Holding a mirror up to the self: egocentric similarity beliefs underlie social projection in cooperation.

Abstract

Similarity between partners entails positive consequences for cooperative interactions. But do people rely on this assumption to construe egocentric judgments about others? Five experiments examined the possibility that people project onto their partners because they believe that similarity to the self leads to success in cooperation. Studies 1a and 1b show that people hold an egocentric similarity belief in cooperation. Studies 2a and 2b test the existence of this belief in more indirect ways. The next three studies manipulate the applicability of the similarity belief and investigate its impact on projection. Study 3 finds that cooperation no longer leads to projection when participants expect a low probability of success. Study 4 replicates this effect in a real cooperative setting. Finally, Study 5 shows that projection occurs only when participants expect their characteristics to be responsible for the success of cooperation. The negative consequences of overestimating similarities in cooperation are discussed.

Full text

Page 1

http://psp.sagepub.com/
Bulletin
Personality and Social Psychology
http://psp.sagepub.com/content/38/10/1259
The online version of this article can be found at:

DOI: 10.1177/0146167212449022
2012 38: 1259 originally published online 13 June 2012Pers Soc Psychol Bull
Claudia Toma, Olivier Corneille and Vincent Yzerbyt
Holding a Mirror Up to the Self : Egocentric Similarity Beliefs Underlie Social Projection in Cooperation


Published by:

http://www.sagepublications.com
On behalf of:




Society for Personality and Social Psychology
can be found at:
Personality and Social Psychology Bulletin

Additional services and information for




http://psp.sagepub.com/cgi/alerts
Email Alerts:

http://psp.sagepub.com/subscriptions
Subscriptions:



http://www.sagepub.com/journalsReprints.nav
Reprints:


http://www.sagepub.com/journalsPermissions.nav
Permissions:
What is This?

- Jun 13, 2012OnlineFirst Version of Record

- Sep 18, 2012Version of Record >>
at Univ Catholique Louvain Bib on September 19, 2012psp.sagepub.comDownloaded from

Page 2

Personality and Social
Psychology Bulletin
38(10) 1259 –1271
© 2012 by the Society for Personality
and Social Psychology, Inc
Reprints and permission:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0146167212449022
http://pspb.sagepub.com
When it comes to social interactions, similarity often
appears to be an asset. Whether a relationship is located
within a professional context or concerns more intimate
aspects, the opportunity to build on common ground often
comes as a blessing. Two potential industrial partners who
share the same views about their business expansion will
have an easier time making a deal. Two spouses coming
from the same socioeconomic background will meet with
fewer problems when making a number of decisions over
the course of their marital life. As a matter of fact, a substan-
tial amount of work suggests that similarity entails a number
of positive consequences in social interactions.
But do social perceivers actually rely on the working
assumption that similarity is beneficial to cooperation when
judging their partners? And, more importantly, what type of
similarity are we talking about? In light of people’s well-
established egocentric tendencies, should we not expect
people to approach this similarity advantage in a rather par-
tisan way, namely, by hoping that their partner will resemble
them? The present article aims to test the hypothesis that
people have an egocentric similarity belief that leads them
to project their characteristics onto their partner when antic-
ipating cooperation. First, we conjecture that people believe
that cooperation is more likely to meet with success when
their partner resembles them rather than when they resemble
their partner. Second, we argue that because people have
this egocentric belief, they tend to see their partner as simi-
lar to themselves in cooperative settings. Below, we discuss
the theoretical rationale and empirical context for our pre-
dictions. We then report five studies that examined these
conjectures.
The Dividends of Similarity
in Cooperative Settings
Empirical evidence suggests that similarity entails a number
of positive consequences. As a matter of fact, cooperation
often occurs as a result of interaction between similar part-
ners (Nowak, 2006). Partner similarity has been found to
induce interpersonal attraction (Byrne, 1971), affiliation
(Lakin & Chartrand, 2003), altruistic behavior (Cunningham,
449022 PSP381010.1177/0146167212449022Toma
et al.Personality and Social Psychology Bulletin
1Université catholique de Louvain, Louvain-la-Neuve, Belgium
2Fonds de la Recherche Scientifique, Brussels, Belgium
Corresponding Author:
Claudia Toma, Université catholique de Louvain, Place Cardinal Mercier 10,
1348 Louvain-la-Neuve, Belgium
Email: claudia.toma@uclouvain.be
Holding a Mirror Up to the Self:
Egocentric Similarity Beliefs Underlie
Social Projection in Cooperation
Claudia Toma1, 2, Olivier Corneille1, and Vincent Yzerbyt1
Abstract
Similarity between partners entails positive consequences for cooperative interactions. But do people rely on this assumption
to construe egocentric judgments about others? Five experiments examined the possibility that people project onto their
partners because they believe that similarity to the self leads to success in cooperation. Studies 1a and 1b show that people
hold an egocentric similarity belief in cooperation. Studies 2a and 2b test the existence of this belief in more indirect ways.
The next three studies manipulate the applicability of the similarity belief and investigate its impact on projection. Study 3
finds that cooperation no longer leads to projection when participants expect a low probability of success. Study 4 replicates
this effect in a real cooperative setting. Finally, Study 5 shows that projection occurs only when participants expect their
characteristics to be responsible for the success of cooperation. The negative consequences of overestimating similarities in
cooperation are discussed.
Keywords
egocentric, similarity belief, interpersonal projection, cooperation, success
Received September 23, 2011; revision accepted March 12, 2012
at Univ Catholique Louvain Bib on September 19, 2012psp.sagepub.comDownloaded from

