Adding Comments and Notes to Your PDF
To facilitate electronic transmittal of corrections, we encourage authors to utilize the
comment/annotations features in Adobe Acrobat. The PDF provided has been comment
enabled, which allows you to utilize the comment and annotation features even if using
only the free Adobe Acrobat reader (see note below regarding acceptable versions).
Adobe Acrobat’s Help menu provides additional details on the tools. When you open
your PDF, the annotation tools are clearly shown on the tool bar (although icons may
differ slightly among versions from what is shown below).
For purposes of correcting the PDF proof of your journal article, the important features to
know are the following:
x

To insert text, place your cursor at a point in text and select the Insert Text tool (
from the menu bar. Type your additional text in the pop-up box.

)

x

To replace text, highlight the text to be changed, select the Replace Text tool (
) from
the menu bar, and type the new text in the pop-up box. Do this instead of deleting and
then reinserting.

x

To delete text, highlight the text to be deleted and press the Delete button on the
keyboard.

x

) to describe changes that need to be made (e.g., changes
Use the Sticky Note tool (
in bold, italics, or capitalization use; altering or replacing a figure) or to answer a
question or approve a change from the editor. To use this feature, click on the Sticky
Note tool in the menu bar and then click on a point in the PDF where you would like to
make a comment. Then type your comment in the pop-up box.

x

Use the Callout tool (
) to point directly to changes that need to be made. Try to put
the callout box in an area of white space so that you do not obscure the text.

x

Use the Highlight and Add Note to Text tool (
) to indicate font problems, bad
breaks, and other textual inconsistencies. Select text to be changed, choose this tool, and
type your comment in the pop-up box. One note can describe many changes.

x

To view a list of changes to the proof or to see a more comprehensive set of annotation
tools, select Comment from the menu bar.

 As with hand-annotated proof corrections, the important points are to communicate
changes clearly and thoroughly, to answer all queries and questions, and to provide
complete information to allow us to make the necessary changes to your article so it is
ready for publication. Do not use tools that incorporate changes to the text in such a way
that no indication of a change is visible. Such changes will not be incorporated into the
final proof.
To utilize the comments features on this PDF you will need Adobe Reader version
7 or higher. This program is freely available and can be downloaded from
http://get.adobe.com/reader/

 APA NLM
tapraid5/emo-emo/emo-emo/emo99917/emo3256d17z xppws S⫽1 1/11/17 10:55 Art: 2016-1427

Emotion
2017, Vol. 0, No. 999, 000

AQ: 1

© 2017 American Psychological Association
1528-3542/17/$12.00 http://dx.doi.org/10.1037/emo0000277

Going Off Script: Effects of Awe on Memory for Script-Typical
and –Irrelevant Narrative Detail
Alexander F. Danvers and Michelle N. Shiota

AQ: 3

People often filter their experience of new events through knowledge they already have; for example,
encoding new events by relying on prototypical event “scripts” at the expense of actual details. Previous
research suggests that positive affect often increases this tendency. Three studies assessed whether
awe—an emotion elicited by perceived vastness, and thought to promote cognitive accommodation— has
the opposite effect, reducing rather than increasing reliance on event scripts. True/false questions on
details of a short story about a romantic dinner were used to determine whether awe (a) reduces the
tendency to impute script-consistent but false details into memory, and/or (b) promotes memory of
unexpected details. Across studies we consistently found support for the first effect; evidence for the
second was less consistent. Effects were partially mediated by subjective awe, and independent of other
aspects of subjective affect. Results suggest that awe reduces reliance on internal knowledge in
processing new events.

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

AQ: au
AQ: 2

AQ: 4

AQ: 8

Arizona State University

Keywords: awe, cognition, memory, positive emotion
Supplemental materials: http://dx.doi.org/10.1037/emo0000277.supp

There are moments in life where one is struck anew by the
vastness, beauty, and complexity of the surrounding world. Standing at the rim of the Grand Canyon, or sitting on the roof of a
friend’s house looking up at the stars, one seems to shed the
perspective and expectations bounded by mundane, day-to-day
experience. At these times, people may feel that they see the world
in a new light, considering deeper questions and opening their
minds to what the universe has to offer. But do moments of awe
change the way people process information at a concrete level?
The three studies presented here build on previous research on the
effects of awe on cognitive processing, asking whether awe helps
suppress people’s tendency to see what they expect to see, and
whether awe facilitates taking in new information from the environment.

Positive Affect and Cognitive Processing
A rich research tradition demonstrates that emotions influence
perception, interpretation, and memory of information from one’s
surroundings, sometimes in subtle ways. Although much of this
research has emphasized implications of emotional stress or distress for cognitive processing, many studies have focused on
characterizing differences between positive- and negative-valence
affect. Much research has suggested that positive and negative
mood states each tend to facilitate a particular processing style:
global, heuristic-driven, and top-down processing for positive
mood versus local, systematic, detail-oriented, and bottom-up processing for negative mood (Schwarz & Bless, 1991; Schwarz &
Clore, 2007; Storbeck & Clore, 2005). In a happy mood people
attend more strongly to global than to local aspects of complex
figures (Fredrickson & Branigan, 2005) and, in a “rose-colored
glasses effect,” find consumer products more appealing (Pham,
2007). Positive affect has also been found to increase individuals’
reliance on stereotypes (e.g., Bodenhausen, Kramer, & Susser,
1994; Huntsinger, Sinclair, & Clore, 2009), judgment heuristics
(Ruder & Bless, 2003), and number rather than quality of persuasive arguments (Bless, Mackie, & Schwarz, 1992; Mackie &
Worth, 1989) in processing information about new people or
making decisions.
Although these cognitive shortcuts are efficient and adaptive in
certain situations, helping people to get the gist of new information
quickly, they tend to impair accurate encoding of detail. For
example, studies have found that people in a positive mood are
more likely to rely on event scripts in encoding details of a novel
story about a commonplace kind of event (Bless et al., 1996). In
studies documenting this effect, participants complete an experimental mood manipulation, then listen to a lengthy story about a
kind of event for which people in the mainstream United States are

Alexander F. Danvers and Michelle N. Shiota, Department of Psychology, Arizona State University.
Michelle N. Shiota developed the study concept, with assistance from
Alexander F. Danvers. Alexander F. Danvers performed data collection
and analysis for all studies. Both authors contributed substantially to data
interpretation and manuscript preparation. Both authors have approved the
final version of this article for submission.
This work was supported by a grant from the John Templeton Foundation to Michelle N. Shiota [Grant No. 23375]. The opinions expressed in
this publication are those of the authors and do not necessarily reflect the
views of the John Templeton Foundation. We thank the members of the
ASU SPLAT lab for their help with this project.
Correspondence concerning this article should be addressed to Michelle
N. Shiota, Department of Psychology, Arizona State University, P.O. Box
871104, Tempe, AZ 85287-1104. E-mail: lani.shiota@asu.edu
1

 APA NLM
tapraid5/emo-emo/emo-emo/emo99917/emo3256d17z xppws S⫽1 1/11/17 10:55 Art: 2016-1427

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

2

DANVERS AND SHIOTA

expected to have a rich mental prototype, such as a couple going
out to a romantic dinner. When later completing true/false questions about whether certain details were present in the story,
participants in a positive mood show a heightened tendency toward
contamination of their memory by the event script—they identify
as true details that you might expect in a romantic dinner, but that
were not in fact present in the story (Bless et al., 1996).
Research documenting these effects typically uses cognitive
tasks whose content is unrelated to the preceding affect induction.
For example, Bless and colleagues (1996) induced positive mood
by asking participants to write about a happy time in their lives, or
to watch clips from the films Dead Poets Society or Flashdance—
affective stimuli unrelated to the “romantic dinner” story or to the
subsequent memory task. This enhances the internal validity of
such studies, controlling cognitive task details and reducing the
likelihood that cognitive effects are driven solely by properties of
the affect-changing stimulus, as distinct from the affect evoked by
that stimulus. This approach also has important implications for
theory, demonstrating “carryover effects”—activation of cognitive
processing styles that are maintained beyond exposure to emotioneliciting or mood-changing stimuli. In this way, researchers can
document cognitive effects of short-lived emotions that can linger
beyond the emotions themselves, as well as effects of moods that
are thought to be inherently longer-lasting and less stimulus bound
(for more on the distinction between emotion and mood, see
Beedie, Terry, & Lane, 2005).
Taken together, the research reviewed above suggests that positive moods commonly encourage activation of and reliance on
internal knowledge structures (heuristics, cognitive schemas and
scripts) relevant to one’s current situation, at the expense of close
attention to details of the actual situation (Fiedler, 2001). Various
theories have been proposed regarding the mechanisms and boundary conditions of this effect. According to affect-as-information
(AAI) theory, for example, affect valence is used to guide the
investment of effort in information processing (Clore, 1992;
Schwarz & Clore, 2007). Positive mood signals that the environment is safe and predictable, so it is okay to rely on heuristics and
internalized knowledge, whereas negative mood signals a problem
requiring more careful attention and systematic processing.
The affect as cognitive feedback (ACF) model, an updating of
AAI theory, holds that instead of preferentially activating a particular processing style, affect valence acts as a metacognitive
signal in which positive affect increases confidence and promotes
continued use of one’s current or default style, whereas negative
affect prompts a switch to a different style (Bless, 2001; Huntsinger, Isbell, & Clore, 2014; Huntsinger, Isbell, & Clore, 2014;
Isbell, Lair, & Rovenpor, 2013). For example, in contrast to
theorizing that positive affect always increases global attentional
focus (Isbell, 2004; Isbell, Burns, & Haar, 2005), positive mood
has been found to increase local focus in subsequent tasks, provided that local focus was primed prior to mood manipulation
(Huntsinger, Clore, & Bar-Anan, 2010). The typical effects of
positive mood highlighted by AAI theory are presumably observed
because, in the absence of a reason to expend effort on systematic,
bottom-up information processing, reliance on heuristics and
stored, internal knowledge is the default cognitive style (Schwarz,
2001).
Hedonic contingency theory (Wegener & Petty, 1994) offers a
different account of the findings above. According to this model,

the core impact of positive affect is to increase motivation to
maintain one’s pleasant mood. Although this decreases one’s motivation to carefully, systematically process content that is unpleasant, serious, or counterattitudinal (common among studies examining mood effects on processing of persuasive messages), it may
increase systematic attention to inherently pleasant stimuli. This
moderating effect was supported by multiple studies (Wegener,
Petty, & Smith, 1995).
Although differing with regard to the theorized mechanism of
affect valence effects, the AAI, ACF, and hedonic contingency
models have a few things in common. First, they assume that the
effects of positive affect/mood are consistent—they do not postulate or test differences among varieties of positive emotion. Assuming that the default cognitive style and information-processing
task are held constant, therefore, different positive-valence emotions should show similar effects. Second, all three theories assume that changing one’s default information processing style is
effortful, and thus requires some form of motivation. Third, in each
theory, motivation to change one’s processing style is altered by
the subjective experience of positive affect, which either warns
that the current style may be problematic (AAI and ACF), or
sensitizes the individual to affective consequences of attending to
nonpleasant information.

