Costs and Benefits of Acting Extraverted: A Randomized Controlled Trial

Abstract

Evidence suggests that extraverted (i.e., bold, agentic) behavior increases positive affect (PA), and could be targeted in wellbeing interventions. However, this evidence is either causally ambiguous or has questionable ecological validity, and the potential costs of sustained extraverted behavior have received minimal attention. To address these limitations, we conducted the first randomized controlled trial (RCT) examining the wellbeing benefits and costs of an extraverted behavior intervention conducted in everyday life. Participants (n = 147) were randomly assigned to an “act-extraverted” intervention or a “sham” (active control) intervention for one week in everyday life. Additional data for a contact control condition were obtained from a previous study (n = 76). Wellbeing outcomes included PA and negative affect (NA), feelings of authenticity, and tiredness—assessed both in the moment and retrospectively. There was a positive overall effect of the acting extraverted intervention on PA and authenticity. However, wellbeing outcomes also depended on dispositional extraversion: more introverted participants had weaker PA increases, experienced increased NA and tiredness, and decreased feelings of authenticity. Implications for wellbeing interventions and personality theory are discussed.

Full text

1
RUNNING HEAD: COSTS AND BENEFITS OF ACTING EXTRAVERTED
Note: This manuscript has been peer-reviewed and accepted for publication in the Journal
of Experimental Psychology: General.
© 2018, American Psychological Association. This paper is not the copy of record and may
not exactly replicate the final, authoritative version of the article. Please do not copy or cite
without authors' permission. The final article will be available, upon publication, via its
DOI: 10.1037/xge0000516
Costs and Benefits of Acting Extraverted: A Randomized Controlled Trial
Rowan Jacques-Hamilton1
Jessie Sun2
Luke D. Smillie1
1University of Melbourne
2University of California, Davis
Author Note
Rowan Jacques-Hamilton, Melbourne School of Psychological Science, University of
Melbourne; Jessie Sun, Department of Psychology, University of California, Davis; Luke
Smillie, Melbourne School of Psychological Science, University of Melbourne.
Portions of these findings were presented at (1) Australian Conference on Personality
and Individual Differences, Melbourne, Australia, November 18-19, 2016; (2) Society of
Australian Social Psychologists, Melbourne, Australia, April 20-22, 2017; (3) European
Conference of Personality, Zadar, Croatia, July 17-21, 2018.
We would like to acknowledge two anonymous reviewers, Anne Aulsebrook,
Kristelle Hudry, and Ted Schwaba for helpful comments on earlier versions of this
manuscript.
The preregistration, data and analysis scripts, and copies of materials used in this
study are available at https://osf.io/gcbxk/

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Correspondence concerning this article should be addressed to Rowan Jacques-
Hamilton, Melbourne School of Psychological Science, University of Melbourne, Australia,
3010, rowan.jacques@unimelb.edu.au.
Abstract
Evidence suggests that extraverted (i.e., bold, agentic) behavior increases positive affect
(PA), and could be targeted in wellbeing interventions. However, this evidence is either
causally ambiguous or has questionable ecological validity, and the potential costs of
sustained extraverted behavior have received minimal attention. To address these limitations,
we conducted a randomized controlled trial (RCT) examining the wellbeing benefits and
costs of an extraverted behavior intervention conducted in everyday life. Participants (n =
147) were randomly assigned to an “act-extraverted” intervention or a “sham” (active
control) intervention for one week in everyday life. Additional data for a contact control
condition were obtained from a previous study (n = 76). Wellbeing outcomes included PA
and negative affect (NA), feelings of authenticity, and tiredness—assessed both in the
moment and retrospectively. There was a positive overall effect of the acting extraverted
intervention on PA and authenticity. However, wellbeing outcomes also depended on
dispositional extraversion: more introverted participants had weaker PA increases,
experienced increased NA and tiredness, and decreased feelings of authenticity. Implications
for wellbeing interventions and personality theory are discussed.
Key Words: Extraversion; Extraverted Behavior; Positive Affect; Wellbeing; Randomized
Controlled Trial.
2

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Costs and Benefits of Acting Extraverted: A Randomized Controlled Trial
Decades of research have shown that personality traits are important predictors of
wellbeing, with the Big Five trait domain of extraversion emerging as a particularly robust
positive predictor of happiness and flourishing (see also Costa & McCrae, 1980; Keyes,
Shmotkin, & Ryff, 2002; Lucas & Diener, 2009; Nave, Sherman, & Funder, 2008; Schmutte
& Ryff, 1997; Smillie, DeYoung, & Hall, 2015; Steel, Schmidt, & Shultz, 2008; Sun,
Kaufman, & Smillie, 2017). Additionally, experience sampling studies and laboratory
experiments suggest that people feel happier in moments when they are acting more
extraverted (e.g., Blackie, Roepke, Forgeard, Jayawickreme, & Fleeson, 2014; Fleeson,
Malanos, & Achille, 2002; McNiel, Lowman, & Fleeson, 2010; Smillie, Wilt, Kabbani,
Garratt, & Revelle, 2015; Sun, Stevenson, Kabbani, Richardson, & Smillie, 2017; Weninger
& Holder, 2013; Wilt, Noftle, Fleeson, & Spain, 2012). This raises the intriguing possibility
that people could reap the wellbeing benefits of being more extraverted simply by acting
more extraverted (Blackie et al., 2014; Fleeson et al., 2002; McNiel et al., 2010; Weninger &
Holder, 2013; Wilt et al., 2012). However, to date, no published research has investigated
whether increasing one’s levels of extraverted behavior throughout daily life could improve
wellbeing. There has also been minimal attention given to the potential negative
consequences of acting extraverted, which might be especially likely for people lower in trait
extraversion (i.e., “introverts”; Cain, 2012; Little, 2008; Zelenski, Sobocko, & Whelan,
2014). Using the first randomized controlled trial on this topic1, we examined the positive and
negative wellbeing consequences of increasing real-world extraverted behavior.
Extraverted Behavior and Positive Affect
The association between trait extraversion and wellbeing is most pronounced for one
component of wellbeing in particular—positive affect (PA). Here, we define positive affect as
1 Although our intervention study appears to be the first of its kind in this literature, we became aware of a
manuscript reporting a similar study being prepared for publication while the present paper was under review
(Margolis & Lyubomirsky, 2018).
3

COSTS AND BENEFITS OF ACTING EXTRAVERTED
being a state with positive valence and high arousal (e.g., “excited”, “lively”), following
Watson and Tellegen (1985). The relation between extraversion and PA has been known for
decades (Costa & McCrae, 1980; Steel et al., 2008), is generalizable across cultures (Fulmer
et al., 2010; Schimmack, Radhakrishnan, Oishi, Dzokoto, & Ahadi, 2002), and holds after
ruling out potential confounding factors such as scale content overlap and common method
variance (Lucas & Fujita, 2000; Smillie, DeYoung, & Hall, 2015). Additionally, studies using
the Experience Sampling Method (ESM; Conner & Lehman, 2012) consistently replicate this
association at the level of momentary states: People report feeling more PA when they act
more extraverted in their everyday lives (Fleeson et al., 2002; Heller, Komar, & Lee, 2007;
Sun, Stevenson, et al., 2017; Wilt, Bleidorn, & Revelle, 2017; Wilt et al., 2012).
Interestingly, none of this research has found evidence that these within-person associations
are moderated by trait extraversion: introverts and extraverts report similarly high levels of
PA following extraverted behavior.
Although it may seem counterintuitive that introverts enjoy acting extraverted as
much as extraverts, this effect is predicted by Fleeson and colleagues’ (2002) principle of
state-trait isomorphism. This theory posits that states and traits share many properties and
consequences, and differ largely in terms of time course: States describe affects, behaviors,
and cognitions over a short period of a time, whereas traits summarize those same affects,
behaviors, and cognitions over a longer period of time (i.e., as the mean of a distribution of
states). From this perspective, an “extravert” is simply someone who acts more extraverted
more often compared to an “introvert”, but there is flexibility in behavior, so even introverts
sometimes behave in an extraverted way (Fleeson, 2001). Indeed, trait levels of extraversion
and other Big Five personality traits correspond well to average levels of their corresponding
states assessed over 1–2 weeks (Fleeson & Gallagher, 2009). State-trait isomorphism
therefore suggests that extraverts may experience higher trait PA than introverts simply
4

COSTS AND BENEFITS OF ACTING EXTRAVERTED
because they more frequently behave in an extraverted way (Ching et al., 2014; Wilt et al.,
2012). If so, it would follow that introverts could increase their experiences of PA by simply
enacting extraverted behaviors more often (Fleeson et al., 2002; Wilt et al., 2012).
Several laboratory-based experimental studies offer causal evidence for the affective
benefits of acting extraverted (Fleeson et al., 2002; Gallagher, Fleeson, & Hoyle, 2011;
McNiel & Fleeson, 2006; McNiel et al., 2010; Smillie, Wilt, et al., 2015; Sun, Stevenson, et
al., 2017; Zelenski, Santoro, & Whelan, 2012; Zelenski et al., 2013). Participants in such
laboratory-based experiments complete short interactive tasks, during which they are
instructed to act more extraverted (e.g. “act bold, talkative, and assertive”), more introverted
(e.g. “act reserved and shy”), or do not receive any behavioral instructions. Consistent with
the findings of correlational ESM studies, these laboratory-based experimental studies show
that participants instructed to act extraverted report greater levels of PA, relative to
participants instructed to act introverted. This effect holds for both self- and informant-ratings
of states and has not been found to depend on trait levels of extraversion—introverts and
extraverts reap similar affective benefits from acting extraverted (Fleeson et al., 2002;
McNiel & Fleeson, 2006; McNiel et al., 2010; Zelenski et al., 2012).
A practical implication of this research is that acting more extraverted could be a
simple tool for enhancing wellbeing. For example, some have suggested that acting
extraverted could be implemented as a positive psychology intervention in healthy
populations (Blackie et al., 2014; Wilt et al., 2012), or as a treatment intervention for
anhedonia in clinical populations (i.e., similar to Behavioral Activation Therapy; Cuijpers,
Van Straten, & Warmerdam, 2007; Zelenski et al., 2013). Proponents of this notion argue that
acting more extraverted could be especially beneficial for introverts, who tend to act
extraverted less often, and therefore experience the resulting hedonic benefits less frequently
than extraverts (Blackie et al., 2014). However, conclusions about the potential therapeutic
5

COSTS AND BENEFITS OF ACTING EXTRAVERTED
applications of extraverted behavior may be premature, due to two critical gaps in the
evidence base. First, the possible costs of sustained extraverted behavior (especially for
introverts) have received scant attention in previous research. Second, a randomized
controlled trial (RCT) which includes both assignment to experimental and control conditions
and the sampling of experiences throughout daily life has never been conducted. Addressing
these two evidence gaps is essential to determine whether acting extraverted in everyday life
could yield wellbeing benefits.
Potential Costs of Acting Extraverted
Little is known about the potential negative consequences of sustained increases in
everyday extraverted behavior. Some have argued that pushing introverts to act more
extraverted could be harmful (e.g., Cain, 2012; Little, 2008). This idea fits well with
evolutionary perspectives on personality, which posit that extraverted behavior is not
uniformly advantageous, but has different costs and benefits in different contexts
(Lukaszewski & von Rueden, 2015; Nettle, 2005). Until we have a well-rounded
understanding of both the positive and negative consequences of extraverted behavior,
advocating any real-world applications of acting extraverted could be premature and
potentially hazardous.
Several theoretical perspectives suggest that acting more extraverted would have
negative impacts on introverts, especially in terms of perceived authenticity and feelings of
tiredness. Authenticity is a subjective judgment that one’s actions express one’s true self
(Fleeson & Wilt, 2010), whereas tiredness is a subjective feeling of lacking energy (Leikas &
Ilmarinen, 2017). The trait-consistency hypothesis (Fleeson & Wilt, 2010) proposes that
people feel most authentic when acting in line with their disposition (i.e. in a manner
consistent with their personality), and that acting counterdispositionally can induce feelings
of inauthenticity. The contra-trait effort hypothesis (Gallagher et al., 2011) proposes that
6

COSTS AND BENEFITS OF ACTING EXTRAVERTED
counterdispositional behaviors are more strenuous than trait-concordant behaviors, which
could result in feelings of tiredness (Zelenski et al., 2012). Similarly, free trait theory (Little,
2008) posits that when people behave against their dispositional tendencies they will
subsequently need to “restore” or “recharge”, otherwise they will suffer physical and
emotional costs, including tiredness. Thus, it’s possible that introverts who try to increase
their levels of extraverted behavior might experience affective benefits at the cost of feeling
inauthentic and tired.
Despite the apparent plausibility of these theoretical arguments, studies to date have
yielded little empirical evidence for costs of acting extraverted. One series of ESM studies
revealed, perhaps surprisingly, that both extraverts and introverts felt more authentic during
moments in which they were acting more extraverted (Fleeson & Wilt, 2010). These results
contradict the trait-consistency hypothesis and suggest that acting extraverted could actually
increase authenticity in introverts. Additionally, two brief laboratory studies have
investigated whether extraverted behavior can increase feelings of tiredness. Gallagher et al.
(2011) found that introverts did not find it significantly more effortful to follow act-
extraverted instructions, relative to act-introverted instructions. Similarly, Zelenski et al.
(2012) found that acting extraverted did not impact introverts’ performance on a cognitive
task, used as a performance-based indicator of cognitive fatigue. They also found that acting
extraverted did not impact on a further possible cost of extraverted behavior—increased
negative affect (NA).
All in all, there is as yet no evidence that acting extraverted causes introverts to feel
inauthentic, tired, or more NA. However, previous studies did not examine the potential costs
of sustained increases in extraverted behavior as part of daily life. For example, introverts
normally engage in some extraverted behavior in their daily life (albeit less frequently than
extraverts; Fleeson et al., 2002), so short bouts of extraverted behavior might not be
7

