on normal cognitive mechanisms. (We will consider examples of
cognitive neuropsychology in Chapters 6 and 9')
On the theoretical side, two concepts have been highly popular: semantic networks and pro-
duction systems. With semantic networks, knowledge is represented as a network of nodes
connected by links. The nodes stand for concepts, and the links stand for relationships berween
concepts. The links are labelled, which means that the type of relationship is explicitly repre-
sented (see Figure 2.1). Semantic netrvorks nicely capture several aspects of memory. One of
 COGNITIVE PSYCHOLOGY COfuIES OF AGE 19

HAS
animal ski n

rs-A

CAN SWrm

Figure 2.1 A simple semantic network

The nodes stand for concepts, and the links stand for relationships between concspts

them is priming, As rve shall see in Chapter 7, one important property of memory is that it is
associative. Thus, rvhen one node in memory is activated, this activation tends to spread to
other nodes (see Box 2.3).
Production systerns consist of rules called 'productions', rvhich are stored in long-terrn
rnemory. A produaion has two parts: a conditiotl (the IF-statement) and an action (the THEN-
statement). lVhenever the condition is satisfied, the action is executed. A simple example of a pro-
duction is: 'lF the light is red, THEN stop,' Figure 2.3 shows a more complicated example. The
content ofworking rnemory (another word for shcirt-term memory) is used to select productions.
This basic explanation omits a number of complications: for example,
',vhat happens when the con-
ditions of several productions are satisfied (see Klahr et el., 1987, for details).

A nice example of the idea that activation flows between nodes in a semantic network is
provided by priming experirnents. ln a typical priming experiment, strings of letters are briefly
presented on a computer screen, and participants have to decide whether they form a word
(e.g. 'fish') or not (e.g. 'xbyp'), by pressing one of two keys as rapidly as possible. (This task is
called a lexical decision task.) The target word is preceded by the very rapid presentation of
another word, called a prime, which might or might not relate to the second word (see Figure
2.2, top). Numerous experiments have shown that the decision is faster when a related word
is presented first (e.g. 'animal' before 'fish', rather than 'chair' before 'fish'). This can be
explained by the idea that when the prime ('anima[', in our example) appears on the screen,
its representation in memory is activated, and activation then spreads to concepts that are
related to it ('fish' and 'skin' in Figure 2.2, bottom). Given that the activation of the node
'fish' is higher than normal when the word is presented on the screen, the time to make a
decision is faster (see Meyer and Schvaneveldt, 1971 , for an example of the application of
this paradigm).
 CHAPTER 2 A BRIEF HISTORY OF COGNITIVE PSYCHOLOGY

animal 500 ms (prime)

500 ms (blank)
fish 5@ ms (target)

HAS

(1) ls-A
swtm

CAN swtm
I
I

i
I
1

i HAS

CAN

I Flgure 2.2 Example of a priming experiment

:l Topr temporal structure ofthe experiment. Bottbm: when the node 'animal, is activated, activation propagates
first to th€ nodes a-djaqent to it, and tften. more weakly, to the nodesone node.away.
ri
il
ri

li
I

1
Production systems capture some important aspects of cognition, and we will
I
meet several
I of them in this book' One useful properry of productions is that they are modular
can add or delete a production without affecting other productions, although
- that is, one
the overall
behaviour of the system may change. Production systems capture rule-like behaviour,
and have
for instance been used for simulating the errors that children make when learning
about
 CONNECTIONISM STRIKES 2L

IF

expression has variable term on right side

THEN

subtract variable term from both sides, and simplify

2x+4: x+2
2x+4= x+2 - x

x+4=2
Figure 2.3 Production
Top: example ofa production used for solving algebra problems. Bottom: example ofthe application ofthis
production. The equation in the first line has a variable on the right side. Therefore, the production is applied (second
line). The third line shows the outcome after simplification.

subtraction (Young and O'Shea, 1981), In general, they have produced good models of learn-
ing, including how people acquire Procedural knowledge (i.e, knowledge about rules).

Stop and Think High school

i:le:r:WOltli:j:haiyOL,Sirdieil 11 irrgir5;r'31.Tlr,nrlat3.\t:'iClr1afl,:l .-:'inSyslef'rliiaa
)j.''ai,:ri: irellvlf ( liesctile wtal ycr,j irave teaiirea?