Different Positive Emotions, Different Effects?
Although most research on emotion and cognition has addressed
implications of positive versus negative valence of subjective
affect, aspects of emotion other than valence can influence cognitive processing as well—a point explicitly acknowledged by ACF
theory (Huntsinger, Isbell, & Clore, 2014). Another rich body of
research documents implications of approach versus avoidance
motivation, even among emotions of the same valence (HarmonJones, Gable, & Price, 2013; Harmon-Jones, Price, & Gable,
2012). For example, high-approach positive emotions, occurring
before a goal is achieved, have been found to promote local
attentional focus and preferential attention to central elements of
the visual field, whereas low-approach “post-goal” positive emotions promote broadened global focus and more distributed visual
attention (Gable & Harmon-Jones, 2008; Gable & Harmon-Jones,
2010a; Gable & Harmon-Jones, 2011). The approach/avoidance
dimension of emotion has been found to predict differences in
attentional focus among negative emotions as well (Gable &
Harmon-Jones, 2010b; Gable, Poole, & Harmon-Jones, 2015).
The studies reported in this paper reflect a third approach,
emphasizing differences among specific discrete emotions rather
than those along a dimension of affect, such as valence or motivational direction. In research from this perspective, hypotheses
often reflect theoretical analyses of particular emotions’ adaptive
functions, proposing a “problem” defined by a prototypical eliciting situation in the ancestral environment, and asking what an
emotional “solution” to that problem might look like in terms of
physiology, nonverbal expression, action tendency, and/or cognitive information processing (Ekman, 1992; Levenson, 1999;
Tooby & Cosmides, 2008). Such models tend not to assume that
cognitive effects of emotion are fully mediated by subjective
feelings or motivation. Rather, emotions are thought to involve
automatic, even unconscious, activation of particular cognitive
mechanisms that are strategically relevant for accomplishing a

 APA NLM
tapraid5/emo-emo/emo-emo/emo99917/emo3256d17z xppws S⫽1 1/11/17 10:55 Art: 2016-1427

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

AWE AND MEMORY FOR NARRATIVE DETAIL

current goal. Several lines of research have shown that specific
emotions of the same valence can have different— even opposing— effects on cognition (e.g., Bodenhausen, Sheppard, &
Kramer, 1994; DeSteno, Petty, Wegener, & Rucker, 2000;
Lerner & Keltner, 2001). Much of this work distinguishes
among negative emotions, showing for example that fear and
anger have different implications for risk seeking (Lerner &
Keltner, 2001), or that anger and sadness have different effects
on stereotyping (Bodenhausen, Sheppard, & Kramer, 1994) and
attribution (Keltner, Ellsworth, & Edwards, 1993).
However, a rapidly growing body of work documents differences among positive emotions as well (Shiota et al., 2014; Shiota
et al., 2017). This research finds that several positive emotions
have distinct effects on outcomes such as autonomic physiology
(Shiota, Neufeld, Yeung, Moser, & Perea, 2011), facial expression
(Campos, Shiota, Keltner, Gonzaga, & Goetz, 2013), vocal expression (Simon-Thomas, Keltner, Sauter, Sinicropi-Yao, & Abramson, 2009), and consumer preference (Griskevicius, Shiota, &
Nowlis, 2010). In the domain of cognitive processing, studies have
demonstrated that different positive emotions have distinct implications for systematic versus heuristic-based processing of persuasive messages, in patterns that are not accounted for by either valence
or motivational direction (Griskevicius, Shiota, & Neufeld, 2010).
The current research builds upon and expands this knowledge by
examining the cognitive effects of a particular positive emotion—
awe.

Awe and Cognitive Accommodation
Awe is an emotion elicited by vast, unfamiliar stimuli such as
panoramic views, great works of architecture and art, and astonishing human accomplishments (Shiota, Keltner, & Mossman,
2007). Functional theories of awe posit that the “problem” posed
by such stimuli is that they are too big and important to ignore, and
may provide valuable information, yet are not adequately accounted for by one’s stored, internal knowledge (e.g., seeing the
view from the Empire State building for the first time). Even when
one has experienced the stimulus or one like it before, it may
strongly evoke salience of the unknown (e.g., looking up into a
vast, brilliant sunset or the night sky). Such stimuli are thought to
challenge the default frame of reference through which one perceives the world, prompting one to put assumptions and stored
knowledge aside in favor of cognitive accommodation—taking in
new information to update understanding (Keltner & Haidt, 2003;
Shiota, Thrash, Danvers, & Dombrowski, 2014).
Although awe can be experienced alongside fear or horror,
depending on the exact stimulus, participants in the United States
nearly always describe awe experiences that are highly pleasant
(e.g., Shiota et al., 2007). Moreover, awe inductions used in prior
experimental research have consistently been found to elicit levels
of positive affect or “happiness” that are high and comparable to
other positive emotion conditions (e.g., pride, enthusiasm, contentment), with very low levels of negative emotions, including fear
(e.g., Griskevicius, Shiota, & Neufeld, 2010; Piff, Dietze, Feinberg, Stancato, & Keltner, 2015; Shiota et al., 2007). Thus, hypotheses emphasizing awe’s affective valence would predict it
should lead to more heuristic, stored knowledge-driven processing,
assuming this is allowed to remain as the default cognitive style,
and the cognitive task is not itself particularly enjoyable. The

3

functional analysis of awe, on the other hand, suggests that awe
should reduce reliance on heuristics and stored knowledge, and
increase intake of new information, even if this more systematic
style is not previously primed.
Consistent with this proposal, prior research has found that
dispositional awe-proneness is associated with high openness to
experience and low need for cognitive closure (Shiota, Keltner, &
John, 2006; Shiota et al., 2007), reflecting awe’s emphasis on
seeking new information and experience, and tolerating uncertainty and ambiguity. Experimentally elicited awe has been found
to increase cautious, systematic processing of persuasive messages, reducing reliance on cognitive shortcuts such as the “number of arguments” heuristic (Griskevicius, Shiota, & Neufeld,
2010). Some theorists have speculated that time perception is
linked to the number of pieces of information being encoded—a
necessary consequence of greater attention to details of stimuli in
the environment (Eagleman et al., 2005; Wearden, 2005), and awe
experiences have been found to alter time perception, expanding
the sense of available time (Rudd, Vohs, & Aaker, 2012).

The Current Research
The present set of studies builds particularly on a previous
finding that, in contrast to most other positive emotions, awe
makes individuals less susceptible to reliance on heuristics in
processing persuasive messages (Griskevicius, Shiota, & Neufeld,
2010). Drawing on the distinction between the central route and
the peripheral route in the elaboration likelihood model of persuasion processing (Cacioppo & Petty, 1989), those studies tested
whether individuals in various positive emotion states were easily
persuaded through the peripheral route— using a “number of arguments” heuristic to judge message quality— or responded more
strongly to the quality of the arguments, indicating systematic
central route processing. Across two studies, individuals in the awe
conditions responded strongly to argument quality, whereas individuals experiencing most other positive emotions were easily
convinced by a large number of arguments regardless of quality.
Griskevicius and colleagues’ (2010) persuasion findings support
the theoretical proposal that awe promotes systematic information
processing, but do not disambiguate the mechanism by which this
occurs. On one hand, awe may reduce the activation of stored
knowledge to use as a filter for current experience—that is, the
tendency to rely on shortcuts such as heuristics and schemas. On
the other hand, awe may directly promote the intake and storage of
new information, improving attention to details that are not predicted by one’s expectations. These two processes are compatible,
but distinct. Importantly, both the “heuristic suppression” and the
“increased attention to detail” effects could account for findings in
previous literature, and Griskevicius and colleagues’ findings do
not tease these explanations apart.
The three studies reported here examined the cognitive effects
of awe in greater detail using an experimental paradigm described
earlier: a test of memory for details of a story with a familiar
script—a couple going out to a romantic dinner. Based on previous
research, we anticipated that participants in most positive emotion
states would tend to filter this story through their conceptual
“event scripts” or prototypical images of such events, as observed
in prior research (Bless et al., 1996). However, we hypothesized
that participants in an awe state would show either or both of two

 APA NLM
tapraid5/emo-emo/emo-emo/emo99917/emo3256d17z xppws S⫽1 1/11/17 10:55 Art: 2016-1427

DANVERS AND SHIOTA

4

effects in processing the story: (a) reduced reliance on internal
knowledge—in this case, the tendency toward false memory for
script-consistent detail (“script independence”); and (b) enhanced
memory for unusual, script-irrelevant details (“intake of unexpected detail”). Across three studies, we attempted to determine
whether awe demonstrated these effects, and how distinct the
awe-related effects were from those of other positive emotions.

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Study 1
In the first study participants watched film clips eliciting awe, a
general positive emotion state, or a neutral state before listening to
a 5-min short story about a couple going out for a romantic dinner.
We expected that awe would lead to greater script independence
and memory for script-irrelevant detail than seen in the general
positive emotion or neutral control conditions.

Method
Participants. Two hundred forty (240) participants completed
the study in return for course credit in an introductory psychology
course at a large Southwestern University. Sample size was based
on a goal of retaining at least 60 participants per cell, providing
just over 90% power to detect a medium effect size (d ⫽ .5), based
upon awe-neutral contrasts seen in our prior research with cognitive dependent variables (e.g., Griskevicius, Shiota, & Neufeld,
2010), and consideration of the effect size that would be reasonable to expect and of theoretical interest. The final sample was
212, with 16 participants removed for spending less than 100
seconds watching the emotion elicitation videos, 10 for reporting
that they did not listen to the going-out-to-dinner story, 1 for not
reporting gender, and 1 for completing the entire study in under 10
min. Mean age of the final sample was 19.40 years (SD ⫽ 1.57);
62% of participants were male; 46% were white, 29% Asian, 13%
Hispanic, and the rest reported another ethnicity.
Procedure. Procedures for all studies were reviewed and approved by the Arizona State University Office of Research Integrity and Assurance. All procedures were completed on Dell laptop
computers with 15“ screens and headphones, in individual workspaces in a lab room. Participants were randomly assigned to
watch an approximately 4-min film clip intended to elicit awe,
general positive emotion, or neutral emotion. The Awe film (n ⫽
68) moves in perspective from the outer edges of the known
universe to the level of the Earth, and down to the subatomic level.
The General Positive Emotion film (n ⫽ 74) shows Olympic figure
skater Sarah Hughes completing her long-form program and unexpectedly winning a gold medal. The Neutral film (n ⫽ 70) gives
instructions on building a cinder-block wall. All film clips included voice-over audio and pleasant instrumental music.
Participants then listened to a 5-min audio clip of a story
describing a couple going out to a romantic dinner, adapted from
Bless and colleagues (1996). After the story participants completed
a brief distractor task: rating preferences for geometric figures.
They then answered a series of true/false questions about whether
or not various details were present in the story.
Measures. Forty true/false questions asked about the presence
of details in the going out to dinner story. Details were initially
written to be either consistent with a middle-class American cultural script for a dinner date (e.g., “did the waiter pour the couple

wine?), or unrelated to the script (e.g. “was the waiter wearing
glasses?”). Half of the items in each category were true (the detail
was present) and half were false. An independent sample (N ⫽ 59)
rated the initial set of 40 items for script typicality; based on these
results, analyses focus on the five most script-typical but false
items, and the five most script-irrelevant but true items.1 The 40
initial items were included in all studies, increasing the length and
difficulty of the memory test.
Following Bless and colleagues (1996), number of correct responses to the five most script-typical but false items was used to
index script independence. An individual relying heavily on a
“going out to dinner” script would be expected to answer these
items incorrectly, imputing script-consistent but absent details into
memory for the story. Number of correct responses to the five best
script-irrelevant but true items was used to index intake of unexpected detail. To answer these questions correctly, an individual
would need to attend to and encode details unlikely to be contained
in their “romantic dinner” event script.
To establish the specificity of this effect, we also analyzed the
effects of emotions on the five most script-typical true items and
the five most script-irrelevant false items. To show that correct
responses on the script-independence index do not represent a
general bias toward answering “false” questions correctly, we
should find that awe does not improve accuracy in the scriptirrelevant false items. Similarly, to show that correct responses on
the intake of novel details index do not represent a general bias
toward answering “true” questions correctly, we should find awe
does not improve accuracy on script-typical true items.
After the true/false items, participants were asked to rate their
emotional experience during the emotion induction film clip as a
manipulation check. On a 0 to 8 scale, they rated the valence (0 ⫽
very negative; 8 ⫽ very positive) and intensity (0 ⫽ no emotion at
all; 8 ⫽ strongest emotion ever felt) of their emotional experience,
as well as how strongly they had felt each of 10 specific emotions.
Analyses. Analysis of variance was used to examine main
effects of Emotion condition on performance for each set of
memory items (Script Independence, Intake of Unexpected Detail,
script-relevant true, and script-irrelevant false items), as well as
pairwise contrasts of the awe condition with the general positive
and neutral controls. Because prior research has documented gender differences in susceptibility to several emotions (e.g., Brody &
Hall, 2008; Simon & Nath, 2004), and implications of gender for
awe are still unknown, we ran analyses both with and without
gender as a potential moderator in all studies reported here; effects
of Emotion were identical in terms of hypothesis-testing outcomes
(i.e., significance of effects) whether or not gender was included,
and in no case was the Gender x Emotion interaction significant.
Results of analyses without gender are reported below unless the
main effect of Gender was significant, in which case results of the
analyses including Gender as a second factor are reported.

