Received: 1 October 2018

| Revised: 28 October 2019

| Accepted: 5 November 2019

DOI: 10.1111/jopy.12524

ORIGINAL ARTICLE

Are awe-prone people more curious? The relationship between

dispositional awe, curiosity, and academic outcomes

Craig L. Anderson1 | Dante D. Dixson2 | Maria Monroy1 | Dacher Keltner1

1
Department of Psychology, University of
California, Berkeley, Berkeley, CA, USA
Abstract
2
Department of Counseling, Educational Objective: Guided by a functional account of awe, we aimed to test the hypothesis
Psychology, and Special Education, that people who often feel awe are also more curious (Studies 1 and 2), and that this
Michigan State University, East Lansing,
relationship in turn relates to academic outcomes (Study 3).
MI, USA
Method: In Study 1 (n = 1,005), we used a self-report approach to test the relation-
Correspondence ship between dispositional awe and curiosity. In Study 2 (n = 100), we used a peer-
Craig L. Anderson, John M. Olin School of
report approach to test if participants’ dispositional awe related to how curious they
Business, Washington University, Campus
Box 1156, One Brookings Drive, St. Louis, were rated by their friends. In Study 3, in a sample of 447 high school adolescents we
MO 63130-4899, USA. tested if dispositional awe related to academic outcomes via curiosity.
Email: [email protected]
Results: We found that dispositional awe was positively related to people's self-
Present address rated curiosity (Study 1) and how curious they were rated by their friends (Study 2).
Craig L. Anderson, Olin Business School, In Study 3, we found that dispositional awe was related to academic outcomes via
Washington University, St. Louis, MO,
USA curiosity.
Conclusions: We conclude that among the seven positive emotion dispositions
Funding information tested, awe was related to unique variance in curiosity, and this link in turn predicted
John Templeton Foundation, Grant/Award
academic outcomes. This work further characterizes awe as an epistemic emotion
Number: 88210 and 95524
and suggests that activities that inspire awe may improve academic outcomes.

KEYWORDS
academic outcomes, awe, curiosity, dispositional positive emotion, peer-report

1 | IN TRO D U C T ION the world and pursue such curiosity in acts of exploration and
innovation. For example, the awe that Marie Curie felt toward
the radioactive subjects of her research, which glowed “like
One cannot help but be in awe when he con- faint fairy light” fueled the curiosity that led to the identifica-
templates the mysteries of eternity, of life, of the tion of new elements and a scientific career distinguished by
marvelous structure of reality. It is enough if two Nobel Prizes (Goldsmith, 2005, p. 96). For John Muir, it
one tries merely to comprehend a little of this was profound feelings of awe elicited by nature that propelled
mystery every day. Never lose a holy curiosity. him across the peaks and valleys of the Sierra Nevada and to
— Albert Einstein (1955, May). Death of a scientific discovery (Muir, 1894). The core themes in these
genius. LIFE, 38(18), p. 65. life narratives, captured by Einstein's quote above, suggest
that awe-prone people are highly curious and motivated to
An examination of people who have pushed the boundaries learn about the world in which we live. However, no empir-
of human knowledge suggests that awe, an emotion felt in ical work has systematically tested the relationship between
the presence of vast stimuli that transcend understanding awe and curiosity at the trait level and whether this rela-
(Keltner & Haidt, 2003), moves people to be curious about tionship holds when controlling for other positive emotions

762 | © 2019 Wiley Periodicals, Inc. wileyonlinelibrary.com/journal/jopy Journal of Personality. 2020;88:762–779.

ANDERSON et al.   
| 763

dispositions. In the present investigation, we aimed to provide & Lopez, 2007; Kashdan et al., 2004, 2009; Neff, Rude, &
the first evidence that dispositional awe is related to unique Kirkpatrick, 2007) and the State Trait Curiosity Inventory
variance in curiosity. Moreover, we tested the importance of (STCI; Kashdan, 2002; Kashdan & Roberts, 2004b;
this relationship by examining how dispositional awe and cu- Spielberger, Peters, & Frain, 1976). Further support for the
riosity relate to academic outcomes. link between curiosity and positive emotions is provided
by work that examined the zero-order correlations between
dispositional discrete positive emotions and the Values in
1.1 | What predicts trait curiosity? Action Inventory of Strengths (VIA; Peterson & Seligman,
2004), which includes curiosity (Güsewell & Ruch, 2012).
Described generally as the desire to learn and acquire new In this study, trait-level awe and five other positive emo-
knowledge (Kang et al., 2009), curiosity has been linked tions—amusement, contentment, compassion, joy, and
to benefits in several domains. For example, in terms of pride—were found to be positively correlated with curios-
intellectual ability, longitudinal studies have shown that ity (Güsewell & Ruch, 2012), although systematic analyses
curiosity predicts future cognitive and problem-solving parsing the unique relationships between awe or other dis-
abilities independently from baseline intellect (Raine, tinct positive emotions and curiosity were not conducted.
Reynolds, Venables, & Mednick, 2002; Reiss & Reiss, In sum, empirical evidence suggests that both Openness to
2004). Moreover, curious people feel more able to cope Experience and positive activation, generally construed,
with highly novel and complex stimuli, such as abstract are related to curiosity. Any attempt, therefore, to docu-
poetry (Silvia, 2008a). Curiosity has also been linked to ment the relationship between dispositional awe and curi-
well-being outcomes such as psychological well-being, osity must account for Openness to Experience and other
subjective happiness, meaning in life, and life satisfac- positive emotions to ascertain whether variance related to
tion (Gallagher & Lopez, 2007; Kashdan & Steger, 2007). awe is unique or part of a more general association between
Finally, curiosity has also been linked to social outcomes. positivity and curiosity.
Evidence suggests that highly curious people are more so-
cially competent (Renner, 2006), have greater social well-
being (Keyes, 1998), and engage in more prosocial behavior 1.2 | Awe and curiosity
such as the initiation of playfulness (Kashdan & Roberts,
2004b) and less interpersonal aggression (Kashdan et al., Awe is produced by a distinct set of appraisals related to
2013). the incongruity between one's knowledge and information
While many studies have documented the benefits as- in the external world. Specifically, the experience of awe
sociated with curiosity, relatively less work has focused on is produced by two central appraisals: perceptual vastness
what factors predict curiosity. One such factor is Openness and need for accommodation (Keltner & Haidt, 2003).
to Experience, a personality trait that describes “the breadth, Vastness signifies a departure from one's typical frame
depth, originality, and complexity of an individual's men- of reference and can be instantiated not only by physical
tal and experiential life” (John & Srivastava, 1999, p. 121). size, but also by temporal, cognitive, or social dimensions.
Openness has been consistently linked to curiosity mea- Empirical work has indeed demonstrated that experiences
sures with correlations ranging between r = .19 and r = .57 of vastness such as sitting beneath a replica Tyrannosaurus
(Kashdan et al., 2009; Kashdan, Rose, & Fincham, 2004; Rex skeleton (Shiota, Keltner, & Mossman, 2007), stand-
Kashdan & Steger, 2007; Mussel, 2010). ing in a grove of tall trees (Piff, Dietze, Feinberg, Stancato,
Positive affect has also been shown to relate to curios- & Keltner, 2015), or taking in the view from the top of a
ity. For example, studies guided by the Broaden and Build tall tower (Stellar, Gordon, Anderson, et al., 2017) reliably
theory (Fredrickson, 1998, 2013) have found that state- elicit awe. Moreover, cross-cultural diary data suggest that
like occurrences of positive emotions (e.g., serenity, joy, in their daily lives people can also feel awe toward stimuli
amusement), relative to negative emotions (e.g., anger, that are not visually vast, such as the great achievements of
sadness) and neutral states, promote behaviors related to other people and the acquisition of new knowledge (Bai et
curiosity such as a wider range of thoughts and behav- al., 2017). Accommodation, the other central appraisal that
iors (Fredrickson & Branigan, 2005; Johnson, Waugh, & leads to awe, is necessary when new information is not ac-
Fredrickson, 2010; Wadlinger & Isaacowitz, 2006). Indeed, counted for by existing mental schema (Piaget, 1970). That
the disposition to experience positively valenced activa- awe-eliciting stimuli need to be accommodated into mental
tion, as measured by the Positive and Negative Activation schema is also supported by previous work. For example,
Schedule (PANAS; Watson, Clark, & Tellegen, 1988), has when describing experiences of awe, people often refer to
also been shown to relate to trait measures of curiosity such stimuli that transcend existing categories, actions of indi-
as the Curiosity and Exploration Inventory (CEI; Gallagher viduals that violate expectations about what humans are
764
|    ANDERSON et al.

capable of, and express the need to understand, to inquire, systematic information processing and reduces reliance on
and to refine their knowledge about the world (Cohen, heuristics, making people more likely to reject weak persua-
Gruber, & Keltner, 2010; Shiota et al., 2007). In short, peo- sive arguments (Griskevicius, Shiota, & Neufeld, 2010) and
ple feel awe when they experience something vast that they less likely to remember false details when remembering a
cannot fully comprehend in the moment. story (Danvers & Shiota, 2016). Other work has shown that
We argue that these appraisals make salient the incongru- state awe leads to greater engagement in explanatory frame-
ence between one's existing knowledge and information in works such as religion and science (Valdesolo & Graham,
the environment, produce awe, and motivate curiosity and 2014; Valdesolo, Park, & Gottlieb, 2016). Such work has
exploration (Keltner & Haidt, 2003; Mcphetres, 2019). This led to recent theorizing that awe fosters scientific learning
functional analysis of awe converges with previous work sug- (Valdesolo, Shtulman, Shtulman, & Baron, 2017), which is
gesting that curiosity is triggered by information that makes supported by trait-level evidence showing that awe is posi-
the individual aware of gaps in existing knowledge (Kashdan, tively related to basic understanding of scientific processes
Sherman, Yarbro, & Funder, 2013; Kashdan, 2004; Kashdan (Gottlieb, Keltner, & Lombrozo, 2018). At the physiological
et al., 2009; Kashdan & Steger, 2007; Loewenstein, 1994; level, compared to other positive emotions like amusement,
Silvia & Kashdan, 2009). Not only do awe-eliciting stim- joyful enthusiasm, and love, awe was shown to be related to
uli represent a gap in existing knowledge by nature of their lower activation of the sympathetic nervous system as mea-
need to be accommodated, but also their vast nature makes sured by pre-ejection period (Shiota, Neufeld, Yeung, Moser,
them especially salient due to their departure from one's typ- & Perea, 2011). This pattern is consistent with the physiolog-
ical frame of reference. When faced with such vast gaps in ical orienting response (Graham & Clifton, 1966) and sup-
knowledge, one action tendency is to seek to close the gap ports that awe is linked to how people gather information.
by curiously gathering more information from the environ- A select literature has compared awe to other epistemic
ment (Loewenstein, 1994). Following from this conceptual or “knowledge” emotions (Silvia, 2010), particularly interest.
analysis, we propose that a primary function of awe is to pro- Interest has been described as a positive emotional state that
duce curiosity. is elicited by stimuli that are appraised as moderately novel
While the link between awe and curiosity has not been or complex, yet able to be understood (Silvia, 2005, 2008b).
rigorously tested to our knowledge, a growing body of work Evidence suggests a link between trait curiosity and state
suggests that awe directs attention away from the self, out- experiences of interest via these appraisals. For example,
wards to one's physical and social environment, a precursor people high in trait curiosity found complex poems easier to
to curiosity. Consistent with its theorized social function understand and thus, more interesting (Silvia, 2008a). Other
(Stellar, Gordon, Piff, et al., 2017), awe has been shown to knowledge emotions have been found to be predicted by the
orient people to the concerns of others, minimizing the self same two-dimensional appraisal space. Evidence suggests
in the process. For example, at the trait level, people high in that confusion, for example, is elicited when novelty and
dispositional awe have been found to be more generous in complexity is high and ability to understand is low (Silvia,
the sharing of resources with others (Piff et al., 2015). At 2010), and surprise is theorized to be elicited by stimuli that
the state level, inductions of awe have been shown to lead are very high in novelty (Silvia, 2009). The appraisals associ-
to more prosocial behavior than inductions of other posi- ated with awe may also fit into this two-dimensional appraisal
tive emotions. For example, after experiencing awe people framework, although to our knowledge this has not been em-
showed more ethical decision-making and generous behavior pirically tested. For example, vastness describes stimuli that
than people in pride and amusement conditions, effects that are high in novelty and complexity and need for accommoda-
were explained by feelings of the diminished sense of self tion can be conceptualized as falling just outside one's ability
(Piff et al., 2015). Similarly, experimental manipulations of to understand, but not far enough to elicit confusion.
awe have been shown to lead to more humble behavior than While awe shares features with other knowledge emotions
neutral conditions (Stellar, Gordon, Anderson, et al., 2017), such as similar appraisals and links to curiosity, empirical stud-
higher willingness to donate time compared to a happiness ies have shown that awe has distinct patterns of expressive sig-
condition (Rudd, Vohs, & Aaker, 2012), increased feelings of naling, core relationship themes, appraisal components, and
the small self and collective orientation compared to amuse- effects on attentional scope. For example, in a study of 2,600
ment and neutral conditions (Bai et al., 2017), and higher facial expressions across five cultures, expressions of awe, and
connection to others versus an amusement condition (Van four other knowledge emotions—interest, surprise, boredom,
Cappellen & Saroglou, 2012). and confusion—were found to have different international core
Aside from social effects, research has also linked awe patterns of facial muscle activation that were conserved across
with how people process information. For example, exper- all cultures (Cordaro et al., 2018). We also note that of the 22
imental work at the state level has shown that brief experi- emotions examined, expressions of interest showed the greatest
ences of awe compared to other positive emotions increases variability between cultures (Cordaro et al., 2018). A study of
ANDERSON et al.   
| 765