Page 3

1260
Personality and Social Psychology Bulletin 38(10)
1986), and trust (Zak & Knack, 2001), and all of these are
important elements responsible for successful cooperation.
Using computer simulations, Riolo, Cohen, and Axelrod
(2001) showed that cooperation can arise between two agents
who are sufficiently similar in some arbitrary characteristics.
In a prisoner’s dilemma game, creating pairs of people who
perceive each other as being similar has been shown to pre-
dict cooperation (Fischer, 2009). In sociobiology, individuals
sharing a gene that leads to a visible tag (e.g., a green beard)
tend to help each other more (Traulsen, 2008).
All forms of similarity have been found to be associated
with favorable social outcomes: Similarity in race increases
helping and cooperation (Gaertner & Dovidio, 1977), simi-
larity in attitudes increases cooperation in social dilemmas
(Van Vugt & Hart, 2003), physiognomic similarity induces
political cooperation (Heschl, 1993), and similarity in person-
ality, values, and goals leads to cooperation in organizations
(Bretz & Judge, 1994). Clearly, thus, people are attracted
more to and cooperate more readily with similar rather than
dissimilar others.
Egocentric Similarity Belief
It may well be that partner similarity leads to positive con-
sequences, but is it the case that people know this and orient
their judgment in light of this belief? People entertain a
wide variety of beliefs, sometimes called lay beliefs, to
make sense of their social environment (Anderson &
Lindsay, 1998). The influence of lay beliefs on people’s
perceptions of others has, in fact, long been a central con-
cern in the study of social judgment (e.g., Heider, 1958;
Ross & Nisbett, 1991). What is particularly important about
these beliefs is that they exert a strong influence on percep-
tion, expectations, and judgments (Dweck, Chiu, & Hong,
1995; Wegener & Petty, 1998). On observing the same
person, social perceivers holding different beliefs do not
“see” exactly the same person.
But how do people elaborate such beliefs? Many lay
beliefs are learned primarily through direct experience
(Anderson & Lindsay, 1998). In the present case, the expo-
sure to different cooperative situations may have led people
to infer that working with similar partners may be highly ben-
eficial. Different individuals remember different situations,
but what is important here is that people usually have in mind
instances that reflect favorably on the self. There is growing
evidence that people entertain self-serving theories about suc-
cess and competence (Dunning, Leuenberger, & Sherman,
1995), that they have self-serving conceptions of traits and
abilities (Dunning & Cohen, 1992), and, more generally, that
they define the social reality in a way that puts themselves
and their conceptions of life under a favorable light (Dunning
& Hayes, 1996). Self-serving biases have been found across
numerous domains: People perceive themselves as more fair
(Messick, Bloom, Boldizar, & Samuelson, 1985), more com-
petent (Yan & Gaier, 1994), more successful (Urban & Witt,
1990), and more socially responsible than others (White &
Plous, 1995). Moreover, people attribute responsibility for
outcomes more to themselves than to others (Ross & Sicoly,
1979; Savitsky, 2007), and they give themselves credit in
cases of success but not in cases of failure (Mark, Mutrie,
Brooks, & Harris, 1984; Miller & Ross, 1975). People are
motivated to think well of themselves, and this motive may
influence their judgments of others. Research has shown that
people tailor their judgments of others to maintain and bolster
their egocentric beliefs about themselves and their self-worth
(Beauregard & Dunning, 1998).
In the research reported here, we propose that another
egocentric tendency people may have is to believe that their
partner’s similarity to themselves predicts successful coop-
eration. Two arguments support this proposal. First, in line
with the idea that the self is a habitual reference point in
judgments, Holyoak and Gordon (1983) found that people
see more similarity of the other with the self than of the self
with the other. Second, and even more importantly, work on
motivated inference (Kunda, 1987) suggests that people
believe that their own attributes, more than those of others,
are desirable to achieve positive outcomes. Therefore, if
people consider their attributes as valuable for cooperation,
they may find it justified to see similarity to the self as a
guarantee of success in this context.
Similarity Beliefs and
Social Projection
A logical consequence of holding beliefs associating simi-
larity with success is that people would prefer to associate
with more similar partners. Moreover, in light of the well-
established fact that human beings are egocentric creatures,
an intriguing consequence of this belief is that people may
see their partner as similar to themselves whenever they
find themselves in a cooperative setting.
It is widely known that people tend to overestimate self–
other similarity when making predictions about other peo-
ple’s behavior (Ross, Greene, & House, 1977), personality
(Krueger, 1998), or attitudes and preferences (Katz &
Allport, 1931). Of interest, research indicates that the latter
tendency is magnified under cooperation, with more social
projection observed for people working in cooperative con-
texts (Riketta & Sacramento, 2008). For instance, Toma,
Yzerbyt, and Corneille (2010) showed that people see their
partner as more similar to themselves when they anticipate
cooperation rather than competition. Of note, however,
little is known about what may drive social projection in
cooperation.
We suggest here that people in a cooperation context project
onto their partner because they hold the belief that interacting
with partners who resemble them leads to success. In other
words, by projecting the self onto the partner, social perceivers
may think (sometimes erroneously) that the cooperation will
turn out to be a success. In the context of a prisoner’s dilemma
at Univ Catholique Louvain Bib on September 19, 2012 psp.sagepub.comDownloaded from

Page 4

Toma et al.
1261
game, if one of the two suspects decides not to confess, he or
she may project the same tendency onto the other suspect in the
hope that this should lead to a release or a minimum punish-
ment. If social projection is enhanced in cooperative contexts
due to the aforementioned egocentric belief, then projection
should be reduced when people no longer expect to reach a
successful outcome in a cooperation setting. Furthermore, pro-
jection should also be reduced when people doubt that their
characteristics may engender successful cooperation.
Overview of the Studies
Past research shows that similarity often facilitates coopera-
tion. We argue that people, building on direct and indirect
cooperative experiences, hold the belief that similarity to the
self is beneficial for cooperative success. As a result, and
combined with the egocentric tendencies that characterize
social perceivers, we predict that this belief will make them
project their own characteristics onto their partner. In Study
1a, we examine whether people hold the belief that similarity
leads to positive outcomes in cooperation, and Study 1b tests
whether this belief is indeed egocentric. In Studies 2a and 2b,
we test the existence of this belief in a more indirect way.
Overall, the goal of these studies is to confirm the existence
of a belief linking similarity to successful outcomes in coop-
eration in a rather egocentric way.
Building on this evidence, Studies 3, 4, and 5 examine the
implications of this belief for social projection. In Study 3,
we manipulated the applicability of this belief by varying the
probability of success in cooperation. We predict that projec-
tion onto partners occurs only when participants expect that
the probability of success is high. In Study 4, we tested
whether this effect is replicable in a real cooperative task.
Finally, in Study 5, we test the possibility that projection
occurs only when people are led to think that their own char-
acteristics are conducive to success in cooperation.
Studies 1 and 2: Do People Hold
an Egocentric Similarity Belief?
Study 1a
People seem to be able to report on their beliefs. As a matter
of fact, measuring such theories has been and can be done
in a rather direct manner (Wegener & Petty, 1998). In this
first study, we asked our participants directly whether they
thought that they would succeed better in cooperation with
a similar than with a dissimilar partner.
Method
Participants. A total of 25 participants (8 females) were
recruited on the campus to participate in this study. Their age
ranged from 19 to 40 years (M = 23.5, SD = 4.88).
Procedure. Participants were asked to imagine a coopera-
tive situation and to choose the partner with whom they
would like to cooperate. Participants responded to five
questions in which they indicated on a 9-point scale ranging
from −4 (very dissimilar) to 4 (very similar) the type of part-
ner who would allow them to succeed in the cooperation
situation and the type of partner that they would like to
choose in this context. Four questions referred to the predic-
tion of performance (“To succeed in cooperation, my part-
ner should be”; “To avoid failure in cooperation my partner
should be”; “Cooperative tasks are difficult to perform if my
partner is”; “I can work better in cooperation if my partner
is”), α = .73. An additional question (“The partner I would
choose in cooperation should be”) referred to participants’
preference between a rather dissimilar and a rather similar
partner in the context of cooperation.
Results and Discussion. We hypothesized that people believe
that success in a cooperative task is better served by having to
face a similar rather than a dissimilar partner. They should
thus be more willing to choose a similar rather than a dissimi-
lar partner for cooperation. Calculating a t test against the
midpoint of the scale for both performance (M = 1.71,
SD =1.08), t(24) = 7.88, p < .001, and choice of the partner
(M = 1.84, SD = 1.90), t(24) = 4.82, p < .001, confirmed
this prediction. A similar study with 25 participants showed
that people do not hold the same belief in other interdepen-
dent situations like competition: M = 0.46 and SD = 1.81,
t(24) = 1.27, p = .22, for performance and M = −0.12 and
SD = 2.55, t(24) = −0.23, ns for the choice of the partner.
Study 1b
The results of Study 1b suggest that people believe that
having a similar partner increases their chances of success in a
cooperative situation. The important question here is whether
people generate this belief in a rather egocentric way so that
they think their partner should resemble them more than
they should resemble their partner. Therefore, this study was
expected to conceptually replicate the results of Study 1a this
time with a manipulation of the target of similarity. Participants
were asked whether their partner should be like them or they
should be like their partner for cooperation to succeed.
Method
Participants and design. A total of 60 participants (9 males)
were recruited on campus to participate in this study. Their
age ranged from 18 to 23 years (M = 19.47, SD = 1.25). Par-
ticipants were randomly assigned to the experimental condi-
tions (self vs. partner).
Procedure. Participants were asked to imagine a coopera-
tive situation and to respond to the same questions as in
Study 1a. The difference was that the target of similarity was
either the self (e.g., “To succeed in cooperation, my partner
should be” on a 9-point scale from −4 [different from me] to
4 [like me]) or the partner (e.g., “To succeed in cooperation,
I should be” on a 9-point scale from −4 [different from my
at Univ Catholique Louvain Bib on September 19, 2012 psp.sagepub.com Downloaded from