Results
Manipulation check. Complete descriptive and statistical test
information for manipulation checks in all studies is available in
the supplementary materials. Participants in the Awe condition
1
Further information regarding the pilot study is available in the online
supplementary materials.

Fn1

 APA NLM
tapraid5/emo-emo/emo-emo/emo99917/emo3256d17z xppws S⫽1 1/11/17 10:55 Art: 2016-1427

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

AWE AND MEMORY FOR NARRATIVE DETAIL

T1

Fn2

AQ: 5

experienced significantly more awe during the film clip (M ⫽
6.01) than those in the Neutral condition (M ⫽ 3.43; t(209) ⫽ 7.34,
p ⬍ .001), and directionally though not significantly more awe
than in the General Positive condition (M ⫽ 5.61; t(209) ⫽ 1.17,
p ⫽ .244). Compared with individuals in the Awe and Neutral
conditions, those in the General Positive condition reported feeling
greater enthusiasm (vs. awe: t(208) ⫽ 6.35, p ⬍ .001; vs. neutral:
t(208) ⫽ 11.78, p ⬍ .001), pride (vs. awe: t(209) ⫽ 7.47, p ⬍ .001;
vs. neutral: t(209) ⫽ 9.54, p ⬍ .001), contentment (vs. awe:
t(209) ⫽ 3.71, p ⬍ .001; vs. neutral: t(209) ⫽ 7.52, p ⬍ .001), and
tenderness/compassion (vs. awe: t(209) ⫽ 5.17, p ⬍ .001; vs.
neutral: t(209) ⫽ 7.93, p ⬍ .001).
Script independence (script-typical false items). Study 1
results are summarized in Table 1. The omnibus main effect of
Emotion approached significance, F(2, 209) ⫽ 2.48, p ⫽ .086,
␩2 ⫽ .023. In planned contrasts testing the main hypotheses, Awe
condition participants scored significantly higher on the Script
Independence index than participants in the General Positive condition, (M difference2 ⫽ .53[0.06 –1.01], t(209) ⫽ 2.22, p ⫽ .028,
d ⫽ 0.37), and the contrast between the Awe and Neutral conditions was not significant, but was in the predicted direction,
t(209) ⫽ 1.32, p ⫽ .188, d ⫽ 0.22. In an additional pairwise test,
the contrast between the Neutral and General Positive conditions
was not significant, t(209) ⫽ ⫺.890, p ⫽ .375, d ⫽ ⫺0.15.
Intake of unexpected detail (script-irrelevant true items).
The omnibus main effect of Emotion was significant, (F(2, 209) ⫽
4.04, p ⫽ .019, ␩2 ⫽ .037). Planned comparisons indicated that
participants in the Awe condition performed significantly better on
intake of script-irrelevant detail than participants in either the
General Positive (M difference ⫽ .38[0.01–0.75], t(209) ⫽ 2.02,
p ⫽ .045, d ⫽ 0.19) or Neutral (M difference ⫽ .52[0.15–0.90],
t(209) ⫽ 2.76, p ⫽ .006, d ⫽ 0.26) conditions. An additional
contrast revealed no significant difference between the Neutral and
General Positive conditions, t(209) ⫽ 0.780, p ⫽ .436, d ⫽ 0.13.
Script-relevant true and script-irrelevant false items. Our
analyses found a significant main effect of gender, therefore we
report the results of 3 ⫻ 2 (Emotion by Gender) between subjects
ANOVAs on the script relevant true items and the script irrelevant
false items. For script irrelevant false items, there were no significant effects of Emotion or Emotion ⫻ Gender interaction, but
there was a main effect of Gender, F(1, 204) ⫽ 4.380, p ⫽ .038;
women answered more of these questions correctly than men (3.63
vs. 3.30). No planned comparisons between Emotion conditions
were significant. For script relevant true items, there were also no
significant effects of Emotion or Emotion ⫻ Gender interaction,
but a main effect of Gender, F(1, 204) ⫽ 6.997, p ⫽ .009; again,
women answered more of these items correctly than men (3.24 vs.

Table 1
Memory for Story Details by Question Category and Emotion,
Study 1
Category

Awe
M (SD)

General positive
M (SD)

Neutral
M (SD)

Script independence
Intake of detail

2.25a (1.48)
3.34a (1.11)

1.72b (1.34)
2.96b (1.12)

1.93a,b (1.48)
2.81b (1.12)

Note. Differing superscripts indicate differences between conditions of at
least the p ⱕ .05 level of significance.

5

2.77). No planned comparisons between Emotion conditions were
significant.

Study 1 Discussion
Results of this study supported both hypothesized cognitive
effects of awe. Individuals in the awe condition answered more
script independence items, and more intake of unexpected detail
items, correctly than those in the general positive condition. The
difference between the awe and neutral conditions was also significant for intake of unexpected detail. Importantly, effects of awe
were specific to these two item categories, suggesting they were
not driven by a global increase in accuracy, or general bias toward
presuming details were true or false.
One limitation of this study was that the general positive emotion condition did not target a particular emotion, and participants
in that condition reported feeling some awe as well as several other
positive emotions. Study 2 addressed this limitation by comparing
awe with three specific positive emotion conditions.

Study 2
In Study 1 experimentally elicited awe was found to facilitate
both independence from the “romantic dinner” event script and
intake of script-irrelevant information when remembering the details of a novel story. Study 2 built upon this finding by contrasting
awe with three specific positive emotions: enthusiasm (a pregoal,
high-approach positive emotion elicited by cues of material reward); pride (a social, self-reflective positive emotion following an
accomplishment); and contentment (a postgoal, low-approach and
-arousal positive emotion), as well as a neutral reference state. This
design differentiates the effects of awe from those of other positive
emotions more cleanly than was possible in Study 1, and asks
whether awe is comparably distinct from several alternative positive emotions, or if the contrast is limited to one or more particular
positive emotion controls. Study 2 also used a different approach
to emotion induction than Study 1.

Method
Participants. In exchange for course credit, 400 undergraduate students enrolled in introductory psychology participated in
this study. This target sample size provides approximately 90%
power to detect emotion condition effects comparable to those
seen in Study 1 (Cohen’s d of .35). Of these participants, 39 were
removed because of inadequate or inappropriate completion of the
relived experience task (e.g., not writing anything at all, or writing
about an event thematically unrelated to the prompt) or skipping a
large number of the memory questions (over 25% of items),
yielding a final sample of 361 participants. Mean age of the final
sample was 19.34 (SD ⫽ 3.02); 66% of participants were male;
50% were white, 16% Asian, and 10% Hispanic, with the remainder reporting another ethnicity.
Procedure and measures. Procedures and measures were
identical to those for Study 1, with two exceptions. First, instead of
watching a video for the emotion induction, participants were
2
Throughout this paper the range in brackets indicates the 95% confidence interval of the effect.

 APA NLM
tapraid5/emo-emo/emo-emo/emo99917/emo3256d17z xppws S⫽1 1/11/17 10:55 Art: 2016-1427

DANVERS AND SHIOTA

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

6

T2

asked to describe and vividly relive a time when they experienced
a particular emotion. Prompts, which have been used in previous
research (Griskevicius, Shiota, & Neufeld, 2010), described a
prototypical emotion eliciting situation for the target emotion:
Awe (n ⫽ 73), “an event when you saw a panoramic view for the
first time” with the Grand Canyon given as an example; Enthusiasm (n ⫽ 76), “when you knew something good was going to
happen to you soon, and you were looking forward to that event”;
Pride (n ⫽ 68), “when you accomplished something important to
you”; Contentment (n ⫽ 72), “when you had just eaten a delicious
meal, and your body felt full and comfortable”; and Neutral (n ⫽
72), “the last time you did the laundry.” Participants wrote about
and relived the event for two minutes. Second, in this condition
participants listened to the story over a set of speakers, as opposed
to via headphones.

Results
Manipulation check. Results of the manipulation check indicated that the Awe condition elicited significantly more selfreported awe than all other conditions (vs. Enthusiasm: t(355) ⫽
3.65, p ⬍ .001; vs. Pride: t(355) ⫽ 5.96, p ⬍ .001; versus
Contentment: t(355) ⫽ 7.23, p ⬍ .001; versus Neutral: t(355) ⫽
9.69, p ⬍ .001). Individuals in the Enthusiasm condition experienced the most enthusiasm, but this difference was only statistically significant in comparison to Contentment, t(355) ⫽ 3.60, p ⬍
.001 and Neutral, t(355) ⫽ 10.18, p ⬍ .001. Individuals in all the
positive emotion conditions reported similarly high levels of contentment, and these were all significantly higher than the Neutral
condition (vs. Awe: t(355) ⫽ 5.62, p ⬍ .001; vs. Enthusiasm:
t(355) ⫽ 6.02, p ⬍ .001; versus Pride: t(355) ⫽ 8.23, p ⬍ .001;
versus Contentment: t(355) ⫽ 5.80, p ⬍ .001); the Contentment
condition also resulted in more contentment than the Pride condition, t(356) ⫽ 2.53, p ⫽ .012. Because of an experimenter error the
term “pride” was not included in the manipulation check. However, participants in the Pride condition were significantly higher
than those in the neutral condition in positive valence, t(354) ⫽
5.38, p ⬍ .001, enthusiasm, t(355) ⫽ 9.32, p ⬍ .001, and contentment, t(355) ⫽ 8.23, p ⬍ .001.
Script independence. Study 2 results are summarized in Table 2. The omnibus main effect of Emotion approached significance, F(4, 356) ⫽ 2.22, p ⫽ .067, ␩2 ⫽ .024. Planned contrasts
comparing Awe to each other condition found significant differences between the Awe and Neutral conditions, t(356) ⫽ 2.68, p ⫽
.007, d ⫽ 0.45, and the Awe and Enthusiasm conditions, t(356) ⫽
2.25, p ⫽ .021, d ⫽ 0.37. The contrasts of Awe with Contentment,
t(356) ⫽ 1.81, p ⫽ .065, d ⫽ 0.31, and with Pride, t(356) ⫽ 1.09,
p ⫽ .277, d ⫽ 0.18 were not significant, but were in the predicted