Stop and lhlnk Semantic networks vs production systems

WnSI a;'e tie itt'e,ences tettv:e r ;:r.ani.,- r:-,/r':j / j a,.a f,r'lilr,cirorr 5v;13r-r;l \A/la: a:t -'rpi- ir-;;;-- t.73

lJVaii.ag:5 3-,r 3i;1;i,rir';2gg5 2

Connectionism strikes
The computational models we have considered so far have used fairly high-level rePresenta-
tions, such as productions, to simulate cognition, Why not develop models using our know-
ledge of the brain, in particular the fact that a large number of neurons 'work' in parallel (see
Chapter 3)? In fact, a stream of research has attempted to do so. For example, Rosenblatt (1958)
carried out a number of experiments where kinds of neural net known as perceptrons learned to
recognise patterns (e,g. letters; see Box 2.4). This type ofresearch was brought to an abrupt end
by Minsky and Papert (1969), who showed that there were serious limitations to what could be
learned by perceptrons. Connectionists had to wait until the 1980s to see the development of
models that overcome these limitations. The magic bullet was a learning algorithm called back'
propagation (Rumelhart et al., 1986), which was more sophisticated than that used by the per-
ceptron. The presence of an algorithm powerful enough to carry out in principle any type of
learning led to a resurgence of connectionism, which has since been used to simulate phenom-
ena in many domains, including perception and the acquisition of language. This has also led to
sometimes heated debates (e.g, Pinker and Nlehler, 1988) as to what is the best way to model
cognition - Nervell and Simon's symbolic approach or connectionism? Increasingly, researchers
have come to the conclusion that both approaches are useful in different lvaysl the symbolic
22 CHAPTER 2 A BRIEF HISTORY OF COGNITIVE PSYCHOLOGY

Box 2.4 RESEARCH CLOSE-UP: Connec,tionist networks

analogy the way

Connectionist networks, also called artificiat neural networks, take as an
(lve will consider biological neurons in the next
neurons transmit information in the brain.
artificial neural networks are much simpler than
chapter, where it will become apparent that
network consists of nodes (also called unlfs or
biological neural networks.) A connectionist
links. Nodes have activation levels, and links have weights, which will
neurins),connected by
weights of the links can
determine how strongly activation will propagate to the next node' The
be changed by learning mechanisms, of which back-propagation is one
of the most successful'
was used by Rosen-
Figure 2.4 illustrates a simple connectionist network, of the type that
UtattitgSa) in his seminal study. The input layer receives activation from the input, which is
The input layer then sends informa-
often coded as binary numbers (i.e. using only 0s and 1s),
that the of activation of the input
tion to the output layer; here, it is important to note level
the input and output nodes' The level
nodes is modulated by the weights ofthe links between
computed: for example, by taking the sum of the
of activation of the output nodes is then
rule is used to transform the activation level ofthe output
activations they receive. Then, a
nodes into an output. For example, in Figure 2.4 the output node with the highest

lnput layers Output layers Output

Target outPut
lnput
I
I I

I I &
ffi

Connections

Figure 2.4 lllustration of a perceptron

propaSates from this layer to the output
The input, coded in a binary fashion, reaches the input layer' Activation
on the activation ofthe input nodes and the weight ofthe links between the input and output
layer, depending
the output. The perceptron then
layers. The node in the output layer with the highest activation determines
compares the output with the target output, and ifthey are different, the weiShts ofthe links betlveen the
r nodes in the input layer and those in the output layer are rnodified'
 THE ASCENT OF COGNITIVE NEUROSCIENCE 23

activation wins and the outputto which the node is attached is selected. More complex net-
works have sets of nodes between input and output nodes. These nodes are called 'hidden
units' and considerably increase the kinds of behaviour the network can simulate.
With many learning algorithms, the output selected by the network is compared with the
correct output, lf it is not the correct one, some weights in the network are changed so that
the correct output is more likely to be selected in the future. The details of the learning alg+'
rithms rapidly become complicated, and you are referred to more advanced texts, such as
those mentioned in the Further reading section, for more information.

approach to explain high-level aspects ofcognition such as problem solving, and connectionism
to explain lower aspects of cognition such as perception. Connectionism, with its emphasis on
neuron-like processing, paved the way to the next upheaval in cognitive psychology and cogni-
tive science: cognitive neuroscience.

The ascent of cognitive neuroscience

While conneclionism questioned the theoretical foundations of cognitive research, cognitive neu-
roscience brought about crucial changes in the way empirical data are collested to study cogaition.
Until the early 1990s cognitive psychologists' main measures were reaction times, percentages of
correst answers, eye movements, r)?es of errors, and verbal protocols. It is true that data about the
brain were used in cognitive neuropsychology, and that some experimentalists were using event-
related potentials. However, data about the brain and its activation were not seen as mainstream
by the majority of cognitive psychologists. All this has changed dramatically in the past two decades.
Staggering technical breakthroughs enabled the development of several brain-irnaging techniques
that make it possible to look at changes in brain activity in real time (see Chapter 3). First, Positron
Emission Tomography (PET), next, funstional Magnetic Resonance Imaging (fuIRI), and then a
number of other techniques have mushroomed, to the point that the use of brain-imaging data has
now become routine when running experiments in cognitive psychology. (Electroencephalogra-
phy (EEG) and event-relatedpotentials, which had been around for more than 50 years, have also
enjoyed renewed interest recently due to the wide availability of computers powerful enough to
analyse the huge amount of data recorded.) The new brain-imaging techniques have also empow-
ered cognitive neuropsychology, as it is now possible to measure the extent of brain lesions pre-
cisely without having to wait for the patient's death to carry out a post-mortem analysis,

