- Email: [email protected]
Egocentrism and vicarious dissonance
Egocentrism and Vicarious Dissonance Shane F. Blackman, Kyle T. Keller, Joel Cooper PII: DOI: Reference:
S0022-1031(15)00103-1 doi: 10.1016/j.jesp.2015.09.001 YJESP 3350
To appear in:
Journal of Experimental Social Psychology
Received date: Revised date: Accepted date:
12 December 2014 31 August 2015 2 September 2015
Please cite this article as: Blackman, S.F., Keller, K.T. & Cooper, J., Egocentrism and Vicarious Dissonance, Journal of Experimental Social Psychology (2015), doi: 10.1016/j.jesp.2015.09.001
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1
SC
RI
PT
Running head: EGOCENTRISM IN VICARIOUS DISSONANCE
NU
Egocentrism and Vicarious Dissonance Shane F. Blackman, Kyle T. Keller, and Joel Cooper
Author Note
AC CE P
Word count: 5284
TE
D
MA
Princeton University
Shane F. Blackman, Department of Psychology, Princeton University Kyle T. Keller, Department of Psychology, Princeton University Joel Cooper, Department of Psychology, Princeton University Correspondences concerning this article should be addressed to Joel Cooper, Department of Psychology, Princeton University, Princeton, NJ 08544. Contact: [email protected]
ACCEPTED MANUSCRIPT 2
EGOCENTRISM IN VICARIOUS DISSONANCE Abstract
Past research on vicarious dissonance (Monin, Norton, Cooper, & Hogg, 2004; Norton,
PT
Monin, Cooper, & Hogg, 2003) has demonstrated that participants experience vicarious
RI
discomfort and change their attitudes when witnessing counterattitudinal behavior from a fellow ingroup member. It has not been shown, however, whether witnessing this counterattitudinal
SC
behavior arouses vicarious dissonance because it contradicts the attitudes of the observed
NU
ingroup member or for egocentric reasons – that is, because it contradicts the attitudes of the observer. In three experiments, we test the hypothesis that the conditions necessary for arousing
MA
vicarious dissonance—namely, observing an ingroup member‘s responsibility for aversive consequences and level of choice—are processed egocentrically by the observer. By
TE
D
manipulating perspective taking in a vicarious induced compliance task, and manipulating choice difficulty in a vicarious free-choice dissonance paradigm, we show that the arousal of vicarious
AC CE P
dissonance crucially depends on the consequences of the ingroup member‘s actions for the observer, and not for the observed ingroup member. Implications for other aspects of vicarious dissonance and vicarious experience in general are discussed. Keywords: attitudes, cognitive dissonance, egocentrism, perspective taking, vicarious dissonance
ACCEPTED MANUSCRIPT 3
EGOCENTRISM IN VICARIOUS DISSONANCE Egocentrism in Vicarious Dissonance In our daily lives, we are motivated to maintain consistency among our beliefs and
PT
behaviors (Festinger, 1957; Heider, 1958; Newcomb, 1957). We strive to maintain and preserve
RI
harmony between our own attitudes and those espoused by important groups and important others within those groups (Cooper & Mackie, 1983; Festinger, 1954; Glasford, Dovidio, &
SC
Pratto, 2009; Glasford, Pratto, & Dovidio, 2008; Hogg & Smith, 2007; Tajfel & Turner, 1986).
NU
As decades of research in the cognitive dissonance tradition have taught us (Cooper, 2007; Harmon-Jones & Mills, 1999), when we perceive contradiction, we feel pressure to resolve that
MA
inconsistency through justification or rationalization, attempting to bring our changeable attitudes in line with our immutable past behavior.
TE
D
Norton, Monin, Cooper, and Hogg (2003) expanded the reach of dissonance theory to examine the impact of witnessing other people behaving in ways that contradict their attitudes
AC CE P
(see also, Monin, Norton, Cooper, & Hogg, 2004). They suggested that people who observe dissonant behavior by another member of a valued ingroup experience dissonance vicariously, and are motivated to change their own attitudes. Tied as we are to an actor by the strong bonds that hold an ingroup together, they predicted that people would experience dissonance vicariously and change their attitudes as a consequence. Simulating a classic induced compliance procedure (Festinger & Carlsmith, 1959), Norton et al. (2003) had participants witness a fellow group member agree to write an essay advocating a position they actually opposed. The results of several experiments showed that participants who were strongly identified with their ingroup changed their attitudes in the direction of the ingroup member‘s advocacy simply by observing the confederate‘s agreement to write a counterattitudinal speech. Personal vs. Vicarious Dissonance
ACCEPTED MANUSCRIPT 4
EGOCENTRISM IN VICARIOUS DISSONANCE A crucial question left open by the set of studies on vicarious dissonance is the process that motivates people to change their attitude. Results have shown that vicarious dissonance is
PT
sensitive to many of the moderators that affect personal cognitive dissonance. For example, the
RI
ingroup member whom people witness acting in a counterattitudinal fashion must have had the freedom to decline the request, the behavior must be counterattitudinal for the writer, and the
SC
there must be a plausible unwanted consequence pursuant to the behavior. Prior research has also
NU
shown that there is discomfort associated with vicarious dissonance: people who change their attitudes in the high dissonance condition report a feeling of greater discomfort.