direction. Additional pairwise tests did not reveal significant differences of any other positive emotion condition from Neutral: for
Enthusiasm, t(356) ⫽ 0.470, p ⫽ .639, d ⫽ 0.08; for Contentment,
t(356) ⫽ 0.838, p ⫽ .402, d ⫽ 0.14; and for Pride, t(356) ⫽ 1.552,
p ⫽ .122, d ⫽ 0.26.
Intake of unexpected detail. The omnibus main effect of
emotion condition was not significant, F(4, 356) ⫽ 0.798, p ⫽
.527, ␩2 ⫽ .009, nor did any pairwise contrasts between Awe
and other conditions reach significance (Awe vs. Enthusiasm:
t(356) ⫽ 0.61, p ⫽ .540, d ⫽ 0.10; Awe vs. Pride: t(356) ⫽ ⫺0.22,
p ⫽ .827, d ⫽ ⫺0.04; Awe versus Contentment: t(356) ⫽ ⫺1.14, p ⫽
.257, d ⫽ ⫺0.19; Awe versus Neutral: t(356) ⫽ ⫺0.07, p ⫽ .945,
d ⫽ ⫺0.01). Unexpectedly, participants in the Awe condition received directionally lower scores on intake of unexpected detail than
those in the Pride, Contentment, and Neutral conditions. In additional
pairwise tests, no other positive emotion differed significantly from
Neutral: for Enthusiasm, t(356) ⫽ ⫺0.680, p ⫽ .497, d ⫽ 0.11; for
Contentment, t(356) ⫽ 1.063, p ⫽ .288, d ⫽ 0.17; and for Pride,
t(356) ⫽ 0.150, p ⫽ .881, d ⫽ 0.03.
Script-relevant true and script-irrelevant false items. There
was no main effect of Emotion on script irrelevant false items, F(4,
356) ⫽ 0.046, p ⫽ .996, and none of the pairwise comparisons
between Emotion conditions were statistically significant. There
was also no main effect of Emotion on script relevant true items,
F(4, 356) ⫽ 0.164, p ⫽ .956, with no significant pairwise comparisons.

Study 2 Discussion
Study 2 results again showed an effect of awe on script independence that differed from other positive emotion states—in this
case, significantly from enthusiasm, and in the hypothesized direction relative to contentment and pride. Awe significantly enhanced script independence in comparison to a neutral control as
well. We again found that the effects of awe on script independence were specific, and did not extend to a general increase in
accuracy. Unlike Study 1, however, Study 2 did not find that awe
enhanced intake of unexpected detail; the effects were in the
opposite direction, though they did not reach significance. It is
noteworthy that the enthusiasm, contentment, and pride conditions
did not differ significantly from neutral on either of our cognitive
processing measures, and in fact, the direction of their effects on
script independence was opposite to that seen in the studies of
positive affect by Bless and colleagues (1996). We will address
this point further in the general discussion.
The difference between the awe and contentment conditions,
which approached significance and was in the hypothesized direction, undermines explanation of the awe effect in terms of ap-

Table 2
Memory for Story Details by Question Category and Emotion, Study 2
Category

Awe
M (SD)

Enthusiasm
M (SD)

Contentment
M (SD)

Pride
M (SD)

Script independence
Intake of detail

2.33a (1.34)
3.08a (1.22)

1.80b (1.53)
2.96a (1.32)

1.90ab (1.30)
3.31a (1.32)

2.07ab (1.36)
3.15a (1.26)

Neutral
M (SD)
1.71b (1.37)
3.24a (1.18)

Note. Differing superscripts indicate differences between conditions of at least the p ⱕ .05 level of significance.

 APA NLM
tapraid5/emo-emo/emo-emo/emo99917/emo3256d17z xppws S⫽1 1/11/17 10:55 Art: 2016-1427

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

AWE AND MEMORY FOR NARRATIVE DETAIL

proach motivation. As noted in the introduction, positive emotions
with differing levels of approach motivation have previously been
found to produce distinct cognitive effects (e.g., Gable & HarmonJones, 2008; Gable & Harmon-Jones, 2010a; Gable & HarmonJones, 2011). Although awe has not been explicitly characterized
in terms of motivational direction, it has little or no appetitive
content, and should be relatively low on approach motivation.
Contentment, as operationalized here, is arguably the prototypical,
postgoal attainment, low-approach motivation positive emotion.
The fact that awe differed from both contentment and enthusiasm
(a prototypical pregoal, high-approach motivation emotion), while
these latter were quite similar to each other, suggests that awe’s
effects are not easily explained in terms of motivational direction.
The contrast with contentment also suggests that awe’s cognitive effects are not easily explained in terms of arousal. Arousal
has long been thought to facilitate automatic cognition at the
expense of controlled cognition (Easterbrook, 1959), and heightened arousal has been shown to increase heuristic processing of
persuasive messages in an effect similar to that previously seen for
most positive emotions (Sanbonmatsu & Kardes, 1988). Because
most positive emotions, including enthusiasm, involve increased
arousal whereas awe has been linked to decreased arousal (Shiota
et al., 2011), an arousal-based mechanism of these contrasts might
be plausible. The contrast between awe and contentment, another
low-arousal positive emotion (Kreibig, 2010) cannot be explained
easily in this manner. This suggests that the facilitation of script
independence, in particular, may be distinctive to awe. Whether
this effect extends to multiple varieties of awe was addressed in
Study 3.

Study 3
Results of Study 2 replicated the Study 1 finding that, relative to
other positive emotions, awe promotes greater independence from
event scripts in encoding the details of a new fictional story. Study
2 also documented a significant difference of awe from a neutral
control condition in this regard, which was not statistically significant in Study 1— but was similar in terms of direction and
magnitude of effect. In Study 3 we turned our attention to different
variants of awe, and asked whether these would have similar or
differing implications for the processing of narrative detail. Although prototypical experiences of awe are consistently described
in positive terms, at least in the United States (Shiota et al., 2007),
terrifying events such as natural disasters and extreme weather
may elicit awe blended with fear (Keltner & Haidt, 2003). Extraordinary prosocial deeds and achievements performed by other
people, which violate our assumptions about human nature and
capabilities, should elicit awe as well as elevation and admiration
(Haidt, 2000; Keltner & Haidt, 2003). Study 3 compared these
awe-related constructs and “pure” awe to nurturant love, a new
positive emotion control condition.
Nurturant love, an emotional response to youth, smallness,
vulnerability, and cuteness (e.g., kittens and puppies), provides a
particularly interesting contrast for awe. On one hand, nurturant
love can be described as a high-approach and high-arousal positive
emotion, whereas awe is lower in both behavioral approachmotivation and sympathetically mediated arousal (e.g., Shiota et
al., 2011). On the other hand, previous research found that awe and
nurturant love share a tendency to promote careful, systematic

7

cognitive processing of persuasive messages (Griskevicius, Shiota,
& Neufeld, 2010), suggesting that it might have effects on script
independence and intake of unexpected detail similar to those of
awe. As in Studies 1 and 2, a neutral reference condition was also
included.

Method
Participants. A total of 327 undergraduate psychology students were recruited to complete this study online for course credit,
with a goal of retaining 300 participants (60 per cell). This target
would provide greater than 90% power to detect an effect of
Cohen’s d ⫽ .40, consistent with the range of awe effect sizes
observed in Studies 1 and 2. As expected given typical response
patterns for online surveys in this population, many participants
had to be excluded for not providing sufficient data: 47 skipped
more than 25% of the memory questions; 20 took more than an
hour to complete the survey, indicating distraction and likely
dissipation of any emotion effects; one took approximately 10 min;
and two reported being unable to hear the romantic dinner story.
These exclusions left a final sample of 257. Mean age of the final
sample was 19.68 (SD ⫽ 2.94); 53% were male; 61% were white,
11% Asian, and 14% Hispanic.
Procedure and measures. Procedures and measures in Study
3 followed those of Studies 1 and 2, except that participants
completed the study online. Emotions were elicited by presenting
a series of 12 photographs for 20 seconds each. Awe photos (n ⫽
53) depicted panoramic nature scenes such as the Grand Canyon
and Aurora Borealis. Awe/fear photos (n ⫽ 51) depicted dangerous natural phenomena, including tornados and lightning strikes.
Elevation/admiration (n ⫽ 50) photos depicted exemplary individuals or actions, such as Martin Luther King, Jr. giving a speech and
firefighters working in the rubble of the World Trade Center after
9/11. Nurturant love (n ⫽ 57) photos depicted baby animals,
including an infant polar bear and a crib full of infant pandas.
Neutral photos (n ⫽ 46) were of prosaic interior and exterior
scenes, such as a bus stop or child’s bedroom.

Results
Manipulation check. Results of the manipulation check indicated that individuals in the Awe condition felt the most awe (vs.
Awe/Fear: t(156) ⫽ 3.40, p ⫽ .001; vs. Elevation/Admiration:
t(156) ⫽ 2.52, p ⫽ .013; versus Nurturant Love: t(156) ⫽ 2.54,
p ⫽ .012; versus Neutral: t(156) ⫽ 4.60, p ⬍ .001). Individuals in
the Nurturant Love condition felt the most tenderness/compassion,
although individuals in the Elevation/Admiration condition felt
similar levels (vs. Awe: t(156) ⫽ ⫺6.08, p ⬍ .001; vs. Awe/Fear:
t(156) ⫽ ⫺5.00, p ⬍ .001; versus Elevation/Admiration:
t(156) ⫽ ⫺1.51, p ⫽ .134; versus Neutral: t(156) ⫽ 5.48, p ⬍
.001). Individuals in the Awe/Fear condition felt significantly
more negative than individuals in all other conditions (vs. Awe:
t(156) ⫽ 8.52, p ⬍ .001; vs. Elevation/Admiration: t(156) ⫽ 5.65,
p ⬍ .001; versus Nurturant Love: t(156) ⫽ 11.11, p ⬍ .001; versus
Neutral: t(156) ⫽ 3.94, p ⬍ .001).
Script independence. Study 3 results are summarized in Table 3. There was a significant omnibus main effect of Emotion
condition, F(4, 252) ⫽ 2.72, p ⫽ .030, ␩2 ⫽ .041. Planned
comparisons revealed that participants in all three awe-related

T3

 APA NLM
tapraid5/emo-emo/emo-emo/emo99917/emo3256d17z xppws S⫽1 1/11/17 10:55 Art: 2016-1427

DANVERS AND SHIOTA

8

Table 3
Memory for Story Details by Question Category and Emotion, Study 3
Category

Awe
M (SD)

Awe/Fear
M (SD)

Elevation/Admiration
M (SD)

Nurturant love
M (SD)

Neutral
M (SD)

Script independence
Intake of detail

3.19a (1.23)
3.75a (1.00)

3.29a (1.15)
3.51ab (1.08)

3.24a (1.02)
3.34ab (1.17)

2.70b (1.16)
3.55ab (.95)

2.89ab (1.06)
3.29b (1.24)

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Note. Differing superscripts indicate differences between conditions of at least the p ⱕ .05 level of significance.