how emotions are communicated via voice also suggests that awe Affective traits have been conceptualized as “stable predispo-
and other knowledge emotion have distinct signaling patterns. sitions toward certain kinds of responding” (Rosenberg, 1998,
Across 10 globalized cultures and one remote village in Bhutan, p. 249). Given that state-level awe has been shown to pro-
people reliably and accurately decoded vocal bursts of awe, in- mote outwards focused attention and information gathering
terest, and surprise (Cordaro, Keltner, Tshering, Wangchuk, & (Danvers & Shiota, 2016; Gottlieb et al., 2018; Griskevicius
Flynn, 2016). Among the 10 positive emotions assessed, vocal et al., 2010; Valdesolo & Graham, 2014; Valdesolo et al.,
bursts of awe were accurately decoded at the highest rate, 71.8%, 2016, 2017), we argue that people who are disposed to feel
and surprise and interest also had high accuracy rates, 59.8% and awe more frequently and intensely would also be higher in
60.2%, respectively (Cordaro, Keltner, et al., 2016). Awe and in- trait curiosity. This is consistent with arguments suggesting
terest have also been shown to be associated with different core that state and trait-level emotion have similar influences upon
relational themes and patterns of appraisals. In terms of core re- cognitive processes (e.g., Lerner & Keltner, 2000, 2001;
lational themes, awe has been shown to make salient the small Oveis, Horberg, & Keltner, 2010; Rosenberg, 1998).
self more than interest, and interest was more strongly related While positive affective states generally have been associ-
to feelings of exploration (Campos, Shiota, Keltner, Gonzaga, ated with curiosity (Fredrickson & Branigan, 2005; Gallagher
& Goetz, 2013). Moreover, in a study of 13 appraisal dimen- & Lopez, 2007; Johnson et al., 2010; Kashdan et al., 2004,
sions, awe and interest were shown to differ on eight of them: 2009; Neff et al., 2007; Wadlinger & Isaacowitz, 2006), a func-
pleasantness, relevance, problems, self-agency, agency over tional analysis of the appraisals of awe suggests that it may be
circumstances, control of self, control of circumstances, and ef- related to unique variance in curiosity, above and beyond the
fort (Tong, 2015). Finally, while there is a dearth of evidence effects of other positive emotions. To stringently test this, in
comparing the effects of different knowledge emotions on atten- the current work, we tested if dispositional awe was related to
tion, one study showed that interest narrowed attention whereas unique variance in curiosity while controlling for six other dis-
awe promoted a more broad scope of attention (Sung & Yih, tinct positive emotion dispositions: amusement, compassion,
2016). Collectively, this evidence suggests that while awe shares contentment, joy, love, and pride. Data suggest that these pos-
features with other knowledge emotions such as interest, it is a itive emotion dispositions are related to personality facets and
distinct affective state. Taken together, the burgeoning science attachment styles in different ways. People who are high in dis-
of awe has linked the emotion to a diverse array of behaviors re- positional awe are prone to appreciating the beauty of vast and
lated to how people engage with their social and physical worlds. challenging stimuli such as nature, and tend to be highly open
to experience (Shiota, Keltner, & John, 2006). Dispositional
amusement reflects the tendency to find humor in situations
1.3 | Current investigation and the acts of other people, and has been found to be moder-
ately associated with Extraversion and Openness to Experience
The central focus of emotion science is to better understand (Shiota et al., 2006). Dispositional compassion describes the
how the emotions that people feel predict how they behave. tendency to notice suffering and feel rewarded by helping oth-
The growing science of awe has identified several behaviors ers and has been found to be strongly related to the personality
that awe seems to promote more than other positive emo- facet of Agreeableness and moderately related to Openness to
tional states such as helping behaviors (Piff et al., 2015; Rudd Experience (Shiota et al., 2006). People high in dispositional
et al., 2012; Stellar, Gordon, Anderson, et al., 2017), better contentment often feel satisfied by their life circumstances
understanding of and interest in science (Gottlieb et al., 2018; (Cordaro, Brackett, Glass, & Anderson, 2016) and tend to
Valdesolo et al., 2016, 2017), and spirituality (Van Cappellen, be higher in Extraversion and Conscientiousness and lower
Saroglou, Iweins, Piovesana, & Fredrickson, 2013). While at in Neuroticism and anxious attachment (Shiota et al., 2006).
face value some of these behaviors seem quite different, or Dispositional joy is conceptualized as the tendency to feel high
even antithetical in the case of religious and scientific think- arousal positive emotion associated with impending reward and
ing, a common thread is that they are all relate to information has been shown to be strongly associated with Extraversion
seeking and how people learn about their physical and social and moderately inversely correlated with Neuroticism (Shiota
world. Grounded in a functional account of distinct emotions, et al., 2006). In the current work, dispositional love describes
our thesis that awe promotes curiosity provides a framework the tendency to have warm and trusting relationships with other
that both unites the extant literature on the behavioral tenden- people generally and is not restricted to romantic relationships.
cies of awe and motivates the present work. Dispositional love has been found to be strongly related to se-
In this work, we aimed to demonstrate the relationship cure attachment style, moderately related to Agreeableness
between awe and curiosity and show that this link in turn is and Openness to Experience, and moderately and inversely
associated with academic outcomes. In pursuit of this aim, associated with Neuroticism (Shiota et al., 2006). The final
we took a trait-level approach that examined people's general positive emotion disposition assessed in the current work was
tendency to feel curiosity and different positive emotions. pride. Dispositional pride reflects the tendency to feel positive
766
|    ANDERSON et al.

self-regard, particularly relating to one's achievements, and has pooled, thus yielding a larage final sample that was rela-
been shown to be strongly associated with Extraversion, mod- tively diverse with regards to age, race, and socioeconomic
erately associated with Conscientiousness, and moderately and status (Buhrmester, Kwang, & Gosling, 2011). In all sam-
inversely related to Neuroticism (Shiota et al., 2006). Together, ples, people first completed demographic and personality
these patterns of findings suggest that these positive emotion measures via online survey software before participating in
dispositions are distinct by demonstrating that they are related an unrelated study. Data from samples were included in the
to personality facets and patterns of attachment in different current study if they assessed all of the following measures:
ways. the Curiosity and Exploration Inventory II (CEI-II; Kashdan
In the current work, we tested two primary hypotheses. et al., 2009), all seven subscales of the dispositional posi-
First, we expected that even when controlling for personality tive emotion scales (DPES; Shiota et al., 2006), the Positive
factors known to be related to curiosity such as Openness to and Negative Activation Scale, short form (I-PANAS-SF;
Experience and trait positive activation, as well as the dis- Thompson, 2007), and Openness to Experience as assessed
positions to feel amusement, compassion, contentment, joy, by the Big Five Inventory (BFI; John & Srivastava, 1999).
love and pride, awe would be related to self-rated (Studies All the data in the possession of the first author at the time
1 and 3) and peer-rated (Study 2) curiosity. Having demon- this manuscript was prepared that met these criteria were in-
strated that dispositional awe is related to unique variance in cluded in the current study. Demographic characteristics of
trait curiosity, we then tested this pattern's association with each individual sample as well as the combined sample are
academic outcomes. We predicted that dispositional awe displayed in Table S1 in the online supporting information.
would be indirectly related to several academic outcomes via
trait curiosity (Study 3).
2.1.2 | Measures

