sociale motions

Research Report
Adaptive Social Emotion or Seventh Sin?
Lisa A. Williams and David DeSteno
Northeastern University
experiment examined the ability of pride
to serve as an adaptive emotion within the context of social
interaction. After an in vivo induction of pride or a neutral
state, participants engaged in a group problem-solving
task. In contrast to a conventional view that pride is often
associated with negative interpersonal outcomes, results
confirmed that proud individuals not only took on a dominant role within the group problem-solving task, but also
were perceived as the most likeable interaction partners.
These findings suggest that pride, when representing an
appropriate response to actual performance (as opposed
to overgeneralized hubris), constitutes a functional social
emotion with important implications for leadership and
the building of social capital.
Views of pride, unlike the majority of emotions, demonstrate a
Janus-faced quality both through time and across cultures, from,
for example, Dante’s Divine Comedy to Shakespeare’s Troilus
and Cressida (Lewis, 2000; Tracy & Robins, 2007a, 2007b).
Unlike love, which is universally admired, or jealousy, which is
universally reviled, pride has been alternately viewed as both
virtue and vice, noble characteristic and deadly sin. The question of pride’s place in the emotion arsenal thus becomes an
intriguing one. Does it engender greater stature or, as suggested
by Emily Bronte¨ in Wuthering Heights, lead to the breeding of
sad sorrows?
In considering this issue, many researchers have suggested
that, although pride holds a somewhat negative connotation in
American culture,1 it may possess a dual nature and, given
certain contingencies, can be beneficial (Fredrickson & Branigan, 2001; Tracy & Robins, 2007a; Williams & DeSteno, 2008).
The pivotal factor appears to be whether pride stems from
objective successes (i.e., authentic, or beta, pride) or from a
Address correspondence to Lisa Williams or David DeSteno, Department of Psychology, Northeastern University, Boston, MA 02115,
e-mail: [email protected] or [email protected]
Synonyms for proud are almost universally negative (e.g., arrogant, haughty).
generalized overly positive assessment of self-worth (i.e.,
hubristic, or alpha, pride; Lewis, 2000; Tangney, 1999; Tracy &
Robins, 2007a, 2007b). To date, however, empirical support for
the beneficial effects of pride has been quite limited.
Working from a functionalist perspective, we endorse the view
that emotions serve as efficient mechanisms that aid individuals
in responding to adaptive challenges. For humans, many challenges involve successful navigation not only of the physical
environment, but also of the social environment and, consequently, necessitate the existence of a class of socially oriented
emotional responses. If pride does represent such an emotion,
then pride derived from appraisals involving actual successes on
definable tasks should lead to distinct behavioral and social
outcomes that serve to benefit individuals vis-a`-vis social
It is our contention that a primary function of pride is to
motivate hedonically costly efforts aimed at acquiring skills that
increase one’s status and value to one’s social group. In essence,
we believe that pride, unlike generalized positive affect (cf.
Wegener & Petty, 1994), should impel one to incur short-term
costs (e.g., expenditure of high effort) for the purpose of reaping
longer-term rewards (e.g., value by a social group). That is, pride
should motivate individuals to acquire and demonstrate abilities, even in the face of initial difficulties, in order to increase
their status and attractiveness with respect to interaction partners. As initial support for this view, we have demonstrated that
pride engenders perseverance on socially valued tasks and have
dissociated this influence from the related factors of self-efficacy,
self-esteem, and generalized positive affect (Williams & DeSteno,
2008). Yet such findings do not speak directly to the social
functions of pride. Simply put, the fact that pride may mediate
increased perseverance on tasks does not necessarily imply
that individuals exhibiting pride are viewed as potential leaders
or experts and, more importantly, are perceived more positively
by peers.
The goal of the present experiment was to put our theory about
the socially adaptive value of pride to the test. Building off our
previous work demonstrating that pride facilitates increased
effort on valued tasks while working individually, the current
Copyright r 2009 Association for Psychological Science
Volume 20—Number 3
Lisa A. Williams and David DeSteno
experiment utilized a group problem-solving task designed to
determine whether proud individuals would not only take on a
dominant role within an interpersonal setting, but also be viewed
positively by their partners. To accomplish this goal, we adapted
the pride induction we developed previously (Williams &
DeSteno, 2008) and had participants engage in a cooperative
group task. Within the context of this interaction, we obtained
two measures of dominance: one based on objective behavior
and one based on subjective impressions. In addition, we obtained
a measure of social value of group members, which we operationalized as interpersonal liking.