conditions scored significantly higher on Script Independence than
participants in the Nurturant Love condition: (vs. Awe, M difference ⫽ .49[0.06 –0.91], t(252) ⫽ 2.25, p ⫽ .025, d ⫽ 0.30; vs.
Awe/Fear, M difference ⫽ .59[0.16 –1.02], t(252) ⫽ 2.72, p ⫽
.007, d ⫽ 0.36; versus Elevation/Admiration, M difference ⫽
.54[0.11–0.97], t(252) ⫽ 2.46, p ⫽ .015, d ⫽ 0.33). Participants in
all three awe conditions also scored directionally higher on script
independence than those in the Neutral condition, although none of
these contrasts were significant (Awe vs. Neutral: t(252) ⫽ 1.31,
p ⫽ .193, d ⫽ 0.26; Awe/Fear vs. Neutral: t(252) ⫽ 1.75, p ⫽
.081, d ⫽ 0.356, Elevation/Admiration versus Neutral: t(252) ⫽
1.51, p ⫽ .132, d ⫽ 0.308). There were no significant differences
among the awe conditions (Awe vs. Awe/Fear: t(252) ⫽ ⫺0.48, p ⫽
.635, d ⫽ ⫺0.09; Awe vs. Elevation/Admiration: t(252) ⫽ ⫺0.23,
p ⫽ .818, d ⫽ ⫺0.05; Awe/Fear versus Elevation/Admiration:
t(252) ⫽ 0.24, p ⫽ .810, d ⫽ 0.048). The Nurturant Love and Neutral
conditions did not significantly differ, t(252) ⫽ ⫺0.85, p ⫽ .398,
d ⫽ ⫺0.17.
Intake of unexpected detail. When analyses were run with
gender as a factor, a significant main effect of gender was observed, F(1, 245) ⫽ 4.64, p ⫽ .032, ␩2 ⫽ .019; women scored
higher on Intake of Unexpected Detail (i.e., script-irrelevant true
items). Thus, analysis used a two-way ANOVA with both Emotion
and Gender as between-subjects factors. Neither the omnibus main
effect of Emotion, F(4, 245) ⫽ 1.17, p ⫽ .323, ␩2 ⫽ .019, nor the
Emotion ⫻ Gender interaction, was significant. No contrast between any of the awe family conditions and either the Nurturant Love
or Neutral condition was significant (Awe vs. Nurturant Love:
t(252) ⫽ 1.02, p ⫽ .308, d ⫽ 0.19; Awe/Fear vs. Nurturant Love:
t(252) ⫽ ⫺0.16, p ⫽ .870, d ⫽ ⫺0.03; Elevation/Admiration versus
Nurturant Love: t(252) ⫽ ⫺0.97, p ⫽ .332, d ⫽ ⫺0.19; Awe versus
Neutral: M difference ⫽ .37[⫺0.07–0.82], t(245) ⫽ 1.66, p ⫽ .098,
d ⫽ 0.34; Awe/Fear versus Neutral: t(252) ⫽ 1.03, p ⫽ .302, d ⫽
0.21; Elevation/Admiration versus Neutral: t(252) ⫽ 0.26, p ⫽ .795,
d ⫽ 0.05). The Nurturant Love and Neutral conditions did not
significantly differ, t(252) ⫽ 1.22, p ⫽ .224, d ⫽ 0.24.
Script-relevant true and script-irrelevant false items. We
conducted 5 ⫻ 2 (Emotion by Gender) between subjects ANOVAs
on the script relevant true items and the script irrelevant false
items. For the script irrelevant false items, there were no significant effects of Emotion or the Emotion ⫻ Gender interaction, but
there was a main effect of Gender, F(1, 256) ⫽ 4.437, p ⫽ .036.
Women answered more of these questions correctly than men
(3.66 vs. 3.40). No planned comparisons between Emotion conditions were significant. For the script relevant true items, there were
also no significant effects of Emotion or the Emotion x Gender
interaction, but a main effect of Gender, F(1, 256) ⫽ 7.817, p ⫽
.006. Women also answered more script relevant true items cor-

rectly than men (3.23 vs. 2.85). No planned comparisons between
Emotion conditions were significant.

Study 3 Discussion
Study 3 contrasted awe with yet another positive emotion—
nurturant love—and again found that awe increased script independent processing of the “romantic dinner” story compared with
that condition. This further extends support for our hypothesis that
awe has unique effects on cognitive processing relative to other
positive-valence emotions. This effect generalized from “pure”
awe to two other awe-related states: an awe-fear blend, and a blend
of elevation and admiration. Notably, nurturant love did not increase script independence; scores in this condition were directionally though not significantly lower than in the neutral condition. As in Study 2, Study 3 again failed to replicate the effect of
awe on intake of unexpected detail seen in Study 1.

Meta-Analysis of Awe Effects, and Mediation by
Subjective Awe
Results of the contrasts between the awe and control conditions
for each of the key dependent variables were statistically significant in some studies, but not in others. To get a more precise
estimate of effect sizes, we conducted meta-analyses comparing
awe (a) to the neutral controls and (b) to the average of the positive
condition controls, separately for script independence and intake of
unexpected detail. Meta-analyses were conducted assuming random effects, since each study differed in its method of emotion
elicitation; this option leads to more conservative results. When
comparing awe to other positive emotions, we combined the enthusiasm, contentment, and pride conditions into an alternative
positive category for Study 2. We combined the awe, awe/fear, and
elevation/admiration conditions into the awe category for Study 3.3
All calculations were done using the Metafor package for R
(Viechtbauer, 2010).
To assess whether effects of awe manipulation were accounted
for by the subjective experience of awe, we also conducted mediation meta-analyses contrasting awe against all control conditions
(neutral and positive emotions) in predicting self-reported awe
across all three studies, and asking whether self-reported awe, in
turn, predicted script independence and intake of unexpected detail. Each mediation analysis also included subjective affect valence and intensity (these two items had been included in the manipulation check, alongside feelings of specific emotions, in all studies;
3
Comparing just the awe condition to the nurturant love condition in
Study 3 led to the same substantive results.

Fn3

 APA NLM
tapraid5/emo-emo/emo-emo/emo99917/emo3256d17z xppws S⫽1 1/11/17 10:55 Art: 2016-1427

AWE AND MEMORY FOR NARRATIVE DETAIL

see Study 1 measures for additional detail) as covariates and additional potential mediators, to ensure that effects on the cognitive
outcomes were accounted for by awe above and beyond these dimensions. The awe, awe/fear, and elevation/admiration conditions from
Study 3 were all coded as awe conditions. We tested the mediated
effect using the product of coefficients method, using bootstrapping to
estimate standard errors (MacKinnon, 2008).

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Script Independence

F1

Fn4

For the script independence index, we found significant differences between awe and general positive (Study 1), enthusiasm
(Study 2), and nurturant love (Study 3) conditions; we also found
a significant difference between awe and neutral in Study 2, but
not Study 1 or 3. In a meta-analysis including all three studies, we
found that the cumulative effect of awe compared with neutral
control was significant (Mdiff ⫽ 0.42, SEdiff ⫽ 0.13, Z ⫽ 3.11, p ⫽
.002; see Figure 1A). There was no evidence for heterogeneity
across the awe versus neutral comparisons (Q(2) ⫽ 1.22, p ⫽
.542). The effect of awe as compared with alternative positive
emotions approached significance (Mdiff ⫽ 0.32, SEdiff ⫽ 0.17,
Z ⫽ 1.84, p ⫽ .066; see Figure 1B). There was evidence of
significant heterogeneity of this effect across studies, likely reflecting the fact that awe was compared with several different
positive emotions across studies (Q (2) ⫽ 7.53, p ⫽ .023).
Probing this possibility further, we conducted a regression analysis on the combined data sets with self-reported awe, enthusiasm,
contentment, love, and tenderness (which were measured as part of
the manipulation check in all studies) entered as simultaneous
predictors of script independence. Self-reported awe showed a
significant positive effect on script independence (B ⫽ 0.184
[0.142, 0.226], t(719) ⫽ 11.595, p ⬍ .001). Self-reported tenderness also showed a significant positive effect (B ⫽ 0.106 [0.058,
0.153], t(719) ⫽ 4.386, p ⬍ .001). Self-reported enthusiasm
(B ⫽ ⫺0.126 [⫺0.181, ⫺0.072], t(719) ⫽ ⫺4.577, p ⬍ .001) and
contentment (B ⫽ ⫺0.110 [⫺0.164, ⫺0.056], t(719) ⫽ ⫺3.982,
p ⬍ .001) both showed significant negative associations with script
independence. Self-reported love did not significantly predict script
independence (B ⫽ ⫺0.005 [⫺0.051, 0.042], t(719) ⫽ ⫺0.194, p ⫽
.847).
In the mediation analysis (see Figure 1C), awe condition (relative to all controls) significantly increased self-reported awe (b ⫽
2.03, Z ⫽ 10.00, p ⬍ .001),4 and self-reported awe significantly
predicted script independence (b ⫽ 0.05, Z ⫽ 2.71, p ⬍ .001),
above and beyond valence and intensity. The test of the mediated
effect through awe was significant (a ⴱ b ⫽ 0.115, Z ⫽ 2.63, p ⫽
.009). Awe condition did not significantly predict either valence or
intensity, although both of these variables predicted script independence (valence: b ⫽ .129, Z ⫽ 4.10, p ⬍ .001; intensity:
b ⫽ ⫺.129, Z ⫽ ⫺3.63, p ⬍ .001), and neither valence nor
intensity significantly mediated the effect of awe condition on
script independence. The direct effect of awe condition was also
significant (c’ ⫽ 1.71, Z ⫽ 14.88, p ⬍ .001), suggesting that the
awe conditions shaped script independence through additional
mechanisms not captured by subjective emotional experience.
Overall, the model explained 32.7% of the variance in our measure
of script independence.