2 | ST U DY 1: D ISP O SIT IONA L Descriptive statistics for Study 1 measures by sample are
AW E A N D C U R IOS IT Y displayed in Table S2 in the online supporting information.
Consistent with best practice, the reliability, and structural
The purpose of Study 1 was to establish the relationship be- validity of all scale scores used in the current work were ex-
tween awe and curiosity at the trait level, above and beyond amined (Furr, 2017; Hussey & Hughes, 2018). Structural va-
other factors known to be associated with curiosity. While lidity was examined via a confirmatory or exploratory factor
initial work showed a zero-order relationship between dis- analysis of scale scores. In accordance with recommended
positional awe and trait curiosity measured by the VIA guidelines (Muthén & Muthén, 1998–2012), a weighted least
(Güsewell & Ruch, 2012), measures known to be linked squares means and variance adjusted (WLSMV) estimator
to both awe and curiosity such as Openness to Experience was used for analyses and multiple fit indices were used to
(Danvers & Shiota, 2016; Kashdan et al., 2004, 2009; assess the fit of the theorized model to the data. Generally,
Shiota et al., 2006) and trait positive activation (Gallagher indicators of good fit include a nonsignificant chi-square, a
& Lopez, 2007; Mussel, 2010) were not accounted for in comparative fit index (CFI) value that is greater than or equal
analyses. This leaves open the possibility that Openness to .90, a Tucker–Lewis index (TLI) value that is greater than
to Experience, for example, fully explains the relationship or equal to .90, a root mean square error of approximation
between awe and curiosity. We thus tested our hypothesis (RMSEA) value below .08, factor loadings greater than .40,
that awe would be related to unique variance in curiosity and reliability estimates above .7 (Costello & Osborne, 2005;
by conducting a series of increasingly stringent models that Dixson, Anderson, Rigney, Niemasik, & Potte, 2018; Kenny,
controlled for Openness to Experience, trait positive acti- 2015; Maccallum, Browne, & Sugawara, 1996; Marsh, Hau,
vation, and six other positive emotion dispositions: amuse- & Wen, 2004; Watkins, 2018).
ment, compassion, contentment, joy, love, and pride (e.g., However, two issues contraindicate strict adherence to
Gottlieb et al., 2018). these general guidelines in the current work. First, previous
research indicates that the chi-square is an inappropriate mea-
sure of fit for sample sizes over 400 because these models
2.1 | Method almost always have a significant chi-square (Kenny, 2015).
Second, previous research shows that the RMSEA will be ar-
2.1.1 | Participants and procedure tificially high for models with low degrees of freedom (i.e.,
<50, as is the case with all the measurement models included
Data from 1,005 people in the US from six separate sam- in this manuscript; Kenny, 2015; Kenny & McCoach, 2003).
ples representing both undergraduate students and people Thus, the RMSEA is not used as an indicator of good fit for
recruited using the Mechanical Turk (mTurk) platform were measurement models throughout this manuscript as they all
ANDERSON et al.   
| 767

have low degrees of freedom, while the chi-square is not used I am generally a contented person (contentment); I am an
as an indicator of good fit throughout the manuscript as all intensely cheerful person (joy); I develop strong feelings
samples are greater than 400. Therefore, also consistent with of closeness to people easily (love); and I am proud of my-
best practice, overall fit was judged holistically on the basis self and my accomplishments (pride). In the pooled data, all
of all of the above factors, with the conclusion being based on seven positive emotion scales were found to be reliable with
a preponderance of evidence. all alpha estimates being > .7 (see Table S3 in the online
supporting information). In addition, the scales were found
Trait curiosity to be structurally valid with a series of 1-factor CFAs reveal-
There are many ways to measure subtypes of curiosity such ing both acceptable factor loadings and fit indices (see Table
as perceptual (Collins, Litman, & Spielberger, 2004), social S4). Although the Love scale had a TLI that was below ac-
(Renner, 2006), epistemic (Litman & Spielberger, 2003), and ceptability, as a result of its CFI, reliability coefficient, and
information deprivation (Litman & Jimerson, 2004; Litman factor loadings, scores on this measure were considered valid
& Silvia, 2006). As this is the first attempt to systematically and reliable. The DPES scales were examined separately, as
test the association between dispositional awe and curiosity opposed to within a single model, in accordance with both
when controlling for other positive emotions, we decided to theory (Shiota et al., 2006) and previous research (Dixson
use a broad conceptualization of curiosity that captures the et al., 2019). We note that one item of the DPES awe scale,
tendency to engage in behaviors such as the external search I seek out experiences that challenge my understanding, is
for new knowledge and understanding. Thus, we selected the very similar to an item on the CEI-II, I am always looking for
CEI-II, a widely used measure of curiosity that operation- experiences that challenge how I think about myself and the
alizes trait curiosity as the tendency to seek out new infor- world. To ensure that findings are not being driven by these
mation and embrace the complexities of life (Kashdan et al., similar items, we also tested our hypotheses using a 5-item
2009). Participants rated how much each of ten items accu- awe scale that omits this item, and in presentation of results
rately reflected their general feelings and behavior. While the note how the results differ from the main findings presented.
authors of the CEI-II presented evidence that it captures two
different dimensions of curiosity, stretching and embracing Trait positive activation
(Kashdan et al., 2009), these dimensions are highly related Unlike the DPES, which assesses distinct positive emo-
and thus, we take an approach commonly used in the litera- tions, the I-PANAS-SF measures the tendency to experi-
ture and form a composite of all 10 items (e.g., Kaczmarek et ence general positively and negatively valenced activation
al., 2013; Kashdan, Afram, Brown, Birnbeck, & Drvoshanov, (Thompson, 2007). Participants were asked to indicate the
2011; Kashdan et al., 2013). Two sample items are: I actively extent to which they generally feel each of five positive and
seek as much information as I can in new situations and eve- five negatively valenced states on a scale from 1 (never) to
rywhere I go, I am out looking for new things or experiences. 5 (always). To measure general positive activation the five
In the current sample, CEI-II scores were found to be reliable items describing positive states were averaged: determined,
with an alpha estimate of .93. CEI-II scores were also found attentive, alert, inspired, and active. In the current sample,
to be structurally valid with the results of a 1-factor confirma- positively valenced I-PANAS-SF scores were found to be
tory factor analysis (CFA; see Table S4 in the online support- both valid and reliable with an alpha estimate of .79 and the
ing information) indicting sufficient factor loading and fit. results of a 1-factor CFA revealing acceptable factor loadings
Only a 1-factor structure of the CEI-II was examined here and fit (see Table S4).
and used in analyses.
Openness to Experience
Dispositional positive emotions Openness to Experience was measured by the BFI which as-
The DPES consists of 38 items that measure people's trait- sesses five broad personality dimensions (John & Srivastava,
level dispositions to experience seven discrete positive emo- 1999). We focused on the 10-item Openness to Experience
tions: awe, amusement, compassion, contentment, joy, love, scale because it has been found to be related to both dispo-
and pride (Shiota et al., 2006). These seven discrete emo- sitional awe (Danvers & Shiota, 2016; Shiota et al., 2006)
tion scales have been shown to be robust and valid (Dixson, and curiosity (Kashdan et al., 2004, 2009; Kashdan & Steger,
Anderson, & Keltner, 2019). Participants were instructed 2007; Mussel, 2010). Participants indicated how well each
to respond on a scale from one (strongly disagree) to seven statement reflected their personality using a scale from 1
(strongly agree) based on how well each item describes their (strongly disagree) to 5 (strongly agree). Example items
own tendencies. Sample items for each scale are: I feel won- from the BFI Openness to Experience scale include: origi-
der almost every day, I see beauty all around me, and I often nal, comes up with new ideas, and likes artistic and creative
feel awe (awe); I find humor in almost everything (amuse- experiences. In the current sample, BFI scores were found
ment); I often notice people who need help (compassion); to be reliable with an alpha estimate of .81. Lastly, although
768
|    ANDERSON et al.

the TLI of BFI scores was slightly below the threshold of compassion, contentment, joy, love, and pride, we concluded
acceptability, BFI scores were concluded to be structurally that awe had a significantly stronger relationship with cu-
valid on the basis of the acceptable CFI, factor loadings, and riosity than these other positive emotions. We note that in
reliability (see Table S4). this model, we also found that pride was positively related
to curiosity. In our third and most stringent model, we found
that when controlling for Openness to Experience, trait posi-
2.2 | Results tive activation, and all other positive emotion dispositions,
awe remained significantly and positively related to trait cu-
2.2.1 | Analytic approach riosity. In this model, both awe and pride were significantly
related to curiosity, but even though the 95% CIs of those
To test the unique variance in trait curiosity related to dispo- relationships overlapped, a z test comparing the coefficients
sitional awe, we ran three models. Given that Openness to (Clogg, Petkova, & Haritou, 1995) indicated that the relation-
Experience and general positive activation have both been ship of awe was significantly stronger, z = 2.60, p = .009.
shown to be related to curiosity (Gallagher & Lopez, 2007; Dispositional amusement, compassion, contentment, joy, and
Kashdan et al., 2004, 2009; Kashdan & Steger, 2007; Mussel, love were not related to curiosity in this model. Analyses run
2010), we first tested the relationship between awe and curi- with the 5-item awe scale omitting the item similar to one on
osity, controlling for both Openness and trait positive activa- the CEI-II yield the same pattern such that in the third model
tion. In the second model, to test the relationship between awe and pride are the only positive emotion dispositions that
awe and curiosity above and beyond the influence of other are related to curiosity. However, we note that in this case the
positive emotion dispositions, we simultaneously regressed relationship between awe and curiosity is weaker, β = .14, SE
all seven scales of the DPES on trait curiosity. Finally, in = .03, t = 4.14, 95% CI [.08, .21].
the third and most stringent model, we entered Openness to In sum, Study 1 provided initial evidence that disposi-
Experience, trait general positive activation, and all seven tional awe is associated with unique variance in trait curiosity
DPES scales as predictors in a regression with trait curiosity by demonstrating that awe was positively related to curiosity
as the outcome.1 This analytic plan distinguishes the current when controlling for Openness to Experience, general pos-
work from other research that has only examined the zero-or- itive activation, and six other discrete positive emotion dis-
der association of dispositional awe and outcomes (Güsewell positions. We also found a relationship between pride and
& Ruch, 2012; Piff et al., 2015) or controlled for positivity curiosity. Given that pride is often felt in response to gain-
using only the PANAS and joy scale of the DPES (Stellar, ing social status and acceptance (Tracy, Cheng, Robins, &
Gordon, Anderson, et al., 2017) by demonstrating the rela- Trzesniewski, 2009; Tracy & Robins, 2007), this pattern is
tionship between awe and curiosity above and beyond the consistent with previous research linking social status en-
influence of the other six positive emotion dispositions and hancement-related appraisals of social situations with trait
relevant traits measured. curiosity (Kashdan, Elhai, & Breen, 2008). However, we
note that the relationship between awe and curiosity was sig-
nificantly stronger than the relationship between pride and
2.2.2 | Main analyses curiosity.

The data and syntax used to generate these results can be

found at https​://osf.io/7ztpu/​. As responses to the CEI-II were 3 | STUDY 2: DISPOSITIONA L
assessed using a different number of scale points in differ- AWE AND PEER RATINGS OF
ent samples (see Table S2), data were z-scored within sam- CURIOSIT Y
ples before being pooled. Zero-order correlations between
measures are displayed in Table S3. We tested if disposi- In Study 1, we demonstrated in a large sample that when
tional awe was related to unique variance in trait curiosity in controlling for traits known to be related to curiosity such as
three increasingly stringent models, the results of which are Openness to Experience and positive activation, as well as
displayed in Table 1. In the first model, we found that after the disposition to experience amusement, compassion, con-
controlling for Openness to Experience and trait positive tentment, joy, love, and pride, dispositional awe was related
activation, awe was related to trait curiosity. In the second to unique variance in curiosity. However, these findings are
model, we found that when controlling for the six other posi- subject to the limitations of self-report methods (Paulhus &
tive emotion dispositions assessed by the DPES, awe was re- Vazire, 2007), most notably including self-perception and
lated to unique variance in curiosity. Moreover, given that the presentation biases. There is also conceptual overlap between
95% confidence interval (CI) of the relationship between awe awe and curiosity. For example, one usage of the English
and trait curiosity did not overlap with that of amusement, term wonder, a synonym for awe, can also be used to signify
ANDERSON et al.   
| 769