COSTS AND BENEFITS OF ACTING EXTRAVERTED
experienced as inauthentic or tiring. Additionally, Leikas and Ilmarinen (2017) found that
both introverts and extraverts felt more tiredness three hours after extraverted moments
occurring in daily life. The authors concluded that extraverted behavior has delayed,
downstream effects on tiredness, and that positive effects of extraverted states (i.e., feelings
of energy and enthusiasm) might obscure feelings of tiredness until after the extraverted
states have dissipated. Because earlier studies (i.e., Gallagher et al., 2011; Zelenski et al.,
2012) only measured tiredness concurrently or immediately after extraverted behavior, any
delayed and cumulative effects on tiredness (and perhaps other wellbeing outcomes) could
have gone unnoticed. Therefore, it is necessary to directly investigate how extended periods
of experimentally manipulated extraverted behavior in daily life impact wellbeing outcomes.
The Value of a Randomized Controlled Trial
Randomized controlled trials (RCTs) are the gold standard for evaluating intervention
effectiveness (Schulz, Altman, & Moher, 2010). Previous research using naturalistic ESM
methods captures extraverted behavior over extended time periods in an ecologically valid
context (i.e., everyday life), but only allows the correlational inference that people tend to
feel happier when they spontaneously act more extraverted in daily life. On the other hand,
laboratory manipulations of extraverted behavior allow better causal inferences but may not
generalize to the wider range of situations that people encounter in everyday life. This is
because such laboratory paradigms are brief (10–30 minutes) and somewhat artificial (e.g.,
deciding on the most useful objects if one were lost on the moon; Fleeson et al., 2002).
Combining these naturalistic and experimental approaches can help us to understand the real-
world, causal implications of acting extraverted.
An important consideration in the design of a RCT is identifying appropriate control
groups. In previous experiments examining the effects of acting extraverted, an “act-
extraverted” group is typically compared to an “act-introverted” group, or a group given no
8

COSTS AND BENEFITS OF ACTING EXTRAVERTED
behavioral instructions at all. A concern with this design is that PA differences between the
act-extraverted and act-introverted groups could be partly attributable to differences in
socially desirable behavior. In Western cultures, where nearly all of this research has been
conducted (cf. Ching et al., 2014), extraverted behaviors (e.g., “bold, sociable”) are generally
more socially desirable than introverted behaviors (e.g., “shy, reserved”; Hart, Ritchie,
Hepper, & Gebauer, 2015; Steenkamp, De Jong, & Baumgartner, 2010). Acting in a socially
desirable way may itself increase PA, confounding the effects of extraverted behavior per se
on PA (Smillie, 2013). PA differences between an act-extraverted group and a no-instructions
group could also be attributable to placebo-type effects, or demand characteristics (Boot,
Simons, Stothart, & Stutts, 2013). Here, we address this issue by comparing the effects of
acting extraverted with both an active control group and a contact control group.
The Present Study
In this study, we randomly assigned participants to a one-week “Act-Extraverted”
intervention (i.e. instructions to engage in more extraverted behaviors) or active control
(“Sham”; i.e., instructions to engage in various non-extraverted behaviors) intervention, and
compared both groups to participants from a previous daily-life study who completed the
same measurement protocol but received no behavioral instructions (“Contact-control”
group). This enabled us to address three primary aims: First, we investigated the
consequences of a novel real-world acting extraverted intervention on a range of positive and
negative wellbeing outcomes (PA, NA, authenticity, and tiredness), assessed in the moment,
retrospectively at the end of the intervention period, and at follow-up two-weeks post-
intervention. Second, we investigated whether the overall intervention effects were mediated
by the hypothesized mechanism, extraverted behavior. Finally, we examined whether the
overall intervention effects were moderated by trait extraversion.
9

COSTS AND BENEFITS OF ACTING EXTRAVERTED
We preregistered the following hypotheses and exploratory research questions at
<anonymized link: https://osf.io/xvr7j/?view_only=a6e25ee9724943969efde1a399682d78>.
First, we hypothesized that participants in the Act-Extraverted condition would have higher
levels of momentary and retrospective extraverted behavior and PA than participants in the
Sham and Contact-control comparison conditions. We also predicted that the effect of
intervention condition on PA would be statistically mediated by momentary extraverted
behavior. Given that trait extraversion has not been found to moderate the effect of acting
extraverted on PA in previous studies, we anticipated that the effect of our acting extraverted
intervention on PA would be similar for introverts and extraverts. We also considered several
exploratory research questions: does the Act-Extraverted intervention influence feelings of
tiredness or subjective authenticity? Are any of the main effects of the intervention mediated
by momentary extraverted behavior, or moderated by trait extraversion? Do any effects of the
intervention persist to a 2-week follow-up? Finally, we addressed a research question that
was not preregistered: Does the Act-Extraverted intervention influence feelings of NA?
Methods
This research received ethical approval from [details masked for review]. Data,
analysis scripts and copies of materials used are provided in the online Supplemental
Materials, available at <anonymized link: https://osf.io/xvr7j/?
view_only=a6e25ee9724943969efde1a399682d78>. As recommended by Simmons, Nelson,
and Simonsohn (2012), we report how we determined our sample size, all data exclusions, all
manipulations, and all measures in the study.
Participants
Participants and Design. Participant flow through the study is shown in Figure 1,
and described below. We recruited 147 participants, aged 18–55 (M = 24.12; 70% female),
through flyers posted at the University of Melbourne and online advertisements. Participants
10

COSTS AND BENEFITS OF ACTING EXTRAVERTED
were randomized to one of two experimental groups (Act-Extraverted or Sham; described
below). To be eligible, participants needed to be 18 years or older, to be fluent English
speakers, and to have access to a mobile device with Android 4.1 (or higher), or iOS 7.0 (or
higher) installed. We aimed to recruit at least 60 participants per experimental group (after
exclusions, described below), based on recommendations for statistical power and unbiased
Level 2 standard errors in multilevel analyses (Maas & Hox, 2005; Scherbaum & Ferreter,
2009). Oversampling was required to replace unusable data (see exclusions below).
Participants were compensated with $15 AUD cash (~$12 USD) after completing baseline
questionnaires (all measures are available in the Supplemental Materials). If they completed
at least 75% of the ESM surveys and the post-intervention questionnaires, they received a
further $20 AUD (cash or gift voucher), entry into a prize draw for $300 AUD cash, and
feedback on their personality and wellbeing based on their baseline and ESM survey
responses.
Insert [Figure 1] here.
Additional comparison group. We used data from a previous study (Sun, Stevenson,
et al., 2017) to form a third comparison group, the Contact-control condition. Participants in
this study were drawn from a similar population and had completed an ESM protocol similar
to that used in the present study (i.e., filling out reports six times per day for seven days), but
the design was naturalistic in that participants did not receive any behavioral instructions.
This can help to detect any placebo effects of the two main experimental protocols resulting
from mere participation in an ESM study. It is important to emphasize, however, that this
dataset was included post-hoc in the present study and differs in a number of respects from
our primary data. Specifically, only a subset of the outcome variables of interest for the
present study was measured by Sun and colleagues (described below), and there were several
other minor differences in study design and procedure that may limit the degree to which this
11

COSTS AND BENEFITS OF ACTING EXTRAVERTED
sample is directly comparable to those randomized to the other two groups (e.g., a different
experimenter, remuneration scheme, and app). A more detailed comparison of the present
study and Sun, Stevenson, et al. (2017) is given in the Supplemental Materials.
Procedure
The following procedures only apply to participants in the newly-recruited
experimental groups. Procedural details for the additional comparison group can be found in
the original paper (Sun, Stevenson, et al., 2017; Study 3).
Introductory meeting. Participants in the experimental groups first attended an
introductory session (Day 1), with up to three participants attending each meeting. During
this session, they provided consent to participate, downloaded an ESM mobile app
(MetricWire Inc., 2017) onto their smartphone, and completed a baseline questionnaire
implemented through QualtricsTM survey software. They were then given the intervention
instructions (see below), and guidance for completing the ESM questionnaires via the mobile
app. To reduce differences in expectation for improvement (i.e., placebo effects) between the
two groups (Boot et al., 2013), all participants were told that the study aimed to investigate
how behavior influences mood and wellbeing in everyday life. They were otherwise kept
naïve to the purpose of the study and were unaware that there were two experimental
conditions.
Randomization. We used a cluster randomization design, with introductory meetings
as the unit of randomization, such that all participants in any given meeting were provided
with the same intervention instructions. To have approximately equal sample sizes in each
condition, meetings were randomized in blocks: for every block of 10 meetings, five
meetings were randomly assigned to each condition. The first author RJH generated the
random allocation sequence, allocated meetings to experimental conditions, enrolled
participants, and conducted the introductory meeting.
12

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Treatment interventions. Participants were randomly assigned to one of two 7-day
intervention conditions (Days 2–8). Participants in the Act-Extraverted condition were
instructed: “in your interactions with other people across the next week, act in a bold,
talkative, outgoing, active, and assertive way, as much as possible”. These terms were drawn
from trait descriptors that mark the high pole of extraversion (e.g., Goldberg, 1992).
Participants in the “Sham” condition were instructed: “in your interactions with other people
across the next week, act in an unassuming, sensitive, calm, modest, and quiet way, as much
as possible”. These terms were markers of both the low pole of extraversion (e.g., “quiet”)
and the high pole of other Big Five traits (e.g., “sensitive”, “modest”), and were selected to
comprise a set of behaviors that seemed coherent and relatively socially desirable. The Sham
condition was designed as an active control group against which to assess any effects of the
Act-Extraverted intervention on wellbeing outcomes, over and above nonspecific effects of
participating in an intervention study. In both conditions, participants were told to ignore the
instructions if they judged the behaviors to be inappropriate for particular situations (e.g.,
being “talkative” in the quiet area of a library, or being “quiet” during a job interview). For
reference, participants were emailed a document summarizing their intervention instructions,
and with definitions of the terms contained in the instructions (see Supplemental Materials).
Participants were shown a reminder of the instructions on their mobile device upon
completing each ESM survey (“Remember to continue acting in [an unassuming, sensitive,
calm, modest, and quiet / a bold, talkative, outgoing, active, and assertive] way in your
interactions with other people :)”).
ESM protocol. During the intervention week (Days 2–8), participants received 6
ESM surveys on their mobile phone every day, for a total of 42 surveys. Each survey was
delivered through the MetricWire application, which would create a phone notification upon
delivery. Surveys were delivered at random times between 9:00am and 10:00pm, with the
13

COSTS AND BENEFITS OF ACTING EXTRAVERTED
restriction that a survey could not be delivered within 90 minutes of the previous survey. If
participants did not complete a new survey within 15 minutes of delivery, a reminder
notification was triggered. If the survey was not opened within 30 minutes, it expired, and
could no longer be accessed. On the third and fifth day of the intervention, the experimenter
contacted each participant to let them know whether they were meeting the 75% completion
rate requirement, and to provide encouragement. At these times, if participants were unlikely
or unable to reach the 75% survey response requirement, they were reminded that they had
the opportunity to withdraw if they wished (i.e., due to being ineligible for further payment).
Retrospective and follow-up questionnaires. At the end of the intervention week
(Day 9), participants were sent the “retrospective questionnaire” via email. This questionnaire
could be completed using a web browser (median completion time = 9.38 minutes), and
participants were requested to complete it on the same day. An additional follow-up
questionnaire was completed two weeks after the conclusion of the study (Day 23). After
completing the follow-up questionnaire, participants were presented with a debrief statement
describing the full aims and design of the study, and received final payment collected as cash
or delivered as a gift voucher via email.
Exclusion criteria. Three preregistered exclusion criteria were used. First, survey
responses were excluded from analyses if they had a large number of identical responses,
indicating inattentive responding. Baseline surveys were excluded if participants responded
to 22 or more of the 25 (i.e., >85%) items in the largest question block with the same value.
Individual ESM reports were excluded if 17 or more of the 20 questions (i.e., 85%)
measured on an 11-point scale (see below) were responded to with the same value. Similarly,
retrospective and follow-up reports were excluded if 17 or more of the 20 items in the largest
survey question block had identical responses. Second, we excluded baseline, retrospective
and follow-up surveys that were deemed to be submitted too quickly. Cutoff times were
14