Page 5

1262
Personality and Social Psychology Bulletin 38(10)
partner] to 4 [like my partner]). All these questions referred
to the prediction of success in cooperation (α = .79).
Results and Discussion. Participants in the similarity to the self
condition had a higher score on the predicted success of
cooperation (M = 1.33, SD = 1.43) than participants in the
similarity to the partner condition (M = 0.07, SD = 1.22),
t(58) = 2.12, p < .001. Moreover, only the score in the simi-
larity to the self condition was different from the midpoint of
the scale, t(29) = 5.08, p < .001 (t < 1 in the similarity to the
partner condition). This suggests that only similarity to the
self is thought to be associated with success in cooperation.
Studies 2a and 2b
Studies 2a and 2b further test the hypothesis that similarity to
the self is believed to be associated with success in coopera-
tion by using an indirect method. Specifically, we provided
participants with a partner whose profile of traits (Study 2a)
or preferences (Study 2b) was similar or dissimilar to their
own and asked them to predict how successful a cooperative
interaction between themselves and their partner would be.
We hypothesized that participants would predict more coop-
eration and expect more success with a similar than with a
dissimilar partner.
Method
Participants and design. A total of 39 participants (22
females) took part in Study 2a, and 35 participants (20
females) took part in Study 2b. They were students in various
majors, and they ranged in age from 19 to 37 years (M = 22.15,
SD = 3.38) in Study 2a and from 18 to 26 years (M = 21.51,
SD = 2.09) in Study 2b. Participants took part in a study of
judgment and cooperation in exchange for 5 Euros. They
were randomly assigned to the experimental conditions, that
is, similarity versus dissimilarity.
Procedure. In Study 2a, participants were first presented
with a list of 18 personality traits selected from the Big Five
personality test, and they were asked to select those 8 traits
that best characterized them. Half of these traits were positive
and half were negative. Participants were then invited to take
part in an unrelated experiment that took about 15 min. In
Study 2b, participants were first presented with a list of 20
activities (e.g., reading, hiking, going to the cinema, cook-
ing), and they were asked to choose 10 activities that they like
most. During this time, the experimenter created for each par-
ticipant a profile of a partner that was designed so as to be
either similar or dissimilar to their own. In the similarity con-
dition, the experimenter filled in a questionnaire by randomly
choosing 6 traits/8 activities among the 8 traits/10 activities
selected by the participants, and 2 traits/activities among the
remaining 10 traits/activities not selected by the participant.
In the dissimilarity condition, the experimenter filled in a
questionnaire by randomly choosing 2 traits/activities among
the 8 traits/10 activities selected by the participants, and 6
traits/8 activities among the remaining 10 traits/activities not
selected by the participant.
In both studies, once participants were done with the unre-
lated experiment, they were presented again with their own
profile and the profile of another participant, who allegedly
had completed the same questionnaire in another session.
Participants were given the opportunity to compare their own
ratings with those of this unknown partner. Participants were
then led to imagine that they would have to engage in a coop-
erative task for which the unknown participant would be their
partner. More specifically, they had to imagine working with
another student for a final year collaborative project, and they
were told that they could both obtain a good grade if they
coordinate their efforts and competences for a high-quality
project. Participants were then asked to answer a series of
questions. These questions referred either to participants’ pre-
dicted performance with their partner (α = .89; “With this per-
son, I will succeed in cooperation”; “With this person,
chances are that I will fail in cooperation [reverse-scored]”; “I
will work efficiently in cooperation with this person”; “The
cooperative task will be difficult to perform with this person
[reverse-scored]”) or to participants’ desire to work with their
partner (α = .89; “I would choose to work in cooperation with
this person”).
Finally, two questions checked whether participants con-
sidered their partner’s profile as being similar or dissimilar
(“This person has a personality profile that is different from
mine [reverse-scored],” “This person is like me”; r = .87).
All questions were answered on a scale ranging from 1 (not
at all) to 9 (very much). Once they had completed the ques-
tionnaire, participants were debriefed, thanked for their par-
ticipation, and dismissed.
Results and Discussion
Manipulation checks. Participants in the similarity condi-
tion perceived their partner as being more similar to them-
selves (M = 6.73, SD = 1.41) than did participants in the
dissimilarity condition (M = 2.65, SD = 1.12), t(37) = 9.99,
p < .001 (Study 2a). The same was found in Study 2b,
t(33) = 13.54, p < .001 (M = 7.47 and SD = 1.23 for the simi-
larity condition and M = 2.32 and SD = 0.99 for the dissimi-
larity condition).
Predicted performance and choice. In Study 2a, participants
in the similarity condition expected to perform better with
their partner (M = 6.56, SD = 1.31) than did participants in the
dissimilarity condition (M = 5.20, SD = 1.93), t(37) = 2.56,
p < .05. Moreover, participants chose to cooperate more with
a partner presented as being similar to themselves (M = 5.84,
SD = 1.95) than with a partner presented as being dissimilar
to themselves (M = 3.60, SD = 2.30), t(37) = 3.27, p < .01.
Similarly, in Study 2b, participants in the similarity condi-
tion expected to perform better with their partner (M = 7.91,
SD = 0.62) than did participants in the dissimilarity condition
(M = 5.85, SD = 1.74), t(33) = 4.71, p < .001. Moreover, par-
ticipants chose to cooperate with a partner presented as being
at Univ Catholique Louvain Bib on September 19, 2012 psp.sagepub.comDownloaded from