TABLE 1 Study 1 Regression models

β SE t p 95%CI
demonstrating the relationship between
Model 1 DPES-awe .35 .03 11.70 <.001 [.29, .40] dispositional awe and curiosity
BFI-open .32 .03 10.10 <.001 [.26, .39]
PANAS-positive .14 .03 5.36 <.001 [.09, .19]
Model 2 DPES-awe .43 .04 12.44 <.001 [.36, .50]
DPES-amusement .04 .03 1.59 .11 [−.01, .10]
DPES-compassion .01 .03 .23 .82 [−.05, .06]
DPES-contentment −.04 .04 −1.24 .22 [−.11, .03]
DPES-joy .04 .04 .95 .35 [−.04, .12]
DPES-love −.05 .03 −1.52 .13 [−.11, .02]
DPES-pride .23 .03 6.80 <.001 [.16, .29]
Model 3 DPES-awe .28 .04 7.76 <.001 [.21, .35]
DPES-amusement .03 .03 1.09 .28 [−.02, .08]
DPES-compassion −.02 .03 −.82 .41 [−.07, .03]
DPES-contentment .01 .03 .21 .84 [−.06, .07]
DPES-joy .02 .04 .50 .62 [−.06, .10]
DPES-love −.02 .03 −.47 .64 [−.08, .05]
DPES-pride .15 .03 4.65 <.001 [.09, .21]
BFI-open .32 .03 9.68 <.001 [.25, .38]
PANAS-positive .09 .03 3.32 .001 [.04, .14]
Note: BFI, Big Five Inventory (John & Srivastava, 1999); DPES, dispositional positive emotion scale
(Shiota et al., 2006); PANAS-positive, positive activation items from International Positive and Negative
Affect Schedule Short Form (I-PANAS-SF; Thompson, 2007). Outcome is trait curiosity as measured by the
Curiosity and Exploration Inventory II (Kashdan et al., 2009). Using a Bonferroni correction to adjust the
p value for multiple comparisons yields a p value of .006.

curiosity. To address these limitations, in Study 2 we used S1). We recruited exclusively college freshmen because the
a peer-report approach to test how participants’ self-ratings first year of college often involves novel intellectual and so-
of awe and other tendencies relate to how curious they were cial experiences, a rich time for both awe and curiosity to
rated by their peers. We used the same analytic technique be experienced and expressed. We sampled from the student
as in Study 1 to test the hypothesis that dispositional awe body at large, with only 8% of participants reporting psy-
would be related to unique variance in peer-rated curiosity by chology or cognitive science as their intended majors. Data
using increasingly stringent analyses to control for Openness presented in the current work overlap with that used by a pre-
to Experience, trait positive activation, and the six other dis- vious report of the relationship between awe and humility
positional positive emotion scales: amusement, compassion, (Stellar, Gordon, Anderson, et al., 2017).
contentment, joy, love, and pride. Peer-report findings that
converge with those from Study 1 would support the validity
of the observed link between dispositional awe and curios- 3.1.2 | Procedure
ity and suggest that the findings from Study 1 are not solely
being driven by participants’ biases or conceptual overlap Participants were recruited at the beginning of the semester
between terms. via flyers posted throughout the university campus. After
enrolling in the study, participants reported demographic in-
formation and completed personality measures. Participants
3.1 | Method then nominated four friends who attended the same univer-
sity to take a brief survey concerning questions about both
3.1.1 | Participants themselves and the participant. We contacted the first two
peers on each participant's list by email and offered them
One hundred and nineteen first-year undergraduate students $5 in exchange for completing the survey. Additional peers
were recruited to participate in a multiphase study on the sub- were contacted as needed if the first two did not complete
ject of awe in exchange for payment (see Sample C in Table the survey. A total of 193 peers submitted reports for 100 of
770
|    ANDERSON et al.

the participants enrolled in the study. When more than one to curiosity, β = .45, SE = .14, t = 3.31, p = .001, 95% CI
peer rated a participant, scores were averaged. The modal [.18, .72]. Moreover, none of the other six positive emotion
response to a question asking peers to report how long they dispositions were related to peer-rated curiosity, |βs| ≤ .14, ps
had known the participant using scale points that represented ≥ .29. Finally, in our third and most stringent model that con-
3-month spans was 6 to 9 months, which corresponded to trolled for Openness to Experience, trait positive activation,
how long the participants had been in college. and six other positive emotion dispositions, awe remained
related to peer-rated curiosity, β = .45, SE = .16, t = 2.90, p
= .005, 95% CI [.14, .76]. Again, no other dispositional posi-
3.1.3 | Measures tive emotion was significantly related to peer-rated curiosity,
|βs| ≤ .12, ps ≥ .3. When adjusting the p value to .006 due to
In addition to demographic information, Openness to multiple comparisons this pattern holds. Moreover, analyses
Experience (BFI), general positive activation (I-PANAS-SF), using the 5-item awe scale yield the same pattern of findings
discrete positive emotion dispositions (DPES), and trait curi- such that awe is the only positive emotion scale to be related
osity (CEI-II) were measured as in Study 1. Descriptive sta- to curiosity, although the effect is slightly weaker, β = .39,
tistics for these measures can be found in Table S2, Sample SE = .15, t = 2.60, p = .011, 95% CI [.09, .69].
C. In the current sample, the scores of all three measures These findings replicate and extend the self-report find-
were found to be reliable with alpha estimates > .7 (BFI ings from Study 1 by using peer-report methodology to show
= .76, I-PANAS-SF = .72, and CEI-II = .81). In addition, that dispositional awe related to peer-reports of curiosity
an exploratory factor analysis (because the sample was too above and beyond the effects of other positive emotion dispo-
small for a CFA; Kenny, 2015) indicated that BFI (.23–.78), sitions, trait positive activation, and Openness to Experience.
I-PANAS-SF (.46–.75), and CEI-II (.27–.76) scores were This converging peer-report evidence makes it less likely that
structurally valid with acceptable ranges of factor loadings. participants’ individual biases were solely driving the results
observed in Study 1. We note that the relationship between
Peer-rated curiosity self and peer-rated curiosity was relatively small compared
Peers reported on the curiosity of the participant using three to other published findings on self-peer agreement in per-
items adapted from the CEI-II (Kashdan et al., 2009): your sonality ratings (for recent examples see Kandler, Riemann,
friend is the type of person who really enjoys the uncertainty & Spinath, 2010; Kolar, Funder, & Colvln, 1996; Mõttus,
of everyday life, everywhere your friend goes, he/she is look- McCrae, Allik, & Realo, 2014; Soto & John, 2017). One ex-
ing for new things or experiences, and your friend views planation of this difference may be that in the current study
challenging situations as an opportunity to grow and learn. most of the peers had known the participants for less than a
Peers indicated how well they thought the statements de- year, whereas in previous work researchers typically recruit
scribed the participants on a scale from 1 (strongly disagree) peers that know participants well, such as romantic partners.
to 7 (strongly agree), and the three items were then averaged
(M = 4.80, SD = .97, α = .81).
4 | STUDY 3: DISPOSITIONAL
AWE, CURIOSIT Y, AND ACADE M I C
3.2 | Results OUTCOM ES

The data and syntax used to generate these results can be Together, Studies 1 and 2 show converging self and peer-
found at https​://osf.io/7ztpu/​. Measures were z-scored prior report evidence that dispositional awe is related to unique
to analysis to facilitate the contextualization of findings with variance in curiosity. In Study 3, we aimed to demonstrate
those from Study 1. Participants’ self-ratings of trait curi- the importance of this relationship by examining its con-
osity were significantly related to peer ratings of curiosity, tribution to academic outcomes in a diverse sample of
r(98) = .25, p = .011. Using the same analytic approach as in adolescents. Positive emotion dispositions in general have
Study 1, we used three increasingly stringent models to test been previously linked to academic outcomes (Dixson et
if participants’ dispositional awe was associated with unique al., 2018), but our aim in the current work was to dem-
variance in how curious they were rated by their peers. In the onstrate that the path by which dispositional awe relates
first model, controlling for Openness to Experience and trait to academic outcomes is via curiosity. Specifically, we
positive activation, we found that dispositional awe was posi- hypothesized that awe would indirectly relate to several
tively related to how curious participants were rated by their academic outcomes—work ethic, behavioral engagement
peers, β = .43, SE = .13, t = 3.36, p = .001, 95% CI [.17, .68]. in school, and academic self-efficacy—via curiosity. To
In the second model, when entering all seven positive emo- demonstrate that this relationship is not part of a broader
tion dispositions as predictors, awe was significantly related association between positivity, curiosity, and academic
ANDERSON et al.   
| 771

outcomes (Gallagher & Lopez, 2007; Kashdan et al., 2004, DPES-joy had a TLI statistic that was slightly below the
2009; Neff et al., 2007) in one model, we included dis- threshold for acceptability, it was concluded to be structur-
positional joy to control for general positivity, and in an- ally valid on the basis of the acceptable CFI, factor load-
other model, we included dispositional pride to account for ings, and reliability.
the close relationship between pride and curiosity found
in Study 1. We hypothesized that the indirect relationship Work ethic
between awe and academic outcomes via curiosity would Work ethnic was measured with the hard work subscale
remain even after controlling for joy and pride. (HWS) of the Work Ethic Scale (Blau & Ryan, 1997). This
6-item scale measures both one's philosophy on hard work
and one's belief in the utility of hard work (e.g., If you work
4.1 | Method hard you will succeed). Response options on this scale
ranged from 1 (Strongly disagree) to 7 (Strongly agree).
4.1.1 | Participants and procedure Higher scores on this scale indicate a higher willingness
to engage in as well as a higher expected payoff from hard
The sample consisted of 447 high school adolescents work. Previous studies have found HWS scores to be re-
(46.7% female) aged 12–19 (Mage = 15.8, SD = 1.28) at- liable (e.g., α = .85; Blau & Ryan, 1997). In the current
tending an urban high school in a Midwestern state. The sample, HWS scores were found to be both reliable and
mean grade point average of the sample was 2.88 (SD = structurally valid with an alpha of .90, and a 1-factor CFA
.75). The self-reported ethnic breakdown of the sample revealing acceptable factor loadings and fit indices (see
was 55.8% European American/White, 25.3% African Table S4).
American/Black, 5.7% Hispanic American/Mexican, 3.0%
Asian American/Pacific Islander, and 10.2% Multi-Ethnic/ Behavioral engagement
Other. In addition, the socioeconomic breakdown was self- Behavioral engagement was measured with a reduced ver-
reported as 9% Poor/Working Class, 15.5% Lower Middle sion of behavioral engagement subscale of the Engagement
Class, 54.2% Middle Class, 16.2% Upper Middle Class, versus Disaffection Scale (Skinner et al., 2008). This
2.1% Lower Upper Class, and 2.1% Wealthy. The current 4-item scale measures students’ willingness to partici-
study's data were a subsection of data that were collected as pate in academic courses (e.g., I pay attention in class).
a part of a school-wide improvement initiative to enhance Response options ranged from 1 (Not true at all) to 7 (Very
the school climate. The school collected data on psycho- true). Higher scores are indicative of both a higher behav-
social factors (e.g., school belonging), positive emotions ioral engagement in the past and a higher willingness to
(e.g., the DPES), academic perceptions (e.g., academic participate in academic courses in the future. Behavioral
self-regulation), and school personnel perceptions (e.g., engagement subscale scores have been found to be reli-
perceptions toward teachers). The survey was administered able in similar samples in previous research (alpha range
during the students’ free period, with an allotted time of of .71–.72; Skinner et al., 2008). In the current sample,
45 min to complete the survey online from a school com- behavioral engagement subscale scores were found to be
puter. All students who attended school on the two desig- both valid and reliable with an alpha estimate of .81, and a
nated survey days participated in the data collection (>95% 1-factor CFA revealing factor loadings > .50 and accept-
of total school population). able fit indices (see Table S4).