Sixty-two individuals (48 female, 14 male; mean age 5 19 years)
participated in exchange for course credit. Participants completed the study in same-sex dyads; one member of each pair was
randomly assigned to the pride condition, and the other to the
neutral condition.
Upon arriving at the lab, participants were informed that they
were one of three individuals completing the experiment. In
reality, the third ‘‘participant’’ was a confederate blind to the
hypotheses of the study. The gender of the confederate matched
the gender of the participants.
At the start of the experiment, all three individuals were
seated at individual personal computers and given an overview
of the tasks they would complete. The experimenter also informed them that the primary purpose of the session was to
obtain scores on and evaluations of related spatial tasks that
were in the process of development. They would complete both
individual and group tasks and subsequently be asked questions
about their performance and the performance of others.
Participants then turned to their computers to complete the
first task, which was presented as a measure of visuospatial
ability. This task consisted of 15 mental rotation exercises.
Participants had to decide whether two images of three-dimensional objects were identical (i.e., were the same object only
oriented differently in space). They were told that the computer
would calculate their scores based on both the accuracy of their
responses and their response times as compared with the speeds
of others. Consequently, participants were not able to estimate
reliably how well they performed.
After all participants had finished this task, the experimenter
announced that individual calibrations of visual acuity were
necessary and, to maintain privacy, would be completed in the
adjoining room individually. Once in the separate room, participants were seated in front of a computer and asked to perform
a 10-s spatial eyesight test. The true purpose of this ‘‘calibration’’ was to allow for the manipulation of the primary inde-
Volume 20—Number 3
pendent variable. At this point, participants assigned to the
pride condition received the pride induction before returning to
the main room; those in the control condition simply completed
the screening before returning. Once all three individuals finished the calibration, they completed an ‘‘evaluation’’ of the
mental rotation exercise. In addition to filler items asking about
image quality and instruction clarity, this evaluation included
the manipulation check for the pride induction.
The experimenter next provided instructions for the second
task. Participants relocated their chairs around a table in the
center of the room. The experimenter informed them that the
task would require the group to work together on a three-dimensional puzzle and that the group would receive a score based
on their progress toward the solution. The puzzle was presented
as a cube, which the experimenter unwound into a single rod of
smaller adjoining cubes. To solve the puzzle, participants would
have to bend and rotate the individual adjoining pieces of the
rod to re-form them back into the large cube. This task appeared
quite similar to the mental rotation tasks that participants had
completed individually; in that task, the presented three-dimensional objects all consisted of adjoining cubes bent and
rotated in different ways. The experimenter also turned on a
video camera located over the confederate’s shoulder and facing
the participants, noting that its use was to get a transcript of the
conversation. Each group was given 6 min to work on the task.
Confederates were trained to interact consistently and to touch
the puzzle for approximately 1 min total. At the end of the allotted
time, participants returned to their computers and completed an
‘‘evaluation’’ of this task, which included subjective ratings of
their partners. Participants were then dismissed.
Manipulations and Measures
Pride Manipulation
As noted, the pride manipulation occurred individually as part
of the supposed calibration for visual acuity. Each participant
was alone with the experimenter at the time of the manipulation.
For participants in the pride condition, the experimenter casually pulled three sheets of paper out of a printer, shuffled through
them, and said, ‘‘I just wanted to show you how you did on that
first task. You got a score in the 94th percentile—great job!
That’s one of the highest scores we’ve seen!’’ As she made these
comments, she pointed to an official-looking score sheet that
included a graphical image indicating the participant’s high
score. Participants in the neutral condition were simply asked to
return to the lab once they had completed the calibration check.
To maintain consistency, confederates entered the control room
with the experimenter, waited the appropriate amount of time,
and then reentered the lab.
Manipulation Checks
Using 7-point scales (1 5 not at all, 7 5 completely), participants indicated how they were currently feeling in reference to
Virtues of Pride
several descriptors. Pride was calculated as the mean response
to the items: proud, confident, satisfied, fulfilled, accomplished,
successful, achieving, productive, and full of self-worth (a 5 .93;
cf. Tracy & Robins, 2007b; Williams & DeSteno, 2008). To
measure differences in perceived subjective ability, participants
also were asked to estimate how well they believed they performed in comparison to others. Participants also rated how
positive, good, content, and happy they felt; these items were
combined to form an index of positive mood (a 5 .91).