COSTS AND BENEFITS OF ACTING EXTRAVERTED
based on how long was required for the first author (RJH) to finish to the questionnaire as
quickly as possible without reading any of the questions. Due to a lack of survey timing
information, we could not implement this exclusion in the archival Contact-control dataset.
Thirdly, participants were excluded if they responded to fewer than 15 valid ESM reports
(after the above exclusions), indicating a lack of engagement with the study. We included a
fourth exclusion criterion which was not preregistered, because we decided that this would
reduce potential biases in the results: reports were excluded if they were not submitted in a
timely fashion. ESM reports were excluded if they were submitted more than 35 minutes
after the survey was triggered, retrospective reports were excluded if they were submitted
after Day 10 (i.e., more than one day late), and follow-up reports if they were submitted after
day 26 (i.e. more than three days late). Final sample sizes for participants included in the
analyses are shown in Figure 1.
Materials
All questionnaires relating to the present hypotheses are reported below. Extraneous
questionnaires are shown in the preregistration document and the Supplemental Materials.
For every scale, responses were averaged to produce a score for that scale.
Baseline questionnaire. We measured trait extraversion using the 20-item
extraversion scale from the Big Five Aspects Scales (BFAS; DeYoung, Quilty, & Peterson,
2007). This scale presents self-descriptions (e.g., “I see myself as a good leader”, “I rarely get
caught up in excitement” [R]) for which respondents indicate their agreement using a 5-point
scale ranging from 1 (strongly disagree) to 5 (strongly agree).
To measure trait PA and trait NA, participants completed the Positive and Negative
Affect Schedule (PANAS; Watson, Clark, & Tellegen, 1988), in which they rated how
accurately adjectives (e.g., “excited”, “irritable”) describe their feelings in general. Responses
were made on a five-point scale ranging from 1 (very slightly) to 5 (extremely).
15

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Trait authenticity was measured with a scale adapted from Fleeson and Wilt (2010), in
which participants rated their agreement with five statements concerning general feelings of
authenticity (e.g., “I act like my true self”, “People would have an accurate impression of me
from the way I act”) on a 7-point scale ranging from 1 (strongly disagree) to 7 (strongly
agree).
Given the absence of a widely-accepted tiredness scale in the literature, trait tiredness
was measured using an 11-item scale constructed for this study. In this scale participants
were asked how accurately each item, either an adjective or statement, described their
feelings in general. Eight of these items (“tired”; “alert” [R]; “fatigued”; “lethargic”;
“energized” [R]; “lively” [R]; “impulsive”; “I nearly always feel alert and awake” [R]) were
adapted from previous scales (Matthews, Jones, & Chamberlain, 1990; Ryan & Frederick,
1997; Thayer, 1986) and three additional items (“At times I feel too tired to complete
everyday activities [e.g. housework, chores, study]”; “I need to spend time restoring &
recharging”; “I do not have energy to do some of the things I would like to do”) were created
for this study. Answers were made on a 5-point scale ranging from 1 (very slightly or not at
all) to 5 (extremely). We excluded one item (“impulsive”) that did not load well onto the
scale according to a confirmatory factor analysis (see Supplemental Materials), resulting in a
10-item scale. Omission of this item did not change the interpretation of the results.
ESM questionnaire. The purpose of the ESM questionnaire was to measure
momentary behaviors and affective experiences during the intervention week. Momentary
extraverted behavior was measured with five items (“in the past hour, how [bold; quiet (R);
gregarious; assertive; reserved (R)] were you?”; Goldberg, 1992). However, we excluded one
item (“gregarious”) that did not overlap with the items used in the archival dataset to ensure
that the extraverted behavior composites were based on the same items for all conditions.
This omission did not substantively influence the obtained results. Momentary PA was
16

COSTS AND BENEFITS OF ACTING EXTRAVERTED
measured with three items drawn from the PANAS-X (“how [excited; lively; enthusiastic] do
you feel right now?”; Watson & Clark, 1999). Momentary NA was initially measured using
two items (“how [irritable; distressed] do you feel right now?”) derived from the PANAS
(Watson et al., 1988). A third NA adjective (“nervous”) was added after 12 participants had
already commenced the study to improve the measurement of this construct. Aggregate NA
scores computed with three items (M = 2.04, SD = 1.44) and scores computed with only two
items (M = 2.04, SD = 1.46) were highly correlated, r = .98, so we computed NA scores from
the mean of all available items. Momentary authenticity was measured with three items (“in
the past hour, [how much were you acting like your true self; how much were you putting on
an act (R)]”; “how accurate an impression would someone have of you from the way you
were acting?]”; Fleeson & Wilt, 2010). Momentary tiredness was measured with two items
formulated for this study (“how [tired do you feel; much do you feel the need to recharge]
right now?”). ESM items were answered on an 11-point integer sliding scale. The extraverted
behavior items had anchors 0 (Not at all) and 10 (Very). PA, NA, and tiredness items had
anchors 0 (Not at all) and 10 (Extremely), and authenticity items had anchors 0 (Not at all)
and 10 (Very much).
Participants were also asked, “In the past hour, how much time have you spent
focused on socially-oriented activities?”. This question was answered using a sliding scale
with 5-minute increments. Participants were instructed to only include interactions focused
on social connection (e.g., casual conversation), and to exclude business-type interactions
(e.g., meetings and business phone calls).
Retrospective and follow-up questionnaire. Participants reported retrospective
extraverted behavior by rating how accurately nine adjectives (“bold”; “quiet” [R];
“gregarious”; “assertive”; “reserved”; “talkative”; “withdrawn” [R]; “extraverted”; “shy”
[R]), drawn from Goldberg (1992), described their behavior during the previous week. Two
17

COSTS AND BENEFITS OF ACTING EXTRAVERTED
of these items (“gregarious”; “shy”) were not measured in the archival dataset, so were
excluded to ensure that the extraverted behavior composites were based on the same items for
all conditions. Responses were made on a 9-point scale (1 = extremely inaccurate, 9 =
extremely accurate), and three items were reverse scored.
Retrospective positive affect, negative affect, authenticity, and tiredness were
measured using the same questionnaires as at baseline; however, these were reworded to be
in the past tense (e.g., “I acted like my true self” rather than “I act like my true self”), and
rated with respect to the previous week, rather than “in general”, or “on average”.
Additionally, one item was added to the retrospective tiredness questionnaire (“It made me
feel tired or fatigued to behave according to the experimenter’s instructions”).
The items included in the follow-up questionnaire were identical to those in the
retrospective questionnaire, except that except that one item was excluded (“It made me feel
tired or fatigued to behave according to the experimenter’s instructions”).
Archival group measures. The Contact-control condition only included a subset of
the above measures, including baseline, momentary, and retrospective measures of
extraverted behavior and PA, and baseline and retrospective measures of NA. Measures of
authenticity, tiredness, and momentary NA were not collected.
Data Analysis
The criterion for statistical significance in all models was set at α = .05, and all
interval estimates were 95% confidence intervals. Any departures from the analyses stated in
the pre-registration are described in the Supplemental Materials.
Descriptive statistics. Descriptive statistics were calculated for all variables using the
dplyr package (version 0.5.0; Wickham & Francois, 2016) in R (version 3.4.0; R
Development Core Team, 2017). Omega (ω) reliability coefficients (Dunn, Baguley, &
Brunsden, 2014) were computed for retrospective variables using the MBESS package in R
18

COSTS AND BENEFITS OF ACTING EXTRAVERTED
(version 4.2.0; Kelley, 2007). For momentary variables, within-person ω coefficients were
computed using Mplus (version 7; Geldhof, Preacher, & Zyphur, 2014; Muthén & Muthén,
2017).
Intervention effects. The effects of intervention condition on each outcome were
tested using multilevel models (for momentary variables) and general linear models (for
retrospective variables). Multilevel models were computed using the nlme package (version
3.1-131; Pinheiro, Bates, DebRoy, Sarkar, & R Development Core Team, 2017), and general
linear models were computed using the base functions in R. A separate model was used for
every outcome, with intervention condition modeled as a predictor. For momentary
outcomes, time since first ESM report was modeled as a Level 1 covariate with random
slopes, following the recommendation of Bolger and Laurenceau (2013), but no other
covariates were included in any model. This is analogous to testing how levels of momentary
outcomes, averaged across the week, differ between groups, but offers more precision than
testing for a difference between aggregated scores. For outcomes that were measured in all
three conditions, pairwise contrasts were conducted. All statistical inferences were drawn
from unstandardized models.
Standardized effect sizes for retrospective measures were expressed in terms of
Hedge’s g, a less biased estimator than Cohen’s d (Lakens, 2013). In the case of momentary
measures, there is currently no consensus in the literature on methods for producing
standardized mean difference effect sizes in multilevel models. We therefore computed
standardized regression coefficients for multilevel models by standardizing all continuous
predictor and dependent variables across all observations. We bootstrapped 95% CIs for
standardized regression coefficients using the lme4 package in R (Bates et al., 2015).
Mediation analyses. We examined whether extraverted behavior was a plausible
mediator of any effects of the Act-Extraverted intervention on momentary wellbeing
19

COSTS AND BENEFITS OF ACTING EXTRAVERTED
outcomes (2-1-1 mediation models) and retrospective wellbeing outcomes (2-1-2 mediation
models). Only the Act-Extraverted vs. Sham and the Act-Extraverted vs. Contact-control
contrasts were considered for mediation due to our interest in the effects of acting
extraverted, and mediation analyses were only conducted where the main effect of the
intervention contrast on the outcome was significant. All mediation analyses were conducted
in Mplus, using syntax adapted from Preacher, Zyphur, and Zhang (2010), and the
MplusAutomation package in R (Hallquist & Wiley, 2016).
We also considered the possibility that enacted extraverted behavior may have
different within-person effects on momentary wellbeing outcomes, compared to naturally
expressed extraverted behavior. To test this possibility, we examined whether the within-
person associations between extraverted behavior and the wellbeing outcomes differed across
conditions using random-slope multilevel models. Momentary extraverted behavior was
person-mean-centered and used to predict momentary outcome variables. Each outcome was
used as a DV in a separate model. These within-person analyses were conducted across the
whole sample, and also separately within each condition.
Moderation analyses. Finally, to investigate whether the effects of the intervention
were moderated by trait extraversion, we conducted further analyses that added the main
effect of trait extraversion, and the condition × trait extraversion interaction terms to the
above “main intervention effects” models. Significant interaction effects were probed by
computing regions of significance analyses using the Johnson-Neyman procedure, following
Preacher, Curran, and Bauer (2006).
Results
Descriptive Statistics and Missing Data
Descriptive statistics for continuous variables by condition are shown in Table 1.
Descriptive statistics for categorical demographic variables, and correlations among
20

COSTS AND BENEFITS OF ACTING EXTRAVERTED
continuous variables, are shown in the Supplemental Materials (Tables S4 and S5,
respectively). There were no significant differences between treatment conditions on any trait
measure (ps > .167). Mean age was significantly lower in the Contact-control condition (M =
21.64, SD = 3.66) relative to both the Act-Extraverted condition (M = 24.51, SD = 7.39),
t(217) = 2.95, p = .004, and the Sham condition (M = 23.77, SD = 6.01), t(217) = 2.24, p
= .026, but did not differ between the latter two conditions t(87) = 0.75, p = .452. Including
age as a covariate did not substantively alter the results of any of our main analyses, and was
therefore excluded from the final reported models for parsimony.
Across the two treatment conditions and the Contact-control group, participants
completed a total of 6,593 ESM reports out of a possible 8,736 (75.5% completion rate).
After exclusions (see Figure 1), this resulted in valid completion rates of 2,177 out of 2,730
(79.7%) in the in the Act-Extraverted condition (Mreports = 33.49, SD = 5.91), 2,456 out of
2,940 (83.5%) in the Sham condition (Mreports = 35.09, SD = 5.15), and 1,818 out of 2,562
(71.0%) in the Contact-control condition (Mreports = 29.80, SD = 5.95). A general linear model
showed that completion rates were significantly higher in females, and negatively predicted
by trait tiredness, but were unrelated to all other baseline trait variables reported in this study
(see Supplemental Materials). Similarly, a logistic regression showed that the probability of
returning to complete the retrospective survey could be predicted by baseline tiredness, but
not by other baseline trait variables, gender, or age (see Supplemental Materials).
Insert [Table 1] here
Manipulation Check
Participants in the Act-Extraverted condition reported significantly higher levels of
extraverted behavior, compared to those in both the Sham and the Contact-control conditions
(which did not differ from each other; see Table 2). This held for both momentary and
retrospective reports, indicating that the intervention successfully increased levels of
21

COSTS AND BENEFITS OF ACTING EXTRAVERTED
extraverted behavior. However, further analyses revealed boundaries to the effectiveness of
our manipulation: trait extraversion significantly moderated the effect of the Act-Extraverted
intervention on momentary extraverted behavior (see Table 3). As depicted in Figure 2, our
attempt to increase momentary extraverted behavior throughout daily life was more
successful for participants higher on trait extraversion, and wholly unsuccessful for extreme
introverts. Specifically, regions of significance analyses revealed that differences in
momentary extraverted behavior between the Act-Extraverted condition and the sham
condition were statistically non-significant for values of trait extraversion below 2.43 (which
is 1.95 SD below the mean level of trait extraversion in this study). Similarly, differences
between the Act-Extraverted condition and the Contact-control condition were statistically
non-significant for values of trait extraversion below 2.80 (1.24 SD below the mean). No
such moderation effects were evident in the retrospective measures (see Table 2).
Effects of the Acting Extraverted Intervention on Wellbeing Outcomes
PA. Results for all main intervention effects are shown in Table 2. Levels of
momentary and retrospective PA were significantly higher in the Act-Extraverted condition
than the Sham condition. Participants in the Act-Extraverted condition also reported higher
levels of PA than those in the Contact-control condition, but this difference was only
statistically significant for retrospective PA. Finally, participants in the Sham condition
reported significantly lower momentary PA than the Contact-control condition, but
retrospective PA did not significantly differ between these groups. Thus, retrospective reports
suggest that the Act-Extraverted intervention boosted levels of PA relative to both control
conditions, whereas momentary reports suggest that this was true only relative to the Sham
condition.
Authenticity, NA, and tiredness. As shown in Table 2, participants in the Act-
Extraverted condition reported significantly higher levels of authenticity than in the Sham
22