Academic self-efficacy
4.1.2 | Measures Academic self-efficacy was measured with the self-efficacy
for academic achievement subscale (ASE) of the Children's
The following measures were used in this study. Multidimensional Self-Efficacy Scale (Bandura, 1990;
Zimmerman & Ban, 1992). This 8-item subscale assesses
Awe, joy, pride, and curiosity how strongly students believe they can achieve in academic
As in Study 1, dispositional awe, joy, and pride were meas- subjects. The subscale is prefaced with the phrase “How
ured with the DPES, and curiosity with the CEI-II. In the well can you learn…” and the items consist of school sub-
current sample, the scores of all four scales were found to jects (e.g., general mathematics, algebra, science, biology,
have acceptable reliability estimates (DPES-awe, α = .76; and social studies). Response options ranged from 1 (Not
DPES-joy, α = .86; DPES-pride α = .82; and CEI-II, α at all) to 6 (Extremely well), with higher scores indicating a
= .90). Further, scores across all scales were found to be stronger belief in one's academic ability. Previous research
structurally valid with a series of 1-factor CFAs indicating in similar samples indicate that ASE scores are reliable with
acceptable factor loadings and fit (see Table S4). Although alpha estimates ranging from .70 to .81 (Dixson, Worrell,
772
|    ANDERSON et al.

Olszewski-Kubilius, & Subotnik, 2016; Zimmerman & Ban, had an acceptable CFI, TLI, and RMSEA, it was concluded
1992). In the current sample, ASE scores were found to be that the model had overall good fit. Further indirect analyses
both reliable and structurally valid with an alpha estimate of revealed that awe was related to academic self-efficacy (β =
.85, and a 1-factor CFA revealing acceptable factor coeffi- .503, SE = .040, p < .001), behavioral engagement (β = .315,
cients and fit indices (see Table S4). SE = .037, p < .001), and work ethic (β = .354, SE = .034, p <
.001), with the model as a whole accounting for 59.4% of cu-
riosity's variance, 3.7% of academic self-efficacy's variance,
4.2 | Results 16.7% of behavioral engagement in school's variance, and
21.1% of work ethic's variance. Taken together these findings
4.2.1 | Preliminary analyses support our hypothesis that dispositional awe is indirectly
related to academic outcomes via curiosity. Finally, the stan-
Means, standard deviations, and intercorrelations are pre- dardized beta loadings for awe items (β range = .433–.833),
sented in Table S5 in the online supporting information. curiosity items (β range = .578–.845), academic self-efficacy
Although no differences across study variables were found items (β range = .677–.838), work ethic items (β range =
for race, several differences were found for gender, all in .806–.879), and behavioral engagement items (β range =
keeping with the extant literature. Females reported signifi- .652–.918) were all acceptable within the model. A similar
cantly higher work ethic (p = .02, g = .23) and behavioral pattern of results was found using the 5-item awe scale.
engagement (p = .01, g = .25) than males, albeit all with To test that these findings are not part of a more general
small effect sizes. pattern between positivity, curiosity, and academic outcomes,
we then tested if this relationship held when controlling for
dispositional joy. We tested a second model that included a
4.2.2 | Primary analyses path from joy to the school outcomes via curiosity. The sec-
ond model started with both awe and joy, continued through
A series of structural equation models (SEMs) were used to curiosity, and concluded with academic self-efficacy, behav-
assess whether awe is indirectly related to the three academic ioral engagement, and work ethnic simultaneously. The results
outcomes assessed—behavioral engagement, academic self- of this model are presented in Table 2. Given that the CFI,
efficacy, and work ethic—via curiosity. Assessing these in- RMSEA, and factor loadings were acceptable and the TLI was
direct relationships via SEM has the advantage of decreasing very close to acceptability, it was concluded that the model ex-
measurement error and the capability to assess more compli- hibited good fit. Follow-up indirect analyses revealed that awe
cated models when compared to multiple regression, the tra- indirectly related to academic self-efficacy (β = .548, SE =
ditional method (see Baron & Kenny, 1986; Gunzler, Chen, .068, p < .001), behavioral engagement (β = .375, SE = .052,
Wu, & Zhang, 2013; Iacobucci, 2008). Consistent with best p < .001), and work ethic (β = .407, SE = .051, p < .001) via
practice, a weighted least squares mean and variance adjusted curiosity even after controlling for the joy to academic self-ef-
(WLSMV) estimator was used for analysis and multiple fit ficacy, behavioral engagement, and work ethic via curiosity
indices were used to assess the fit of the theorized model to pathways in the model. Importantly, in this model joy was not
the data (Muthén & Muthén, 1998–2012). Generally, indica- indirectly related to academic self-efficacy (β = −.035, SE =
tors of good fit include a nonsignificant chi-square, a com- .057, p = .544), behavioral engagement (β = −.024, SE = .039,
parative fit index (CFI) value that is greater than or equal to p = .543), and work ethic (β = −.026, SE = .042, p = .542)
.90, a Tucker-Lewis Index (TLI) value that is greater than or via curiosity. The model accounts for 58.6% of curiosity's
equal to .90, and a Root Mean Square Error of Approximation variance, 2.7% of academic self-efficacy's variance, 21.6% of
(RMSEA) value below .08 (Kenny, 2015; Maccallum et al., behavioral engagement's variance, and 25.5% of work ethic's
1996; Marsh et al., 2004). However, Kenny (2015) asserted variance. These results support our hypothesis that awe, and
that the chi-square is an inappropriate measure of fit for sam- not positive emotion in general, indirectly relates to academic
ple sizes over 400 because these models almost always have self-efficacy, behavioral engagement, and work ethic via cu-
a significant chi-square. Thus, the chi-square was not used as riosity. Finally, the standardized beta loadings for awe items
an indicator of good fit in the current analysis. (β range = .437–.766), joy items (β range = .637–.835), cu-
The first model started with awe, continued through cu- riosity items (β range = .561–.834), academic self-efficacy
riosity, and ended with academic self-efficacy, behavioral items (β range = .680–.837), work ethic items (β range =
engagement in school, and work ethic simultaneously (see .812–.879), and behavioral engagement items (β range =
Figure 1). Consistent with previous research (Dixson et al., .667–.917) were all acceptable within the model. A similar
2018) curiosity, work ethic, behavioral engagement, and aca- pattern of results was found using the 5-item awe scale.
demic self-efficacy were correlated in the model. Fit indices Due to the results of Study 1 indicating that awe and pride
for this model are presented in Table 2. Given that the model have similar relationships to curiosity, a third model was
ANDERSON et al.   
| 773

FIGURE 1 Panel A depicts standardized pathways from awe to academic outcomes via curiosity. Panel B depicts these effects controlling
for joy. *p < .001

TABLE 2 Fit indices for indirect models derived from structure equation modeling (WLSMV)

Model χ2 df CFI TLI RMSEA (90% CI)

Awe-curiosity-academic outcomes 1,781.49 519 .918 .911 .074 (.070, .078)
Awe/joy-curiosity-academic outcomes 2,382.57 730 .900 .894 .071 (.068, .074)
Awe/pride-curiosity-academic outcomes 2,457.56 692 .890 .882 .076 (.072, .079)
Note: Academic outcomes = academic self-efficacy, behavioral engagement, and work ethic; All p values < .001.

examined. The purpose of this model was to examine if the started with awe and pride, continued through curiosity,
indirect relationship between awe and the school outcomes and ended with academic self-efficacy, behavioral engage-
via curiosity held even after controlling for pride. This model ment, and work ethic simultaneously. Results of this model
774
|    ANDERSON et al.

are presented in Table 2. As can be seen, only the RMSEA enhancement behaviors and curiosity (Kashdan et al., 2008)
reached the threshold of acceptability. Given that more fit in- and consistent with the notion that people with high status
dices indicated poor fit than good fit, it was concluded that have more physical and social capital to support exploratory
the model had poor overall fit and no further interpretations behavior.
of this model were carried out. While we strived to demonstrate the unique variance in
trait curiosity that is related to dispositional awe by strin-
gently controlling for Openness to Experience, trait positive
5 | G EN E R A L D ISC U S S ION activation, and six other distinct positive emotion disposi-
tions, there are limitations to this approach. First, there may
The tendency to be curious and explore has been linked to be other meaningful predictors of trait curiosity that are also
better intellectual (Raine et al., 2002; Reiss & Reiss, 2004), related to dispositional awe. The absence of such predictors
personal well-being (Gallagher & Lopez, 2007; Kashdan & in the models tested could result in the relationship between
Steger, 2007), and social (Kashdan & Roberts, 2004a, 2004b; awe and curiosity being overestimated in the current work
Keyes, 1998; Renner, 2006) outcomes. However, beyond per- (Carlson & Wu, 2012). For example, traits that describe how
sonality factors such as Openness to Experience and general people seek out and persevere through challenging experi-
positive activation, little is known about what predicts curios- ences such as Sensation Seeking (Byman, 2005; Zuckerman,
ity. Guided by a functional account of emotions and analyses 1994), boredom aversion (Kashdan, McKnight, Fincham, &
of the appraisals that lead to awe, we hypothesized that among Rose, 2011; Loewenstein, 1994), and persistence in goal-di-
the seven distinct positive emotions examined, awe would be rected behavior (Kashdan et al., 2009) have been associated
uniquely related to curiosity. In the current work we used a with curiosity and may also be related to dispositional awe.
personality approach to test this hypothesis by examining the Emotion regulation ability has also been linked to curiosity
unique variance in trait curiosity explained by dispositional (Kashdan et al., 2009) and may help people high in disposi-
awe when controlling for Openness to Experience, trait posi- tional awe cope with the challenging nature of awe-inspiring
tive activation, and the disposition to experience amusement, stimuli (Gordon et al., 2017). While future work that assesses
compassion, contentment, joy, love, and pride. these measures may reveal a more nuanced understanding of
In Study 1, using a self-report approach, we showed that how dispositional awe relates to curiosity, such an approach
people who were higher in dispositional awe were on average has an inherent limitation that also applies to the current
higher in curiosity. In Study 2, we replicated and extended the work. Namely, interpretation of the relationship between two
findings from Study 1 using peer-report methodology, showing variables can become obscured as more covariates are added
that the higher participants were in dispositional awe the more (Carlson & Wu, 2012). While it is unclear what the disposi-
curious they were judged to be by their friends. Importantly, tional awe variable conceptually represents after Openness
the findings from these two studies held when controlling for to Experience, trait positive activation, and six other positive
Openness to Experience, trait positive activation, and the six emotion dispositions are partialled out, that awe is consis-
other positive emotion dispositions assessed, confirming our tently and strongly related to curiosity at the zero-order and
hypothesis that dispositional awe would be related to unique using several different covariate approaches gives us confi-
variance in trait curiosity. Having demonstrated this relation- dence in our overall interpretation of the findings.
ship, in Study 3, we showed its importance by testing its asso- While the theoretical framing of this work suggests a di-
ciation with academic outcomes such as work ethic, behavioral rectional influence of awe on curiosity, we note that current
engagement in school, and academic self-efficacy. Using an data cannot speak to causality. Experimental work at the state
SEM approach, we found that dispositional awe was indirectly level may be one way to address the limitations noted above
related to all three academic outcomes via trait curiosity even and speak to the casual effect of awe on curiosity. Such experi-
when controlling for dispositional joy. Taken altogether, these mental research would also afford the opportunity to compare
findings demonstrate the roles that awe and curiosity play in the effect of awe on curiosity with that of other emotions.
how people learn about the world. Given that in Study 1 dispositional pride was also related to
While we consistently demonstrated that awe was related trait curiosity, it may be a stringent comparison condition for
to unique variance in curiosity, this does not suggest that awe work testing if awe uniquely promotes curiosity. A state-level
is the only predictor of curiosity. Consistent with previous experimental approach would also afford the opportunity to
work (Gallagher & Lopez, 2007; Güsewell & Ruch, 2012; compare awe to other knowledge emotions such as interest,
Kashdan et al., 2004, 2009) we found that both Openness to confusion, and surprise, which have been shown to relate
Experience and trait positive activation were strongly related to how people engage with and evaluate information. Such
to curiosity. In terms of other positive emotion dispositions, comparisons are needed to test if the effect of state awe on
pride also emerged as a predictor of curiosity in Study 1. curiosity is stronger than that of other positive and knowledge
This is consistent with previous work linking social status emotions.
ANDERSON et al.   
| 775