9

Intake of Unexpected Detail
For the intake of unexpected detail index, we found a significant
difference between awe and the general positive and neutral conditions in Study 1, and the comparison between awe and neutral
approached significance in Study 3. In meta-analysis, the cumulative effect comparing awe to neutral control was nonsignificant
(Mdiff ⫽ 0.28, SEdiff ⫽ 0.22, Z ⫽ 1.26, p ⫽ .208), but with
significant heterogeneity across studies (Q(2) ⫽ 7.12, p ⫽ .029).
From the forest plot (Figure 2A), it appears that the comparison
between awe and neutral had a similar effect in Studies 1 and 3, but
a very different effect in Study 2. The cumulative effect comparing
awe to alternative positive emotions was small and nonsignificant
(Mdiff ⫽ 0.08, SEdiff ⫽ 0.11, Z ⫽ 0.70, p ⫽ .486; Figure 2B), with
no significant heterogeneity across studies (Q(2) ⫽ 4.18, p ⫽
.124).
As for script independence, we conducted a regression analysis
on the combined data sets with self-reported awe, enthusiasm,
contentment, love, and tenderness entered as simultaneous predictors of intake of unexpected detail. Self-reported awe showed a
significant positive effect (B ⫽ 0.215 [0.171, 0.259], t(719) ⫽
9.576, p ⬍ .001), as did self-reported tenderness (B ⫽ 0.144 [0.094,
0.194], t(719) ⫽ 5.666, p ⬍ .001). Self-reported enthusiasm
(B ⫽ ⫺0.131 [⫺0.188, ⫺0.074], t(719) ⫽ ⫺4.489, p ⬍ .001) and
contentment (B ⫽ ⫺0.129 [⫺0.186, ⫺0.072], t(719) ⫽ ⫺4.409, p ⬍
.001) both showed significant negative associations with intake of
unexpected detail. Self-reported love did not significantly predict
intake of unexpected detail (B ⬍ .001 [⫺0.049, 0.050], t(719) ⫽
0.037, p ⫽ .970).
The mediation analysis (Figure 2C) provided stronger support
for an effect of awe manipulation on intake of unexpected detail.
In this analysis, awe (vs. all controls) condition significantly
increased self-reported awe (b ⫽ 1.98, Z ⫽ 9.88, p ⬍ .001), which
in turn significantly predicted intake of novel detail (b ⫽ 0.08, Z ⫽
3.84, p ⬍ .001). The test of the mediated effect through awe was
significant (a ⴱ b ⫽ 0.157, Z ⫽ 3.44, p ⫽ .001). This mediated
effect was observed controlling for the implications of valence and
intensity, and indeed, controlling for these dimensional aspects of
subjective affect appears to have revealed a significant effect of
awe condition to emerge that was not present in the meta-analyses
without mediation, reported above. Being in an awe versus control
condition was not a significant predictor of valence (b ⫽ ⫺.237,
Z ⫽ ⫺1.46, p ⫽ .143) or intensity (b ⫽ ⫺.011, Z ⫽ ⫺0.07, p ⫽
.943), but valence and intensity were significant predictors of
intake of novel detail (valence: b ⫽ .173, Z ⫽ 5.87, p ⬍ .001;
intensity: b ⫽ ⫺.123, Z ⫽ ⫺3.72, p ⬍ .001). The mediated effects
through valence (a ⴱ b ⫽ ⫺.041, Z ⫽ ⫺1.49, p ⫽ .162) and
intensity (a ⴱ b ⫽ 0.001, Z ⫽ 0.07, p ⫽ .945) were not significant.
In addition to the effect mediated by subjective awe, the direct
effect of awe condition on intake of novel detail was significant as
well (c’ ⫽ 1.73, Z ⫽ 15.78, p ⬍ .001), suggesting that the awe
conditions influenced this outcome through mechanisms not captured by subjective emotional experience. Overall, the model explained 33.6% of the variance in our measure of intake of unexpected detail.
4
Unstandardized regression coefficients are presented in the text. Standardized regression coefficients are presented in figures.

F2

 APA NLM
tapraid5/emo-emo/emo-emo/emo99917/emo3256d17z xppws S⫽1 1/11/17 10:55 Art: 2016-1427

DANVERS AND SHIOTA

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

10

AQ: 9

Figure 1. Meta-analyses of awe effects, script independence. A. Forest plot, awe versus neutral. B. Forest
plot, awe versus alternative positive emotions. C. Mediation analysis. In forest plots (A & B), “RE
Model” ⫽ random effects model. For each study, size of the box representing mean effect size estimate
indicates weighting of that study in the meta-analysis. Numeric values in each row indicate the mean and
95% confidence interval of effect size estimates in bootstrapping analyses; positive values indicate greater
accuracy in awe conditions as compared with control conditions. In the mediation path diagram (C), values
are standardized coefficients. ⴱ p ⬍ .05, ⴱⴱ p ⬍ .01, ⴱⴱⴱ p ⬍ .001.

General Discussion
Functional accounts of awe suggest that this emotion should
promote increased cognitive accommodation, inhibiting people’s
default tendency to filter current experience through the lens of
prior knowledge and expectations (Keltner & Haidt, 2003; Shiota
et al., 2007). Previous research has found evidence for this effect
in the context of persuasive message processing, finding that
participants in an awe state responded more strongly to argument

quality than those in neutral or most other positive emotion states;
the latter were easily persuaded by a large number of arguments
regardless of quality (Griskevicius, Shiota, & Neufeld, 2010).
Awe’s effects in the persuasion studies could have been accounted
for by either or both of two, specific cognitive processes: suppressing the influence of heuristics and stored knowledge; and/or systematically encoding and using novel information from the environment. The studies reported here were designed to extend

 APA NLM
tapraid5/emo-emo/emo-emo/emo99917/emo3256d17z xppws S⫽1 1/11/17 10:55 Art: 2016-1427

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

AWE AND MEMORY FOR NARRATIVE DETAIL

11

Figure 2. Meta-analyses of awe effects, intake of unexpected detail. A. Forest plot, awe versus neutral. B.
Forest plot, awe versus alternative positive emotions. C. Mediation analysis. In forest plots (A & B), “RE
Model” ⫽ random effects model. For each study, size of the box representing mean effect size estimate indicates
weighting of that study in the meta-analysis. Numeric values in each row indicate the mean and 95% confidence
interval of effect size estimates in bootstrapping analyses; positive values indicate greater accuracy in awe
conditions as compared with control conditions. In the mediation path diagram (C), values are standardized
coefficients. ⴱ p ⬍ .05, ⴱⴱ p ⬍ .01, ⴱⴱⴱ p ⬍ .001.

evidence linking awe to cognitive accommodation, while also
teasing apart these two more subtle aspects of accommodation, and
assessing the effect of awe on each.
In three experiments, each using a different approach to eliciting
awe, we found that awe consistently enhanced script independence. Overreliance on event scripts in processing new information is one example of filtering through stored knowledge. In these
studies, awe reduced people’s tendency to falsely impute details

from the mainstream U.S. “romantic dinner” script into memory of
a specific story, heard for the first time. This effect of awe set it
apart from a variety of other, specific positive emotions including
enthusiasm, nurturant love, and a blend of positive emotions.
Meta-analysis revealed a significant difference of awe from neutral
affect in this regard, with a consistent effect size across studies, as
well as a difference in the hypothesized direction, though more
variable, from alternative positive emotions (presumably reflecting

 APA NLM
tapraid5/emo-emo/emo-emo/emo99917/emo3256d17z xppws S⫽1 1/11/17 10:55 Art: 2016-1427

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

12

DANVERS AND SHIOTA

the variety of positive emotion controls). Mediation analysis revealed that, across all studies, subjective feelings of awe partially
mediated the effect of awe manipulation on script independence.
However, a substantial direct effect of awe condition (i.e., not
mediated by subjective feelings) was significant as well.
Evidence for the “intake of detail” effect was more variable
across studies. Awe showed this effect compared with a general
positive manipulation in one study, and compared with neutral in
two of the three studies, but not in the other contrasts. In metaanalysis, there was no consistent finding relating awe condition to
intake of unexpected detail as compared with neutral controls
(although there was significant heterogeneity in effect sizes across
studies), or other positive emotions considered as a group. It is
possible that differences in the methods used to evoke target
emotions account for the inconsistent findings for this aspect of
cognition. Studies 1 and 3, which showed stronger awe effects,
used novel visual stimuli (either films or photos) to elicit emotion
states, whereas Study 2 used a relived emotion task. It may be that
exposure to novel awe stimuli primed greater intake of detail than
did the more familiar awe memories. Alternatively, it may be that
the awe-eliciting movie and photos primed greater visualization of
the “romantic dinner” story, which in turn facilitated memory for
detail. A significant effect of awe condition did emerge after
controlling for subjective affect valence and intensity in the overall
mediation analysis. This effect was mediated, in part, by selfreported feelings of awe, though a significant direct effect of awe
condition remained as well. Future studies clarifying the mechanisms and boundary conditions of awe effects on intake of detail
will be valuable.
Our results are not consistent with a view of positive emotions
as universally increasing reliance on heuristic processing (Fiedler,
2001). In fact, no positive emotion other than awe differed significantly from neutral in these studies, although effects of general
positive emotion (Study 1) and nurturant love (Study 3) were in the
heuristic-facilitating direction predicted by much prior research.
Moreover, in meta-analyses more positive ratings of subjective
affect valence significantly predicted greater script independence
and intake of novel detail, suggesting reduced reliance on heuristics. We recommend some caution in interpreting the affectvalence effects as they include the various awe conditions, most of
which were experienced by participants as highly pleasant in
general as well as awe-inducing in particular; this may account in
part for the unexpected direction of the effect. Affect valence and
intensity were included in the mediation models to assess whether
subjective awe, specifically, mediated the cognitive effects of
experimental awe manipulation, above and beyond these dimensional aspects of subjective affect. Notably, controlling for valence
in the mediation analyses also ruled out the possibility that awe’s
cognitive effects differed from those of other positive emotions
because it was experienced as more negative. The lack of a
significant pathway from emotion condition (awe vs. other conditions) to valence indicates that awe did not typically have a
negative tinge.
The affect-as-cognitive-feedback (ACF) model’s account of the
effects of positive mood (Isbell, Lair, & Rovenpour, 2013) also
does not explain the present findings. The ACF model suggests
that positive mood will encourage individuals to continue using
whatever cognitive processing style is either the automatic default,
or previously primed. A great deal of research indicates that

reliance on heuristics and stored knowledge is a default among
mainstream North American participants such as those in the
present studies (Schwarz, 2001). In no study reported above was a
systematic cognitive style actively primed prior to emotion manipulation, yet awe clearly and consistently reduced participants’
reliance on event scripts. It is possible that simply being in the
experimental session primed more careful attention and systematic
processing, which was then amplified by positive affect. However,
this would not explain why some positive emotion manipulations
led to directional reduction in script independence (Studies 1 and
3), nor why awe manipulation produced cognitive effects over and
above those of subjective valence. The effect of awe on cognitive
processing also cannot parsimoniously be explained by motivational direction (approach/avoidance) or arousal. In Study 2, the
effect of awe on script independence differed from that of contentment—a low-approach, low-arousal emotion—as well as that
of enthusiasm—a high-approach, high-arousal emotion. These two
conditions produced effects that were similar to each other, but
distinct from awe. Hedonic contingency theory (Wegener & Petty,
1994) also fails to account for the differences between awe’s
effects and the effects of other positive emotions, which emerged
even when the focus of attention (romantic dinner story) and
resulting implications for subsequent mood were held constant.
Rather, the findings across studies are consistent with theoretical
perspectives emphasizing differences among “discrete” emotions
in general, and the cognitive function of awe in particular (Keltner
& Haidt, 2003; Shiota et al., 2007). Supplemental meta-analyses
revealed that, across studies, subjective feelings of awe and tenderness significantly predicted reduced reliance on the event script
heuristic (i.e., greater script independence and intake of novel
detail), whereas subjective enthusiasm and contentment predicted
greater reliance on event scripts, and love had no effect. Although
these patterns are consistent with our expectation that different
positive emotions have differing effects, we caution against assuming that subjective feelings necessarily drive all observable
cognitive outcomes. One important finding to emerge from the
mediation analyses is that the effects of awe condition on cognitive
processing were only partially mediated through subjective awe.
The significant direct effects that remained indicate that awe
increased both script independence and intake of unexpected detail
through cognitive mechanisms that were not consciously accessible, at least in the form of subjective emotional experience.
People seek and value experiences of awe. They expend considerable time, effort, and resources traveling to places of great
natural beauty, like the Grand Canyon and Niagara Falls; creating
and visiting vast products of human ingenuity, like the Taj Mahal
and the Eiffel Tower; and experiencing works of art that help one
see the world through new eyes, like Van Gogh’s Starry Night and
Picasso’s Guernica. For many, these experiences can feel transcendent—as if one’s worldview is transformed. Yet psychological
science has only recently begun to document how awe affects
cognition and behavior in concrete, objective, quantifiable terms.
Results of this program of research suggest that, consistent with
theory as well as phenomenological experience, awe helps individuals open their minds. It does this specifically by suppressing
the usual expectations through which people filter experience of
the world, and enhancing people’s attention to unexpected details.
Although not large, the awe effects documented in these studies
are consistent with theoretical definitions of this emotion and