COSTS AND BENEFITS OF ACTING EXTRAVERTED
condition, and there were no significant differences between the two conditions on NA or
feelings of tiredness. There was also no evidence that participants in the Act-Extraverted
intervention experienced wellbeing costs in terms of increased NA or tiredness.
Insert [Table 2] here
Follow-up effects. Participants in the Act-Extraverted condition reported higher
levels of weekly extraverted behavior than participants in the Sham condition at 2 weeks
post-intervention. However, there were no significant between-group differences in any other
wellbeing outcome at follow-up (see Supplemental Materials). This suggests that there was a
return-to-baseline effect after withdrawal of the intervention.
Mechanisms Underlying Effects of the Intervention
Next, we examined three potential mechanisms that could explain the effect of the
Act-Extraverted intervention on PA and authenticity.
The mediating role of extraverted behavior. The most obvious explanation for the
observed effect of intervention condition on wellbeing outcomes is differences in extraverted
behavior. In fact, it would be puzzling if the increased wellbeing observed in the Act-
Extraverted condition was unrelated to increases in extraverted behavior. We conducted
multilevel mediation analyses to investigate this. As shown in Table 3, average extraverted
behavior was a statistically significant mediator of the effect of intervention condition on
average momentary PA, retrospective PA, and retrospective authenticity. However, there was
no significant indirect effect via average momentary extraverted behavior for average
momentary authenticity. In the Act-Extraverted vs. Sham comparison, the indirect effect via
momentary extraverted behavior accounted for 67% of the total intervention effect on
momentary PA, 66% of the total effect on retrospective PA, and 54% of the total effect on
retrospective authenticity. In the Act-Extraverted vs. Contact-control comparison, the indirect
23

COSTS AND BENEFITS OF ACTING EXTRAVERTED
effect via momentary extraverted behavior accounted for 70% of the total effect on
retrospective PA.
Insert [Table 3] here
Moderation of momentary associations by intervention condition. Another
plausible mechanism is that the intervention condition changed the momentary association
between extraverted behavior and the wellbeing outcomes. For example, successful
compliance with instructions to act more extraverted could increase PA due to a sense of
accomplishment, or as a result of study demand characteristics. This could result in a stronger
association between momentary extraverted behavior and PA in the Act-Extraverted group,
contributing to between-condition differences in average momentary PA. We therefore
examined the momentary associations between extraverted behavior and the four wellbeing
outcomes across the full sample and in each subgroup (see Table 4). For concision, we will
only describe the momentary associations for PA and authenticity (given that there were no
between-condition differences in NA and tiredness). As reported in previous studies,
participants tended to report more PA and authenticity during moments in which they had
recently been acting more extraverted. The strength of the association between extraverted
behavior and PA was weaker in the Sham condition than the Contact-control condition (see
Table 5). However, there was no evidence that the strength of the association between
extraverted behavior and PA or authenticity was different between the Act-Extraverted and
Sham condition, or between the Act-Extraverted and Contact-control condition. Thus, it
seems more likely that the effects of the Act-Extraverted intervention on momentary
wellbeing outcomes were due to increased levels of extraverted behavior, rather than by
strengthening the relation between extraverted behavior and wellbeing outcome.
Insert [Table 4] here
Insert [Table 5] here
24

COSTS AND BENEFITS OF ACTING EXTRAVERTED
The mediating role of social activity. Finally, it is possible that the Act-Extraverted
intervention could have influenced wellbeing by increasing the amount of time participants
spent in social situations (see Lucas, Le, & Dyrenforth, 2008). However, participants in the
Act-Extraverted and Sham conditions did not report spending different amounts of time in
social situations throughout the intervention study, β = 0.04 [CI: –0.12, 0.21], t(133) = 0.51, p
= .614. This suggests that the intervention changed how people acted in social situations (i.e.,
how much extraversion they enacted), rather than the quantity of time that people spent in
social situations.
Were the Intervention Effects Conditional on Trait Extraversion?
So far, we have shown that the Act-Extraverted intervention increased PA and
authenticity and had no effects on NA and tiredness for the average participant. However, it
is possible that the Act-Extraverted intervention could have different effects for more
extraverted people relative to more introverted people. In our final set of analyses, we
examined trait extraversion as a moderator of the intervention effects reported above. Where
trait extraversion was a significant moderator of an intervention effect, we used the Johnson-
Neyman procedure to compute regions of significance. Regions of significance indicate the
range of values on trait extraversion for which intervention effects were significant. Results
are depicted in Table 6 and Figure 2.
PA. Contrary to all previous laboratory experiments and naturalistic ESM studies in
this literature, we found that trait extraversion significantly moderated the effect of the Act-
Extraverted intervention on retrospective (but not momentary) PA, relative to the Sham
condition. Regions of significance analyses showed that only participants with trait
extraversion scores exceeding 3.46 (0.05 SD above the mean) reported higher levels of
retrospective PA in the Act-Extraverted condition than the Sham condition.
25

COSTS AND BENEFITS OF ACTING EXTRAVERTED
NA. Trait extraversion also moderated the effects of the Act-Extraverted intervention
on momentary (but not retrospective) NA, relative to the sham condition. Regions of
significance analyses showed that participants with trait extraversion levels below 3.07 (0.71
SD below the mean) reported higher levels of momentary NA in the Act-Extraverted
condition relative to the Sham condition.
Authenticity. Trait extraversion moderated the effects of the intervention on both
momentary and retrospective measures of authenticity. Regions of significance analyses
showed that participants who scored above 3.34 (0.19 SD below the mean) on trait
extraversion reported higher momentary authenticity in the Act-Extraverted condition than in
the Sham condition. In addition, we found a cross-over interaction effect for retrospective
authenticity. Those who scored above 3.36 (Z = 0.16 SD below the mean) on trait
extraversion reported higher levels of retrospective authenticity in the Act-Extraverted
condition, but those who scored below 2.70 (1.43 SD below the mean) on trait extraversion
reported lower levels of retrospective authenticity in the Act-Extraverted condition (relative
to the Sham condition).
Tiredness. Although there were no main effects of the Act-Extraverted intervention
on tiredness, we found that Trait extraversion moderated the effects of the intervention on
both momentary and retrospective feelings of tiredness. Regions of significance analyses
indicated that there was a cross-over interaction effect for momentary tiredness: those who
scored above 3.79 (0.68 SD above the mean) on trait extraversion reported lower levels of
momentary tiredness in the Act-Extraverted condition, whereas those who scored below 2.72
(1.39 SD below the mean) on trait extraversion reported higher levels of momentary tiredness
in the Act-Extraverted condition (relative to the Sham condition). In addition, those who
scored above 3.83 (0.77 SD above the mean) on trait extraversion reported lower levels of
retrospective tiredness in the Act-Extraverted condition than the Sham condition.
26

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Insert [Table 6] here
Insert [Figure 2] here
Summary
Table 7 summarizes the findings from this study. Overall, we found that participants
in the Act-Extraverted intervention, relative to comparison conditions, had higher levels of
extraverted behavior, PA, and authenticity. Mediation analyses suggested that extraverted
behavior was a plausible mediator of the overall Act-Extraverted intervention effect on PA,
whereas the evidence that extraverted behavior mediated effects on authenticity was mixed.
Finally, moderation analyses revealed that the effects of the Act-Extraverted intervention
varied as a function of trait extraversion: although more extraverted participants showed all
of the above effects, plus lower levels of tiredness, more introverted participants reported
smaller increases in extraverted behavior and PA, increased NA, reduced authenticity, and
increased tiredness.
Discussion
A maturing literature shows that people feel happier when they act more extraverted,
even if they are dispositional introverts (Fleeson et al., 2002; Sun, Stevenson, et al., 2017).
However, no published study has yet investigated whether people can improve their
wellbeing by deliberately acting more extraverted during their everyday lives. To address this
critical gap in the evidence, we conducted a randomized controlled trial of the potential
wellbeing benefits and costs of acting extraverted across a week of daily life. On average, our
Act-Extraverted intervention increased participants’ reports of PA and authenticity, and these
effects could be explained by increases in average levels of extraverted behavior reported
across the intervention period. These wellbeing benefits did not appear to be offset by any
costs in terms of increased NA or tiredness, or reduced feelings of authenticity. However, in
sharp contrast to virtually every previous study in this literature, we found that these effects
27

COSTS AND BENEFITS OF ACTING EXTRAVERTED
depended on one’s level of trait extraversion. Specifically, the intervention was successful for
participants who had at least average levels of trait extraversion, but extreme introverts
experienced wellbeing costs in terms of increased NA, tiredness, and reduced feelings of
authenticity. These are novel findings, supported by what is arguably the most stringent test
yet conducted of the impact of acting extraverted on wellbeing.
Was the Act-Extraverted Intervention Beneficial, and Why?
We found that acting extraverted over one week in everyday life increased levels of
PA in the average participant. This finding is consistent with the growing body of
correlational ESM research (e.g., Fleeson et al., 2002; Sun, Stevenson, et al., 2017; Wilt et
al., 2017) and experimental laboratory research (e.g., McNiel et al., 2010; Smillie, Wilt, et al.,
2015; Zelenski et al., 2012) showing links between extraverted behavior and PA. However,
there was one complicating finding: When compared to the Sham condition, the Act-
Extraverted intervention appeared unequivocally successful for increasing both momentary
and retrospective reports of PA. However, compared to the Contact-control group, the Act-
Extraverted intervention only increased levels of retrospective, but not momentary, PA. This
lack of difference in momentary PA is similar to the results of previous laboratory studies,
most of which have shown that participants instructed to act extraverted during a group
discussion task do not significantly differ in PA compared to a comparison group given no
behavioral instructions (McNiel & Fleeson, 2006; Sun, Stevenson, et al., 2017; Zelenski et
al., 2012, Study 2; Zelenski et al., 2013, Studies 2 & 5; cf. Zelenski et al., 2012, Study 1;
Zelenski et al., 2013, Study 3). However, it should be noted that the momentary effect was in
the same direction as the retrospective effect, albeit of smaller magnitude. Therefore, it is
possible that we lacked statistical power to detect this small effect. Nevertheless, the evidence
overall suggests that following instructions to act more extraverted might be beneficial for
28

COSTS AND BENEFITS OF ACTING EXTRAVERTED
PA relative to following sham instructions, but less decisively so in comparison to how one
would naturally behave in the absence of any instructions.
We also examined potential costs of acting extraverted on perceptions of authenticity,
NA, and tiredness. On average, people did not feel less authentic by following instructions to
act extraverted: To the contrary, instructions to act extraverted increased feelings of
authenticity for most participants. This result is consistent with the finding that naturalistic
extraverted behavior is associated with higher levels of perceived authenticity (Fleeson &
Wilt, 2010), and shows for the first time that this effect extends to experimentally
manipulated extraverted behavior. From an intervention acceptability standpoint, this
suggests that most participants find it natural to act more extraverted, and do not feel like
they are “putting on an act”. We also found no overall cost of the Act-Extraverted
intervention in terms of feelings of tiredness or NA. This was despite recent correlational
evidence indicating that extraverted behavior is associated with increased levels of tiredness
three hours later (Leikas & Ilmarinen, 2017)2. Thus, the main effects of the intervention were
wholly positive, and no costs of extraverted behavior were detected for the average
participant.
What might explain the effects of the Act-Extraverted intervention on PA and
authenticity? We hypothesized that the Act-Extraverted intervention should have effects on
wellbeing outcomes via increased extraverted behavior across the week. Although this seems
like it would be a foregone conclusion, it was important to put it to the test, as it would be
concerning (suggesting demand characteristics or placebo effects) if extraverted behavior was
not found to mediate the effects of the Act-Extraverted intervention on wellbeing outcomes.
Supporting this mechanism, we found that extraverted behavior accounted for a large
proportion of the overall effect of the intervention on PA. On the other hand, the evidence
2 We attempted to conceptually replicate Leikas and Ilmarinen’s (2017) finding of an association between
extraverted behavior and later feelings of tiredness, but no such lagged effects emerged as significant in our data
(see Supplemental Materials).
29

COSTS AND BENEFITS OF ACTING EXTRAVERTED
that the effect of the intervention on authenticity was mediated by extraverted behavior was
more mixed: retrospective authenticity was mediated by extraverted behavior, but average
momentary authenticity was not. Nevertheless, and although we acknowledge the limitations
of statistical mediation (Bullock, Green, & Ha, 2010), these analyses suggest that extraverted
behavior offers a plausible explanation for at least some of the intervention effects.
The extraverted behavior mechanism also appears to be more plausible than two
alternative explanations. First, it seemed possible that the intervention instructions could
influence the consequences of extraverted behavior. For example, successful compliance with
instructions to act more extraverted could lead to increased PA, due either to a sense of
accomplishment or, more trivially, to demand characteristics. Such an effect could contribute
to the observed differences between conditions in PA. However, we found that extraverted
behavior had similar associations with momentary positive affect and authenticity in both of
the experimental conditions. This implies that the Act-Extraverted intervention affected
wellbeing outcomes by increasing levels of extraverted behavior, rather than by changing the
consequences of that extraverted behavior. Second, although some studies have found that the
association between trait extraversion and PA can be partially explained in terms of social
activity (Srivastava, Angelo, & Vallereux, 2008; cf. Lucas et al., 2008), we found no
evidence that the Act-Extraverted intervention affected how much time people spent in social
interactions. This indicates that the intervention effects are likely to be attributable to
processes beyond the quantity of social interactions they experience; rather, changes in
qualitative aspects of social experiences resulting from increased extraverted behavior may
be more relevant (see Smillie, Wilt, et al., 2015; Sun, Stevenson, et al., 2017). However, we
acknowledge that other unmeasured processes could also explain the intervention effects
(e.g., demand characteristics).
30