More work is also needed to explain the relationship be- how awe interventions can promote curiosity should include
tween awe and curiosity. Research showing that the appraisals people across the lifespan.
of novelty/complexity and the ability to understand predict In conclusion, we provided the first evidence that awe-
feelings of interest (Silvia, 2005, 2008b; Silvia, Henson, prone people are more curious, a finding that is consistent
& Templin, 2009) may inform such work. Given that awe- with the life narratives of scientists and explorers that have
inspiring stimuli are vast and transcend current understand- changed human history through their exploration. Our find-
ing, it may be that successfully accommodating experiences ings suggest that experiences of awe may play an import-
of awe will increase people's perceived ability to understand ant role in sparking the curiosity of future generations of
complex stimuli, which has been found to predict how inter- Einsteins and Curies.
esting people find challenging stimuli such as abstract poems
(Silvia, 2008a). Such a pattern may explain how awe-prone
people who frequently and intensely experience awe, become ACKNOWLEDGEMENTS
more curious over time. We note that a bidirectional relation-
ship is also possible, such that people who are more curious The author(s) disclosed receipt of the following financial sup-
feel awe more frequently because their exploration exposes port for the research, authorship, and/or publication of this ar-
them to more vast and novel experiences. ticle: Preparation of this manuscript was supported by Grants
Another future direction suggested by the current work is 95524 and 88210 from the John Templeton Foundation.
testing the potential of awe interventions to bolster curiosity
and thus academic outcomes. Previous research has shown CONFLICT OF INTEREST
that dispositional awe is related to science learning (Gottlieb The author(s) declared no potential conflicts of interest with
et al., 2018), but the results from Study 3 suggest that awe and respect to the research, authorship, and/or publication of this
curiosity relate to broader academic outcomes such as work article.
ethic and behavioral engagement in school. Together with
emerging evidence that awe promotes creativity (Chirico, ORCID
Glaveanu, Cipresso, Riva, & Gaggioli, 2018; Zhang et al., Craig L. Anderson https://orcid.org/0000-0003-0931-1617
2018) and systematic processing of information (Danvers
& Shiota, 2016; Griskevicius et al., 2010), the current work ENDNOTE
raises the possibility that experiences of awe may enhance a 1
We note that our analytic strategy which includes only first-order rela-
suite of cognitive capabilities that foster learning. Empirical tionships leaves the possibility that the association of an untested inter-
evidence along these lines would highlight the importance of action between dispositional awe and a covariate which was correlated
with dispositional awe would be incorrectly attributed to DPES awe (see
providing real world experiences of awe to youth in schools
Chapter 7, Cohen, Cohen, West, & Aiken, 2003). To examine if such an
through nature experiences (Anderson, Monroy, & Keltner,
interaction were inflating the reported effect size of the relationship be-
2018a, 2018b; Piff et al., 2015), trips to museums, and cul- tween dispositional awe and trait curiosity, we tested an additional eight
tural exchanges. models that included all the predictors used in Model 3 (see Table 1)
One strength of the current work is that the relationship and an interaction effect between dispositional awe and one of the co-
between awe and curiosity was found to have a consistently variates. In all cases the effect size of dispositional awe on trait curiosity
strong relationship across samples that were diverse with re- was found to be within .01 of the result reported in the main text, sug-
gards to race and age. High school and college are periods gesting that it is an accurate estimate of the unique variance in curiosity
explained by dispositional awe in the model tested.
when people in late adolescence and emerging adulthood,
are presented with many novel and challenging experiences
and, as such, are a highly relevant to the study of awe, cu- R E F E R E NC E S
riosity, and academic outcomes. An open question, how- Anderson, C. L., Monroy, M., & Keltner, D. (2018a). Emotion in the
ever, is whether the association between awe and curiosity wilds of nature: The coherence and contagion of fear during threat-
changes as people age. To our knowledge, there has been no ening group-based outdoors experiences. Emotion, 18(3), 355–368.
https​://doi.org/10.1037/emo00​00378​
investigation of the frequency and intensity of experiences
Anderson, C. L., Monroy, M., & Keltner, D. (2018b). Awe in nature
of awe in older adults, and there is limited evidence that cer-
heals: Evidence from military veterans, at-risk youth, and college
tain types of curiosity decrease over the lifespan (Giambra, students. Emotion, 18(8), 1195–1202. https​ ://doi.org/10.1037/
Camp, & Grodsky, 1992). In the current work we did not find emo00​00442​
any evidence that the relationship between awe and curiosity Bai, Y., Maruskin, L. A., Chen, S., Gordon, A. M., Stellar, J. E., McNeil,
differed by age, although the correlation was descriptively G. D., … Keltner, D. (2017). Awe, the diminished self, and col-
higher in the older mTurk sample. Given that curiosity has lective engagement: Universals and cultural variations in the small
been shown to be associated with longer and healthier lives self. Journal of Personality and Social Psychology, 113, 185–209.
https​://doi.org/10.1037/pspa0​000087
in older adults (Swan & Carmelli, 1996) future research on
776
|    ANDERSON et al.

Bandura, A. (1990). Multidimensional scales of perceived self-efficacy. Danvers, A. F., & Shiota, M. (2016). Going off script: Effects of awe on
Stanford, CA: Stanford University. memory for script-typical and –irrelevant narrative detail. Emotion,
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable 17, 938–952. https​://doi.org/10.1037/emo00​00277​
distinction in social the moderator-mediator variable distinction in Dixson, D. D., Anderson, C. L., & Keltner, D. (2019). Measuring pos-
social psychological research: Conceptual, strategic, and statistical itive emotions: Validation of scores on the seven dispositional pos-
considerations. Journal of Personality, 51, 1173–1182. https​://doi. itive emotions scales. Journal of Well-Being Assessment. Advance
org/10.1037/0022-3514.51.6.1173 online publication.
Blau, G., & Ryan, J. (1997). On measuring work ethic: A neglected Dixson, D. D., Anderson, C. L., Rigney, A. M., Niemasik, M. A., &
work commitment facet. Journal of Vocational Behavior, 51, 435– Potte, A. (2018). Positivity in school: How positive traits relate to
448. https​://doi.org/10.1006/jvbe.1996.1568 productive school attributes. Manuscript Submitted to Journal of
Buhrmester, M., Kwang, T., & Gosling, S. D. (2011). Amazon’s me- Happiness Studies.
chanical turk: A new source of inexpensive, yet high-quality, Dixson, D. D., Worrell, F. C., Olszewski-Kubilius, P., & Subotnik, R. F.
data? Perspectives on Psychological Science, 6, 3–5. https​://doi. (2016). Beyond perceived ability: The contribution of psychosocial
org/10.1177/17456​91610​393980 factors to academic performance. Annals of the New York Academy
Byman, R. (2005). Curiosity and sensation seeking: A conceptual and of Sciences, 1377, 67–77. https​://doi.org/10.1111/nyas.13210​
empirical examination. Personality and Individual Differences, 38, Fredrickson, B. L. (1998). What good are positive emotions?
1365–1379. https​://doi.org/10.1016/j.paid.2004.09.004 Review of General Psychology, 2, 300–319. https​ ://doi.
Campos, B., Shiota, M. N., Keltner, D., Gonzaga, G. C., & Goetz, org/10.1037//1089-2680.2.3.300
J. L. (2013). What is shared, what is different? Core relational Fredrickson, B. L. (2013). Positive emotions broaden and build. In P.
themes and expressive displays of eight positive emotions. Devine & A. Plant (Eds.), Advances in experimental social psychol-
Cognition & Emotion, 27, 37–52. https​://doi.org/10.1080/02699​ ogy (1st ed., Vol. 47, pp. 1–53). San Diego, CA: Elsevier. https​://doi.
931.2012.683852 org/10.1016/B978-0-12-407236-7.00001-2
Carlson, K. D., & Wu, J. (2012). The illusion of statistical control: Fredrickson, B., & Branigan, C. (2005). Positive emotions broaden
Control variable practice in management. Organizational Research the scope of attention and thought-action repertoires. Cognition
Methods, 15, 413–435. https​://doi.org/10.1177/10944​28111​428817 & Emotion, 19, 313–332. https​://doi.org/10.1080/02699​93044​
Chirico, A., Glaveanu, V. P., Cipresso, P., Riva, G., & Gaggioli, A. (2018). 1000238
Awe enhances creative thinking: An experimental study. Creativity Furr, R. M. (2017). Psychometrics: An introduction. Los Angeles, CA:
Research Journal, 30, 123–131. https​ ://doi.org/10.1080/10400​ Sage.
419.2018.1446491 Gallagher, M. W., & Lopez, S. J. (2007). Curiosity and well-be-
Clogg, C. C., Petkova, E., & Haritou, A. (1995). Statistical methods ing. The Journal of Positive Psychology, 2, 236–248. https​://doi.
for comparing regression coefficients between models. American org/10.1080/17439​76070​1552345
Journal of Sociology, 100, 1261–1293. Retrieved from http://www. Giambra, L. M., Camp, C. J., & Grodsky, A. (1992). Curiosity and stim-
jstor.org/stabl​e/2782277 https​://doi.org/10.1086/230638 ulation seeking across the adult life span: Cross-sectional and 6- to
Cohen, A. B., Gruber, J., & Keltner, D. (2010). Comparing spiritual 8-year longitudinal findings. Psychology and Aging, 7, 150–157.
transformations and experiences of profound beauty. Psychology https​://doi.org/10.1037/0882-7974.7.1.150
of Religion and Spirituality, 2, 127–135. https​://doi.org/10.1037/ Goldsmith, B. (2005). Obsessive genius: The inner world of Marie
a0019126 Curie. New York, NY: WW Norton & Company.
Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multi- Gordon, A. M., Stellar, J. E., Anderson, C. L., Mcneil, G. D., Loew, D.,
ple regression/correlation analysis for the behavioral sciences (3rd & Keltner, D. (2017). The dark side of the sublime: Distinguishing
ed.). Mahwah, NJ: Erlbaum. a threat- based variant of awe. Journal of Personality and Social
Collins, R. P., Litman, J. A., & Spielberger, C. D. (2004). The measure- Psychology, 113, 310–328. https​://doi.org/10.1037/pspp0​000120
ment of perceptual curiosity. Personality and Individual Differences, Gottlieb, S., Keltner, D., & Lombrozo, T. (2018). Awe as a scientific
36, 1127–1141. https​://doi.org/10.1016/S0191-8869(03)00205-8 emotion. Cognitive Science, 42, 2081–2094. https​://doi.org/10.1111/
Cordaro, D. T., Brackett, M., Glass, L., & Anderson, C. L. (2016). cogs.12648​
Contentment: Perceived completeness across cultures and tradi- Graham, F. K., & Clifton, R. K. (1966). Heart-rate change as a compo-
tions. Review of General Psychology, 20, 221–235. https​ ://doi. nent of the orienting response. Psychological Bulletin, 65, 305–320.
org/10.1037/gpr00​00082​ https​://doi.org/10.1037/h0023258
Cordaro, D. T., Keltner, D., Tshering, S., Wangchuk, D., & Flynn, L. Griskevicius, V., Shiota, M. N., & Neufeld, S. L. (2010). Influence
M. (2016). The voice conveys emotion in ten globalized cultures of different positive emotions on persuasion processing: A func-
and one remote village in Bhutan. Emotion, 16, 117–128. https​://doi. tional evolutionary approach. Emotion, 10, 190–206. https​://doi.
org/10.1037/emo00​00100​ org/10.1037/a0018421
Cordaro, D. T., Sun, R., Keltner, D., Kamble, S., Huddar, N., & McNeil, Gunzler, D., Chen, T., Wu, P., & Zhang, H. (2013). Introduction to
G. (2018). Universals and cultural variations in 22 emotional ex- mediation analysis with structural equation modeling. Shanghai
pressions across five cultures. Emotion, 18, 75–93. https​ ://doi. Archives of Psychiatry, 25, 390–394. https​ ://doi.org/10.3969/j.
org/10.1037/emo00​00302​ issn.1002-0829.2013.06.009
Costello, A. B., & Osborne, J. W. (2005). Best practices in exploratory Güsewell, A., & Ruch, W. (2012). Are only emotional strengths emo-
factor analysis: Four recommendations for getting the most from tional? Character strengths and disposition to positive emotions.
your analysis. Practical Assessment, Research & Education, 10, Applied Psychology: Health and Well-Being, 4, 218–239. https​://
1–9. https​://doi.org/10.1.1.110.9154 doi.org/10.1111/j.1758-0854.2012.01070.x
ANDERSON et al.   
| 777