Page 6

Toma et al.
1263
similar to themselves more (M = 7.44, SD = 1.09) than with a
partner presented as being dissimilar to themselves (M =
3.00, SD = 1.22), t(33) = 4.44, p < .001.
These results indicate that participants tend to believe
that a similar, as opposed to a dissimilar, partner would
allow them to better succeed in cooperation. Moreover, they
expressed a preference for a similar rather than a dissimilar
person to engage with in the cooperative task. This was the
case for profiles based on similarity of traits or similarity of
preferences.
We decided to use both traits and preferences because the
use of personality traits may raise some questions with
respect to perceived similarity. Some of the personality
traits used here appeared highly desirable in the context of
future cooperation (e.g., sociable, conscious, creative, trust-
ful), and therefore it was possible that participants chose
these traits more than other traits because of this reason and
not because these traits were similar to those that partici-
pants perceived in themselves. Moreover, although half of
the traits presented to the participants were positive and half
were negative, participants predominantly chose the posi-
tive traits for their own profile. Among the eight traits
selected for their own profile, participants chose an average
of 4.54 (SD = 1.21) positive traits, t(38) = 2.78, p < .01. For
this reason, a profile designed to be similar was perceived
not only as similar but also as more desirable and positive.
Because the similarity in terms of preferences was also
shown to shape real-word relationships (Jamieson, Lydon,
& Zanna, 1987), the convergence of results on both studies
provides a strong test of our hypothesis.
As a set, the studies presented so far show that people
believe that similarity is beneficial in cooperation and that
this belief is egocentric. People may therefore project their
characteristics onto their partner in cooperation because they
believe that their own attributes are more likely than other
attributes to facilitate desired outcomes (Kunda, 1987). This
specific hypothesis is tested in the following studies.
Studies 3 to 5: Implications
of the Similarity Belief for
Social Projection
Building on the above findings, our second central question
deals with the consequences of people’s egocentric belief
that similarity to the self leads to success in cooperation.
Specifically, the question is whether people are also more
willing to perceive their partners as similar when they expect
to succeed in cooperation. Our next three studies address this
issue by manipulating the contextual conditions that disrupt
(or not) the applicability of the similarity belief and examin-
ing its impact on interpersonal projection. This manipulation
is in line with methodological approaches suggesting that, to
test a mechanism hypothesis, one should compare a condition
that is expected to disrupt the proposed mechanism with a
condition that is expected not to (Spencer, Zanna, & Fong,
2005). In the case of the present research, the use of the simi-
larity belief should be disrupted if participants expect to fail in
cooperation or if the partner’s characteristics rather than their
own characteristics are expected to be responsible for the suc-
cess. We therefore manipulated the expected outcome of
cooperation either in an imagined scenario (Study 3) or in a
real cooperative learning task (Study 4). Study 5 manipulated
participants’ expectations that their characteristics (or those of
the partner) facilitate the success of cooperation.
Study 3
This study examines whether people will project the self onto
their partner when they doubt the success of cooperation.
Clearly, if people are convinced that the cooperation will not
lead to success, the similarity belief is of no use and therefore
participants should not be tempted to see their partner as
similar to themselves. To examine this question, we manipu-
lated the expected outcome of cooperation by varying the
probability of success. We hypothesized that projection
should be observed only when participants are told that there
is a high, as opposed to low, likelihood of success.
Method
Participants and design. A total of 46 participants (35
females), all students in various disciplines, took part in a
study of spontaneous impression formation. They ranged in
age from 18 to 24 years (M = 20.20, SD = 1.58). Participants
were randomly assigned to one of two experimental condi-
tions (low vs. high probability of success).
Procedure. On the first page of the questionnaire, partici-
pants rated themselves on a list of 16 personality traits (8
positive and 8 negative). These traits were borrowed from
Riketta and Sacramento (2008), who showed them to be
unrelated to cooperation in a pretest. Participants had to indi-
cate the extent to which each of the traits (e.g., progressive,
silent) characterized them on a 9-point rating scale ranging
from 1 (not at all) to 9 (very much).
On the second page, participants read a scenario in which
they had to imagine being employed in a new software com-
pany. In this company, they would have to team up with an
unknown person (the target). They learned that the top man-
ager of this company would offer a bonus trip to the Caribbean
if they managed to sell more than 10,000 copies of computer
software during the first 6 months. The relationship with the
target was presented as being cooperative. Participants were
told that both they and their partner could win the trip, so that
they should help each other to sell, together, the 10,000 cop-
ies of the computer software. Depending on the condition,
participants were told that by working this way, 95% (vs. 5%)
of the teams previously succeeded in selling more than 10,000
copies of computer software during the first 6 months.
On the third page of the questionnaire, participants were
asked to rate the target on the same list of traits that they had
at Univ Catholique Louvain Bib on September 19, 2012psp.sagepub.com Downloaded from

Page 7

1264
Personality and Social Psychology Bulletin 38(10)
used to rate the self. Because the valence of the traits could
vary as a function of their attribution to the self (Krueger,
1998; Sinha & Krueger, 1998), participants were also asked
to rate the valence of each personality trait, using a 9-point
scale ranging from −4 (rather negative) to +4 (rather
positive).
Finally, as a manipulation check, participants were asked
whether their chances to succeed in cooperation were high
or low. All participants answered correctly. When all tasks
had been completed, participants were debriefed, thanked,
and dismissed.
Results. Because the ratings of personality traits were nested
within participants, the data were analyzed by means of mul-
tilevel modeling. The traits were our Level 1 variable, and the
expected outcome was our Level 2 variable. The analysis
used the ratings of the target as our criterion and the self-
ratings and valence of traits as predictors at Level 1. The
expected outcome of cooperation (low vs. high probability of
success) was included as a moderator variable at Level 2.
Our hypothesis was that the variation of the Level 1 slopes
(i.e., the relationship between self-ratings and target ratings)
would be influenced by our Level 2 variable (expected out-
come) and that this effect would occur independently of trait
valence. In other words, we predicted the presence of a sig-
nificant cross-level interaction between self-ratings and the
expected outcome. To test our prediction, we implemented
the following model:
Level 1 model:
Targetij = π0i + π1i Self + π2i Valence + π3i Self × Valence + εij
Level 2 model:
π0i = β00 + β01 Outcome + u0i
π1i = β10 + β11 Outcome + u1i
π2i = β20 + β21 Outcome + u2i
π3i = β30 + u3i
with π0i, β00, β10, β20, β30 as intercepts; π1i, π2i, β01, β11, β21 as
slopes; and εij, u0i, u1i, u2i, u3i as residuals. Targetij is the
response variable of individual i measured for the trait j.
Expected outcome was coded −1 for low probability and +1
for high probability. β01 refers to the extent to which the target
is judged differently as a function of the expected outcome.
β10 denotes the extent to which the self is used to judge the
target. The critical parameter for our hypothesis is β11, because
it denotes the extent to which the tendency of the self-ratings
to predict the target ratings (i.e., π1i) varies as a function of
expected outcome. β20 refers to the extent to which the
valence ratings affect target ratings. β21 is the parameter that
allows us to control for the impact of valence on the
interaction between self and expected outcome. This is
because the interaction between self-ratings and expected
outcome is not adjusted for valence simply by controlling for
trait valence. As a matter of fact, one may also need to control
for the interaction between the moderator (expected out-
comes) and the variable that covaries with trait valence (self;
see Yzerbyt, Muller, & Judd, 2004). Self-ratings were cen-
tered at the mean of each participant’s ratings. It is important
to note that parameters in this model (especially β10 and β11)
reflect the importance of self–target covariance rather than
correspondence. The method of estimation is restricted maxi-
mum likelihood, and the covariance matrix is unstructured.
These specifications also apply to the subsequent models.
Confirming the presence of projection, participants pro-
jected the self onto the target (B10 = .12, SE = .05, t = 2.35,
p < .05). As predicted, the relationship between self and target
ratings (self–target projection) depended on the expected
probability of success, B11 = .11, SE = .05, t = 2.28, p < .05,
after controlling for the main effect of valence as well as for
the Self × Valence and Condition × Valence interactions.
Importantly, the relationship between self-ratings and target
ratings was positive and significant when the probability of
success was high, B11 = .23, SE = .07, t = 3.33, p < .01, and
nonsignificant when the probability of success was low,
B11 = .001, SE = .07, t < 1. In other words, participants pro-
jected their personality traits onto the target when they had a
high chance of succeeding in cooperation but not when they
had a low chance of succeeding (see Figure 1). Valence did
not affect projection, B20 = −.001, SE = .01, t < 1. This study
shows that people used their similarity belief to judge their
partner as similar to themselves only when they expect to suc-
ceed in cooperation. When the chances of success of coopera-
tion seem slim, people do not infer any similarity between
themselves and their potential partner.
Study 4
The aim of Study 4 was to replicate the findings of Study 3,
this time using an actual cooperative task. In this study,
participants faced a real partner (a confederate) with whom
they expected to succeed or to fail in a cooperative learning
task. If participants project their characteristics more in a
complex cooperative task requiring coordination with a real
partner when the likelihood of success is high rather than
low, then this finding would rule out the possibility that the
previous results were solely due to the absence of a concrete
partner in the task.
Method
Participants and design. A total of 41 students in psychol-
ogy (34 females) took part in a study investigating the effect
of cooperation on performance. They ranged in age from 18
to 23 years (M = 19.79, SD = 1.34). Participants were ran-
domly assigned to one of two experimental conditions (low
vs. high probability of success).
at Univ Catholique Louvain Bib on September 19, 2012psp.sagepub.comDownloaded from