COSTS AND BENEFITS OF ACTING EXTRAVERTED
If changes in extraverted behavior are the most likely explanation for the effects of the
intervention, a final consideration is whether differences in wellbeing outcomes between the
Act-Extraverted and Sham groups were due to the Act-Extraverted instructions increasing
extraverted behavior, or due to the sham instructions decreasing extraverted behavior.
Comparing levels of extraverted behavior in the Act-Extraverted and Sham groups to the
uninstructed Contact-control group indicates that both of these processes may have occurred.
The Act-Extraverted group reported higher levels of extraverted behavior relative to the
Contact-control group, whereas the Sham group reported somewhat lower levels of
retrospective (but not momentary) extraverted behavior relative to the Contact-control group.
Differences in PA between the three groups showed a similar pattern: The Act-Extraverted
group reported higher levels of PA relative to the Contact-control group, whereas the sham
group reported lower levels of momentary (but not retrospective) PA relative to the Contact-
control group. Overall, these findings suggest that relative to the Contact-control group, the
Act-Extraverted intervention increased levels of extraverted behavior, whereas the Sham
intervention decreased levels of extraverted behavior, and that differences in levels of
extraverted behavior between these groups can potentially explain the observed differences in
wellbeing outcomes.
For Whom Did the Intervention Work Best?
We found positive effects and no costs of an Act-Extraverted intervention on average,
but an understanding of the effectiveness of an intervention is incomplete without considering
the moderating role of individual differences. Models of person-activity fit (Lyubomirsky &
Layous, 2013) suggest that different activities are more successful at increasing happiness for
different people. In the context of acting extraverted, some have suggested that introverts
may find it draining and inauthentic to act more extraverted (e.g., Cain, 2012; Little, 2008).
Although previous research has generally not supported for these claims, we found that the
31

COSTS AND BENEFITS OF ACTING EXTRAVERTED
consequences of extraverted behavior did indeed depend on dispositional extraversion.
Specifically, although the Act-Extraverted intervention had multiple benefits and no costs for
people who had at least average levels of extraversion, more introverted people experienced
fewer benefits and more costs of acting extraverted.
To begin with, the effect of the Act-Extraverted intervention on PA was weaker for
more introverted people. Although the Act-Extraverted intervention appeared to increase
momentary PA, regardless of dispositional extraversion, no benefit in retrospective PA was
detected in relatively introverted participants. This dependence on trait extraversion may not
have emerged in previous daily-life studies because such studies have not attempted to
constrain individuals’ naturally occurring behavior. On the other hand, lab experiments in
which participants have been instructed to act counterdispositionally have typically involved
a short (~20min), structured interactive task involving one or two strangers. Unlike in
previous studies, introverted participants assigned to our Act-Extraverted intervention were
instructed to sustain increased levels of extraverted behavior for multiple days across
repeated interactions in daily life. Given the lack of previous support for the moderating role
of dispositional extraversion, and the inconsistencies between state and retrospective reports,
this finding should be interpreted cautiously until replicated. It should also be noted that this
study was powered to detect our hypotheses regarding main effects, but was not highly
powered to detect moderation effects. However, if our results prove to be robust, they suggest
that acting extraverted can increase experiences of PA in most individuals, but that these
benefits could be more limited for relatively introverted people, at least as recalled in
retrospective reports.
Next, although higher levels of authenticity were reported by participants in the Act-
Extraverted intervention, this effect was moderated by trait extraversion: participants of
average extraversion and above who were instructed to act more extraverted reported
32

COSTS AND BENEFITS OF ACTING EXTRAVERTED
increased feelings of momentary and retrospective authenticity, but more introverted
participants did not. These findings conceptually diverge from previous ESM research on this
topic (Fleeson & Wilt, 2010), which showed that extraverted moments were associated with
greater feelings of authenticity for introverts and extraverts alike3. It is again possible that the
extended experimental nature of the present study can account for this difference in findings:
more introverted people may feel authentic in naturally-expressed extraverted moments, but
less authentic when intentionally attempting to increase and sustain extraverted behavior. If
robust, these findings support previous cautions that acting extraverted may make introverts
feel less authentic.
Finally, the Act-Extraverted intervention also increased retrospective tiredness and
retrospective NA among more introverted people. This finding lends support to the contra-
trait effort hypothesis (Gallagher et al., 2011), and notions that introverts may need to
“restore” after behaving in an extraverted way, or else suffer affective costs (Little, 2008).
Again, it is possible that previous laboratory-based experiments (Gallagher et al., 2011;
Zelenski et al., 2012) have not detected this effect due to their short duration, and that acting
extraverted for short durations of time is insufficient to induce subjective feelings of tiredness
and NA. However, as the tiredness measures developed for this study require further
validation, and the corresponding effects on momentary tiredness and NA were non-
significant, these conclusions are tentative.
Recalled Versus Experienced Wellbeing
There were several differences between momentary reports and retrospective reports
in our results. These differences could merely be due to measurement error, or could reflect
more substantive differences, in line with the idea that retrospective and momentary reports
measure different types of conscious “selves” (Conner & Barrett, 2012). As retrospective
3 We instead found that the association between momentary extraverted behavior and authenticity was
only positive in more extraverted people, see Supplemental Materials.
33

COSTS AND BENEFITS OF ACTING EXTRAVERTED
reports are more distant from actual experiences and, due to imperfections of memory, are
more prone to distortion by recent events, semantic knowledge, and beliefs (Robinson &
Clore, 2002), retrospective reports may have been influenced by beliefs about extraverted and
counterdispositional behavior. For example, introverts tend to under-estimate how much PA
and overestimate how much NA they will experience after acting extraverted (Zelenski et al.,
2013). Additionally, people tend to believe that they feel most authentic when they act in line
with their Big Five traits, and that introverts will feel less authentic when acting in an
extraverted way (Fleeson & Wilt, 2010). The influence of such lay beliefs could explain why
retrospective reports seemed less favorable than momentary reports in our study.
However, these biases do not necessarily make retrospective reports inferior to
momentary reports. Another process that could cause retrospective and momentary reports to
differ is that retrospective reports often manifest larger degrees of change relative to
momentary reports (e.g., Luong, Wrzus, Wagner, & Riediger, 2016), possibly due to
momentary effects accumulating over time into larger effects within global retrospective
reports. Additionally, in some circumstances retrospective reports can be better predictors of
future behavior and other criteria than momentary experiences (Conner & Barrett, 2012;
Redelmeier, Katz, & Kahneman, 2003). For example, Wirtz, Kruger, Scollon, and Diener
(2003) found that students’ recollections of feelings during a spring vacation, relative to
momentary reports of those feelings, were better predictors of whether those students
intended to repeat the trip in the future. Therefore, it is not obvious whether we should
evaluate the success of the Act-Extraverted intervention on the basis of effects on
retrospective or momentary reports. When the two differ, we must balance the evidence from
both momentary and retrospective measures, and consider the possible implications of
differences between the two. For example, if participants consistently reported feeling more
momentary PA during the Act-Extraverted intervention, but do not remember feeling any
34

COSTS AND BENEFITS OF ACTING EXTRAVERTED
happier when retrospecting at the end of the intervention week, feedback could be provided
reminding participants that the intervention actually did make them feel happier in the
moment, potentially correcting their beliefs over the long term.
Limitations and Future Directions
Our RCT provides a critical, direct test of the claim that acting more extraverted in
everyday life might be an effective strategy for increasing wellbeing, and one of the most
comprehensive examinations of both the benefits and costs of extraverted behavior. However,
there are many ways that future studies could build off this initial effort.
Comparison group. Here, the effects of the Act-Extraverted intervention were
compared to a sham intervention that included behavioral instructions drawn from across the
Big Five domains. The sham instructions were intended to control for placebo-type effects
associated with participation in an intervention, while accounting for potential social
desirability confounds. However, the sham instructions may have produced their own specific
intervention effects that confounded the effects of the Act-Extraverted intervention, as
discussed previously. For example, it is possible that instructions to act “sensitive” and
“calm” may lead to wellbeing benefits specifically for introverts4. Therefore, future research
could explore alternative active comparison groups to better understand the specific effects of
an Act-Extraverted intervention. For example, comparison groups adopting other sets of
instructions (e.g. “Act in a polite and respectful way”) could be used.
Effectiveness of the manipulation for introverts. According to momentary reports,
the present intervention was less effective at increasing levels of extraverted behavior among
more introverted people, and was wholly ineffective at increasing extraverted behavior in
extreme introverts. This could be because increasing extraverted behavior for extended
periods of time is unsustainable for introverts (e.g., due to the negative consequences that are
incurred), which would establish the boundary conditions of earlier conclusions that
4 We are grateful to an anonymous reviewer for this suggestion.
35

COSTS AND BENEFITS OF ACTING EXTRAVERTED
introverts and extraverted are equally capable of engaging in extraverted behavior (e.g.,
Fleeson et al., 2002). However, another possibility is that the present intervention was too
difficult and/or insufficiently engaging for introverts. Importantly, the reduced effectiveness
of the manipulation for more introverted participants may help explain why such individuals
did not benefit as much as their extraverted counterparts in terms of wellbeing outcomes.
Acting more extraverted may yet be shown to be beneficial for introverts if an alternative
implementation of an Act-Extraverted intervention could more successfully increase
extraverted behavior for such individuals.
There are numerous modifications that could be made to the present intervention to
attempt to increase its potency for introverts. In particular, reducing the perceived intensity of
the intervention could be beneficial. Instead of being instructed to act extraverted “as much as
possible”, participants could be asked to act extraverted in a small number of social
interactions per day. Brief field experiments showing that even minimal social interactions
(e.g., talking to strangers during a commute, Epley & Schroeder, 2014; having a genuine
interaction with a cashier, Sandstrom & Dunn, 2014) can influence wellbeing. This suggests
that a less intensive intervention than the one evaluated here might still be effective. A less
intensive intervention might also be less intimidating for introverts, who tend to predict that
acting extraverted will result in negative affective consequences (Zelenski et al., 2013). By
allowing more freedom to return to an introverted “restorative niche” (Little, 2008), a less
intensive intervention might also result in fewer costs to NA, authenticity, and tiredness. It
might also prove more effective for participants to establish specific implementation
intentions (a series of “if-then” statements; e.g., “If I encounter situation X, I will do Y”),
which has previously been shown to help people achieve their personality change goals
(Hudson & Fraley, 2015). Finally, the intensity of the intervention could even be tailored to
an individual’s level of trait extraversion and motivation to participate in the intervention.
36

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Long-term effects. Another critical area to be investigated is how to sustain the
positive wellbeing outcomes of an Act-Extraverted intervention over the longer term. We
found that there were no consequences of acting extraverted on wellbeing outcomes at two-
week follow-up. This was true even though participants in the Act-Extraverted intervention
still reported somewhat higher levels of extraverted behavior at two-week follow-up. This
suggests that once the intervention is withdrawn, its positive consequences also cease.
Nevertheless, it would obviously be desirable if the benefits of an intervention could be
sustained beyond the intervention period.
Longer-term effects of the intervention might have been achieved if participants were
encouraged and felt motivated to continue acting in an extraverted way after the conclusion
of the intervention. This might be achieved if participants were informed that the intervention
was thought to benefit wellbeing. This is seen in positive psychology interventions which
often have an overt focus on improving wellbeing, and often show effects on wellbeing that
persist for at least three to six months, potentially because participants continued engaging in
these positive activities after the intervention period (Bolier et al., 2013). Alternatively,
personality-change intervention approaches (Hudson & Fraley, 2015; Roberts et al., 2017)
could facilitate longer-term change in extraverted behavior and wellbeing (Hudson & Fraley,
2016). Such longer-term interventions may also help make extraverted behavior more
habitual, which could reduce the negative consequences of the intervention (as implied by the
contra-trait effort hypothesis; Gallagher et al., 2011) while retaining its positive effects.
Generalizability. This paper investigated several outcome measures to obtain a broad
view of the effects of acting extraverted on wellbeing. However, there are many other aspects
of wellbeing that we did not consider here (e.g., meaning, engagement, positive relationships;
Ryff, 1989; Seligman, 2012). Previous laboratory-based experiments have shown that acting
extraverted impacts on multiple positive affective states and measures of wellbeing (McNiel
37