Stop and Think Link between cognitive psychology and technology

rave 5een !-nary exarr,ples of low iew Ceveiopinenis ir :ngirreerilg arJ technOlogy tave 3.iii'ed
\y'/e
icrwar4:he fiei,J oi cogn i',ive ps;,cl.oiogy. r-r.osi r-,otably the acvenr cf lne c isrral corn ouler an'i b iai n'
,r.agrng merhods. is ,ia gcod ^'. 516 1l-rirg? Whartecnnoiogy'rviil or'ing iie next
'ievol!tion' ir ccgliii'le
rsycnoiogy?

In spite of these impressive technological advances and the large quantity of experiments
that have been carried out on cognition using these techniques, the impact on cognitive theories
has been less clear. The dream ofcognitive neuroscience is to realise a double integration (see
Figure2.5): horizontally,betweenthedifferenttypesofcognitivefunctions(perception,memory,
24 CHAPTER 2 A BRIEF HISToRY oF CoGNITIVE PSYcHoLoGY

Neurotransm itters

Neurons

Clusters of neurons

Erain areas

c>0J
.o39P '=u
= lq
ddgo
cEh
u-J E6
u'6
o-E c
uo
(u
+i
o
.c
ut

Figure 2.5 Double integration, the dream of cognitive neuroscience

A general theory should explain both our diveFe cognitive abilities (horizontal axis) and the way these abilities are
implemented in the brain, not only at the level of broad regions, but also at finer levels of granularity (neurons and
neurotransmitters).

language, thinking and so on) and the brain areas they engage; and vertically, by linking proc-
esses at the neuronal level (or even the levels below) to high-level processes such as recognising
a Picasso or understanding language. It is likely that such integration will require much more
powerful theories and computational models than those that are currently available. This issue
of integration will be taken up in Chapter 3.

The search for unified theories of cognition

One possible way forward to address the issue of large-scale theoretical integration is offered by
the idea of unified theories of cognition (UTC), originally proposed by Newell (1990). As you
can see in Box 2.5, one important properry of good theories is that they can be refuted by experi-
ments. Newell noted that cognitive psychologists have focused on falsification, by designing
experiments aimed at refuting current theories, and, unlike physicists, have not put much empha-
sis on building theories that account for a large amount of data. This has resulted in a large
number of small-scale theories, which sometimes explain only the results of a specific experi-
mental paradigm. For example, a theory might be developed to explain how boys (as opposed to
girls) learn to read in Chinese (as opposed to English). Newell argued that cognitive psychologists
should try to construct more ambitious theories that account for a large number of phenomena.
Anomalous data not explained by the theory should be used as cues suggesting how to improve
it. Newell's central insight was that, as a theory explains an increasingly larger number of phe-
nomena, these phenomena offer so many constraints that the theorist is almost forced to develop
a correct theory. Given the large amount of data to simulate, and presumably the complexity of
the processes involved, such a theory must be implemented as a computer program.
 THE SEARCH FOR UNIFIED THEORIES OF COGNITION 25

Box 2.5 lN FOCUS: What is a good theory?

Scientists generally agree that a number of properties should be met for a theory to be con-
sidered as viable. First, a theory should betestable, or'falsifiable'(Popper, 1959) - that is,
experiments should exist that in principle could lead to results showing that the theory is
incorrect. Second, a theory should be internally cons,stent - it should not contain any contra-
dictory statements. As an example of theory that is not consistent, consider a theory claiming
both that (a) short-term memory is limited to seven items, and (b) short-term memory is not
limited in its capacity. The problem is that an inconsistent theory can generate any prediction
and its opposite, which makes it in fact non-testable. Third, a theory should be fully specified:
that is, all key terms and processes should be defined. When this requirement is not satisfied,
it is too easy - but scientifically unproductive - to explain any data inconsistent with the
predictions by adding ad hoc hypotheses, which in fact means, again, that the theory cannot
be tested. Now, the bad news is that most theories in (cognitive) psychology are informal and
rather vague, and thus flout this third requirement. The final requirement is that a theory
should be generalisable, and thus should be applicable beyond a specific set of experiments.
A weakness of cognitive psychology is that numerous theories cannot really be generalised,
as they aim to explain only the results of a narrow set of experiments (Newell, 1990).

Stop and Think Good theory

, ]j raf:/',,,.'a\i,r,l :::i''i.lr)i'irSi-r3aa-''r,';1 1-313,:.1',t; -.1 ' 3)l.a-11 :.1 ,a::::-jtr1/ia:::(::'. -

..i: ;i;e . -;.: C: ::r ,i :. t:.';it' iy. aJ'. j, jI:t-r:ri St:i- ;la-.i ti i:.,1 gei'eia', .:'.: i:1
-ar
:lj \r/i-rai eiija: altr:j a>-i-;iirqy :a. r'y l::j:,-iri3iia?