Page 8

Toma et al.
1265
Procedure. The cooperative learning procedure was
adapted from Buchs and Butera (2009). Before the experi-
mental session, participants registered on a participation list
that required two persons per slot. Participants thus expected
to meet another person for the cooperative task. In reality,
one of the two registered participants was always the confed-
erate. The confederate’s name was written by the experi-
menter, who used a different name each time.
Each participant arrived at the laboratory at the same
time as the confederate. The confederate was always a
25-year-old female student dressed the same way and with
the same hairstyle throughout the entire experiment. First,
participants were invited to take place in different rooms
where they were asked to evaluate themselves on the same
list of traits as in Study 3. Then, the naive participant joined
the confederate in the next room where they were sitting
face to face for about 5 min. During this time, the experi-
menter explained the cooperative task in detail. The two
partners were informed that they would work together on a
cooperative learning task in which they needed to try their
best to promote their own learning and that of their partner.
The two partners received two social psychology texts: one
text on conformity for the naive participant and one text
on cognitive dissonance for the confederate. They were
informed that each of them would read his or her own text
for 10 min and then he or she would summarize the content
to their partner for 8 min. The partners were told that the
efficiency of the cooperative learning will be evaluated
using a multiple-choice test, including questions from the
two texts studied during the learning session. They were also
informed that they would have to complete the multiple-
choice test together and that only the joint performance
would be taken into account. Depending on the condition,
participants were told that this test was very easy/difficult
and that 95% (vs. 5%) of the previous teams succeeded/
failed in answering all the questions correctly.
Next, participants were again invited to sit in different
rooms and were asked to rate the partner on the same list of
traits that they had used to rate the self, and they also rated
trait valence. Finally, as a manipulation check, participants
were asked whether their chance of succeeding on the test
was low, from 1 (not at all) to 9 (very much). When all tasks
had been completed, participants were debriefed, thanked,
and dismissed.
Results. Two participants were excluded from the analyses
due to incomplete responses for self-ratings.
Manipulation check. Participants in the low probability of
success condition more readily agreed that their chance of suc-
ceeding in cooperation would be low (M = 4.21, SD = 1.84)
than participants in the high probability of success condition
(M = 3.00, SD = 1.52), t(37) = 2.24, p < .05. This result con-
firms the success of our manipulation.
Projection. As in Study 3, we predicted that the relation-
ship between the self-ratings and the target ratings would
be influenced by the probability of success, and that this
effect would emerge regardless of trait valence. As pre-
dicted, the relationship between self-ratings and target rat-
ings (self–target projection) depended on the probability of
success, B11 = .12, SE = .05, t = 2.29, p < .05, after control-
ling for the main effect of valence as well as for the Self ×
Valence and Condition × Valence interactions. Importantly,
the relationship between self-ratings and target ratings was
significantly positive in the high probability of success
condition, B11 = .15, SE = .07, t = 2.06, p < .05, and nonsig-
nificantly negative in the low probability of success condi-
tion, B11 = −.08, SE = .07, t = −1.15, p = .25. Overall, the
self was not projected onto the target, B10 = .03, SE = .05,
t < 1. Also, valence only marginally affected the projection
of self-ratings, B20 = −.02, SE = .01, t = −1.89, p = .06, with
participants tending to project more on negative than on
positive traits.
Discussion. Studies 3 and 4 strongly suggest that the egocen-
tric similarity belief may have led people to project their
self-view onto their partner. Participants projected their
characteristics onto the partner only when they expected
cooperation to be successful. This effect is even clearer in
Study 4 where participants were facing a real partner and
were expecting to collaborate on a real cooperative learning
task. Study 4 minimizes the possibility that participants in
Study 3 projected for the sole reason that they received no
concrete information about their partner and were thus con-
strained by the artificial nature of the method.
Another important message emerging from these results
is that, although some studies observed projection mainly on
positive traits (Stathi & Crisp, 2008), we found projection to
occur independently of the valence of the traits.
Figure 1. Relationship between self-ratings and target ratings by
expected outcome, controlling for trait valence (Study 3)
at Univ Catholique Louvain Bib on September 19, 2012 psp.sagepub.comDownloaded from

Page 9

1266
Personality and Social Psychology Bulletin 38(10)
Study 5
Studies 3 and 4 provide consistent evidence that the anticipa-
tion of success is necessary for social projection to be
observed in cooperation. However, given that people gener-
ally infer that cooperation is likely to meet with success, one
might argue that people always project when anticipating
cooperation. Therefore, the aim of Study 5 is to show that the
success of cooperation is not a sufficient condition for social
projection to occur in cooperation. People may well use their
similarity belief as an antecedent for projection but only
when they expect to succeed and are led to believe that it is
their own characteristics that engender this success.
Study 1b showed that people believe that the similarity of
the partner to the self, more than the similarity of the self to
the partner, leads to positive outcomes in cooperation. This
suggests the possibility that people are prone to believe that
their personal characteristics are conducive to success, and so
they generally want their personal characteristics to be shared
by their partner. If this is the case, then social projection may
be disrupted when it is made clear that the characteristics of
the partner rather than those of the self led to success.
Study 5 addresses this possibility by keeping constant the
anticipated success across conditions. However, this time,
we manipulated whether participants expected that success
in the task was due to both the self and the partner, only to
the self, or only to the partner. Our prediction was that social
projection should occur when participants are led to believe
that their characteristics are conductive to success but
should be disrupted when participants are led to believe that
their partner’s characteristics are conductive to success.
A second aim of this study was to examine whether the
latter projective effects are observed for attributes that are
irrelevant to cooperation. We know from Study 2b that peo-
ple readily consider the similarity of general preferences to
be a good predictor of people’s success in cooperation. It
would be most important if we could show that people proj-
ect general preferences that are irrelevant to success in the
task onto their partner, at least when they believe it is their
characteristics that are responsible for the success of
cooperation.
Method
Participants and design. A total of 69 participants (51
females), all students in various disciplines, took part in a
study of spontaneous impression formation. They ranged in
age from 18 to 37 years (M = 21.49, SD = 3.07). Participants
were randomly assigned to one of three experimental condi-
tions (success both, success self, success partner).
Procedure. On the first page of the questionnaire, partici-
pants rated themselves on a list of 20 activity preferences.
Participants had to indicate the extent to which they liked
each of the 20 activities (e.g., reading, hiking, going to the
cinema, cooking) on a 9-point rating scale ranging from 1
(not at all) to 9 (very much).
On the second page, participants read a scenario in which
they had to imagine cooperating with another student (the
target). They learned that they were about to finish a final
year project with a target student. Participants were told that
obtaining a very good grade for this project was of utmost
importance because they would like to pursue an MA pro-
gram. Therefore, both they and their partner combined their
efforts, shared their entire information, and spent an equal
amount of time preparing this project. Participants were
then informed that the final project was highly likely to be
successful and were assigned to one of the three conditions
(success both, success self, or success partner). In the
success-both condition, participants were told that the pro-
fessor in charge of the evaluation of the project praised their
qualities, and he or she publicly affirmed that both partners’
qualities were expected to contribute to the success of coop-
eration. In the success-self condition, participants were told
that the professor in charge of the evaluation of the project
only praised the participant’s own qualities, and he or she
publicly affirmed that the participant’s qualities, more than
those of their partner, were expected to contribute to the
success of cooperation. In the success-partner condition,
participants were told that the professor in charge of the
evaluation of the project only praised their partner’s quali-
ties, and he or she publicly affirmed that their partner’s
qualities, more than their own qualities, were expected to
contribute to the success of cooperation. Participants were
then asked to imagine the partner with whom they would
cooperate in this situation.
Next, participants were asked to rate the target on the
same list of activity preferences that they had used to rate
the self. Finally, as a manipulation check, participants were
asked whether the success in the situation described was
due to both partners, only to the self, or only to their part-
ner. All participants answered correctly. When all tasks had
been completed, participants were debriefed, thanked, and
dismissed.
Results. Our hypothesis held that the relationship between
the self-ratings and the target ratings should be influenced
by our Level 2 variable (attribution of success). The attribu-
tion of success variable was represented by two orthogonal
contrasts (C1 and C2). A contrast analysis is particularly
recommended here because it allows a powerful and unam-
biguous test of specific hypotheses (see Rosenthal, Rosnow,
& Rubin, 2000). In the present context, the C1 contrast (0,
+1, −1) compared the condition in which the self versus
only the partner was expected to be responsible for the suc-
cess, whereas the C2 contrast opposed the success-both con-
dition to the other two conditions (2, −1, −1). The order of
conditions for these two contrasts was success both, success
self, and success partner. As we used C1 and C2 to describe
the attribution of success variable, and because activity
preferences were not characterized by valence (as in previ-
ous studies), we estimated the following multilevel model.
at Univ Catholique Louvain Bib on September 19, 2012 psp.sagepub.com Downloaded from