COSTS AND BENEFITS OF ACTING EXTRAVERTED
& Fleeson, 2006; Smillie et al., 2015), and that trait extraversion is linked with multiple
dimensions of wellbeing (Sun, Kaufman, & Smillie, 2017). Nevertheless, future work that
examines the effects of an acting extraverted intervention on other aspects of wellbeing
would be helpful to determine the breadth and boundaries of the effects reported here.
Final Thoughts and Conclusion
Given the robust links that extraverted behavior has with wellbeing, many have asked
the naturally occurring question: Should more introverted people act extraverted more often?
Our goal is not to deliver any prescriptive recommendations about whether to act more
extraverted. Instead, we have provided evidence that acting extraverted has merit as a
potential self-regulatory tool, which could be strategically deployed in order to yield affective
benefits. However, we also found that those who were more introverted, relative to those who
were more extraverted, experienced fewer benefits of sustained increases in extraverted
behavior. Introverts might “feel good” after naturally expressing extraverted behaviors, or
when enacting extraversion for short bouts, but attempts to enact extraverted behavior for
longer durations of time seem to have less pronounced benefits, and possible costs. Thus,
rather than attempting to act more extraverted, introverts may benefit more from pursuing
other wellbeing pathways (e.g., mindfulness and other positive psychology interventions;
Bolier et al., 2013). We also note that there are individual differences in ideal levels of affect
(Tsai, Knutson, & Fung, 2006; Tamir, 2009), and that introverts may not desire the
experience of PA as strongly as extraverts. Therefore, the assumption that experiencing
increased levels of PA is “beneficial” and desirable may be misguided (e.g., Ford, Mauss, &
Gruber, 2015; Gruber, 2011; Gruber, Mauss, & Tamir, 2011).
In conclusion, the present study aimed to provide new insights into the debate around
whether acting more extraverted is beneficial for wellbeing, even for introverts. In the first
randomized controlled trial to be published on this topic, we found that an Act-Extraverted
38

COSTS AND BENEFITS OF ACTING EXTRAVERTED
intervention increased PA and feelings of authenticity in those who were at or above average
on trait extraversion. The effect of the intervention on PA was statistically mediated by
extraverted behavior. However, potential benefits of the Act-Extraverted intervention in
terms of PA were less apparent for dispositional introverts, and these participants did not
increase in feelings of authenticity. Highly introverted participants may even suffer costs in
terms of increased NA and tiredness, and reduced feelings of authenticity. Because these
findings may hinge on specific features of our RCT, such as our choice of control groups,
future research should explore alternative implementations of an Act-Extraverted
intervention, especially those that may be more accessible and beneficial for introverts.
Context of the Research
The present research was inspired by fundamental questions about what personality is,
how it impacts on our lives, and whether we can intervene in these processes to promote
human flourishing. In 1990, Nancy Cantor observed that, whereas we often think about
personality in terms of what people ‘are’ or ‘have’, we might also think about personality in
terms of what people ‘do’. A more recent, growing literature on counterdispositional
behavior, which has primarily focused on manipulations of extraverted behavior, seems to
reinforce this perspective. Specifically, several studies suggest that the wellbeing benefits
enjoyed by extraverted people can be reaped by all people when they behave in a more
extraverted way. This has profound theoretical implications for understanding why extraverts
have higher wellbeing, as well as potential practical implications for promoting wellbeing.
We planned our Randomized Controlled Trial of the wellbeing effects of extraverted
behavior specifically to inform these practical implications, given that previous studies had
comprised either laboratory-based experiments (which have limited ecological validity) or
correlational daily life studies (the results of which are causally ambiguous). Our results
provide both encouragement and caution to the idea that people can boost their wellbeing by
39

COSTS AND BENEFITS OF ACTING EXTRAVERTED
engaging in extraverted behavior. Perhaps our most significant finding was the observation
that dispositional introverts may reap fewer wellbeing benefits, and perhaps even incur some
wellbeing costs, from acting more extraverted. Our future research priorities in this area
include confirming the present findings, and identifying wellbeing pathways that are more
effective for introverted people.
References:
Augie, B. (2016). gridExtra: Miscellaneous functions for “grid” graphics (R package version
2.2.1). https://CRAN.R-project.org/package=gridExtra
Baumeister, R. F., Vohs, K. D., & Tice, D. M. (2007). The strength model of self-control.
Current Directions in Psychological Science, 16(6), 351-355.
https://doi.org/10.1111/j.1467-8721.2007.00534.x
Bates, D., Maechler, M., Bolker, B., & Walker, S. (2015). Fitting Linear Mixed-Effects
Models Using lme4. Journal of Statistical Software, 67(1), 1-48.
https://doi.org/10.18637/jss.v067.i01.
Blackie, L. E., Roepke, A. M., Forgeard, M. J., Jayawickreme, E., & Fleeson, W. (2014). Act
Well to Be Well: The Promise of Changing Personality States to Promote Well Being.‐
The Wiley Blackwell Handbook of Positive Psychological Interventions, 462-474.
https://doi.org/10.1002/9781118315927.ch27
Bolger, N., & Laurenceau, J. (2013). Fundamentals of intensive longitudinal data. In N.
Bolger & J. Laurenceau (Eds.), Intensive longitudinal methods: An introduction to
diary and experience sampling research (pp. 27-39). New York: Guilford Press.
Bolier, L., Haverman, M., Westerhof, G. J., Riper, H., Smit, F., & Bohlmeijer, E. (2013).
Positive psychology interventions: a meta-analysis of randomized controlled studies.
BMC public health, 13(1), 1. https://doi.org/10.1186/1471-2458-13-119
40

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Boot, W. R., Simons, D. J., Stothart, C., & Stutts, C. (2013). The pervasive problem with
placebos in psychology: Why active control groups are not sufficient to rule out
placebo effects. Perspectives on Psychological Science, 8(4), 445-454. https://doi.org/
10.1177/1745691613491271
Bullock, J. G., Green, D. P., & Ha, S. E. (2010). Yes, but what’s the mechanism? (Don’t
expect an easy answer). Journal of personality and social psychology, 98(4), 550.
https://doi.org/10.1037/a0018933
Cain, S. (2012). Quiet: the power of introverts in a world that can't stop talking. New York:
Crown Publishers.
Ching, C. M., Church, A. T., Katigbak, M. S., Reyes, J. A. S., Tanaka-Matsumi, J., Takaoka,
S., . . . Ortiz, F. A. (2014). The manifestation of traits in everyday behavior and affect:
A five-culture study. Journal of Research in Personality, 48, 1-16.
http://dx.doi.org/10.1016/j.jrp.2013.10.002
Cantor, N. (1990). From thought to behavior: “Having” and “doing” in the study of
personality and cognition. American Psychologist, 45, 735–750.
Conner, T. S., & Barrett, L. F. (2012). Trends in ambulatory self-report: the role of
momentary experience in psychosomatic medicine. Psychosomatic medicine, 74(4),
327. https://doi.org/10.1097/psy.0b013e3182546f18
Conner, T. S., & Lehman, B. J. (2012). Getting started: Launching a study in daily life. In M.
R. Mehl & T. S. Conner (Eds.), Handbook of research methods for studying daily life.
New York, New York: Guilford Press.
Costa, P. T., & McCrae, R. R. (1980). Influence of extraversion and neuroticism on
subjective well-being: happy and unhappy people. Journal of personality and social
psychology, 38(4), 668. https://doi.org/10.1037//0022-3514.38.4.668
41

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Cuijpers, P., Van Straten, A., & Warmerdam, L. (2007). Behavioral activation treatments of
depression: A meta-analysis. Clinical Psychology Review, 27(3), 318-326.
https://doi.org/10.1016/j.cpr.2006.11.001
DeYoung, C. G., Quilty, L. C., & Peterson, J. B. (2007). Between facets and domains: 10
aspects of the Big Five. Journal of personality and social psychology, 93(5), 880.
https://doi.org/10.1037/e633942013-310
Epley, N., & Schroeder, J. (2014). Mistakenly seeking solitude. Journal of Experimental
Psychology: General, 143(5), 1980. https://doi.org/10.1037/e578192014-009
Fleeson, W. (2001). Toward a structure-and process-integrated view of personality: Traits as
density distributions of states. Journal of personality and social psychology, 80(6),
1011. https://doi.org/10.1037//0022-3514.80.6.1011
Fleeson, W., Malanos, A. B., & Achille, N. M. (2002). An intraindividual process approach
to the relationship between extraversion and positive affect: is acting extraverted as"
good" as being extraverted? Journal of personality and social psychology, 83(6),
1409. https://dx.doi.org/10.1037/a0016786
Fleeson, W., & Wilt, J. (2010). The Relevance of Big Five Trait Content in Behavior to
Subjective Authenticity: Do High Levels of Within-Person Behavioral Variability
Undermine or Enable Authenticity Achievement? Journal of Personality, 78(4),
1353-1382. https://doi.org/10.1111/j.1467-6494.2010.00653.x
Ford, B. Q., Mauss, I. B., & Gruber, J. (2015). Valuing happiness is associated with bipolar
disorder. Emotion, 15(2), 211. https://doi.org/10.1037/emo0000048
Fox, J., & Hong, J. (2009). Effect displays in R for multinomial and proportional-odds logit
models: Extensions to the effects package. Journal of Statistical Software, 32(1), 1-
24. https://doi.org/10.18637/jss.v032.i01
42

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Fulmer, C. A., Gelfand, M. J., Kruglanski, A. W., Kim-Prieto, C., Diener, E., Pierro, A., &
Higgins, E. T. (2010). On “feeling right” in cultural contexts how person-culture
match affects self-esteem and subjective well-being. Psychological Science.
https://doi.org/10.1177/0956797610384742
Gallagher, P., Fleeson, W., & Hoyle, R. H. (2011). A self-regulatory mechanism for
personality trait stability: Contra-trait effort. Social Psychological and Personality
Science, 2(4), 335-342. https://doi.org/10.1177/1948550610390701
Geldhof, G. J., Preacher, K. J., & Zyphur, M. J. (2014). Reliability estimation in a multilevel
confirmatory factor analysis framework. Psychological methods, 19(1), 72.
https://doi.org/10.1037/e698612011-001
Goldberg, L. R. (1992). The development of markers for the Big-Five factor structure.
Psychological assessment, 4(1), 26. https://doi.org/10.1037//1040-3590.4.1.26
Gruber, J. (2011). Can feeling too good be bad? Positive emotion persistence (PEP) in bipolar
disorder. Current Directions in Psychological Science, 20(4), 217-221. https://doi.org/
10.1177/0963721411414632
Gruber, J., Mauss, I. B., & Tamir, M. (2011). A dark side of happiness? How, when, and why
happiness is not always good. Perspectives on Psychological Science, 6(3), 222-233.
https://doi.org/10.1177/1745691611406927
Hagger, M. S., Wood, C., Stiff, C., & Chatzisarantis, N. L. (2010). Ego depletion and the
strength model of self-control: a meta-analysis. Psychological bulletin, 136(4), 495.
https://doi.org/10.1037/a0019486
Hallquist, M., & Wiley, J. (2016). MplusAutomation: Automating Mplus model estimation
and interpretation (R package version 0.6-4).
https://CRAN.R-project.org/package=MplusAutomation
43

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Hart, C. M., Ritchie, T. D., Hepper, E. G., & Gebauer, J. E. (2015). The Balanced Inventory
of Desirable Responding Short Form (BIDR-16). SAGE Open, 5(4).
https://doi.org/10.1177/2158244015621113
Heller, D., Komar, J., & Lee, W. B. (2007). The dynamics of personality states, goals, and
well-being. Personality and Social Psychology Bulletin, 33(6), 898-910.
https://doi.org/10.1177/0146167207301010
Hudson, N. W., & Fraley, R. C. (2015). Volitional personality trait change: Can people
choose to change their personality traits? Journal of personality and social
psychology, 109(3), 490. https://doi.org/10.1037/pspp0000021
Hudson, N. W., & Fraley, R. C. (2016). Changing for the Better? Longitudinal Associations
Between Volitional Personality Change and Psychological Well-Being. Personality
and Social Psychology Bulletin, 0146167216637840.
https://doi.org/10.1177/0146167216637840
Kelley, K. (2007). Methods for the behavioral, educational, and social sciences: An R
package. Behavior Research Methods, 39(4), 979-984.
https://doi.org/10.3758/bf03192993
Keyes, C. L. M., Shmotkin, D., & Ryff, C. D. (2002). Optimizing well-being: the empirical
encounter of two traditions. Journal of Personality and Social Psychology, 82(6),
1007-1022. https://doi.org/10.1037/0022-3514.82.6.1007
LaHuis, D. M., Hartman, M. J., Hakoyama, S., & Clark, P. C. (2014). Explained variance
measures for multilevel models. Organizational Research Methods, 17(4), 433-451.
https://doi.org/10.1177/1094428114541701
Lakens, D. (2013). Calculating and reporting effect sizes to facilitate cumulative science: a
practical primer for t-tests and ANOVAs. Frontiers in Psychology, 4, 863.
https://doi.org/10.3389/fpsyg.2013.00863
44

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Leikas, S., & Ilmarinen, V. J. (2017). Happy now, tired later? Extraverted and Conscientious
behavior are related to immediate mood gains, but to later fatigue. Journal of
Personality, 85, 603-615. https://doi.org/10.1111/jopy.12264
Little, B. R. (2008). Personal projects and free traits: Personality and motivation
reconsidered. Social and Personality Psychology Compass, 2(3), 1235-1254.
https://doi.org/10.1111/j.1751-9004.2008.00106.x
Lucas, R. E., & Diener, E. (2009). Personality and Subjective Well-Being. In E. Diener (Ed.),
Science of Well-Being (Vol. 37, pp. 75-102). https://doi.org/10.1007/978-90-481-
2350-6_4
Lucas, R. E., & Fujita, F. (2000). Factors influencing the relation between extraversion and
pleasant affect. Journal of Personality and Social Psychology, 79(6), 1039-1056.
https://doi.org/10.1037/0022-3514.79.6.1039
Lucas, R. E., Le, K., & Dyrenforth, P. S. (2008). Explaining the extraversion/positive affect
relation: Sociability cannot account for extraverts' greater happiness. Journal of
Personality, 76(3), 385-414. https://doi.org/10.1111/j.1467-6494.2008.00490.x
Lukaszewski, A. W., & von Rueden, C. R. (2015). The extraversion continuum in
evolutionary perspective: A review of recent theory and evidence. Personality and
Individual Differences, 77, 186-192. https://doi.org/10.1016/j.paid.2015.01.005
Luong, G., Wrzus, C., Wagner, G. G., & Riediger, M. (2016). When bad moods may not be
so bad: Valuing negative affect is associated with weakened affect–health
links. Emotion, 16(3), 387-401. https://doi.org/10.1037/emo0000132
Lyubomirsky, S., & Layous, K. (2013). How do simple positive activities increase well-
being? Current Directions in Psychological Science, 22(1), 57-62.
https://doi.org/10.1177/0963721412469809
45