 APA NLM
tapraid5/emo-emo/emo-emo/emo99917/emo3256d17z xppws S⫽1 1/11/17 10:55 Art: 2016-1427

AWE AND MEMORY FOR NARRATIVE DETAIL

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

AQ: 6

associated predictions regarding its impact on cognitive processing. These studies add to a new and rapidly expanding body of
evidence on awe. In just the last decade, strong programs of
research have documented nonverbal expressions of awe (e.g.,
Campos et al., 2013; Simon-Thomas et al., 2009), physiological
aspects of awe (Shiota et al., 2011), and effects of awe on prosocial behavior (e.g., Piff, Dietze, Feinberg, Stancato, & Keltner,
2015; Prade & Saroglou, 2016), language use (Darbor, Lench,
Davis, & Hicks, 2016), absorption with a stimulus (Van Elk,
Karinen, Specker, Stamkou, & Baas, 2016), and agency detection
(Valdesolo & Graham, 2014), all of which are distinct from the
effects of other positive emotions.
This study was conducted with a population of undergraduate
college students, so further research is needed to determine
whether it generalizes to populations of different ages, education
levels, and cultural backgrounds. It is also important to note that,
relative to real-life experiences of awe (i.e., those not involving
laptop computer screens and relived memories), the emotions
elicited by our awe manipulations were likely on the weak side.
We would anticipate that more intense awe experiences, such as
standing at the rim of the Grand Canyon, would lead to stronger
cognitive effects. Future research is needed to compare these
effects of awe with those of other epistemological emotions (those
whose primary adaptive functions are thought to involve learning),
such as interest and amusement.
Future studies are also needed to assess specific mechanisms of
these effects. For example, appraisals of smallness (Shiota et al.,
2007) or uncertainty (Valdesolo & Graham, 2014) might account
for our observed effects of awe stimuli on cognition. With respect
to the exact cognitive process modulated by awe, one possibility is
that awe directly inhibits activation of heuristics and facilitates
accommodation (Griskevicius, Shiota, & Neufeld, 2010; Keltner &
Haidt, 2003). Another possibility is suggested by attention restoration theory, which posits that stimuli commanding intense involuntary attention—such as awe stimuli— give effortful attention
control mechanisms a brief respite, allowing them to rebound in a
subsequent task (Berman, Jonides, & Kaplan, 2008). Third, an
extension of the ACF model may account for the effects of awe
observed here. Awe may promote rejection of one’s default or
current cognitive processing style—whatever that style may
be—in much the same way previously observed for negative
mood. The ACF account allows for specific emotions to have such
effects (Huntsinger, Isbell, & Clore, 2014). Future research is
needed to determine which of these specific processes is facilitated by awe. Another important future direction for this research will be extending these basic findings to more applied
contexts, such as eyewitness memory, evaluating scientific evidence, and creative problem solving. Cultivating awe in people’s everyday lives may help them approach more tasks in an
open-minded way.
Beyond advancing knowledge about awe, this research also adds
to the growing body of evidence that that positive emotions can
and should be differentiated based on their effects on cognition.
Although most previous literature on cognitive effects has characterized emotion broadly as either positive or negative, we found
here that taking a more fine-grained approach to positive emotion— one that more fully accounts for its variability—led to
theoretical predictions borne out by empirical tests. All positive
emotions are not the same. Theorizing about the diversity of

13

positive emotions, and using a functional perspective to generate
hypotheses, can help affective scientists better understand how the
mind works.

References
Beedie, C., Terry, P., & Lane, A. (2005). Distinctions between emotion and
mood. Cognition and Emotion, 19, 847– 878. http://dx.doi.org/10.1080/
02699930541000057
Berman, M. G., Jonides, J., & Kaplan, S. (2008). The cognitive benefits of
interacting with nature. Psychological Science, 19, 1207–1212. http://dx
.doi.org/10.1111/j.1467-9280.2008.02225.x
Bless, H. (2001). The relation between mood and the use of general
knowledge structures. In L. L. Martin & G. L. Clore (Eds.), Theories of
mood and cognition: A user’s handbook (pp. 9 –29). Mahwah, NJ:
Erlbaum.
Bless, H., Mackie, D. M., & Schwarz, N. (1992). Mood effects on attitude
judgments: Independent effects of mood before and after message elaboration. Journal of Personality and Social Psychology, 63, 585–595.
http://dx.doi.org/10.1037/0022-3514.63.4.585
Bless, H., Schwarz, N., Clore, G. L., Golisano, V., Rabe, C., & Wölk, M.
(1996). Mood and the use of scripts: Does a happy mood really lead to
mindlessness? Journal of Personality and Social Psychology, 71, 665–
679. http://dx.doi.org/10.1037/0022-3514.71.4.665
Bodenhausen, G. V., Kramer, G. P., & Susser, K. (1994). Happiness and
stereotypic thinking in social judgment. Journal of Personality and
Social Psychology, 66, 621– 632. http://dx.doi.org/10.1037/0022-3514
.66.4.621
Bodenhausen, G. V., Sheppard, L. A., & Kramer, G. P. (1994). Negative
affect and social judgment: The differential impact of anger and sadness.
European Journal of Social Psychology, 24, 45– 62. http://dx.doi.org/10
.1002/ejsp.2420240104
Brody, L. R., & Hall, J. A. (2008). Gender and emotion in context. In M.
Lewis, J. M. Haviland-Jones, & L. F. Barrett (Eds.), Handbook of
emotions (3rd ed., pp. 395– 408). New York, NY: Guilford.
Cacioppo, J. T., & Petty, R. E. (1989). The elaboration likelihood model:
The role of affect and affect-laden information on processing in persuasion. In A. Tybout & P. Caferata (Eds.), Cognitive and affective responses to advertising (pp. 69 – 89). Lexington, MA: Lexington Books.
Campos, B., Shiota, M. N., Keltner, D., Gonzaga, G. C., & Goetz, J. L.
(2013). What is shared, what is different? Core relational themes and
expressive displays of eight positive emotions. Cognition and Emotion,
27, 37–52. http://dx.doi.org/10.1080/02699931.2012.683852
Clore, G. L. (1992). Cognitive phenomenology: Feelings and the construction of judgment. In L. L. Martin & A. Tesser (Eds.), The construction
of social judgments (pp. 133–163). Hillsdale, NJ: Erlbaum.
Darbor, K. E., Lench, H. C., Davis, W. E., & Hicks, J. A. (2016).
Experiencing versus contemplating: Language use during descriptions of
awe and wonder. Cognition and Emotion, 30, 1188 –1196.
DeSteno, D., Petty, R. E., Wegener, D. T., & Rucker, D. D. (2000). Beyond
valence in the perception of likelihood: The role of emotion specificity.
Journal of Personality and Social Psychology, 78, 397– 416. http://dx
.doi.org/10.1037/0022-3514.78.3.397
Eagleman, D. M., Tse, P. U., Buonomano, D., Janssen, P., Nobre, A. C., &
Holcombe, A. O. (2005). Time and the brain: How subjective time
relates to neural time. The Journal of Neuroscience, 25, 10369 –10371.
http://dx.doi.org/10.1523/JNEUROSCI.3487-05.2005
Easterbrook, J. A. (1959). The effect of emotion on cue utilization and the
organization of behavior. Psychological Review, 66, 183–201. http://dx
.doi.org/10.1037/h0047707
Ekman, P. (1992). An argument for basic emotions. Cognition and Emotion, 6, 169 –200. http://dx.doi.org/10.1080/02699939208411068
Fiedler, K. (2001). Affective states trigger processes of assimilation and

 APA NLM
tapraid5/emo-emo/emo-emo/emo99917/emo3256d17z xppws S⫽1 1/11/17 10:55 Art: 2016-1427

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

14

AQ: 7

DANVERS AND SHIOTA

accommodation. In L. L. Martin & G. L. Clore (Eds.), Theories of mood
and cognition: A user’s guidebook. Mahwah, NJ: Erlbaum.
Fredrickson, B. L., & Branigan, C. (2005). Positive emotions broaden the
scope of attention and thought-action repertoires. Cognition and Emotion, 19, 313–332. http://dx.doi.org/10.1080/02699930441000238
Gable, P. A., & Harmon-Jones, E. (2008). Approach-motivated positive
affect reduces breadth of attention. Psychological Science, 19, 476 – 482.
http://dx.doi.org/10.1111/j.1467-9280.2008.02112.x
Gable, P. A., & Harmon-Jones, E. (2010a). The effect of low versus high
approach-motivated positive affect on memory for peripherally versus
centrally presented information. Emotion, 10, 599 – 603. http://dx.doi
.org/10.1037/a0018426
Gable, P., & Harmon-Jones, E. (2010b). The blues broaden, but the nasty
narrows: Attentional consequences of negative affects low and high in
motivational intensity. Psychological Science, 21, 211–215. http://dx.doi
.org/10.1177/0956797609359622
Gable, P. A., & Harmon-Jones, E. (2011). Attentional consequences of
pregoal and postgoal positive affects. Emotion, 11, 1358 –1367. http://
dx.doi.org/10.1037/a0025611
Gable, P. A., Poole, B. D., & Harmon-Jones, E. (2015). Anger perceptually
and conceptually narrows cognitive scope. Journal of Personality and
Social Psychology, 109, 163–174. http://dx.doi.org/10.1037/a0039226
Griskevicius, V., Shiota, M. N., & Neufeld, S. L. (2010). Influence of
different positive emotions on persuasion processing: A functional evolutionary approach. Emotion, 10, 190 –206. http://dx.doi.org/10.1037/
a0018421
Griskevicius, V., Shiota, M. N., & Nowlis, S. M. (2010). The many shades
of rose-colored classes: An evolutionary approach to the influence of
different positive emotions. Journal of Consumer Research, 37, 238 –
250. http://dx.doi.org/10.1086/651442
Haidt, J. (2000). The positive emotion of elevation. Prevention & Treatment, 3. http://dx.doi.org/10.1037/1522-3736.3.1.33c
Harmon-Jones, E., Gable, P. A., & Price, T. F. (2013). Does negative affect
always narrow and positive affect always broaden the mind? Considering the influence of motivational intensity on cognitive scope. Current
Directions in Psychological Science, 22, 301–307. http://dx.doi.org/10
.1177/0963721413481353
Harmon-Jones, E., Price, T. F., & Gable, P. A. (2012). The influence of
affective states on cognitive broadening/narrowing: Considering the
importance of motivational intensity. Social and Personality Psychology
Compass, 6, 314 –327. http://dx.doi.org/10.1111/j.1751-9004.2012
.00432.x
Hunsinger, M., Isbell, L. M., & Clore, G. L. (2012). Sometimes happy
people focus on the trees and sad people focus on the forest: Contextdependent effects of mood in impression formation. Personality and
Social Psychology Bulletin, 38, 220 –232. http://dx.doi.org/10.1177/
0146167211424166
Huntsinger, J. R., Clore, G. L., & Bar-Anan, Y. (2010). Mood and globallocal focus: Priming a local focus reverses the link between mood and
global-local processing. Emotion, 10, 722–726. http://dx.doi.org/10
.1037/a0019356
Huntsinger, J. R., Isbell, L. M., & Clore, G. L. (2014). The affective control
of thought: Malleable, not fixed. Psychological Review, 121, 600 – 618.
http://dx.doi.org/10.1037/a0037669
Huntsinger, J. R., Sinclair, S., & Clore, G. L. (2009). Affective regulation
of implicitly measured stereotypes and attitudes: Automatic and controlled processes. Journal of Experimental Social Psychology, 45, 560 –
566. http://dx.doi.org/10.1016/j.jesp.2009.01.007
Isbell, L. M. (2004). Not all happy people are lazy or stupid: Evidence of
systematic processing in happy moods. Journal of Experimental Social
Psychology, 40, 341–349. http://dx.doi.org/10.1016/j.jesp.2003.06.003
Isbell, L. M., Burns, K. C., & Haar, T. (2005). The role of affect on the
search for global and specific target information. Social Cognition, 23,
529 –552. http://dx.doi.org/10.1521/soco.2005.23.6.529