Page 10

Toma et al.
1267
Level 1 model:
Targetij = π0i + π1i Self + εij
Level 2 model:
π0i = β00 + β01 C1 + β02 C2 + u0i
π1i = β10 + β11 C1 + β12 C2 + u1i
The parameters critical to our hypothesis are β11 and β12.
β11 tests whether there are differences in the projection of
self onto the target between the success-self condition and
the success-partner condition. β12 tests the differences in the
projection of self onto the target between the success both
and the other two conditions. This analysis revealed that self
influenced target ratings overall, B10 = .24, SE = .02, t = 11.97,
p < .001. Moreover, the interaction between self and the C1
contrast was significant suggesting that projection of the self
onto the target was different in the success-self condition and
in the success-partner condition, B11 = .05, SE = .02, t = 2.16,
p < .05. The interaction between self and the C2 contrast was
also significant, B12 = .06, SE = .01, t = 4.92, p < .001, sug-
gesting that the projection of the self onto the target was dif-
ferent in the success-both condition and in the other two
conditions.1
To probe the interaction between self-ratings and suc-
cess attribution on target ratings, we tested the simple
slopes representing self–target projection in each success
attribution condition. The critical slope refers to self–target
projection in the success-self and the success-partner con-
ditions. If projection occurred only when the success was
due to the self, but not when the success was due to the
partner, this would imply that participants projected only
when they thought that their characteristics were conduc-
tive to success. The results revealed that the relationship
between self-ratings and target ratings was positive and
significant when the success was expected to be due to both
partners, B = .43, SE = .07, t = 6.42, p < .001, positive and
significant when the success was expected to be due only to
the self, B = .22, SE = .06, t = 3.28, p < .01, and nonsignifi-
cant when the success was attributed only to the partner,
B = .10, SE = .07, t = 1.45, p = .16. These results are pre-
sented in Figure 2.
Discussion. Together, the results of Study 5 suggest that even
when cooperation leads to a positive outcome, projection of
the self onto the partner is disrupted if the partner is held
responsible for the success. In this situation, people no lon-
ger have reasons to believe that the similarity of the other to
the self is conductive to success in cooperation, and so they
stop projecting onto their partner. In our study, participants
only projected onto their partner when they found them-
selves in a situation reinforcing the belief that their personal
characteristics were conducive to success.
These results are provocative to the extent that partici-
pants projected on preference activities, namely, on charac-
teristics irrelevant for cooperation. This pattern is consistent,
however, with research showing that regardless of the type
of attributes under consideration, people may come to
believe that these attributes can help them to achieve desired
outcomes (Kunda, 1987).
What is also noteworthy about this study is that ratings of
the self were more strongly associated with ratings of the
target when the success was expected to be due to both part-
ners than when the success was expected to be due only to
the self, suggesting that the common fate of success contrib-
uted to projection. Although this is consistent with research
showing that the initial shared similarity enhances projection
(Ames, 2004), shared similarity cannot entirely account for
the effects found here.
General Discussion
A substantial amount of work suggests that similarity
between partners entails a number of positive consequences
for cooperation. The present research aimed to provide evi-
dence that beyond this objective reality, people hold simi-
larity beliefs that strongly influence their egocentric
judgments about potential partners. Past work on social
projection has shown that people tend to project their own
characteristics onto others, especially in situations of coop-
eration (Riketta & Sacramento, 2008; Toma et al., 2010).
However, until now, it remained unclear why people do so.
In the present research, we propose that people judge tar-
get persons to be similar to them in cooperation because
they believe that similarity to the self leads to positive out-
comes in this situation. Our results supported this idea. Two
studies provided direct (Studies 1a and 1b) and indirect
(Studies 2a and 2b) evidence that people hold an egocentric
Figure 2. Relationship between self-ratings and target ratings by
attribution of success (Study 5)
at Univ Catholique Louvain Bib on September 19, 2012psp.sagepub.com Downloaded from

Page 11

1268
Personality and Social Psychology Bulletin 38(10)
belief according to which similarity to the self leads to suc-
cess in cooperation. These studies also revealed that partici-
pants believe that both similarity in terms of personality
traits (Study 2a) or general preferences (Study 2b) is benefi-
cial in cooperation.
The next three studies (Studies 3, 4, and 5) investigated the
impact of the similarity belief on social projection. Study 3
manipulated the applicability of this belief and provided evi-
dence that participants who were convinced that the coopera-
tion was going to lead to success projected their self-view
onto their partner, whereas participants who doubted the suc-
cess of cooperation stopped projecting. Importantly, Study 4
showed that this effect also occurs in a real cooperative set-
ting in which participants actually faced a confederate part-
ner. Finally, Study 5 showed that the anticipated success of
cooperation is a necessary but not a sufficient condition for
social projection. People use their similarity belief to project
onto others only when they expect to possess those character-
istics that are conductive to the success of cooperation.
Egocentric Beliefs in Perceived Similarity
These findings may have important implications for the
domain of egocentric inferences. Past research has shown
that people define social traits, categories, and abilities in ego-
centric ways, emphasizing those characteristics that put the
self in a favorable light (Dunning & Cohen, 1992; Dunning,
Leuenberger, & Sherman, 1995). People have egocentric mod-
els and beliefs about social domains because they create them
in line with their own image and interests (Kunda, 1987).
The present research suggests that such egocentric infer-
ences also occur when people think that similarity between
partners facilitates cooperation. People seem to know that
similarity is an asset for cooperation, but they approach this
similarity in a rather partisan way, namely, by inferring that
their partner should resemble them rather than they should
resemble their partner. What could explain this effect? One
reason could simply be that the self is an easily available
example and thus a privileged reference point in judgments
of similarity. In line with this idea, research done by Holyoak
and Gordon (1983) showed that people tend to see others as
being more similar to the self than vice versa.
Interestingly, in our studies, participants do not simply
perceive asymmetric similarity, but they also associate simi-
larity to the self with more positive outcomes than similarity
to the partner. Therefore, another reason could be related to
people’s belief about their own abilities. It is widely known
that people posses inflated views of their attributes and abili-
ties (Alicke, 1985). At the same time, no matter what people’s
attributes are, they all believe that their attributes are more
likely than another’s attributes to facilitate the desired out-
come (Kunda, 1987). Given these circumstances, people’s
egocentric judgment of similarity in cooperation may be a
rational consequence of people’s belief about their own abili-
ties. Although Study 5 suggests this possibility, future studies
should directly investigate whether people think that their
attributes are more important and are more likely to facilitate
effective cooperation than those of others
Social Projection, a Motivated Process?
The present empirical evidence may also have important
implications for research on social projection. So far, research
suggested that a focus on similarity is the default process in
cooperation. That is, cooperation has been found to activate
an integration mind-set in which similarities between the self
and others are emphasized (Carnevale & Probst, 1998), with
a resulting assimilation of the self into the other and also of
the other into the self. According to the egocentric compari-
son model of social prediction (Mussweiler, 2003), when
people find themselves in a context of cooperation, they may
engage in a process of similarity testing, yielding target judg-
ments that are consistent with the self.
The present results may suggest that social projection is a
motivated process. Because our participants projected onto
their partner only when they expected their cooperation to
lead to success, this may imply that the participants’ goal to
succeed was responsible for the projection. In his review of
social projection, Krueger (2007) stated that although “pro-
jection can emerge with minimal cognitive contribution . . .
projection can also be engaged and suspended strategically”
(p. 2). Therefore, it is reasonable to conclude that people in
our studies appraised their partner in a strategic way by rely-
ing on their own characteristics when it was worthwhile to
do so (see also Toma, Yzerbyt, & Corneille, 2012).
The present research is consistent with prior work suggest-
ing that projection and perceived similarity may be goal
dependent. Participants activated with a mate-search goal per-
ceived more sexual arousal in attractive opposite-sex targets
than did participants not activated with this goal (Maner et al.,
2005). In romantic relationships, the desire to be close to the
partner has been found to increase projection and perceived
self–other similarity (Slotter & Gardner, 2009). Perceived
similarity more than actual similarity has been found to be
functional in that it predicts such positive outcomes as high
self-esteem and relationship satisfaction (Murray, Holmes,
Dolderman, & Griffin, 2000). People infer partner similarity
because they believe this may contribute to their own growth
and to the acquisition of new skills (Rusbult, Kumashiro,
Kubacka, & Finkel, 2009). Research on social dilemmas also
suggests that people’s projection of their choices onto others
(whether cooperative or competitive) may have a motivational
source: “Subjects may feel the need to justify their decisions—
defectors in order to assuage possible guilt over their decision,
cooperators to avoid feeling duped” (Dawes, McTavish, &
Shaklee, 1977, p. 8). Future studies should directly manipulate
participants’ need to justify their qualities for cooperation and
participants’ motivation to succeed in this situation, and to test
whether these motivations would increase projection of the
self onto the partner.
at Univ Catholique Louvain Bib on September 19, 2012psp.sagepub.com Downloaded from