MA
The current research is designed to shed light on the basis of people‘s motivation to change their own attitudes after observing a group member‘s behavior. One plausible view is that
D
people take the perspective of someone to whom they feel close (e.g., a member of their social
TE
group) and experience the discrepant situation from that person‘s perspective. The tendency of
AC CE P
group members to fuse their identities with those of other members of their group (Hogg & Smith, 2007) gives credence to this possibility. Adopting my fellow group member‘s perspective, I may experience what she experiences and change my attitudes to reduce the discomfort.
On the other hand, research has shown that adopting the perspective of others – even those to whom we feel close -- is not an easy task (Epley, Keysar, van Boven & Gilovich, 2004; Savitsky, Keysar, Epley, Carter & Swanson, 2011). People seem to begin their attempt at perspective taking by adopting an egocentric point of view, often making insufficient corrections from there. This prompts us to consider the possibility that the process that leads to attitude change in a group situation may be fundamentally an egocentric one. It is possible that it is more about ‗me‘ (the observer) and less about taking the perspective of the actor. That is, it is more
ACCEPTED MANUSCRIPT 5
EGOCENTRISM IN VICARIOUS DISSONANCE about how the actor would be affected by the group member‘s actions. Instead of viewing observers as changing their attitudes because of their fused identity with the group member who
PT
is in the throes of dissonance, the egocentric view is that vicarious dissonance feels
RI
uncomfortable because it is the way I think I would feel were I in the shoes of the actor. Prior research suggests that the egocentric view of vicarious dissonance is a viable approach. In one of
SC
Norton et al. (2003) studies, participants were asked about their discomfort after witnessing the
NU
counterattitudinal behavior of their fellow group member. They found that the magnitude of attitude change was not related to how they thought their fellow group member felt, but rather
MA
how they thought they would feel if they were in their partner‘s shoes. Rather than sharing their partner‘s discomfort, the participants seemed motivated by how they thought they would feel if
TE
D
they were in the essay writer‘s situation.
The present research seeks to clarify the mechanism underlying vicarious dissonance by
AC CE P
assessing two possible processes. We differentiate other-oriented perspective taking from egocentric simulation. In the first view, people experience the dissonance situation from the perspective of the actor, feeling or imagining what the actor feels. Adopting the tension state of cognitive dissonance on behalf of the actor, the observer reduces dissonance in the same way as the actor. In the second view, ingroup members cognitively simulate the dissonance situation. They imagine what they would experience if they were in the actor‘s situation. In the egocentric view, observers make the actor‘s situation their own and respond accordingly. We first report the results of two studies using the previously validated vicarious induced compliance paradigm and directly manipulate participants‘ perspective. In the third experiment, we report the first use of a new paradigm, adding converging evidence to the notion that people take an egocentric approach when observing a fellow group member engage in a dissonant act.
ACCEPTED MANUSCRIPT 6
EGOCENTRISM IN VICARIOUS DISSONANCE Study 1 Study 1 is designed to systematically investigate the role of perspective taking in
PT
vicarious dissonance. Participants observed a member of their ingroup engage in behavior that
RI
contradicted their attitude. In order to facilitate the shared connection between the participant and group member, some of our participants were instructed to view the situation from the
SC
perspective of the group member. By contrast, in order to facilitate egocentric simulation, we
NU
asked other participants to adopt their own perspective, imagining how they would feel if they were in the group member‘s situation. Although both processes are viable, our prediction—based
MA
on the clue provided by Norton et al.‘s (2003) discomfort findings—is that vicarious dissonance is facilitated when taking one‘s own egocentric perspective and diminished when trying to take
Method
AC CE P
Participants
TE
D
the perspective of the ingroup member.
Participants were recruited online via Amazon‘s Mechanical Turk service. Because no vicarious dissonance experiment has been reported using an on-line procedure, we chose to recruit at least 20 participants per condition to provide sufficient power. Our strategy was to leave mTurk open for 48 hours with the constraint that a minimum of 120 workers participate. 156 people volunteered for the experiment. Of these, 144 completed the entire survey. Three were ultimately excluded for failing to follow all of the instructions, resulting in a sample of N=141. Participants ranged in age from 18 to 74 (M = 33.9). 59.9% of the sample were women. Procedure Participants gave their consent to participate in an experiment comparing the writing styles of people in different countries. They believed that they were participating at the same
ACCEPTED MANUSCRIPT 7
EGOCENTRISM IN VICARIOUS DISSONANCE time as another American MTurk Worker (i.e., a member of their ingroup)—making their ingroup identity as an American salient through highlighting metacontrast (Brewer & Gardner,
PT
1996; Tajfel & Turner, 1979)—and were told that they had been assigned to read an essay
RI
written by that Worker. They then saw instructions indicating that their partner was either told (low choice condition) or asked (high choice condition) to write an essay in favor of government
SC
influence in the media. Specifically, in the high choice condition, the partner was asked, ―Are
NU
you willing to write the essay?‖ and the participant was permitted to decline to write. In the low choice condition, there was no permission to decline. Participants were also informed that the
MA
other Worker did not personally agree with the position they wrote about. While their partner was ostensibly writing the essay, participants in the experimental conditions were given a
D
perspective-taking priming manipulation. In the egocentric condition, participants were told,
TE
―While your partner is writing his/her essay, we would like you to take this moment to write a
AC CE P
few sentences about what you imagine you would be feeling if you were writing this essay. What would be going through your head?‖ In the other-oriented perspective taking conditions, they were asked to imagine what their partner is feeling while writing the essay. In a control condition, no instructions about perspective taking were given and participants went directly to the dependent measures.