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Maas, C. J., & Hox, J. J. (2005). Sufficient sample sizes for multilevel modeling.
Methodology, 1(3), 86-92. https://doi.org/10.1027/1614-2241.1.3.86
Margolis, S., & Lyubomirsky, S. (2018). Experimental manipulation of extraverted and
introverted behavior and its effects on well-being. Manuscript submitted for
publication.
Matthews, G., Jones, D. M., & Chamberlain, A. G. (1990). Refining the measurement of
mood: The UWIST mood adjective checklist. British Journal of Psychology, 81(1),
17-42. https://doi.org/10.1111/j.2044-8295.1990.tb02343.x
McNiel, J. M., & Fleeson, W. (2006). The causal effects of extraversion on positive affect
and neuroticism on negative affect: Manipulating state extraversion and state
neuroticism in an experimental approach. Journal of Research in Personality, 40(5),
529-550. https://doi.org/10.1016/j.jrp.2005.05.003
McNiel, J. M., Lowman, J. C., & Fleeson, W. (2010). The effect of state extraversion on four
types of affect. European Journal of Personality, 24(1), 18-35.
https://doi.org/10.1002/per.738
MetricWire Inc. (2016). Metricwire (Version 2.2.10 to Version 3.2.0) [Mobile application
software]. Retrieved from https://play.google.com/store/apps/details?
id=com.metricwire.android3&hl=en or
https://itunes.apple.com/au/app/metricwire/id1117108543?mt=8
Muthén, L. K., & Muthén, B. O. (2017). Mplus User's Guide (Eighth ed.). Los Angeles, CA:
Muthén & Muthén.
Nave, C. S., Sherman, R. A., & Funder, D. C. (2008). Beyond self-report in the study of
hedonic and eudaimonic well-being: Correlations with acquaintance reports, clinician
judgments and directly observed social behavior. Journal of Research in Personality,
42(3), 643-659. https://doi.org/10.1016/j.jrp.2007.09.001
46

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Nettle, D. (2005). An evolutionary approach to the extraversion continuum. Evolution and
Human Behavior, 26(4), 363-373.
https://doi.org/10.1016/j.evolhumbehav.2004.12.004
Nezlek, J., (2008). An introduction to multilevel modeling for social and personality
psychology. Social and Personality Psychology Compass, 2, 842-860. https://doi.org/
10.1111/j.1751-9004.2007.00059.x
Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D., & R Development Core Team (2017). nlme:
Linear and Nonlinear Mixed Effects Models (Version 3.1-131). https://CRAN.R-
project.org/package=nlme
Preacher, K. J., Curran, P. J., & Bauer, D. J. (2006). Computational tools for probing
interactions in multiple linear regression, multilevel modeling, and latent curve
analysis. Journal of Educational and Behavioral Statistics, 31(4), 437-448.
https://doi.org/10.3102/10769986031004437
Preacher, K. J., Zyphur, M. J., & Zhang, Z. (2010). A general multilevel SEM framework for
assessing multilevel mediation. Psychological Methods, 15(3), 209.
https://doi.org/10.1037/a0020141
R Development Core Team (2017). R: A language and environment for statistical computing
(Version 3.4.2) [Computer software]. Vienna, Austria. http://www.R-project.org
Redelmeier, D. A., Katz, J., & Kahneman, D. (2003). Memories of colonoscopy: A
randomized trial. Pain, 104(1-2), 187-194. http://dx.doi.org/10.1016/S0304-
3959(03)00003-4
Richardson, B. (2015). Instant Survey (iOS Version 1.3; Android Version 1.1) [Mobile
application software]. Retrieved from
https://itunes.apple.com/us/app/instant-survey/id955226674?mt=8 or
47

COSTS AND BENEFITS OF ACTING EXTRAVERTED
https://play.google.com/store/apps/details?
id=au.edu.deakin.psychology.surveyframework
Robinson, M. D.,Clore, G. L. (2002). Belief and feeling: Evidence for an accessibility model
of emotional self-report. Psychological Bulletin, 128(6), 934-960.
https://doi.org/10.1037//0033-2909.128.6.934
Roberts, B. W., Luo, J., Briley, D. A., Chow, P. I., Su, R., & Hill, P. L. (2017). A systematic
review of personality trait change through intervention. Psychological Bulletin,
143(2), 117-141. http://dx.doi.org/10.1037/bul0000088
Rosseel, Y. (2012). lavaan: An R package for structural equation modeling. Journal of
Statistical Software, 48(2), 1-36. https://doi.org/10.18637/jss.v048.i02
Ryan, R. M., & Frederick, C. (1997). On energy, personality, and health: Subjective vitality
as a dynamic reflection of well-being. Journal of Personality, 65(3), 529-565. https://
doi.org/10.1111/j.1467-6494.1997.tb00326.x
Ryff, C. D. (1989). Happiness is Everything, or is it? Journal of Personality and Social
Psychology, 57(6), 1069–1081. https://doi.org/10.1037//0022-3514.57.6.1069
Sandstrom, G. M., & Dunn, E. W. (2014). Is efficiency overrated? Minimal social
interactions lead to belonging and positive affect. Social Psychological and
Personality Science, 5(4), 437-442. https://doi.org/10.1177/1948550613502990
Scherbaum, C. A., & Ferreter, J. M. (2009). Estimating statistical power and required sample
sizes for organizational research using multilevel modeling. Organizational Research
Methods, 12(2), 347-367. https://doi.org/10.1177/1094428107308906
Schimmack, U., Radhakrishnan, P., Oishi, S., Dzokoto, V., & Ahadi, S. (2002). Culture,
personality, and subjective well-being: integrating process models of life satisfaction.
Journal of Personality and Social Psychology, 82(4), 582.
https://doi.org/10.1037/0022-3514.82.4.582
48

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Schmutte, P. S., & Ryff, C. D. (1997). Personality and well-being: reexamining methods and
meanings. Journal of Personality and Social Psychology, 73(3), 549.
https://doi.org/10.1037/0022-3514.73.3.549
Schulz, K. F., Altman, D. G., & Moher, D. & the CONSORT Group (2010). CONSORT
2010 statement: updated guidelines for reporting parallel group randomised trials.
BMC medicine, 8(1), 18. https://doi.org/10.1186/1745-6215-11-32
Seligman, M. E. P. (2011). Flourish: A visionary new understanding of happiness and well-
being. New York, NY, US: Free Press.
Simmons, J. P., Nelson, L. D., & Simonsohn, U. (2012). A 21 word solution. SPSP Dialogue,
1–4. http://doi.org/10.2139/ssrn.2160588
Smillie, L. D. (2013). Why does it feel good to act like an extravert? Social and Personality
Psychology Compass, 7(12), 878-887. https://doi.org/10.1111/spc3.12077
Smillie, L. D., DeYoung, C. G., & Hall, P. J. (2015). Clarifying the relation between
extraversion and positive affect. Journal of Personality, 83(5), 564-574.
https://doi.org/10.1111/jopy.12138
Smillie, L. D., Wilt, J., Kabbani, R., Garratt, C., & Revelle, W. (2015). Quality of social
experience explains the relation between extraversion and positive affect. Emotion,
15(3), 339. https://doi.org/10.1037/emo0000047
Srivastava, S., Angelo, K. M., & Vallereux, S. R. (2008). Extraversion and positive affect: A
day reconstruction study of person–environment transactions. Journal of Research in
Personality, 42(6), 1613-1618. https://doi.org/10.1016/j.jrp.2008.05.002
Steel, P., Schmidt, J., & Shultz, J. (2008). Refining the relationship between personality and
subjective well-being. Psychological Bulletin, 134(1), 138-161.
https://doi.org/10.1037/0033-2909.134.1.138
49

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Steenkamp, J.-B. E., De Jong, M. G., & Baumgartner, H. (2010). Socially desirable response
tendencies in survey research. Journal of Marketing Research, 47(2), 199-214. https://
doi.org/10.1509/jmkr.47.2.199
Sun, J., Kaufman, S. B., & Smillie, L. D. (2017). Unique associations between big five
personality aspects and multiple dimensions of well-being. Journal of Personality.
https://doi.org/10.1111/jopy.12301
Sun, J., Stevenson, K., Kabbani, R., Richardson, B., & Smillie, L. D. (2017). The pleasure of
making a difference: Perceived social contribution explains the relation between
extraverted behavior and positive affect. Emotion, 17(5), 794-810.
https://doi.org/10.1037/emo0000273
Solt, F., & Hu, Y. (2015). interplot: Plot the effects of variables in interaction terms.
Available at The Comprehensive R Archive Network (CRAN) https://CRAN.R-
project.org/package=interplot
Tamir, M. (2009). Differential Preferences for Happiness: Extraversion and Trait-Consistent
Emotion Regulation. Journal of Personality, 77, 447–470.
https://doi.org/10.1111/j.1467-6494.2008.00554.x
Thayer, R. E. (1986). Activation-deactivation adjective check list: Current overview and
structural analysis. Psychological Reports, 58(2), 607-614.
https://doi.org/10.2466/pr0.1986.58.2.607
Tsai, J. L., Knutson, B., & Fung, H. H. (2006). Cultural variation in affect valuation. Journal
of Personality and Social Psychology, 90(2), 288.
Watson, D., & Clark, L. A. (1999). The PANAS-X: Manual for the positive and negative
affect schedule-expanded form. Unpublished Manuscript. University of Iowa.
50

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief
measures of positive and negative affect: the PANAS scales. Journal of Personality
and Social Psychology, 54(6), 1063. https://doi.org/10.1037//0022-3514.54.6.1063
Watson, D., & Tellegen, A. (1985). Toward a consensual structure of mood. Psychological
Bulletin, 98(2), 219-235.
Weninger, R. L., & Holder, M. D. (2013). Extraversion and subjective well-being.
International Journal of Psychology Research, 8(3), 141-172. ISSN: 1932-6092
Wickham, H. (2009). ggplot2: Elegant graphics for data analysis. New York: Springer.
Wickham, H., & Francois, R. (2016). dplyr: A Grammar of Data Manipulation. R package
version 0.5.0. https://CRAN.R-project.org/package=dplyr
Wilt, J. A., Bleidorn, W., & Revelle, W. (2017). Velocity explains the links between
personality states and affect. Journal of Research in Personality, 69, 86-95.
https://doi.org/10.1016/j.jrp.2016.06.008
Wilt, J. A., Noftle, E. E., Fleeson, W., & Spain, J. S. (2012). The dynamic role of personality
states in mediating the relationship between extraversion and positive affect. Journal
of Personality, 80(5), 1205-1236. https://doi.org/10.1111/j.1467-6494.2011.00756.x
Wirtz, D., Kruger, J., Scollon, C. N., & Diener, E. (2003). What to do on spring break? The
role of predicted, on-line, and remembered experience in future choice. Psychological
Science, 14(5), 520-524. https://doi.org/10.1111/1467-9280.03455
Zelenski, J. M., Santoro, M. S., & Whelan, D. C. (2012). Would introverts be better off if
they acted more like extraverts? Exploring emotional and cognitive consequences of
counterdispositional behavior. Emotion, 12(2), 290-303.
https://doi.org/10.1037/a0025169
Zelenski, J. M., Sobocko, K., & Whelan, D. C. (2014). Introversion, solitude, and subjective
well-being. The handbook of solitude: Psychological perspectives on social isolation,
51

COSTS AND BENEFITS OF ACTING EXTRAVERTED
social withdrawal, and being alone, 184-201.
https://doi.org/10.1002/9781118427378.ch11
Zelenski, J. M., Whelan, D. C., Nealis, L. J., Besner, C. M., Santoro, M. S., & Wynn, J. E.
(2013). Personality and affective forecasting: Trait introverts underpredict the hedonic
benefits of acting extraverted. Journal of Personality and Social Psychology, 104(6),
1092-1108. https://doi.org/10.1037/a0032281
52