Dominance was measured in three ways. The first measure was
based on objective behavior. Given that solving the puzzle required active manipulation of its pieces, a coder viewed each
video clip and recorded the total amount of time each participant
manipulated the puzzle. The other two measures involved subjective assessments of dominance. For the first, we utilized participants’ ratings of each other. As part of the second evaluation,
each person completed ratings of his or her two interaction
partners with respect to items measuring subjective dominance,
leadership, contributions, ability, and relative time spent talking
and listening during the puzzle completion (cf. Anderson, Srivastava, Beer, Spataro, & Chatman, 2006; Dovidio, Ellyson,
Keating, Heltman, & Brown, 1988). These items were averaged
to form an index of subjective dominance (a 5 .80). Finally, we
obtained a second measure of subjective dominance from two
individuals who did not take part in the experiment and,
therefore, would not be subject to any self-presentation concerns
about rating others as more dominant than the self. These individuals were drawn from the same population as the primary
participants and were unaware of both the purpose of the study
and the experimental conditions of the participants. Each rater
was asked to view each 6-min video clip and provide a rating
(using a 7-point scale: 15 not at all dominant, 7 5 extremely
dominant) for the dominance level of each individual in the
respective dyads (a 5 .78).
As part of the postpuzzle evaluation measure, participants also
completed ratings of their two interaction partners with respect
to their level of liking for each. Specifically, they were asked how
much they liked the person, would want to work with the person
again, and enjoyed the interaction with the person. These items
were averaged to form an index of liking (a 5 .84).
Given that participants were nested within dyads, all statistical
analyses utilized techniques appropriate for nested data. All
reported t values stem from paired t tests, and all reported regression coefficients stem from multilevel models in which
participant-level data were nested within dyadic groupings.
As expected, the pride manipulation proved successful; participants in the pride condition reported feeling more pride (M 5 4.78)
than did those in the neutral condition (M 5 3.73), t(30) 5 3.47,
p 5 .002, prep 5 .98, d 5 1.00. Confirming predictions, proud
participants also evidenced greater dominance within the context of the group problem-solving task. They spent more time
manipulating the puzzle (M 5 192 s) than did neutral participants (M 5 145 s), t(30) 5 2.18, p 5 .04, prep 5 .92, d 5 0.66.
They were also perceived as more dominant by their partners
(M 5 4.61) than were neutral participants (M 5 4.30), t(29) 5
1.86, p 5 .07, prep 5 .86, d 5 0.50,2 and, confirming this perception, were also judged to be more dominant by third-party
observers (M 5 4.31) than were neutral participants (M 5 3.48),
t(30) 5 2.56, p 5 .02, prep 5 .92, d 5 0.66. Of import, dominance was attributable specifically to pride; time spent manipulating the puzzle was directly predicted by pride intensity, b 5
16.31, t(60) 5 2.06, p 5 .04, prep 5 .89, but not by positive mood
or self-efficacy (i.e., participants’ subjective assessments of their
level of relative performance on the task).3 Thus, although both
judgments of self-efficacy and positive mood were elevated in
the pride condition (ts > 2.60, preps > .95), these constructs
cannot explain the predicted finding in dominance behavior.
Finally, our primary hypothesis regarding the social attractiveness of an individual exhibiting pride also received support.
As shown in Figure 1, individuals experiencing pride were more
liked by their partners (M 5 5.25) than were individuals in the
neutral condition (M 5 4.74), t(30) 5 2.53, p 5 .02, prep 5 .98,
d 5 0.93. To demonstrate that this effect stemmed from differential perceptions by participants in the neutral condition, we
also compared liking for participants with liking for the confederate. As depicted in Figure 1, neutral participants demonstrated enhanced liking for the proud participant (M 5 5.25) as
compared to the confederate (M 5 4.89), t(30) 5 2.71, p 5 .01,
prep 5 .95. However, proud participants did not evidence any
differential liking between neutral participants (M 5 4.74) and
confederates (M 5 4.82), t < 1, thereby confirming that differential liking reflects an enhanced attraction of neutral participants to partners experiencing pride.4 This finding also
demonstrates that liking is not a simple function for effort
expended on the task, as proud individuals did not show any
differential liking for neutral participants and confederates, even
though confederates exerted less effort than neutral participants.