Let us consider an example. A first researcher develops a theory of memory, and manages to
estimate two parameters: the capacify of working memory, and the time to create a nerv node in
long-term memory. A second researcher develops a theory of problem solving in arithmetic.
Although she uses a parameter for the capaciry of rvorking memory, she is not arvare of the vrork
of the first researcher, and can essentially freely vary the value of this parameter to fit the data.
Now, consider a third researcher, who is following the UTC approach, and rvho is interested in
both domains. Once the capacity of working memory has been estimated with the simulations
on memory, it is not a free parameter any more when doing the simulations on arithmetic, If
the first simulations have shotvn that the capacity of working memory is seven, then this value
should be used in the simulations about arithmetic, and indeed in all subsequent simulations.
This is a porverful idea, but not an easy one to carry out in practice. At the moment, there are
just half a dozen UTCs, and r,ve briefly review three of them. Newell (1990) developed a program
called Soar, rvhere cognition is essentially seen as problem solving, Soar is implemented as a
production system, and has simulated numerous phenomena about learning, problem solving,
categorisation and natural language understanding. CHREST (Gobet et i,tl,,200l) emphasises
the role of perception in cognition; knorvledge is implemented as the grorvth of a network of
perceptuai chunks, which are linked to possible actions. Simulations inciude ho.,r'people become
experts in chess and physics, and how children acquire their first language. The most sophistic-
ated UTC is currently ACT-R, developed by Anderson and his group (Anderson er ol,, 2004;
 26 CHAPTER 2 A BRIEF HISTORY OF COGNITIVE PSYCHOLOGY

Anderson and Lebitre, 1998). ACT-R is a production system, and the essential aspects of cogni-
tion are seen as the activation of production rules and declarative elements in long-tenn
memory. The simulated phenomena range from simple perceptual and memory tasks to solving
algebra problems and driving a car. A particularly interesting feature o[Anderson's work is that,
tvhile Soar and CHREST have mostly simulated data about performance, errors and response
times, ACT-R has also simulated brain-imaging data, thus building a fascinating link with cog-
nitive neuroscience.

Chapter summary
lVhile psychology as a scientific field was founded in 1879, modern cognitive psychology was
founded in 1956. It is now a mature field of research, and there is no doubt that it can boast
tremendous achievements. Sophisticated experimental techniques have been developed, and
key cognitive concepts, such as production systems and semantic netlvorks, have been identi-
fied, Powerful theories have been developed, such as Anderson's ACT-R, which can make
detailed predictions about behaviour in numerous domains. Finatly, cognitive psychology has
led to the development of practical applications in fields such as education and even
psychotherapy.
In recent years, there has been a trend towards an increased partnership between cognitive
psychology and neuroscience, as exemplified by simulations using artificial neural networks and
the use of brain-imaging techniques. As is perhaps typical of mature sciences, cognitive psychol-
ogy has split into numerous subfields. Interestingly, certain issues have shown little progress. In
spite of several attempts, cognitive psychology still lacks a unified theory, and some very old
questions are still debated today. These include the distinction between primary and secondary
memory, originally proposed by Iames in 1890, and the nature of consciousness, to which
Wundt devoted considerable attention in 1874.

Further reading
Gardner (1987) provides a lively history of the cognitive revolution, Broadbent (1958) wrote
one of the first books where the computer analogy is used systematically, Classic books on the
standard information processing approach include Newell and Simon (1972) and Lachman ef
al. (1979). The standard source for the connectionist approach is offered by the two volumes
edited by McClelland and Rumelhart (McClelland and Rumelhart, 1986; Rumelhart and
McClelland, 1986). Bechtel andAbrahamsen (1991) and ivlcleod, Plunkett and Rolls (199S)
provide a non-technical introduction. The recent developments in cognitive neuroscience are
covered by Bear et el. (2007) andGazzaniga et aI. (2009). Newell (1990) is the key reference
about the idea of unified theories of cognition. Christopher Green's Classics in the History of
Psychology (http://psychclassics.yorku.calindex.htm) contains many of the early works men-
tioned in this chapter.

It
 A brief introduction
to neuroscience
d
d There are two main types of cell in the brain: neurons and glial cells. Neurons process
and transmit information, and glial cells provide support for the neurons.

d tVithin neurons, information is transmitted electrically, through action potentials.