After the priming manipulation (or control), participants filled out the study‘s dependent measures. Attitudes towards government influence in the media were assessed by asking participants, ―To what extent are you in favor of government influence in the media?‖, assessed on a 15 point scale. We also collected data on participants‘ perceptions of their own and their partner‘s representativeness and typicality, their identification with the ingroup (Americans), their liking and perceived similarity to their partner and their level of personal and vicarious
ACCEPTED MANUSCRIPT 8
EGOCENTRISM IN VICARIOUS DISSONANCE discomfort. In addition, we collected measures of their perceptions of the attitude of the average American and participants‘ perceptions of the importance of the issue. Participants were led to
PT
believe that they would read their partner‘s essay after hearing some information about it in the
RI
instructions, but after filling out the study‘s dependent measures the participants were instead
Results
NU
Manipulation check
SC
thanked and debriefed.
Our manipulation check indicated that our manipulation of choice in the study was
MA
successful, as participants in the high-choice conditions (M = 10.38, SD = 4.88) had significantly higher perceptions of their alleged partner‘s level of choice, F(1, 135) = 16.02, p < .001, η2p =
Attitude measures
TE
D
.10, than those in the low-choice conditions (M = 6.71, SD = 5.45).
AC CE P
To see whether our manipulations of choice and perspective taking had an effect on participants‘ attitudes towards government influence in the media, we conducted a 2 x 3 factorial ANOVA on the data. Our analyses revealed a significant interaction of choice and perspective taking, F(2, 135) = 3.37, p = .04 p2 = .06. To explore this interaction, we conducted analyses to explore the simple effect of choice within each perspective taking condition. In line with our hypotheses, there was no effect of choice in the no perspective-taking condition, p = .38, nor in the other perspective-taking condition, p = .87. When participants were instructed to take their own ―vicarious‖ perspective, however, participants in the high-choice condition (M = 4.21, SD = 2.88) exhibited significantly more positive attitudes towards government influence in the media, F(1, 62) = 7.05, p = .01, η2p = .16, than did participants in the low-choice condition (M = 2.14, SD = 2.01; see Figure 1).
ACCEPTED MANUSCRIPT 9
EGOCENTRISM IN VICARIOUS DISSONANCE
Affect measures
PT
We assessed participants‘ self-reported discomfort and their vicariously experienced
RI
discomfort. For participants‘ own discomfort, a 2 x 3 factorial ANOVA showed that there was not a significant effect for either perspective taking, p = .87, or choice, p = .84, on participants‘
SC
levels of discomfort, nor was there a significant interaction, p = .87. Similarly, for participants‘
NU
reported vicarious discomfort, neither perspective taking, p = .37, nor choice, p = .36, nor was the interaction significant, p = .29.
MA
Discussion
This experiment sought to investigate the importance of perspective taking in the process
TE
D
of vicarious dissonance. As predicted, the pattern of results followed the previously reported vicarious dissonance findings (Norton et al., 2003) only when participant were given instructions
AC CE P
designed to enhance an egocentric simulation of the fellow group member‘s situation. When given instructions to vicariously imagine writing the essay rather than to empathize with the essay-writer, participants in the high-choice condition expressed more favorable attitudes towards government influence in the media than participants in the low-choice condition. These canonical dissonance effects of choice on attitude change were not present, however, when participants were told to take their partner‘s perspective or received no instructions at all. One aspect of our study that merits further attention is that we found no differences across conditions in participants’ reported self or vicarious discomfort. This is consistent with prior research in both personal (Elliot &Devine, 1994) and vicarious dissonance (Norton et al, 2003) that did not find differences in discomfort when discomfort was measured after people had the opportunity to change their attitudes. When Norton et al
ACCEPTED MANUSCRIPT 10
EGOCENTRISM IN VICARIOUS DISSONANCE
(2003) systematically manipulated the order of measurement in the vicarious dissonance paradigm, they found that when discomfort was measured first, people reported
PT
differences in their level of discomfort but did not show attitude change. On the other
RI
hand, when discomfort was measured following the attitude measure, there was attitude
SC
change but no reported discomfort. In the Norton et al study, as in the current study, it is likely that when participants report their attitudes towards the issue before reporting their
NU
discomfort, the discomfort is reduced by the change in attitude.
In addition, the current study was not designed to address one of our key theoretical
MA
assumptions – that strong identification with the ingroup is important for vicarious dissonance to occur. Based on prior research that showed that strong group identification is necessary for
TE
D
vicarious dissonance, we elected to run Study 1 in a way that maximized identification with the in-group for all participants. We used participants‘ nationality (American citizens) as the
AC CE P
ingroup, adopting Brewer & Gardner‘s (1996) method of metacontrasts to keep identification high. We assessed the degree of identification by asking participants how strongly they identified with being an American citizen. There was very little variance in response to the identification question, and we were not in a position to show that the magnitude of identification facilitated vicarious dissonance in the egocentric conditions. In order to provide evidence for this key assumption, we conducted a second study for the purpose of showing that egocentric vicarious dissonance depends on strong in-group identification.