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Invited to Participate (n = 210)
Not enrolled (n = 63)
 Did not schedule meeting with
researchers (n = 61)
 Incompatible device (n = 1)
 Under 18 years of age (n =1)
Participants Analyzed (n = 61)
Participant exclusions:
 Did not complete
questionnaire (n = 2)
 Completed questionnaire
too late (n = 2)
Participants analyzed (n = 65)
2,177 ESM reports analyzed
Participant exclusions:
 Completed fewer than 15
valid ESM reports (n = 6)
ESM report exclusions:
 6 reports submitted late
 2 reports where ≥85% of
items had the same
response.
Allocated to
“Act Extraverted”
intervention (n = 71)
Participants analyzed (n = 70)
2,456 ESM reports analyzed
Participant exclusions:
 Completed fewer than 15
valid ESM reports (n = 6)
Report exclusions
 19 reports submitted late
Allocated to
“Sham”
intervention (n = 76)
Participants Analyzed (n = 62)
Participant exclusions:
 Did not complete
questionnaire (n = 3)
 Completed questionnaire
too late (n = 3)
 ≥85% of items had the
same response (n = 2)
Introductory
Meeting
(Day 1)
Retrospective
Questionnaires
(Day 9)
Intervention
and ESM
(Days 2-8)
Randomized (n = 147)
Recruitment
Archival dataset:
“Contact-control”
intervention (n = 76)
Participants analyzed (n = 61)
1,818 ESM reports analyzed
Participant exclusions:
 Completed fewer than 15
valid ESM reports (n = 15)
Report exclusions
 11 reports submitted late
 3 reports where ≥85% of
items had the same
response.
Participants Analyzed (n = 59)
Participant exclusions:
 Completed questionnaire
too late (n = 2)
Follow-up
Questionnaires
(Day 23)
Participants Analyzed (n = 56)
Participant exclusions:
 Did not complete
questionnaire (n = 3)
 Completed questionnaire
too late (n = 1)
 Completed questionnaire
too quickly (n = 1)
Participants Analyzed (n = 58)
Participant exclusions:
 Completed questionnaire
1 week too early (n = 1)
 Completed questionnaire
too late (n = 3)
Figure 1. Consolidated Standards of Reporting Trials (CONSORT) flow diagram.
53

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Figure 2. Predicted values for study outcomes at different scores on trait extraversion, by
experimental condition. Standard error (±1) is shaded. Boundaries of the regions of
significance are shown by vertical dashed red lines. E Behavior = extraverted behavior. For
clarity, only one region of significance line is shown for momentary extraverted behavior,
representing the Act-Extraverted vs Contact-control comparison; the Act-Extraverted vs
Sham comparison is excluded. Plots were generated using the ggplot2 package in R
(Wickham, 2009)
54

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Table 1
Descriptive statistics by condition
Act-Extraverted condition Sham condition Contact-control condition
Variable ω M SDBP SDWP ICC(1) M SDBP SDWP ICC(1) M SDBP SDWP ICC(1)
Trait measures
Extraverted behavior .85 3.50 0.55 3.45 0.53 3.37 0.46
Positive affect .83 3.48 0.63 3.51 0.61 3.55 0.49
Negative affect .87 1.95 0.62 2.14 0.76 2.11 0.61
Authenticity .89 5.14 1.18 5.10 1.16 — —
Tiredness .86 2.65 0.68 2.72 0.68 — —
ESM measures
Extraverted behavior .74 5.69 1.26 1.65 .32 3.97 1.13 1.22 .42 4.37 0.94 1.88 .16
Positive affect .81 5.18 1.64 1.53 .49 4.40 1.75 1.60 .52 5.01 1.24 1.52 .36
Negative affect .63 2.13 1.55 1.24 .56 1.99 1.29 1.28 .48 — — — —
Authenticity .66 7.18 1.77 1.40 .57 6.49 1.47 1.43 .48 — — — —
Tiredness .62 3.87 1.83 1.93 .44 4.03 1.34 1.94 .30 — — — —
Minutes in social
interactions 21.02 9.48 17.23 .20 20.29 9.12 16.60 .20 — — — —
Retrospective measures
Extraverted behavior .81 6.40 1.13 4.74 1.12 5.15 1.03
Positive affect .89 3.61 0.63 3.33 0.68 3.23 0.60
Negative affect .85 1.83 0.59 1.89 0.58 2.03 0.61
Authenticity .84 5.43 1.03 4.87 0.88 — —
Tiredness .83 2.37 0.68 2.52 0.55 — —
Note. ω = omega reliability coefficient; within-person omegas and aggregated means are reported for ESM measures; SDBP = between-person
standard deviation; SDWP = average within-person standard deviation; ICC(1) = intra-class correlation coefficient, which represents the
proportion of total variation due to variation between-persons. Blank cells indicate non-applicable measures, whereas dashes indicate that data
were not available.
Table 2
Intervention Effects on Wellbeing Outcomes

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Act-Extraverted vs. Sham Act-Extraverted vs. Contact-control Sham vs. Contact-control

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Dependent variable b (SE)ES [95% CI] b (SE)ES [95% CI] b (SE)ES [95% CI] R2
Momentary
Extraverted
behavior
1.70*** (0.19) 0.80 [0.61, 0.98] 1.32*** (0.20) 0.62 [0.42, 0.82] –0.38 (0.20) –0.18 [–0.38, +0.03] .12
Positive affect 0.76** (0.27) 0.33 [0.08, 0.55] 0.20 (0.28) 0.09 [–0.15, 0.34] –0.56* (0.27) –0.25 [–0.47, –0.01] .02
Negative affect 0.11 (0.24) 0.06 [–0.18, 0.32] — — — — .00
Authenticity 0.77** (0.28) 0.33 [0.11, 0.58] — — — — .03
Tiredness –0.19 (0.27) –0.07 [–0.31, 0.15] — — — — .00
Retrospective
Extraverted
behavior
1.66*** (0.20) 1.52 [1.16, 1.87] 1.26*** (0.20) 1.15 [0.79, 1.51] –0.42* (0.20) –0.37 [–0.73, –0.01] .30
Positive affect 0.28* (0.11) 0.43 [0.08, 0.79] 0.38** (0.12) 0.60 [0.24, 0.96] 0.05 (0.12) 0.16 [–0.19, 0.52] .06
Negative affect –0.06 (0.11) –0.09 [–0.45, 0.26] –0.20 (0.11) –0.33 [–0.69, 0.03] –0.14 (0.11) -0.24 [–0.60, 0.12] .02
Authenticity 0.57** (0.17) 0.59 [0.23, 0.95] — — — — 08
Tiredness –0.15 (0.11) –0.24 [–0.59, 0.12] — — — — .01
Note. * p < .05, ** p < .01, *** p < .001. b = the effect of intervention condition (unstandardized regression coefficient); SE = standard error; ES =
standardized effect size (Hedges g for retrospective measures, and standardized regression coefficients for momentary measures); 95%
confidence around the effect size is shown in brackets; the effects of covariates are not shown (full model output is presented in the
Supplemental Materials); R2 for multilevel models was computed following LaHuis, Hartman, Hakoyama, and Clark (2014). Dashes indicate
that data were not available.

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Table 3
Mediation of Intervention Effects on Wellbeing Outcomes by Extraverted Behavior
Act-Extraverted vs. Sham Act-Extraverted vs. Contact-control
Dependent
Variable
IV → M
(a path)
M → DV
(b path)
Direct effect
(c’ path)
Indirect effect
(a x b path)
95% CI of
indirect effect
IV → M
(a path)
M → DV
(b path)
Direct effect
(c’ path)
Indirect effect
(a x b path)
95% CI of
indirect effect
Momentary
(2-1-1)
PA 1.67*** 0.98*** -0.8 1.64 [0.98, 2.29]
Authenticity 1.67*** –0.19 1.01 –0.32 [–1.03, 0.39] — — — — —
Retrospective
(2-1-2)
PA 1.67*** 0.33*** –0.28*0.55 [0.38, 0.73] 1.32*** 0.22*** 0.07 0.28 [0.15, 0.42]
Authenticity 1.67*** 0.18** 0.26 0.30 [0.07, 0.53] — — — — —
Note. * p < .05, ** p < .01, *** p < .001. Indirect effects in bold denote that the 95% CI does not capture zero. Blank spaces indicate that mediation
analyses were not conducted, whereas dashes indicate that data were not available.
58

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Table 4
Associations Between Momentary Extraverted Behavior and Momentary Wellbeing
Outcomes by Condition
Dependent
variable
Full sample
b (SE)
Act-Extraverted
b (SE)
Sham
b (SE)
Contact-control
b (SE)
Positiv
e affect
0.40*** (0.02) 0.39*** (0.03) 0.34*** (0.04) 0.45*** (0.03)
Negativ
e affect
–0.01 (0.02) –0.04* (0.02) 0.03 (0.03) —
Authen
ticity
0.09** (0.03) 0.10* (0.04) 0.09* (0.05) —
Tiredne
ss
–0.20*** (0.03) –0.24*** (0.04) –0.13* (0.05) —
Note. *p < .05, **p < .01, ***p < .001. b = the fixed effect of extraverted behavior
(unstandardized regression coefficient). SE = standard error. Dashes indicate that data were
not available.

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Table 5
Moderation of Momentary Associations Between Extraverted Behavior and Wellbeing
Outcomes by Intervention Condition
Act-E vs. Sham Act-E vs. Contact Sham vs. Contact
Dependent
variable b (SE)b (SE)b (SE)
Positiv
e affect
0.05 (0.05) –0.07 (0.05) –0.12* (0.05)
Negativ
e affect
–0.06 (0.04) — —
Authen
ticity
0.01 (0.06) — —
Tiredne
ss
–0.12 (0.06) — —
Note. *p < .05, **p < .01, ***p < .001. Act-E = Act-Extraverted. b = the cross-level interaction
effect of extraverted behavior × intervention condition (unstandardized regression
coefficient). SE = standard error. The effects of covariates are not shown (full model output is
presented in the Supplemental Materials). Dashes indicate that data were not available.

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Table 6
Moderation of Main Intervention Effects by Trait Extraversion
Act-Extraverted vs. Sham Act-Extraverted vs. Contact-control Sham vs. Contact-control
Dependent variable bCondition (SE)bCondition × Trait E (SE)bCondition (SE)bCondition × Trait E (SE)bCondition (SE)bCondition × Trait E (SE)R2
Momentary
Extraverted
behavior –1.16 (1.24) 0.81* (0.35) –2.16 (1.37) 0.99* (0.39) –1.00 (1.36) 0.18 (0.40) .14
Positive affect –0.11 (1.74) 0.24 (0.50) –2.37 (1.92) 0.72 (0.55) –2.26 (1.91) 0.49 (0.56) .05
Negative affect 3.47* (1.50) –0.95* (0.43) — — — — .07
Authenticity –4.23* (1.71) 1.42** (0.48) — — — — .10
Tiredness 4.78** (1.77) –1.42** (0.50) — — — — .03
Retrospective
Extraverted
behavior –0.00 (1.21) 0.44 (0.34) 0.44 (1.26) 0.17 (0.36) 0.45 (1.26) –0.27 (0.37) .46
Positive affect –1.78* (0.74) 0.58** (0.21) –0.45 (0.77) 0.21 (0.22) 1.33 (0.78) –0.37 (0.23) .19
Negative affect –0.32 (0.73) 0.08 (0.21) –0.37 (0.76) 0.06 (0.22) –0.05 (0.76) –0.02 (0.22) .06
Authenticity –4.14*** (1.05) 1.33*** (0.30) — — — — .30
Tiredness 1.46* (0.72) –0.44* (0.20) — — — — .17
Note. * p < .05, ** p < .01, *** p < .001. bcondition = the main effect of intervention condition (unstandardized regression coefficient). bCondition × Trait E =
the intervention condition × trait extraversion interaction effect. SE = standard error. Effects of trait extraversion and covariates are not shown
(full model output is presented in the Supplemental Materials). R2 for multilevel models, based on multivariance partitioning (MVP) was
calculated following LaHuis et al. (2014, Equation 14). Dashes indicate that no data were available.

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Table 7
Summary of Key Results
Outcome
Variable
Main Intervention
Effect
Main Intervention Effect
Mediated by Extraverted
Behavior?
Main Intervention Effect Moderated by Trait E?
Extraverted Behavior
Momentary
Act-E > Sham Act-E = Sham for extreme introverts (ZTrait E less than –1.95)
Act-E > Contact Act-E = Sham for introverts (ZTrait E less than –1.24)
Contact = Sham No
Retrospective
Act-E > Sham No
Act-E > Contact No
Contact > Sham No
PA
Momentary
Act-E > Sham Yes No
Act-E = Contact No
Contact > Sham No
Retrospective
Act-E > Sham Yes Act-E = Sham for those below average on Trait E (ZTrait E less than 0.05)
Act-E > Contact Yes No
Contact = Sham No

COSTS AND BENEFITS OF ACTING EXTRAVERTED
Table 7 Continued
Outcome
Variable
Main Intervention
Effect
Main Intervention Effect
Mediated by Extraverted
Behavior?
Main Intervention Effect Moderated by Trait E?
Authenticity
Momentary Act-E > Sham No Act-E = Sham for those below average on Trait E (ZTrait E less than –0.19)
Retrospective Act-E > Sham
Yes Act-E = Sham for those below average on Trait E (ZTrait E less than –0.16)
Sham > Act-E in introverts (ZTrait E < –1.43)
Tiredness
Momentary Act-E = Sham Act-E < Sham for those above average on Trait E (ZTrait E greater than 0.68)
Act-E > Sham for introverts (ZTrait E less than –1.39)
Retrospective Act-E = Sham Act-E < Sham for those above average on Trait E (ZTrait E greater than 0.77)
NA
Momentary Act-E = Sham Act-E > Sham for those below-average on Trait E (ZTrait E less than –0.71)
Retrospective
Act-E = Sham None
Act-E = Contact None
Contact = Sham None
Note. Act-E = Act-Extraverted. > indicates significantly greater, < indicates significantly less, = indicates ns difference. Z indicates the number
of standard deviations from the mean of trait E. Blank cells indicate that analyses were not conducted.