Between neurons, information is transmitted chemically by releasing
neurotransmitters in synapses, the small gaps separating neurons.
d the two hemispheres of the brain cornprise three main parts: the brain stem, the
midbrain and the forebrain. These parts are further divided into smaller structures.
d the cerebral cortex, a component of the forebrain, is the place where most high-level
functions take place. lt consists of four lobes: the frontal lobe, the temporal lobe, the
parietal lobe and the occipital lobe,
d lNhile its expianations must respect the constraints offered by biology, cognitive
psychology studies mental behaviour at a specific level - the cognitive band - which
has its own laws and principles.

lntroduction
The brain is one of the most complex objects in the known universe. It has 100 billion neurons,
and each neuron is on average connected to 7000 other neurons - that makes a stunning 700
trillion connectionsl These numbers are staggering, but a moment's thought will convince you
that all this complexity is needed. Not only is the brain at the basis of essential lorv-level biologi-
cal functions, such as monitoring heart rate, respiration and feeding, it is also required for

27
?8 CHAPTER 3 A BRIEF INTRODUCTION TO NEUROSCIENCE

behaviours key for sulival, such as perceir.ing, learning and making r.apid clecisions. It is also
the organ underpinning the abilities that are (perhaps) unique to humans: thinking, emotions,
consciousness and love. Given all this, it is not surprising that the brain is extremely demanding
with respect to energy: whilst accounting for only 2 per cent of boclyweight, the brdin consumes
about 20 per cent ofthe glucose used by the body.
The 1990s were designated the decade of the brain by former US president George H. W.
Bush, which was recognition of the incredible advances that had been made in our understand-
ing of the brain and the potential for future progress. Converging efforts from biochemistry,
biology' neuroscience and psychology, among other sciences, have led to stunning discoveries
about the workings of the brain and the mind. This has been made possible by numerous
technological breakthroughs, most notably with computers, microscopy and brain imaging.
Cognitive psychologists want to understand the mechanisms underpinning mental proc-
esses, and some of the answers inevitably lie in the brain. It is, therefore, important to have at
least some knowledge of the biology of the brain in order to understand how the mind works.
This is becoming increasingly necessary as the gap between cognitive psychology and neuro-
science is diminishing, as witnessed by the new discipline of cognitive neuroscience. This chapter
provides a brief introduction to neuroscience, with an emphasis on themes that are important
for the study of cognition.
The complexity of the brain not only resides in the number of neurons and the intricate
biochemical mechanisms that allow them to function, but also in how these neurons are organ-
ised in clusters, these clusters are organised in yet larger clusters, and so forth. It is probably the
case that, while the different levels of this hierarchy are fairly well understood, we still barely
know the mechanisms that relate the various levels together. In this chapter, we consider two
important levels in the hierarchical architecture of the brain: the cell level at the microscopic
level, and the structural level describing the main regions of the brain at the macroscopic level.
(A more advance treatment of vision will be provided in Chapter 16.)

The neuron
There are two main rypes of cell in the brain: glial cells and neurons. Glial cells, which include
astrocFtes and oligodendrocytes, have a number of functions: they supply nutrition to neurons,
repair them, provide structural support, help maintain homeostasis, produce myelin (see below),
and facilitate signal transmission within and between neurons. Glial cells constitute about 90 per
cent of the cells in the human brain, while neurons make the remaining l0 per cent.
Neurons come in several types, depending on their structure and their function. Among
many others, one can distinguish between sensory, motor and pyramidal neurons (see Figure
3' I ). In spite of their variety, all neurons have the same overall function: to process and transmit
information. As shown in Figure 3.2, which depicts the structure of the typical human neuron,
most neurons have three main components: body, dendrites and axons.
The body, also called soma,is between 5 and 100 microns in diameter. (A micron is one-
thousandth of a millimetre.) It is surrounded by a membrane and contains the cytoplasm, the
nucleus and a number of organelle s.The cytopla.srn is a jelly-like substance that is in continuous
movement, The nucleus contains the genetic code of the neuron, which is used for protein
synthesis' The organelle.s are essentiai for the metabolism of the neuron: they carry out chemical

l.- -: iilz*.:1 -:.9:i-::aj'

 THE NEURON 29

Dendrites

Peri
process

Axon
Direction
of impulse
process

Axon

(a) Multipolar (b) Bipolar (c) Unipolar

Figure 3.1 Different types of neuron

(a) Muttipolar neurons have many extensions from their body: one axon and several dendrites. Most neurons in the
brain are multipolar. (b) Bipolar neurons have two extensions: one consisting ofdendrites and one axon. They are
typical ofspecialised sensory pathways (e.g. vision, smell, sight and hearing). (c) Unipolar cells have a single extension
(an axon) from their body. Many somatosensory (touch, Pain, temperature) neurons are unipolar.
Sou/ce: Shier, Butler and Lewis, 2010.

synthesis, generate and store energy, and provide the structural support (a sort of'skeleton') for
the neuron.
Dendrites are branching structures that receive information from other neurons' Note that
dendrites only receive information from other neurons and cannot transmit any information to
them. Instead, they propagate information to the soma, which in turn propagates information
to the a,ron.
Axons, which can be from a ferv millimetres to 1 metre long, transmit information from the
soma to other neurons. They end with terminal buttons, which store chemicals that are used for
inter-neuron communication, as we shall see below. There are trvo t'?es of axon. Axons of the
first type - myelinated axons are covered by a fatty, white substance called myelin. Crucially,
-
as we shall see in the next section, there are gaps in the sheath of myelin at places called the
nodes of Ranvier. Myelin insulates the axon from its environment, and therefore makes electric
transmission more effrcient and faster. lvith this kind of axon, myelin is essential for proper
30 CHAPTER 3 A BRIEF INTRODUCTION TO NEUROSCIENCE