Data from one dyad were removed because of an aberrantly low dominance
The b coefficients refer to slopes from two-level hierarchical linear models
regressing time on the potential predictors while controlling for dyad membership.
To verify the presence of an interaction between emotion and target identity,
we utilized a three-level hierarchical linear model analysis in which likeability
ratings for individuals (i.e., partner vs. confederate) were nested within participants, which were nested within dyads. As expected, results confirmed that
neutral individuals, as compared with proud individuals, evidenced greater
differences in liking between the two targets, or, put differently, evidenced a
steeper slope for the Level 1 variable indexing the effect of target identity on
liking, b 5 0.51, t(120) 5 3.39, p 5 .001, prep 5 .99.
Volume 20—Number 3
Lisa A. Williams and David DeSteno
Liking for Interaction Partner
Neutral State,
Proud Target
Neutral State,
Proud State,
Neutral Target
Proud State,
Fig. 1. Mean ratings of liking for interaction partners as a function of the participant’s emotional
state (neutral or proud) and the interaction partner’s identity (proud target, neutral target, or
confederate). Higher numbers indicate increased liking.
Taken together, we believe these findings are noteworthy. They
are the first to show functional outcomes of pride within the
context of actual social behavior, and thereby constitute an
important cornerstone for a view of pride as an adaptive emotion.
Although pride is certainly associated with the constructs
of self-efficacy and self-esteem, the experience of the emotion
itself appears to be a principle motivator of adaptive behavior
and social perception (Williams & DeSteno, 2008). However,
illumination of the mechanisms underlying increases in perceived attractiveness requires future investigation. Heightened
attractiveness cannot stem solely from increased behavioral
dominance, yet it is nonetheless likely that some interaction
between behaviors and expressive signals associated with authentic pride serves to indicate social value (cf. Tracy & Robins,
2007a).5 Indeed, it may be just such a link between expression
and context that serves as a mechanism to dissociate adaptive
versus hubristic pride. Over time, the disconnect between
pride expressions and behavioral evidence may hone individuals’ accuracies with respect to the ‘‘signal value’’ of pride
expressed by particular actors.
In our study, liking for others was associated with their perceived dominance, b 5 0.44, t(60) 5 3.01, p 5 .004, prep 5 .99, but was not predicted by
their level of general positive mood.
Volume 20—Number 3
At present, though, it appears certain that pride does not always
breed sad sorrows. Rather, when experienced under appropriate
conditions, it may play an essential role in the development of
leadership and social capital.
Anderson, C., Srivastava, S., Beer, J.S., Spataro, S.E., & Chatman, J.A.
(2006). Knowing your place: Self-perceptions of status in face-toface groups. Journal of Personality and Social Psychology, 91,
Dovidio, J.F., Ellyson, S.L., Keating, C.F., Heltman, K., & Brown, C.E.
(1988). The relationship of social power to visual displays of
dominance between men and women. Journal of Personality and
Social Psychology, 54, 233–242.
Fredrickson, B.L., & Branigan, C. (2001). Positive emotions. In T.J.
Mayne & G.A. Bonanno (Eds.), Emotions: Current issues and
future directions (pp. 123–151). New York: Guilford Press.
Lewis, M. (2000). Self-conscious emotions: Embarrassment, pride,
shame, and guilt. In M. Lewis & J.M. Haviland-Jones (Eds.),
Handbook of emotions (2nd ed., pp. 623–636). New York: Guilford Press.
Tangney, J.P. (1999). The self-conscious emotions: Shame, guilt, embarrassment and pride. In T. Dalgleish & M.J. Power (Eds.),
Handbook of cognition and emotion (pp. 541–568). New York:
John Wiley & Sons.
Tracy, J.L., & Robins, R.W. (2007a). Emerging insights into the nature
and function of pride. Current Directions in Psychological Science,
16, 147–150.
Virtues of Pride
Tracy, J.L., & Robins, R.W. (2007b). The psychological structure of
pride: A tale of two facets. Journal of Personality and Social
Psychology, 92, 506–525.
Wegener, D.T., & Petty, R.E. (1994). Mood management across
affective states: The hedonic contingency hypothesis. Journal of
Personality and Social Psychology, 66, 1034–1048.
Williams, L.A., & DeSteno, D. (2008). Pride and perseverance: The
motivational role of pride. Journal of Personality and Social
Psychology, 94, 1007–1017.
Volume 20—Number 3