Hussey, I., & Hughes, S. (2018, November 19). Hidden invalidity social threat situation. Cognitive Therapy and Research, 28, 119–
among fifteen commonly used measures in social and personality 141. https​://doi.org/10.1023/B:COTR.00000​16934.20981.68
psychology. https​://doi.org/10.31234/​osf.io/7rbfp​ Kashdan, T. B., & Roberts, J. E. (2004b). Trait and state curiosity in
Iacobucci, D. (2008). Mediation analysis. Thousand Oaks, CA: Sage. the genesis of intimacy: Differentiation from related constructs.
John, O. P., & Srivastava, S. (1999). The Big-Five Trait taxonomy: Journal of Social and Clinical Psychology, 23, 792–816. https​://doi.
History, measurement, and theoretical perspectives. In L. Pervin & org/10.1521/jscp.23.6.792.54800​
O. P. John (Eds.), Handbook of personality: Theory and research Kashdan, T. B., Rose, P., & Fincham, F. D. (2004). Curiosity and ex-
(pp. 102–138). New York, NY: Guilford. ploration: Facilitating positive subjective experiences and personal
Johnson, K. J., Waugh, C. E., & Fredrickson, B. L. (2010). Smile to see growth opportunities. Journal of Personality Assessment, 82, 291–
the forest: Facially expressed positive emotions broaden cognition. 305. https​://doi.org/10.1207/s1532​7752j​pa8203_05
Cognition & Emotion, 24, 299–321. https​://doi.org/10.1080/02699​ Kashdan, T. B., & Steger, M. F. (2007). Curiosity and pathways to
93090​3384667 well-being and meaning in life: Traits, states, and everyday behav-
Kaczmarek, L. D., Kashdan, T. B., Kleiman, E. M., Baczkowski, B., iors. Motivation and Emotion, 31, 159–173. https​://doi.org/10.1007/
Enko, J., Siebers, A., … Baran, B. (2013). Who self-initiates grat- s11031-007-9068-7
itude interventions in daily life? An examination of intentions, Kashdan, T. B., Sherman, R. A., Yarbro, J., & Funder, D. C. (2013).
curiosity, depressive symptoms, and life satisfaction. Personality How are curious people viewed and how do they behave in social
and Individual Differences, 55, 805–810. https​://doi.org/10.1016/j. situations? From the perspectives of self, friends, parents, and unac-
paid.2013.06.013 quainted observers. Journal of Personality, 81, 142–154. https​://doi.
Kandler, C., Riemann, R., & Spinath, F. M. (2010). Sources of vari- org/10.1111/j.1467-6494.2012.00796.x
ance in personality facets: A multiple-rater twin study of self-peer, Keltner, D., & Haidt, J. (2003). Approaching awe, a moral, spiritual, and
peer-peer, and self-self (dis) agreement. Journal of Personality, 78, aesthetic emotion. Cognition & Emotion, 17, 297–314. https​://doi.
1565–1594. https​://doi.org/10.1111/j.1467-6494.2010.00661.x org/10.1080/02699​93030​2297
Kang, M. J., Hsu, M., Krajbich, I. M., Loewenstein, G., McClure, S. Kenny, D. A. (2015). Measuring model fit. Retrieved from http://www.
M., Wang, J. T., & Camerer, C. F. (2009). The wick in the can- david​akenny.net/cm/fit.htm
dle of learning: Epistemic curiosity activates reward circuitry and Kenny, D. A., & McCoach, D. B. (2003). Effect of the number of vari-
enhances memory. Psychological Science, 20, 963–973. https​://doi. ables on measures of fit in structural equation modeling. Structural
org/10.1111/j.1467-9280.2009.02402.x Equation Modeling, 10, 333–351. https​ ://doi.org/10.1207/S1532​
Kashdan, T. B. (2002). Social anxiety dimensions, neuroticism, and the 8007S​EM1003
contours of positive psychological functioning. Cognitive Therapy Keyes, C. (1998). Social well-being. Social Psychology Quarterly,
and Research, 26, 789–810. https​ ://doi.org/10.1023/A:10212​ 61(2), 121–140. https​://doi.org/10.2307/2787065
93501345 Kolar, D. W., Funder, D. C., & Colvin, C. R. (1996). Comparing the
Kashdan, T. B. (2004). Curiosity. In C. Peterson & M. E. P. Seligman accuracy of personality judgments by the self and knowledge-
(Eds.), Character strengths and virtues: A handbook and classifica- able others. Journal of Personality, 64, 311–337. https​ ://doi.
tion (pp. 125–141). New York, NY: Oxford University Press. org/10.1111/j.1467-6494.1996.tb005​13.x
Kashdan, T. B., Afram, A., Brown, K. W., Birnbeck, M., & Drvoshanov, Lerner, J. S., & Keltner, D. (2000). Beyond valence: Toward a model of
M. (2011). Curiosity enhances the role of mindfulness in reduc- emotion-specific influences on judgement and choice. Cognition
ing defensive responses to existential threat. Personality and & Emotion, 14, 473–493. https​://doi.org/10.1080/02699​93004​
Individual Differences, 50, 1227–1232. https​ ://doi.org/10.1016/j. 02763​
paid.2011.02.015 Lerner, J. S., & Keltner, D. (2001). Fear, anger, and risk. Journal of
Kashdan, T. B., DeWall, C. N., Pond, R. S., Silvia, P. J., Lambert, N. M., Personality and Social Psychology, 81, 146–159. https​ ://doi.
Fincham, F. D., … Keller, P. S. (2013). Curiosity protects against org/10.1037/0022-3514.81.1.146
interpersonal aggression: Cross-sectional, daily process, and be- Litman, J. A., & Jimerson, T. L. (2004). The measurement of curiosity
havioral evidence. Journal of Personality, 81, 87–102. https​://doi. as a feeling of deprivation. Journal of Personality Assessment, 82,
org/10.1111/j.1467-6494.2012.00783.x 147–157. https​://doi.org/10.1207/s1532​7752j​pa8202_3
Kashdan, T. B., Elhai, J. D., & Breen, W. E. (2008). Social anxiety Litman, J. A., & Silvia, P. J. (2006). The latent structure of trait cu-
and disinhibition: An analysis of curiosity and social rank ap- riosity: Evidence or interest and deprivation curiosity dimensions.
praisals, approach-avoidance conflicts, and disruptive risk-taking Journal of Personality Assessment, 86, 318–328.
behavior. Journal of Anxiety Disorders, 22, 925–939. https​://doi. Litman, J. A., & Spielberger, C. D. (2003). Measuring epistemic curios-
org/10.1016/j.janxd​is.2007.09.009.Social ity and its diversive and specific components. Journal of Personality
Kashdan, T. B., Gallagher, M. W., Silvia, P. J., Winterstein, B. P., Breen, Assessment, 80, 75–86. https​://doi.org/10.1207/S1532​7752J​PA8001
W. E., Terhar, D., & Steger, M. F. (2009). The curiosity and explo- Loewenstein, G. (1994). The psychology of curiosity: A review and
ration inventory-II: Development, factor structure, and psychomet- reinterpretation. Psychological Bulletin, 116, 75–98. https​ ://doi.
rics. Journal of Research in Personality, 43, 987–998. https​://doi. org/10.1037/0033-2909.116.1.75
org/10.1016/j.jrp.2009.04.011 Maccallum, R. C., Browne, M. W., & Sugawara, H. M. (1996). Power
Kashdan, T. B., McKnight, P. E., Fincham, F. D., & Rose, P. (2011). When analysis and determination of sample size for covariance struc-
curiosity breeds intimacy: Taking advantage of intimacy opportuni- ture modeling. Psychological Methods, 3, 130–149. https​ ://doi.
ties and transforming boring conversations. Journal of Personality, org/10.1037/1082-989X.1.2.130
79, 1067–1099. https​://doi.org/10.1111/j.1467-6494.2010.00697.x Marsh, H. W., Hau, K.-T., & Wen, Z. (2004). In search of golden rules:
Kashdan, T. B., & Roberts, J. E. (2004a). Social anxiety’s impact on Comment on hypothesis-testing approaches to setting cutoff values
affect, curiosity, and social self-efficacy during a high self-focus for fit indexes and dangers in overgeneralizing Hu and Bentler’s
778
|    ANDERSON et al.