Dendrites

Nucleus
sheath
Cell
mem
Soma
(cell body) Node of Axon
Ranvier terminals
Axon

Figure 3.2 The components of a typical neuron

Source:5hier, Butler and Lewis, 2010.

electric transmission, and its loss leads to serious neurological diseases such as multiple scler-
osis' Axons of the second type are not covered by myelin and their electric transmission is
slower,
It is important to note that neurons are active even when they do not receive information
from other neurons. Neurons must feed themselves (through blood vessels), maintain physio-
logical Parameters within a certain range (homeostasis), and maintain their electrical equilib-
rium, which is essential for them to be able to transmit information.

Communication within and between neurons

Within-neuron electrical communication
Electric signals are used for transmitting information from the soma to the end of the axon. The
action potential, also known as nerve impulse or spike, is a brief electrical impulse that moves
along the membrane of an axon. It is initiated at the place where the axon leaves the soma, called
the 'axon hillock'. The potential of a neuron membrane is normally at
-70 millivolts. Signal
transmission is achieved by changing this potential. By altering the equilibrium of chemicals
(mostly sodium and potassium) within and outside the membrane, there is first a depolarisa-
tion, where the voltage is increased. This is followed by a hyperpolarisation, where the potential
is more negative than normal, and then a return to the
-70 millivolts resting potential. This
sequence lasts about 2 milliseconds.
lVith myelinated axons, the action potential travels continuously down the axon once
started, until it reaches a node ofRanvier. Then, the sequence ofdepolarisation, hyperpolarisa-
tion and back to resting potential starts again, initiating a new impulse. The electrical impulse is
thus propagated through the myelin from one node of Ranvier to the next one, Given that
transmission 'jumps' from one node of Ranvier to the next one, this method of conduction is
called'saltatory conduction' (from the Latin saltqre,to hop).
 COMMUNICATIOi\I Ui ITI-IIN AND BETVVEEI'I NEURONS 3I

Box 3.1 lN FOCUS: Reticularists vs neuronists

The idea that neurons are the building blocks of the brain and that they communicate th rough
synapses is now so well known that it may come as a surprise that it took decades of research
and bitter arguments to reach this conclusion. Two competing views were advanced during
the nineteenth century. The'reticularisls' (reticulum in Latin means network) proposed that
neurons formed a continuous network and were fused together by their neurites (axons and
dendrites). The analogy \,vas the blood-circulation system, in which arteries and veins form a
continuous network. This was seen as an important exception to the cell theory proposed in
1838, which saw cells as distinct elements. By contrast, the'neuronists'argued that neurons
were indeed distinct functional units, and that there were gaps between them. lt was difficult
to reach a clear-cut conclusion, as the then-available microscopes did not have the necessary
resolution. ln addition, neural tissues appear as a whitish, jelly-like substance when placed
under the microscope, and it was not possible to decide with any degree of certainty whether
there were gaps or not between neurons.
Critical progress was made by the development of stains - substances that colour different
parts of the neural tissue differently. One of the key contributions \^ras made by the ltalian
Camillo Golgi. The stain he developed, which is still used nowadays and appropriately bears his
name, enabled histologists (biologists specialising in the microscopic analysis of cells and tissues)
to clearly see the two main components of neuronsr the soma and the neurites. But the key data
were collected by the Spanish histologist Santiago Ram6n y Cajal, who patiently used Golgi stain
for over 25 years to identifi/ different types ofneurons and how neurons were connected, bring-
ing considerable evidence for the presence of gaps between them. This conclusion was later
unambiguously supported by electron microscopy in the '1950s. Golgi and Ram6n y Cajal shared
the Nobel prize in physiology or medicine in 1906, but the irony is that, while of course Ram6n
y Cajal supported the neurone doctrine, Golgi remained during his entire life a champion of the
reticular hypothesis. (For detail about this controversy, see Bear, Connors and Paradiso, 2007.)

irVith myelinated axons, transmission can reach 100 metres per second, and there is little
decay in the strength of the signal, In non-myelinated axons, the transmission loses its strength
due to interferences and leaks, and is also relatively slow. Note that the signal (impulse) is either
on or off; therefore, rvhat is being used to transmit information is not the strength of the signal,
but the rate of firing. Neurons typically can fire up to 100 impulses per second, although some
can fire up to 500 impulses per second.