ACCEPTED MANUSCRIPT 11
EGOCENTRISM IN VICARIOUS DISSONANCE
PT
Study 2
RI
METHOD Participants and Procedure
SC
We recruited 84 new mTurk workers for the second study who ranged in age from 19 to
NU
66 (M= 36.8). 46.5% were female. As in the first experiment, the study was introduced as a study on the writing styles of people from different countries and participants believed their
MA
partner was another American mTurk worker. We precisely repeated the instructions of the egocentric conditions of Study 1, with half of the participants believing their partner chose – and
media.
TE
D
the other half believing their partner was told – to write in favor of government influence in the
AC CE P
Attitudes were assessed with the same 15-point scale used in Study 1. Participants were then asked about their level of group identification. Instead of asking about their identification as Americans, we asked participants about how much they identified with the specific group, ―mTurk workers.‖ Following the dependent measures, participants were thanked and debriefed. Results and Discussion
Consistent with the assumption underlying vicarious cognitive dissonance, level of identification had a significant effect on reported attitudes in the high choice condition. When participants learned that their fellow mTurk worker agreed to make a speech advocating increased influence of the government in media, those who identified strongly with their group (i.e., were above the median) came to agree with that position (M = 4.62) more than those who were below the median on identification (M = 1.18) (F (1, 82) = 4.32, p <.02, p2= .019. The
ACCEPTED MANUSCRIPT 12
EGOCENTRISM IN VICARIOUS DISSONANCE
interaction between choice and level of identity was also significant (F(1, 82) = 9.93, p <.01, p2 = .108.
PT
Figure 2 depicts the attitude scores regressed on choice and level of identification
RI
(standardized). The choice x identification interaction was significant, b=.5472, t(82) = 3.21, p
SC
= .002. As predicted, higher levels of identification led to higher attitude scores for participants in the high choice condition (b= .51; t (41) = 3.77; p <.01). We had no a priori prediction about
NU
the effect of identification in the low choice conditions. The data show a slight reversal of the effect of identification on attitude change in the low choice condition, but that effect was not
MA
significant (p<.40).
Overall, the results of the first two studies offer strong support for the notion that
TE
D
vicarious dissonance is a function of egocentric identification. The first study showed that people change their attitudes in the direction advocated by a fellow group member when they
AC CE P
adopt an egocentric position, imagining how they would feel if they were in their partner‘s position. As predicted, attitudes became more positive to the partner‘s counterattitudinal statement if the partner chose, rather than was told, to write the position. Study 2 was designed to clarify an ambiguity in the first study. In that study, we had asked participants about their identification with their nationality and found very little variance in the degree of identification. In the second study, we asked about participants‘ identification with a more specific and less nationalistic identity, asking about their identification with the group ―mTurk workers.‖ We found that the effect for choice in the egocentric condition depended on people feeling highly identified with that in-group. Study 3
ACCEPTED MANUSCRIPT 13
EGOCENTRISM IN VICARIOUS DISSONANCE
In our third experiment, we approached the egocentric model of vicarious dissonance in a completely novel way. We adopted the classic free choice paradigm of dissonance (Brehm,
PT
1956) in which participants witnessed a fellow group member make a choice between two items.
RI
In a typical study that arouses personal cognitive dissonance (e.g., Kitayama, Snibbe, Markus, & Suzuki, 2004; Harmon-Jones, Harmon-Jones, Fearn, Sigelman, & Johnson, 2008), participants
SC
are asked to rate a number of objects and are then asked to make a choice between two of the
NU
items. Because choosing one object over another necessarily entails accepting the unwanted elements of the chosen object and losing the desirable elements of the unchosen alternative,
MA
those consequences create dissonance. The dissonance is resolved by spreading the attractiveness of the alternatives following the choice; the more difficult the choice (i.e., the closer the
TE
D
attractiveness prior to the choice), the greater the dissonance. The extant literature has not used free choice as a method for studying vicarious
AC CE P
dissonance. One purpose of the current study is to demonstrate that people, after observing an ingroup member make a difficult choice, also spread the choice alternatives. Second, the free choice instantiation of dissonance allows us converging methodology to examine the egocentric simulation vs. other-oriented perspective taking process that underlies the effect. In this study, people observed a fellow group member make a choice that was either easy or difficult for that group member. Orthogonal to that manipulation, we systematically manipulated whether the choice would have been difficult or easy for the observer. The question is: Whose choice difficulty drives the spreading of the alternatives? We hypothesized that participants‘ attitudes would respond egocentrically to the choice difficulty—by exhibiting spreading of the alternatives when the observed choice would have been difficult for them. By contrast, we expected that participants would be unaffected by the fellow group member‘s actual difficulty in
ACCEPTED MANUSCRIPT 14
EGOCENTRISM IN VICARIOUS DISSONANCE making their choice. We predicted that it is the way the choice difficulty affects the perceiver
(the egocentric perspective)—not the fellow group member—that drives the vicarious dissonance
PT
effect.
RI
Method Participants
SC
Based on prior laboratory research on vicarious cognitive dissonance (Norton et
NU
al<,2003; Monin et al, 2004), we sought 60 participants for Study 2. 70 undergraduate students were recruited for an in-lab study and compensated either with $8 or partial course credit. Ten
MA
participants were eliminated from the analyses due to suspicion. This left us with a final valid sample of N = 60. 65% of the sample was women; the mean age was 19.5 years.