Page 12

Toma et al.
1269
Implications of Social Projection
in Cooperation
One reason to care about social projection in cooperation is
that, when people expect others to be similar to them, this
not only fosters prosocial action but, ironically enough,
may also entail distortions, unwanted outcomes, and con-
flict. To be sure, the bulk of the literature focuses on the
benefits of similarity in cooperation. Perceived similarity
increases liking and attraction (Byrne, 1971), and is related
to compassion (Oveis, Horberg, & Keltner, 2010) and
altruistic behavior (Cunningham, 1986). Perceived self–
other similarity induces willingness to act at a cost to the
self (Krebs, 1975) and to forgo personal rewards (Batson,
Turk, Shaw, & Klein, 1995). Still, our findings also point
to the possibility that misguided perceptions of similarity
may sometimes be harmful in cooperation. That is, people
who overestimate the similarity between themselves and
their partner may fall prey to some specific mode of solv-
ing problems or interacting with others that may prove less
beneficial than if some room for divergence had been
acknowledged.
For example, in decision-making settings, people who
perceive others as being similar to them may limit their
search for information because they wrongly assume that
others possess the same information as they do. As for
problem-solving situations, people who readily focus on
similarities may fail in cooperative tasks that require com-
plementary points of view. In line with these intuitions,
Yzerbyt, Woltin, Corneille, and Bourgeois (2011) recently
argued that a focus on similarities should not always help
reaching successful negotiation outcomes. In two studies,
these authors primed their participants with a dissimilarity
mind-set rather than a similarity mind-set as a means of
manipulating egocentric projective tendencies and found
that dissimilarity rather than similarity was more effective in
allowing participants to reach a mutual agreement.
Conclusion
When people anticipate cooperation, they tend to project
their own characteristics onto their partner. Previous research
considered that a focus on similarities is the default process
in cooperative settings (Carnevale & Probst, 1998;
Mussweiler, 2003). Our findings suggest that the egocentric
similarity belief may well be the key factor in explaining
projective tendencies in this situation. Because it remains
unclear how well this conjecture of similarity fares in the
context of real interactions, future work would do well to
examine these issues in more detail. To be sure, self-
fulfilling prophecies may enter the picture and lead to a
happy conclusion for the interaction. At the same time, one
should not expect projection to be associated only with
positive dividends. Clarifying the negative consequences of
overestimating similarities definitely stands as an important
item on our research agenda.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with
respect to the research, authorship, and/or publication of this
article.
Funding
The author(s) disclosed receipt of the following financial support
for the research, authorship, and/or publication of this article: The
research was supported by Grant ARC06/11-337 from the
Communauté Française de Belgique and Fonds de la Recherche
Scientifique (FNRS).
Note
1. We also tested the interactions between the self and the linear
contrast C1 = (+1, 0, −1), and between the self and the qua-
dratic contrast C2 = (+1, −2, +1). The Self × C1 interaction was
significant, B = .13, SE = .02, t = 5.75, p < .001, while the Self
× C2 interaction was not, B = .01, SE = .01, t < 1.
References
Alicke, M. D. (1985). Global self-evaluation as determined by the
desirability and controllability of trait adjectives. Journal of
Personality and Social Psychology, 49, 1621-1630.
Ames, D. (2004). Strategies for social inference: A similarity con-
tingency model of projection and stereotyping in attitude preva-
lence estimates. Journal of Personality and Social Psychology,
87, 573-585.
Anderson, C. A., & Lindsay, J. J. (1998). The development, persever-
ance and change of naive theories. Social Cognition, 16, 8-30.
Batson, C. D., Turk, C. L., Shaw, L. L., & Klein, T. R. (1995). Infor-
mation function of empathic emotion: Learning that we value
the other’s welfare. Journal of Personality and Social Psychol-
ogy, 68, 300-313.
Beauregard, K. S., & Dunning, D. (1998). Turning up the contrast:
Self-enhancement motives prompt egocentric contrast effects
in social judgments. Journal of Personality and Social Psychol-
ogy, 74, 606-621.
Bretz, R. D., Jr., & Judge, T. A. (1994). Person-organization fit and the
theory of work adjustment: Implications for satisfaction, tenure,
and career success. Journal of Vocational Behavior, 44, 32-54.
Buchs, C., & Butera, F. (2009). Is a partner’s competence threat-
ening during dyadic cooperative work? It depends on resource
interdependence. European Journal of Psychology of Educa-
tion, 24, 145-154.
Byrne, D. (1971). The attraction paradigm. New York, NY: Aca-
demic Press.
Carnevale, P. J., & Probst, T. M. (1998). Social values and social
conflict in creative problem solving and categorization. Journal
of Personality and Social Psychology, 74, 1300-1309.
Cunningham, M. R. (1986). Levites and brother’s keepers: A socio-
biological perspective on prosocial behavior. Humboldt Jour-
nal of Social Relations, 13, 35-67.
at Univ Catholique Louvain Bib on September 19, 2012 psp.sagepub.comDownloaded from