Between-neu ron com mu n ication

The gap between the end ofthe axon ofa neuron and the dendrites ofanother neuron is called
lhe synapse, ot s)tn(tptic cleft. It is about 10-50 nanometres wide. (A nanometre is one-millionth
of a millimetre,) Once the action potential reaches the end of an axon, it releases chemical sub-
stances called 'neurotransmitters' in the synaptic cleft. Neurotransmitters can either e.tclte the
postsynaptic neuron (i.e, they increase the likelihood that this neuron generates an action
potential) or inhibit it (i.e. they decrease this likelihood). By extension, synapses can be either
excitatorT or inhibitory. There are over I 00 different t1'pes of neurotransmitter, which belong to
thlee main categories: an-rino acids, peptides and monoamines.

I
12 CHAPTER 3 A BRIEF INTRODUCTION TO NEUROSCIENCE

Two particularly important amino acids are glutamate and GABA (gamma-aminoburyric
acid). Glutamate modulates the threshold of excitation of neurons, and is implicated in learning
and memory. GABA inhibits action at receptors. The human brain uses about 100 different
types of peptides. Endogenous opioids, an important class of peptides, have two main functions:
they reduce pain and are implicated in reward/reinforcement mechanisms, which of course are
essential for learning. Note that drugs from the opiate family, such as opium, morphine and
heroin, stimulate endogenous opioids receptors, with the double effect that they act as analge-
sics and lead to dependence due to their strong effect on reward systems.
Some monoamines are particularly important for cognition . Dopamine modulates attention,
learning and movement. Serotonin affects arousal. Finally, adrenalin and noradrenalin (also
known as epinephrine and norepinephrine, respectively) play a role in alertness. Other neuro-
transmitters do not fit into the categories of amino acids, peptides and monoamines. An import_
ant example is acetylcholine' which plays an excitatory role in the brain, and is linked to arousal,
reward and learning.
Learning is thought to result from changes in the synapses between neurons, by a mechan-
ism called long-term Potentiation (LTP). The chemical details are complex, but LTP is essen-
tially the strengthening of the connection between two neurons by changing the chemicals in
the synapse. A key principle for LTP is Hebbian learning'neurons that fire together, wire
together' (Hebb, 1949). Recent studies also suggest that some learning is underpinned by the
growth ofnew synapses.

Stop and Think Brain design

imagine a (oerhaps not too) distanl future, where it is possibte to deveioo 'smart drugs' - irugs that
learning, memory and ihinki;rg You are the crrief,scientist of a company cJsrgning ir-lch drugs.
lmOrgye
wnat kind ciapproach co ycr: choose: (a) drugs that increase the nryelinisaticn of axonl; (b)d,-Lrgs tnai
tnci'ease ihe size of neurons; ft) Crugs lhat increase the number of neui'ons, oi-{d) drugs tnat affjct tre
quantity of soecific neurolransrniitei's? Deiend your choice(s) using what ycr; have leained in this
chaoter.

Brain anatomy
The brain consists of two hemispheres separated by the longitudinal fissure and connected by
the corpus callosum, a large bundle of more than 250 million axons. It can be further divided
into three broad parts: the brain stem, the midbrain and the forebrain. In turn, each part can be
divided into subparts, which again can be analysed more finely, and so on until one reaches the
level of fairly small networks of neurons, called nuclei.Indeed,, a quick look at an advanced text-
book on brain anatomy will reveal a dazzlingnumber of structures and substructures. We will
not go into such detail here, but rather present a simplified roadmap to the brain (see Figures
3.4 and 3.5 for a quick overview of the structures of the brain). When reading the subsequent
pages, keep in mind the following property of the brain: in general, the left part of the brain is
linked to the right part of the body and vice versa. So, for example, if you suffer damage to the
part of the leftbrain dealing with finger movements, you might lose control of the fingers of the
righthand.
 BRAIi\i ANATOMY 33 :

'Box 3.? RESEARCH CLOSE-UP: Brain-imaging techniques

ln the following chapters, we will occasionally come across experiments where imaging
methods are used for studying what parts of the brain implement specific cognitive functions,
or how rapidly the brain processes information. lt might therefore be useful to give a brief
introduction to these techniques here.
Electroencephalography(EEG), invented in 1929, has been used since the'1960s forstudy-
ing cognitive processes (Luck, 2005). Electrodes are placed on the scalp and record the stlm
of all the electrical activity of the millions of neurons located on the other side of the skull. A
variation of EEG, called event-related potentials (ERP), measures electrical activity while
stimuli are repeatedly presented. ERPs are useful for psychologists as they produce more reli-
able data than EEG alone. Finally, there is a long tradition of research in neuroscience where
electrodes are inserted in the brain of animals for measuring the activity of a single neuron
- a technique called single unit recording (Huettel et a\.,20041.
Functional Magnetic Resonance lmaging (fMRl) is used in psychotogy and cognitive neuro-
sciences to measure blood-related responses to neural activity in the brain. Because nellrons
need energy to work properly, we can measure btood activity, which itself reflects the supply
ofsugar and oxygen to neurons, to know which neurons are active,
Positron Emission Tomography (PET) is a technique based on the detection of a radioac-
tive tracer (positron-emitting particle), which constructs a threedimensional image of the
level of activity for the tracer. The tracer is a biologically active molecule that emits gamma
rays for a brief period of time. The molecule is selected with the aim of targeting a specific
physiological function. The advantage d this technique is that we can see where a specific
molecule acts in the body and pinpoint the extent to which a particular location is active.
Magnetoencephalography (MEG) is the magnetic equivalent of EEG. While EEG records elec-
trical activity on the surface of the scalp, MEG records the magnetic fields around the scalp.
Figure 3.3 shows the spatial and temporal resolution of the four techniques we have
described. You can see for instance that fMRl has a very good spatial resolution (less than