TE
D
Design
In this experiment, participants witnessed a partner (actually a confederate) make a
AC CE P
choice between two candies. The perceived difficulty of this choice (easy or hard), however, was manipulated for the participant and the partner independently. This created a 2 (participant choice difficulty: easy, hard) by 2 (partner choice difficulty: easy, hard) design. Procedure
Consistent with prior free choice studies, participants were asked to rate a number of items, two of which were to be featured in a choice later in the study. Before taking part in the experiment, participants completed an online pre-measure. In this pre-measure survey, participants were asked to rate the desirability of fourteen candies using a 1-9 Likert scale with anchors assessing the desirability of the items. Upon arrival, the experimenter explained that the study was, ostensibly, a replication of a previous study originally run at a different, rival university. Because vicarious dissonance relies
ACCEPTED MANUSCRIPT 15
EGOCENTRISM IN VICARIOUS DISSONANCE
on high levels of group identification, this explanation referencing a neighboring, rival university served to elicit identification of the participant with their university identity (Brewer & Gardner,
PT
1996), ensuring a near-uniform level of identification across all participants in the study. The experimenter further explained that the study, originally run in dyads, was split up due to room
RI
constraints. Participants were then told that one participant, the ‗chooser,‘ would be video
SC
recorded during the experiment and this video would be shown to their partner, the ‗judger,‘ at a
NU
later time. In actuality, all participants were assigned to the ‗judger‘ role. The experimenter loaded a video on the computer. In the video, participants watched a
MA
gender-matched ‗partner‘ seated in the same room go through a free choice experiment. The partner in the video was first asked to rate all items (candies) before being asked to make a
D
choice between two of them. When asked to make their choice, a box was placed between the
TE
partner and the camera, clearly labeled ‗A‘ and ‗B.‘ Because it was important to make the videos
AC CE P
identical for all participants, the actual candy within the box was not seen. Matching the labeled boxes in the video to a key previously provided to the participant allowed participants to know what choice of candies was available to their partner without actually seeing the items themselves. The perceived difficulty of the choice for the participant and the perceived difficulty of the choice for the partner were manipulated. In order to manipulate the perception that their partner was making an easy or a difficult choice, we manipulated the amount of deliberation time that the partner exhibited before selecting one of the candies in the video (cf. Liberman & Förster, 2006). For an easy choice, the partner made his/her decision instantly, whereas they hesitated for 10 seconds before deciding to indicate a difficult choice. To create choices that the participant perceived to be easy, participants believed their partner was choosing between one candy that the participant had rated high and one candy that the participant had rated low on the
ACCEPTED MANUSCRIPT 16
EGOCENTRISM IN VICARIOUS DISSONANCE pretest, with a minimum separation of three points on the scale. In the hard choice condition, participants believed their partner was choosing between two candies that the participant had
PT
rated highly on the pretest and that also had scores that were identical or separated by only one
RI
point on the scale. This operationalization of deliberation time as a proxy for explicit choice difficulty has the additional benefit of implying that the partner holds similar attitudes towards
SC
the two alternatives when the decision takes a long time, echoing our manipulation of choice
NU
difficulty for the actual participant. Finally, when it came time for the partner to select a candy in the video, the partner always chose the candy that the participant rated more positively1.
MA
Following the video, participants completed the post-measure that asked participants to rate the desirability of two candies on the same 1-9 Likert scale used in the pre-measure. In
TE
D
Brehm‘s (1956) study and subsequent replications, dissonance is measured using the degree to which attitudes towards the chosen and unchosen alternatives grow more polarized following the
AC CE P
choice—that is, spreading of alternatives. Our dependent measure of attitude change, therefore, was the difference in attitude change (i.e., spreading of alternatives) for the chosen and unchosen candies. This measure was created in two steps. First, two difference scores were calculated by subtracting the pre-measure attitude score from the post-measure attitude score for both the chosen and unchosen items. Finally, the spreading of the two choice alternatives was computed by subtracting attitude change towards the unchosen candy from attitude change towards the chosen candy. In addition to this attitude change measure, participants were asked to rate the perceived degree of difficulty their partner demonstrated when making the choice as a manipulation check. Upon completion of the questionnaire, participants were debriefed and compensated. Results
ACCEPTED MANUSCRIPT 17
EGOCENTRISM IN VICARIOUS DISSONANCE Manipulation check
To assess whether our manipulation of the partner‘s choice difficulty was successful, we
PT
conducted a 2 (participant choice difficulty) by 2 (partner choice difficulty) ANOVA on
RI
participants‘ perceptions of their partner‘s difficulty making the choice in the video. As expected, participants who witnessed their partner deliberate on the choice longer (M = 3.97, SD
SC
= 1.51) believed that their partner‘s choice was more difficult than participants whose partner
NU
made the choice instantly (M = 1.39, SD = 0.57), F(1, 56) = 70.16, p < .001, η2p = .56. Spreading of alternatives
MA
We hypothesized that participants who witnessed their partner choose between two items rated highly by the participant would change their attitudes about those items more than
TE
D
participants who witnessed their partner choose between a highly-rated and lowly-rated item, regardless of the amount of difficulty their partners appeared to have in making the choice. To
AC CE P
investigate this, we conducted a 2 (participant choice difficulty) by 2 (partner choice difficulty) ANOVA on our measure of spreading of alternatives. For this measure, a higher score translates to an increasing distance between participants‘ ratings for the chosen and unchosen items after witnessing their partner‘s choice (relative to pretest ratings). In line with our predictions, we found a main effect of the participant‘s own choice difficulty, but not that of their partner, on spreading of alternatives. In other words, participants who witnessed their partner make a choice between two items that the participant had previously rated highly (M =0.88, SD = 2.49) exhibited greater spreading in their ratings of the items relative to participants who witnessed their partner choose between one high- and one low-ranked item (M = -0.77, SD = 2.29), F(1, 56) = 7.20, p = .01, η2p = .11 (see Figure 3). In contrast, the effect of the partner‘s choice difficulty, and the interaction, did not reach statistical significance, ps > .14.