Isbell, L. M., Lair, E. C., & Rovenpor, D. R. (2013a). Affect-as- information about processing styles: A cognitive malleability approach. Social
and Personality Psychology Compass, 7, 93–114. http://dx.doi.org/10
.1111/spc3.12010
Keltner, D., Ellsworth, P. C., & Edwards, K. (1993). Beyond simple
pessimism: Effects of sadness and anger on social perception. Journal of
Personality and Social Psychology, 64, 740 –752. http://dx.doi.org/10
.1037/0022-3514.64.5.740
Keltner, D., & Haidt, J. (2003). Approaching awe, a moral, spiritual, and
aesthetic emotion. Cognition and Emotion, 17, 297–314. http://dx.doi
.org/10.1080/02699930302297
Kreibig, S. D. (2010). Autonomic nervous system activity in emotion: A
review. Biological Psychology, 84, 394 – 421. http://dx.doi.org/10.1016/
j.biopsycho.2010.03.010
Lerner, J. S., & Keltner, D. (2001). Fear, anger, and risk. Journal of
Personality and Social Psychology, 81, 146 –159. http://dx.doi.org/10
.1037/0022-3514.81.1.146
Levenson, R. W. (1999). The intrapersonal functions of emotion. Cognition
and Emotion, 13, 481–504. http://dx.doi.org/10.1080/026999399379159
Mackie, D. M., & Worth, L. T. (1989). Processing deficits and the mediation of positive affect in persuasion. Journal of Personality and Social
Psychology, 57, 27– 40. http://dx.doi.org/10.1037/0022-3514.57.1.27
MacKinnon, D. P. (2008). Introduction to statistical mediation analysis.
Mahwah, NJ: Erlbaum.
Pham, M. T. (2007). Emotion and rationality: A critical review and
interpretation of empirical evidence. Review of General Psychology, 11,
155–178. http://dx.doi.org/10.1037/1089-2680.11.2.155
Piff, P. K., Dietze, P., Feinberg, M., Stancato, D. M., & Keltner, D. (2015).
Awe, the small self, and prosocial behavior. Journal of Personality and
Social Psychology, 108, 883– 899. http://dx.doi.org/10.1037/
pspi0000018
Rudd, M., Vohs, K. D., & Aaker, J. (2012). Awe expands people’s
perception of time, alters decision making, and enhances well-being.
Psychological Science, 23, 1130 –1136. http://dx.doi.org/10.1177/
0956797612438731
Ruder, M., & Bless, H. (2003). Mood and the reliance on the ease of
retrieval heuristic. Journal of Personality and Social Psychology, 85,
20 –32. http://dx.doi.org/10.1037/0022-3514.85.1.20
Sanbonmatsu, D. M., & Kardes, F. R. (1988). The effects of physiological
arousal on information processing and persuasion. Journal of Consumer
Research, 15, 379 –385. http://dx.doi.org/10.1086/209175
Schwarz, N. (2001). Feelings as information: Implications for affective
influences on information processing. In L. L. Martin & G. L. Clore
(Eds.), Theories of mood and cognition: A user’s guidebook (pp. 159 –
176). Mahwah, NJ: Erlbaum.
Schwarz, N., & Bless, H. (1991). Happy and mindless, but sad and smart?
The impact of affective states on analytical reasoning. In J. P. Forgas
(Ed.), Emotion and social judgments (pp. 55–72). New York, NY:
Wiley.
Schwarz, N., & Clore, G. L. (2007). Feelings and phenomenal experiences.
In E. T. Higgins & A. Kruglanski (Eds.), Social psychology: A handbook
of basic principles (2nd ed.). New York, NY: Guilford Press.
Shiota, M. N., Campos, B., Oveis, C., Hertenstein, M., Simon-Thomas, E.,
& Keltner, D. (2017). Beyond happiness: Toward a science of discrete
positive emotions. [Manuscript accepted for publication in American
Psychologist.]
Shiota, M. N., Keltner, D., & John, O. P. (2006). Positive emotion dispositions differentially associated with Big Five personality and attachment
style. The Journal of Positive Psychology, 1, 61–71. http://dx.doi.org/
10.1080/17439760500510833
Shiota, M. N., Keltner, D., & Mossman, A. (2007). The nature of awe:
Elicitors, appraisals, and effects on self-concept. Cognition and Emotion, 21, 944 –963. http://dx.doi.org/10.1080/02699930600923668

 APA NLM
tapraid5/emo-emo/emo-emo/emo99917/emo3256d17z xppws S⫽1 1/11/17 10:55 Art: 2016-1427

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

AWE AND MEMORY FOR NARRATIVE DETAIL
Shiota, M. N., Neufeld, S. L., Danvers, A. F., Osborne, E. A., Sng, O., &
Yee, C. I. (2014). Positive emotion differentiation: A functional approach. Social and Personality Psychology Compass, 8, 104 –117. http://
dx.doi.org/10.1111/spc3.12092
Shiota, M. N., Neufeld, S. L., Yeung, W. H., Moser, S. E., & Perea, E. F.
(2011). Feeling good: Autonomic nervous system responding in five
positive emotions. Emotion, 11, 1368 –1378. http://dx.doi.org/10.1037/
a0024278
Shiota, M. N., Thrash, T., Danvers, A. F., & Dombrowski, J. T. (2014).
Transcending the self: Awe, elevation, and inspiration. In M. Tugade,
M. N. Shiota, & L. Kirby (Eds.), Handbook of positive emotion (pp.
362–395). New York, NY: Guilford Press.
Simon, R. W., & Nath, L. E. (2004). Gender and emotion in the United
States: Do men and women differ in self-reports of feelings and expressive behavior? American Journal of Sociology, 109, 1137–1176. http://
dx.doi.org/10.1086/382111
Simon-Thomas, E. R., Keltner, D. J., Sauter, D., Sinicropi-Yao, L., &
Abramson, A. (2009). The voice conveys specific emotions: Evidence
from vocal burst displays. Emotion, 9, 838 – 846. http://dx.doi.org/10
.1037/a0017810
Storbeck, J., & Clore, G. L. (2005). With sadness comes accuracy; with
happiness, false memory: Mood and the false memory effect. Psychological Science, 16, 785–791. http://dx.doi.org/10.1111/j.1467-9280
.2005.01615.x
Tooby, J., & Cosmides, L. (2008). The evolutionary psychology of the
emotions and their relationship to internal regulatory variables. In M.
Lewis, J. M. Haviland-Jones, & L. F. Barrett (Eds.), Handbook of
emotions (pp. 114 –137). New York, NY: Guilford Press.

15

Valdesolo, P., & Graham, J. (2014). Awe, uncertainty, and agency detection. Psychological Science, 25, 170 –178. http://dx.doi.org/10.1177/
0956797613501884
Van Elk, M., Karinen, A., Specker, E., Stamkou, E., & Baas, M. (2016).
‘Standing in awe’: The effects of awe on body perception and the
relation with absorption. Collabra, 2, 4. [Advance online publication.]
http://dx.doi.org/10.1525/collabra.36
Viechtbauer, W. (2010). Conducting meta-analyses in R with the metafor
package. Journal of Statistical Software, 36, 1– 48. http://dx.doi.org/10
.18637/jss.v036.i03
Wearden, J. H. (2005). The wrong tree: Time perception and time experience in the elderly. In J. Duncan, L. Phillips, & P. McLeod (Eds.),
Measuring the mind: Speed, age, and control (pp. 137–158). Oxford,
UK: Oxford University Press. http://dx.doi.org/10.1093/acprof:oso/
9780198566427.003.0006
Wegener, D. T., & Petty, R. E. (1994). Mood management across affective
states: The hedonic contingency hypothesis. Journal of Personality and
Social Psychology, 66, 1034 –1048. http://dx.doi.org/10.1037/0022-3514
.66.6.1034
Wegener, D. T., Petty, R. E., & Smith, S. M. (1995). Positive mood can
increase or decrease message scrutiny: The hedonic contingency view of
mood and message processing. Journal of Personality and Social Psychology, 69, 5–15. http://dx.doi.org/10.1037/0022-3514.69.1.5

Received February 15, 2016
Revision received November 11, 2016
Accepted December 13, 2016 䡲

 JOBNAME: AUTHOR QUERIES PAGE: 1 SESS: 2 OUTPUT: Wed Jan 11 10:56:02 2017
/tapraid5/emo-emo/emo-emo/emo99917/emo3256d17z

AUTHOR QUERIES
AUTHOR PLEASE ANSWER ALL QUERIES

1

AQau—Please confirm the given-names and surnames are identified properly by the colors.
⫽ Given-Name, ⫽ Surname
The colors are for proofing purposes only. The colors will not appear online or in print.
AQ1—Author: This article has been lightly edited for grammar, style, and usage. Please compare
against your original document and make changes on these paged proofs. If no change is
required in response to a question, please write “OK as set” in the margin.
AQ2—Author: Please be sure to provide the name of the department(s) with which you and your
coauthors are affiliated at your respective institutes if you have not already done so. If you are
affiliated with a governmental department, business, hospital, clinic, VA center, or other
nonuniversity-based institute, please provide the city and U.S. state (or the city, province, and
country) in which the institute is based. Departments should be listed in the author footnote
only, not the byline. If you or your coauthors have changed affiliations since the article was
written, please include a separate note indicating the new department/affiliation: [author’s
name] is now at [affiliation].
AQ3—Author: Please indicate the source(s) of any facial subjects used in your manuscript. If your
photos have been taken from a noncommercial database, please provide the name of that
database. If your photos have been taken from a commercially copyrighted database, please
provide the name of the database AND indicate that you have received permission to reprint
and provide the name of the copyright holder(s). If you are using original photos, please
confirm that the subjects granted permission to have their likenesses reprinted.
AQ4 —Author: Your article contains supplemental material. If you have queries for references that
are not cited in the text, these can be ignored if the references are found in the supplemental
material.
AQ5—Author: Please review the typeset table(s) carefully against your original table(s) to verify
accuracy of editing and typesetting.
AQ6 —Author: Prade & Saroglou (2016) is not in your references. Please add to the reference list or
delete from the text.
AQ7—Author: Hunsinger et al. (2012) is not cited in your text. Please cite in the text or delete from
the references.

 JOBNAME: AUTHOR QUERIES PAGE: 2 SESS: 2 OUTPUT: Wed Jan 11 10:56:02 2017
/tapraid5/emo-emo/emo-emo/emo99917/emo3256d17z

AUTHOR QUERIES
AUTHOR PLEASE ANSWER ALL QUERIES
AQ8 —Author: Please confirm that all authors’ institutional affiliations (including
city/state/country locations) and correspondence information are correct as shown in the
affiliations footnote.
AQ9 —Author: Please confirm figure(s) and legend(s) as set.

2