O.
Ll EEG
(L,
E PET
d
=E
c

ffi
6
E
o fMR
f 4
o
a, MEG
(! 2
(!
o. 0
0-3 00 102 103
Temporal resolution (in seconds)

Figure 3.3 Temporal and spatial resolution of common brain-imaging techniques

The horizontal axis shows the temporal resolution in seconds, and the vertical a-\is shows the spatial resolution
in millimetres.
 34 CHAPTER 3 A BRIEF INTRODUCTION TO NEUROSCIENCE

4 mm) compared to the other techniques, and thus is the best if one is interested in locating
\
) cognitive functions. On the other hand, EEG and MEG have good temporal resolution, even
though they lack special precision.
Researchers have recently developed methods to use EEG and fMRl simultaneously, and
so take advantage ofthe high temporal resolution of EEG and the high spatial resolution of
fMRI.

cotd
/,Spinal
Eain stem
y'uedulta
\Pons
\cerebettum

*-."**,(ld,J*fl'
Central
nervoW
system formation
-Midbrain Tectum Superior colliculi

lnferior colliculi

Figure 3.4 The main components of the central nervous system

See Figure 3.5 for the details of the telencephalon.

Caudate nucleus

ene$^1 Putamen
\ Globus pallidus

Hippocampus
/

Telencephalon Limbic system L^::*

- uuru
\otraaorv
\
\ uammitary bodies
\ cinru,.,u .on.,

./ . Frontal lobe

Cerebral cortex 4r"^roo,,oo"

tooe
\ruietat
\ Occipital lobe
Figure 3.5 The main components of the telencephalon (or cerebrum)

tt
 BRAIN ANATOMY 35

The brain stem

As its name indicates, the brain stem is located at the bottom of the brain (see Figure 3.6). It
connects other parts of the brain lvith the spinal cord" which sends motor commands to the
body, and receives internal and external se11sory information fiont it. Nloving up, the next
structure isthe rnedulla, rvhich relays information from and to the spinal cord, and is in charge
of vital autonomic functions such as respiration and swallowing. The pons relays messages tiom
the ntedulla further up in the brain, and controls sleep and arousal. it also serves to transmit
information betrveen the cerebellum and other parts of the brain. The cerebelilurn is a large
structure (about 10 per cent of the volume of the brain) that contains about half of the neurons
in the brain. Its role is to integrate various types of information, and it is in particularly impli-
cated in classical conditioning (a simple form of learning; see Chapter 7), motor control and
coordination, as well as language production.

The midbrain
The midbrain, also known as ntesencephalon, consists of hvo main ltructures - the tegmentum and
the tectum - together with a number of smaller structures. The tegmenturn comprises three main
parts. The red nucleus and the sttbstantia r;igrafulfiI important functions in the motor system. The
retiurlar fonntttion regulates arousal, attention and sleep. Damage to the reticular formation leads
to coma. The tectum consists of two main structures. The ntperior collicdiplay a role in reflexive eye
movements to sudden visual changes, and the infeior colliailiplay a role in auditory reflexes,

Midbrain

Brainstem Pons
Cerebellum
Medulla
oblongata
Spinal cord

Figure 3.5 The major components of the brain stem, with their main functions
Ihe spinat cotd t(ansmits sensory.motor information between the brain and the body. Tne medulla is a relay station
between the brain and the spinal cord, and is responsible for respiration and swallowing. The pons is in charge of sleep
and arousal, and is a relay station between the cerebellum and other parts ofthe brain. The cerebellum plays an
important role in classical conditioning, motorcontrol and coordination, and language production.
Source: Shier, Butler and Lewis, 2010.
Another random document with
no related content on Scribd:
Project Gutenberg™ eBooks are often created from several printed
editions, all of which are confirmed as not protected by copyright in
the U.S. unless a copyright notice is included. Thus, we do not
necessarily keep eBooks in compliance with any particular paper
edition.

Most people start at our website which has the main PG search
facility: www.gutenberg.org.

This website includes information about Project Gutenberg™,

including how to make donations to the Project Gutenberg Literary
Archive Foundation, how to help produce our new eBooks, and how
to subscribe to our email newsletter to hear about new eBooks.