ACCEPTED MANUSCRIPT 18
EGOCENTRISM IN VICARIOUS DISSONANCE
PT
[Insert Figure 2]
RI
In order to deconstruct whether this effect stemmed from attitude change towards the chosen vs. unchosen item, we conducted two other 2x2 ANOVAs on the attitude change scores
SC
for both the chosen and unchosen candies. While scores for the chosen items were not affected
NU
by either of our choice difficulty variables, we did observe a significant main effect of participant choice difficulty for the unchosen alternative, F(1, 56) = 16.47, p < .001, ηp2 = .23, such that
MA
participants for whom the partner‘s choice was difficult (M = -1.12, SD = 1.86) derogated the partner‘s unchosen item more than participants for whom the partner‘s choice was easy (M =
TE
D
0.80, SD = 1.90). As expected, there was no main effect for the partner‘s choice difficulty or for the interaction between participant and partner difficulty.
AC CE P
Discussion
Adapting Brehm‘s free choice paradigm to Norton et al.‘s (2003) vicarious dissonance paradigm, the present study orthogonally manipulated the difficulty of a choice for both an actor (a confederate) and an observer (the participant). Participants watched an ostensible partner make a choice between two items ideographically tailored to the participants‘ preferences. We predicted that, after witnessing another ingroup member make a choice between two items, participants would change their attitudes in an egocentric fashion—in other words, that they would be sensitive only to the amount of difficulty the choice would have presented for themselves, and not for the observed ingroup other. Our data support these predictions. It is worth pointing out the result that participants‘ spreading of alternatives appears to be mainly driven by the derogation of the unchosen alternative rather than any increase in the
ACCEPTED MANUSCRIPT 19
EGOCENTRISM IN VICARIOUS DISSONANCE
desirability of the chosen alternative. This is consistent with results of prior researh using the free choice paradigm In particular, Shultz, Léveillé, and Lepper (1999) found that participants in a
PT
free choice paradigm reduced their dissonance by derogating the unchosen alternative when the
RI
alternatives were highly desirable, but reduced their dissonance by increasing desirability of the chosen alternative when the alternatives were initially undesirable.
SC
General Discussion
NU
Vicarious dissonance opens a new window into a classic psychological process. The experience of cognitive dissonance has typically been seen as a state of personal discomfort that
MA
prompts justification or rationalization through attitude change, but recent research has demonstrated that dissonance can be aroused vicariously through witnessing inconsistency on the
TE
D
part of a fellow ingroup member. Previous research to date, however, has not conclusively addressed whether the inconsistent behavior of the observed other is arousing because people are
AC CE P
sensitive to the consistency of other ingroup members (cf. McKimmie, Terry, & Hogg, 2009) and feel the other‘s discomfort or because people are uncomfortable imagining themselves behaving inconsistently in the same situation. In two experiments, one of which that constitutes the first use of a vicarious free choice paradigm, we provide evidence for the latter. Our results suggest that people process two crucial variables for dissonance arousal—choice and consequences—egocentrically. When watching another ingroup member make a choice between two items, participants only showed spreading of alternatives—i.e., attitude change—when those items would have constituted a difficult choice for them, and not when the ingroup member found the decision visibly difficult themselves. In other words, participants only processed information about consequences when those consequences mattered from their perspective. Similarly, when participants were either instructed to take the perspective of the ingroup other,
ACCEPTED MANUSCRIPT 20
EGOCENTRISM IN VICARIOUS DISSONANCE told to imagine their own ‗vicarious‘ perspective, or given no perspective-taking instructions, participants only showed sensitivity to information about the other‘s level of choice when they
PT
egocentrically imagined themselves in the other‘s shoes. Taken together with previous evidence
RI
for the role of vicarious discomfort in vicarious dissonance, these findings point to the conclusion that vicarious dissonance is fundamentally an egocentric process.
SC
Support for an egocentric model has strong implications for the experience of vicarious
NU
dissonance. Perspective taking has in the past been implicated in vicarious experience (Goldstein & Cialdini, 2007; Eisenberg et al., 1991; Smith, 2006), yet a clear linkage to vicarious
MA
dissonance has not been clearly established in previous literature. Consistent with the work of Epley et al (2004; Savitsky et al, 2011), the present experiments suggest that simulation, rather
TE
D
than true perspective taking, is the driving force behind vicarious dissonance. Vicarious dissonance may additionally have untapped potential as a tool for changing
AC CE P
attitudes for the better. In cases where people possess positive attitudes towards destructive behaviors, such as smoking or drug abuse, or negative attitudes towards positive behaviors, such as flossing, leveraging the behavior of an ingroup member could be a productive way to change people‘s attitudes when other more direct methods (e.g. persuasion) are ineffective. Vicarious free choice adds a new dimension in broadening the types of attitudes targetable through vicarious dissonance and the ease with which they can be changed. One might leverage vicarious free choice to get children to choose healthier brands of snacks or beverages over other attractive but unhealthy ones, for instance, where merely witnessing another ingroup member choose a healthier brand on its own might not be aversive or inconsistent enough to arouse dissonance. The knowledge that vicarious dissonance functions egocentrically—i.e., that choice and consequences involved in the dissonant act seem to matter only for the observer—is essential for
ACCEPTED MANUSCRIPT 21
EGOCENTRISM IN VICARIOUS DISSONANCE designing interventions that can be tailored to any situation. With a thorough understanding of
the mechanisms by which vicarious dissonance functions, researchers have yet one more tool to
AC CE P
TE
D
MA
NU
SC
RI
PT
effect attitude change beyond classic means such as persuasion and social influence.