Page 13

1270
Personality and Social Psychology Bulletin 38(10)
Dawes, R., McTavish, J., & Shaklee, H. (1977). Behavior, commu-
nication, and assumptions about other people’s behavior in a
commons dilemma situation. Journal of Personality and Social
Psychology, 35, 1-11.
Dunning, D., & Cohen, G. L. (1992). Egocentric definitions of
traits and abilities in social judgment. Journal of Personality
and Social Psychology, 63, 341-355.
Dunning, D., & Hayes, A. F. (1996). Evidence for egocentric com-
parison in social judgment. Journal of Personality and Social
Psychology, 71, 213-229.
Dunning, D., Leuenberger, A., & Sherman, D. A. (1995). A new
look at motivated inference: Are self-serving theories of suc-
cess a product of motivational forces? Journal of Personality
and Social Psychology, 59, 58-68.
Dweck, C. S., Chiu, C., & Hong, Y. (1995). Implicit theories and
their role in judgments and reactions: A world from two per-
spectives. Psychological Inquiry, 6, 267-285.
Fischer., I. (2009). Friend or foe: Subjective expected relative simi-
larity as a determinant of cooperation. Journal of Experimental
Psychology, General, 138, 341-350.
Gaertner, S. L., & Dovidio, J. F. (1977). The subtlety of white rac-
ism, arousal and helping behavior. Journal of Personality and
Social Psychology, 35, 691-707.
Heider, F. (1958). The psychology of interpersonal relations.
New York, NY: Wiley.
Heschl, A. (1993). Physiognomic similarity and political coopera-
tion: An exploratory investigation. Politics and the Life Sci-
ences, 12, 61-68.
Holyoak, K. J., & Gordon, P. C. (1983). Social reference points.
Journal of Personality and Social Psychology, 44, 881-887.
Jamieson, D., Lydon, J., & Zanna, M. (1987). Attitude and activity
preference similarity: Differential bases of interpersonal attrac-
tion for low and high self-monitors. Journal of Personality and
Social Psychology, 53, 1052-1060.
Katz, D., & Allport, F. (1931). Students’ attitudes. Syracuse, NY:
Craftsman.
Krebs, D. L. (1975). Empathy and altruism. Journal of Personality
and Social Psychology, 32, 1134-1146.
Krueger, J. (1998). On the perception of social consensus. Advances
in Experimental Social Psychology, 30, 163-240.
Krueger, J. I. (2007). From social projection to social behavior.
European Review of Social Psychology, 18, 1-35.
Kunda, Z. (1987). Motivated inference: Self-serving generation and
evaluation of causal theories. Journal of Personality and Social
Psychology, 53, 636-647.
Lakin, J., & Chartrand, T. L. (2003). Using nonconscious behav-
ioral mimicry to create affiliation and rapport. Psychological
Science, 14, 334-339.
Maner, J. K., Kenrick, D. T., Neuberg, S. L., Becker, D. V., Robertson, T.,
Hofer, B., . . . Schaller, M. (2005). Functional projection: How
fundamental social motives can bias interpersonal perception.
Journal of Personality and Social Psychology, 88, 63-78.
Mark, M. M., Mutrie, N., Brooks, D. R., & Harris, D. V. (1984).
Causal attributions of winners and losers in individual competi-
tive sports: Towards a reformulation of the self-serving bias.
Journal of Sport Psychology, 6, 184-196.
Messick, D. M., Bloom, S., Boldizar, J. P., & Samuelson, C. D.
(1985). Why we are fairer than others. Journal of Experimental
Social Psychology, 21, 480-500.
Miller, D. T., & Ross, M. (1975). Self-serving biases in the attribution
of causality: Fact or fiction? Psychological Bulletin, 82, 213-225.
Murray, S. L., Holmes, J. G., Dolderman, D., & Griffin, D. W.
(2000). What the motivated mind sees: Comparing friends’ per-
spectives to married partners’ views of each other. Journal of
Experimental Social Psychology, 36, 600-620.
Mussweiler, T. (2003). When egocentrism breeds distinctness-
comparison processes in social prediction: Comment on Kar
niol (2003). Psychological Review, 110, 581-584.
Nowak, M. A. (2006). Five rules for the evolution of cooperation.
Science, 314, 1560-1563.
Oveis, C., Horberg, E. J., & Keltner, D. (2010). Compassion, pride,
and social intuitions of self-other similarity. Journal of Person-
ality and Social Psychology, 98, 618-630.
Riketta, M., & Sacramento, C. (2008). “They cooperate with us, so
they are like me”: Perceived intergroup relationship moderates
projection from self to outgroups. Group Processes and Inter-
group Relations, 11, 115-131.
Riolo, R. L., Cohen, M. D., & Axelrod, R. (2001). Evolution of
cooperation without reciprocity. Nature, 414, 441-443.
Rosenthal, R., Rosnow, R. L., & Rubin, D. B. (2000). Contrasts and
effect sizes in behavioral research: A correlational approach.
Cambridge, UK: Cambridge University Press.
Ross, L., Greene, D., & House, P. (1977). The false consensus
phenomenon: An attributional bias in self-perception and
social-perception processes. Journal of Experimental Social
Psychology, 13, 279-301.
Ross, L., & Nisbett, R. E. (1991). The person and the situation:
Perspectives of social psychology. New York, NY: McGraw-
Hill.
Ross, M., & Sicoly, F. (1979). Egocentric biases in availability and
attribution. Journal of Personality and Social Psychology, 37,
322-336.
Rusbult, C. E., Kumashiro, M., Kubacka, K. E., & Finkel, E. J.
(2009). “The part of me that you bring out”: Ideal similarity
and the Michelangelo phenomenon. Journal of Personality and
Social Psychology, 96, 61-82.
Savitsky, K. (2007). Egocentric bias. In R. F. Baumeister &
K. D. Vohs (Eds.), Encyclopedia of social psychology. Thou-
sand Oaks, CA: SAGE.
Sinha, R. R., & Krueger, J. (1998). Idiographic self-evaluation and
bias. Journal of Research in Personality, 32, 131-155.
Slotter, E. B., & Gardner, W. L. (2009). Where do “You” end and
“I” begin? Pre-emptive self-other inclusion as a motivated
process. Journal of Personality and Social Psychology, 96,
1137-1151.
Spencer, S. J., Zanna, M. P., & Fong, G. T. (2005). Establishing a
causal chain: Why experiments are often more effective than
mediational analyses in examining psychological processes.
Journal of Personality and Social Psychology, 89, 845-851.
Stathi, S., & Crisp, R. J. (2008). Imagining intergroup contact pro-
motes projection to outgroups. Journal of Experimental Social
Psychology, 44, 943-957.
at Univ Catholique Louvain Bib on September 19, 2012psp.sagepub.com Downloaded from

Page 14

Toma et al.
1271
Toma, C., Yzerbyt, V., & Corneille, O. (2010). Anticipated coopera-
tion vs. competition moderates interpersonal projection. Jour-
nal of Experimental Social Psychology, 46, 375-381.
Toma, C., Yzerbyt, V., & Corneille, O. (2012). Nice or smart? Task
relevance of self-characteristics moderates interpersonal projec-
tion. Journal of Experimental Social Psychology, 48, 335-340.
Traulsen, A. (2008). Mechanisms for similarity based cooperation.
European Physical Journal B, 63, 363-371.
Urban, M. S., & Witt, L. A. (1990). Self-serving bias in group mem-
ber attributions of success and failure. Journal of Social Psy-
chology, 130, 417-418.
Van Vugt, M., & Hart, C. M. (2003). Social identity as social glue:
Fostering loyalty and stability in groups. Unpublished manu-
script, University of Southampton.
Wegener, D. T., & Petty, R. E. (1998). The naive scientist revisited:
Naive theories and social judgment. Social Cognition, 16, 1-7.
White, J. A., & Plous, S. (1995). Self-enhancement and social
responsibility: On caring more, but doing less, than others.
Journal of Applied Social Psychology, 25, 1297-1318.
Yan, W., & Gaier, E. L. (1994). Causal attributions for college suc-
cess and failure: An Asian-American comparison. Journal of
Cross-Cultural Psychology, 25, 146-158.
Yzerbyt, V., Woltin, K.-A., Corneille, O., & Bourgeois, S.
(2011). Making deals out of differences: A dissimilarity
mindset facilitates joint negotiation outcomes. Unpublished
Manuscript.
Yzerbyt, V. Y., Muller, D., & Judd, C. M. (2004). Adjusting
researchers’ approach to adjustment: On the use of covariates
when testing interactions. Journal of Experimental Social Psy-
chology, 40, 423-431.
Zak, P. J., & Knack, S. (2001). Trust and growth. Economic Jour-
nal, 111, 295-321.
at Univ Catholique Louvain Bib on September 19, 2012psp.sagepub.comDownloaded from