(1999) findings. Structural Equation Modelling, 11, 320–341. https​ attachment style. The Journal of Positive Psychology, 1, 61–71.
://doi.org/10.1207/s1532​8007s​em1103_2 https​://doi.org/10.1080/17439​76050​0510833
Mcphetres, J. (2019). Oh, the things you don’t know: Awe promotes Shiota, M. N., Keltner, D., & Mossman, A. (2007). The nature of
awareness of knowledge gaps and science interest gaps and science awe: Elicitors, appraisals, and effects on self-concept. Cognition
interest. Cognition & Emotion, 1–17. Advance online publication. & Emotion, 21, 944–963. https​://doi.org/10.1080/02699​93060​
https​://doi.org/10.1080/02699​931.2019.1585331 0923668
Mõttus, R., McCrae, R. R., Allik, J., & Realo, A. (2014). Cross-rater Shiota, M. N., Neufeld, S. L., Yeung, W. H., Moser, S. E., & Perea,
agreement on common and specific variance of personality scales E. F. (2011). Feeling good: Autonomic nervous system respond-
and items. Journal of Research in Personality, 52, 47–54. https​:// ing in five positive emotions. Emotion, 11, 1368–1378. https​://doi.
doi.org/10.1016/j.jrp.2014.07.005 org/10.1037/a0024278
Muir, J. (1894). The mountains of California (Reprint). New York, NY: Silvia, P. J. (2005). What is interesting? Exploring the ap-
American Museum of Natural History. praisal structure of interest. Emotion, 5, 89–102. https​ ://doi.
Mussel, P. (2010). Epistemic curiosity and related constructs: org/10.1037/1528-3542.5.1.89
Lacking evidence of discriminant validity. Personality and Silvia, P. J. (2008a). Appraisal components and emotion traits:
Individual Differences, 49, 506–510. https​ ://doi.org/10.1016/j. Examining the appraisal basis of trait curiosity. Cognition &
paid.2010.05.014 Emotion, 22, 94–113. https​://doi.org/10.1080/02699​93070​1298481
Muthén, L. K., & Muthén, B. O. (1998–2012). Mplus user’s guide (7th Silvia, P. J. (2008b). Interest—The curious emotion. Current
ed.). Los Angeles, CA: Author. Directions in Psychological Science, 17, 57–60. https​ ://doi.
Neff, K. D., Rude, S. S., & Kirkpatrick, K. L. (2007). An examination org/10.1111/j.1467-8721.2008.00548.x
of self-compassion in relation to positive psychological function- Silvia, P. J. (2009). Looking past pleasure: Anger, confusion, disgust,
ing and personality traits. Journal of Research in Personality, 41, pride, surprise, and other unusual aesthetic emotions. Psychology
908–916. https​://doi.org/10.1016/j.jrp.2006.08.002 of Aesthetics, Creativity, and the Arts, 3, 48–51. https​ ://doi.
Oveis, C., Horberg, E. J., & Keltner, D. (2010). Compassion, pride, org/10.1037/a0014632
and social intuitions of self-other similarity. Journal of Personality Silvia, P. J. (2010). Confusion and interest: The role of knowledge emo-
and Social Psychology, 98, 618–630. https​ ://doi.org/10.1037/ tions in aesthetic experience. Psychology of Aesthetics, Creativity,
a0017628 and the Arts, 4, 75–80. https​://doi.org/10.1037/a0017081
Paulhus, D. L., & Vazire, S. (2007). The self-report method. In R. W. Silvia, P. J., Henson, R. A., & Templin, J. L. (2009). Are the sources of inter-
Robins, R. C. Fraley, & R. F. Krueger (Eds.), Handbook of research est the same for everyone? Using multilevel mixture models to explore
methods in personality psychology (pp. 224–239). New York, NY: individual differences in appraisal structures. Cognition & Emotion,
Guilford. 23, 1389–1406. https​://doi.org/10.1080/02699​93090​2850528
Peterson, C., & Seligman, M. E. P. (2004). Character strengths and Silvia, P. J., & Kashdan, T. B. (2009). Interesting things and curious
virtues: A handbook and classification. New York, NY: Oxford people: Exploration and engagement as transient states and enduring
University Press. strengths. Social and Personality Psychology Compass, 3, 785–797.
Piaget, J. (1970). Piaget’s theory. In P. H. Mussen (Ed.), Carmichael’s https​://doi.org/10.1111/j.1751-9004.2009.00210.x
manual of child psychology (3rd ed., pp. 703–720). New York, NY: Skinner, E., Furrer, C., Marchand, G., Kindermann, T., Sherwood, H., &
John Wiley and Sons Inc. Usinger, P. (2008). Engagement and disaffection in the classroom:
Piff, P. K., Dietze, P., Feinberg, M., Stancato, D. M., & Keltner, D. Part of a larger motivational dynamic? Journal of Educational
(2015). Awe, the small self, and prosocial behavior. Journal of Psychology, 100, 765–781. https​://doi.org/10.1037/a0012840
Personality and Social Psychology, 108, 883–899. https​ ://doi. Soto, C. J., & John, O. P. (2017). The next Big Five Inventory (BFI-
org/10.1037/pspi0​000018 2): Developing and assessing a hierarchical model with 15 facets to
Raine, A., Reynolds, C., Venables, P. H., & Mednick, S. A. (2002). enhance bandwidth. Fidelity, and Predictive Power, 113, 117–143.
Stimulation seeking and intelligence: A prospective longitudinal https​://doi.org/10.1037/pspp0​000096
study. Journal of Personality and Social Psychology, 82, 663–674. Spielberger, C. D., Peters, R. A., & Frain, F. (1976). The state-trait curi-
https​://doi.org/10.1037/0022-3514.82.4.663 osity inventory (Unpublished manual), University of South Florida,
Reiss, S., & Reiss, M. (2004). Curiosity and mental retarda- Tampa, FL.
tion: Beyond IQ. Mental Retardation, 42, 77–81. https​ ://doi. Stellar, J. E., Gordon, A. M., Anderson, C. L., Piff, P. K., Mcneil, G.
org/10.1352/0047-6765(2004)42<77:CAMRB​I>2.0.CO;2 D., & Keltner, D. (2017). Awe and humility. Journal of Personality
Renner, B. (2006). Curiosity about people: The development of a social and Social Psychology, 114, 258–269. https​://doi.org/10.1037/str00​
curiosity measure in adults. Journal of Personality Assessment, 87, 00014​
305–316. https​://doi.org/10.1207/s1532​7752j​pa8703 Stellar, J. E., Gordon, A. M., Piff, P. K., Cordaro, D., Anderson, C. L.,
Rosenberg, E. L. (1998). Levels of analysis and the organization of Bai, Y., … Keltner, D. (2017). Self-transcendent emotions and their
affect. Review of General Psychology, 2, 247–270. https​ ://doi. social functions: Compassion, gratitude, and awe bind us to others
org/10.1037/1089-2680.2.3.247 through prosociality. Emotion Review, 9(3), 200–207.https​://doi.
Rudd, M., Vohs, K. D., & Aaker, J. (2012). Awe expands peo- org/10.1177/17540​73916​684557
ple’s perception of time, alters decision making, and enhances Sung, B., & Yih, J. (2016). Does interest broaden or narrow atten-
well-being. Psychological Science, 23, 1130–1136. https​ ://doi. tional scope? Cognition & Emotion, 30, 1485–1494. https​://doi.
org/10.1177/09567​97612​438731 org/10.1080/02699​931.2015.1071241
Shiota, M. N., Keltner, D., & John, O. P. (2006). Positive emotion dis- Swan, G. E., & Carmelli, D. (1996). Curiosity and mortality in
positions differentially associated with Big Five personality and aging adults: A 5-year follow-up of the Western Collaborative
ANDERSON et al.   
| 779

Group Study. Psychology and Aging, 11, 449–453. Retrieved Watkins, M. W. (2018). Exploratory factor analysis: A guide to best
from http://www.ncbi.nlm.nih.gov/pubme​d/8893314 https​://doi. practice. Journal of Black Psychology, 44, 219–246. https​://doi.
org/10.1037/0882-7974.11.3.449 org/10.1177/00957​98418​771807
Thompson, E. R. (2007). Development and validation of an internation- Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and
ally reliable short-form of the positive and negative affect sched- validation of brief measures of positive and negative affect: The
ule (PANAS). Journal of Cross-Cultural Psychology, 38, 227–242. PANAS scales. Journal of Personality and Social Psychology, 54,
https​://doi.org/10.1177/00220​22106​297301 1063–1070. Retrieved from http://www.ncbi.nlm.nih.gov/pubme​
Tong, E. M. W. (2015). Differentiation of 13 positive emotions d/3397865 https​://doi.org/10.1037/0022-3514.54.6.1063
by appraisals. Cognition & Emotion, 29, 484–503. https​ ://doi. Zhang, J. W., Anderson, C. L., Razavi, P., Mello, Z., Shaban-Azad, H.,
org/10.1080/02699​931.2014.922056 Monroy, M., … Keltner, D. (2018). Trait and state based experience
Tracy, J. L., Cheng, J. T., Robins, R. W., & Trzesniewski, K. H. of awe promotes creativity. Manuscript invited for resubmission.
(2009). Authentic and hubristic pride: The affective core of self-es- Zimmerman, B. J., Bandura, A., & Martinez-Pons, M. (1992). Self-
teem and narcissism. Self and Identity, 8, 196–213. https​ ://doi. motivation for academic attainment: The role of self-efficacy be-
org/10.1080/15298​86080​2505053 liefs and personal goal setting. American Educational Research
Tracy, J. L., & Robins, R. W. (2007). Emerging insights into the nature Journal, 29, 663–676. https​://doi.org/10.3102/00028​31202​
and function of pride. Current Directions in Psychological Science, 9003663
16, 147–150. https​://doi.org/10.1111/j.1467-8721.2007.00493.x Zuckerman, M. (1994). Behavioral expressions and biosocial bases
Valdesolo, P., & Graham, J. (2014). Awe, uncertainty, and agency of sensation seeking. New York, NY: Oxford University Press.
detection. Psychological Science, 25, 170–178. https​ ://doi. Retrieved from http://www.scirp.org/(S(vtj3f​a45qm​1ean4​5vvff​
org/10.1177/09567​97613​501884 cz55))/refer​ence/Refer​ences​Papers.aspx?Refer​enceI​D=1135472
Valdesolo, P., Park, J., & Gottlieb, S. (2016). Awe and scientific expla-
nation. Emotion, 16, 937–940. https​://doi.org/10.1037/emo00​00213​
Valdesolo, P., Shtulman, A., & Baron, A. S. (2017). Science is awe- SUPPORTING INFORMATION
some: The emotional antecedents of science learning. Emotion Additional supporting information may be found online in
Review, 9, 215–221. https​://doi.org/10.1177/17540​73916​673212 the Supporting Information section.
Van Cappellen, P., & Saroglou, V. (2012). Awe activates religious and
spiritual feelings and behavioral intentions. Psychology of Religion
and Spirituality, 4, 223–236. https​://doi.org/10.1037/a0025986 How to cite this article: Anderson CL, Dixson DD,
Van Cappellen, P., Saroglou, V., Iweins, C., Piovesana, M., & Fredrickson, Monroy M, Keltner D. Are awe-prone people more
B. L. (2013). Self-transcendent positive emotions increase spiritu- curious? The relationship between dispositional awe,
ality through basic world assumptions. Cognition & Emotion, 27,
curiosity, and academic outcomes. Journal of
1378–1394. https​://doi.org/10.1080/02699​931.2013.787395
Wadlinger, H. A., & Isaacowitz, D. M. (2006). Positive mood broad-
Personality. 2020;88:762–779. https​://doi.
ens visual attention to positive stimuli. Motivation and Emotion, 30, org/10.1111/jopy.12524​
87–99. https​://doi.org/10.1007/s11031-006-9021-1