ACCEPTED MANUSCRIPT 22
EGOCENTRISM IN VICARIOUS DISSONANCE Acknowledgments
AC CE P
TE
D
MA
NU
SC
RI
PT
We appreciate the help of Ms. Lauren Feldman who assisted in the data collection.
ACCEPTED MANUSCRIPT 23
EGOCENTRISM IN VICARIOUS DISSONANCE References Brehm, Jack W. ―Postdecision changes in the desirability of alternatives.‖ The Journal of
PT
Abnormal and Social Psychology 52.3 (1956): 384. Brewer, Marilynn B., and Wendi Gardner. ―Who is this ―We‖? Levels of collective identity and
RI
self representations.‖ Journal of Personality and Social Psychology 71.1 (1996): 83.
SC
Cooper, J. (2007). Cognitive dissonance: 50 years of a classic theory. London: Sage.
NU
Cooper, J., & Mackie, D. (1983). Cognitive dissonance in an intergroup context. Journal of Personality and Social Psychology, 44(3), 536-544.
MA
Eisenberg, N., Fabes, R. A., Schaller, M., Miller, P., Carlo, G., Poulin, R., & Shell, R. (1991). Personality and socialization correlates of vicarious emotional responding. Journal of
TE
D
Personality and Social Psychology, 61(3), 459. Elliot, A. J., & Devine, P. G. (1994). On the motivational nature of cognitive dissonance:
AC CE P
Dissonance as psychological discomfort. Journal of Personality and Social Psychology, 67(3), 382.
Epley, N., Keysar, B., Van Boven, L., & Gilovich, T. (2004). Perspective taking as egocentric anchoring and adjustment. Journal of Personality and Social Psychology, 87(3), 327. Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7(2), 117-140. Festinger, L. (1957). A theory of cognitive dissonance. Evanston, IL: Row, Peterson. Festinger, L., & Carlsmith, J. M. (1959). Cognitive consequences of forced compliance. The Journal of Abnormal and Social Psychology, 58(2), 203. Glasford, G. E., Dovidio, J. F., & Pratto, F. (2009). I continue to feel so good about us: In-group identification and the use of social identity-enhancing strategies to reduce intragroup dissonance. Personality and Social Psychology Bulletin, 35(4), 415-427.
ACCEPTED MANUSCRIPT 24
EGOCENTRISM IN VICARIOUS DISSONANCE Glasford, G. E., Pratto, F., & Dovidio, J. F. (2008). Intragroup dissonance: Responses to in-
group violation of personal values. Journal of Experimental Social Psychology, 44, 1057-
PT
1064.
RI
Harmon-Jones, E., & Mills, J. (eds) (1999). Cognitive dissonance: Progress on a pivotal theory in social psychology, Washington, DC: APA.
SC
Harmon-Jones, E., Harmon-Jones, C., Fearn, M., Sigelman, J. D., & Johnson, P. (2008). Left
NU
frontal cortical activation and spreading of alternatives: tests of the action-based model of dissonance. Journal of Personality and Social Psychology, 94(1), 1-15.
MA
Heider, F. (1958). The psychology of interpersonal relations. NY: Wiley. Hogg, M. A., & Smith, J. R. (2007). Attitudes in social context: A social identity perspective.
TE
D
European Journal of Social Psychology, 18(1), 89-131. Shultz, T. R., Léveillé, E., & Lepper, M. R. (1999). Free choice and cognitive dissonance
AC CE P
revisited: Choosing ―lesser evils‖ versus ―greater goods.‖ Personality and Social Psychology Bulletin, 25(1), 40-48. Liberman, N., & Förster, J. (2006). Inferences from decision difficulty. Journal of Experimental Social Psychology, 42(3), 290-301. McKimmie, B. M., Terry, D. J., & Hogg, M. A. (2009). Dissonance reduction in the context of group membership: The role of metaconsistency. Group Dynamics: Theory, Research, and Practice, 13(2), 103-119. Monin, B., Norton, M. I., Cooper, J., & Hogg, M. A. (2004). Reacting to an assumed situation vs. conforming to an assumed reaction: The role of perceived speaker attitude in vicarious dissonance. Group Processes & Intergroup Relations, 7(3), 207-220.
ACCEPTED MANUSCRIPT 25
EGOCENTRISM IN VICARIOUS DISSONANCE Newcomb, T. M., & American Council on Public Affairs. (1957). Personality and Social Change. Dryden.
PT
Norton, M. I., Monin, B., Cooper, J., & Hogg, M. A. (2003). Vicarious dissonance: Attitude
RI
change from the inconsistency of others. Journal of Personality and Social Psychology, 85(1), 47-62.
SC
Savitsky, K., Keysar, B., Epley, N., Carter, T. & Swanson, A. (2011). The closeness-
NU
communication bias: Increased egocentrism among friends versus strangers. Journal of Experimental Social Psychology, 47, 269-273.
MA
Smith, A. (2006). Cognitive empathy and emotional empathy in human behavior and evolution. Psychological Record, 56(1).
TE
D
Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. The Social Psychology of Intergroup Relations, 33(47), 74.
AC CE P
Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behavior. Psychology of Intergroup Relations, 7-24.
ACCEPTED MANUSCRIPT 26
EGOCENTRISM IN VICARIOUS DISSONANCE Footnotes
AC CE P
TE
D
MA
NU
SC
RI
PT
1. In the case where alternatives in the participant hard choice condition were tied in terms of desirability, the actor‘s ostensible choice was randomly determined.
ACCEPTED MANUSCRIPT 27
EGOCENTRISM IN VICARIOUS DISSONANCE
Figure Captions
PT
Figure 1. Attitude towards government influence in the media in Study 1 as a function of
RI
perspective taking instructions.
SC
Figure 2. Attitudes toward government influence for egocentrically primed participants as a
NU
function of choice and level of in-group identification.
MA
Figure 3. Post-decision spread of alternatives as a function of observer and actor choice
AC CE P
TE
D
difficulty.
ACCEPTED MANUSCRIPT 28
PT
EGOCENTRISM IN VICARIOUS DISSONANCE
Low Choice
6
RI
High Choice
SC NU
4
MA
3
1 No PT
TE
D
2
Other PT Egocentric PT Perspective Taking Condition
AC CE P
Attitude
5
Figure 1
ACCEPTED MANUSCRIPT 29
AC CE P
TE
D
MA
NU
SC
RI
PT
EGOCENTRISM IN VICARIOUS DISSONANCE
Figure 2
ACCEPTED MANUSCRIPT 30
EGOCENTRISM IN VICARIOUS DISSONANCE
Easy Choice
Difficult Choice
PT
1
RI
0.5
SC
0 -0.5 -1
NU
Spreading of Alternatives
1.5
MA
-1.5
AC CE P
TE
D
Actor Condition
Figure 3
Observer Condition
ACCEPTED MANUSCRIPT 31
EGOCENTRISM IN VICARIOUS DISSONANCE Highlights
AC CE P
TE
D
MA
NU
SC
RI
PT
Tested an egocentric model for the underlying process of vicarious dissonance. For the first time, showed effects of vicarious dissonance in both free choice and induced compliance. Vicarious dissonance was facilitated when participants adopted an egocentric perspective. Results show that vicarious dissonance is “about me” rather than about a group member.
S0022-1031(15)00103-1 doi: 10.1016/j.jesp.2015.09.001 YJESP 3350
To appear in:
Journal of Experimental Social Psychology
Received date: Revised date: Accepted date:
12 December 2014 31 August 2015 2 September 2015
Please cite this article as: Blackman, S.F., Keller, K.T. & Cooper, J., Egocentrism and Vicarious Dissonance, Journal of Experimental Social Psychology (2015), doi: 10.1016/j.jesp.2015.09.001
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1
SC
RI
PT
Running head: EGOCENTRISM IN VICARIOUS DISSONANCE
NU
Egocentrism and Vicarious Dissonance Shane F. Blackman, Kyle T. Keller, and Joel Cooper
Author Note
AC CE P
Word count: 5284
TE
D
MA
Princeton University
Shane F. Blackman, Department of Psychology, Princeton University Kyle T. Keller, Department of Psychology, Princeton University Joel Cooper, Department of Psychology, Princeton University Correspondences concerning this article should be addressed to Joel Cooper, Department of Psychology, Princeton University, Princeton, NJ 08544. Contact: [email protected]
ACCEPTED MANUSCRIPT 2
EGOCENTRISM IN VICARIOUS DISSONANCE Abstract
Past research on vicarious dissonance (Monin, Norton, Cooper, & Hogg, 2004; Norton,
PT
Monin, Cooper, & Hogg, 2003) has demonstrated that participants experience vicarious
RI
discomfort and change their attitudes when witnessing counterattitudinal behavior from a fellow ingroup member. It has not been shown, however, whether witnessing this counterattitudinal
SC
behavior arouses vicarious dissonance because it contradicts the attitudes of the observed
NU
ingroup member or for egocentric reasons – that is, because it contradicts the attitudes of the observer. In three experiments, we test the hypothesis that the conditions necessary for arousing
MA
vicarious dissonance—namely, observing an ingroup member‘s responsibility for aversive consequences and level of choice—are processed egocentrically by the observer. By
TE
D
manipulating perspective taking in a vicarious induced compliance task, and manipulating choice difficulty in a vicarious free-choice dissonance paradigm, we show that the arousal of vicarious
AC CE P
dissonance crucially depends on the consequences of the ingroup member‘s actions for the observer, and not for the observed ingroup member. Implications for other aspects of vicarious dissonance and vicarious experience in general are discussed. Keywords: attitudes, cognitive dissonance, egocentrism, perspective taking, vicarious dissonance
ACCEPTED MANUSCRIPT 3
EGOCENTRISM IN VICARIOUS DISSONANCE Egocentrism in Vicarious Dissonance In our daily lives, we are motivated to maintain consistency among our beliefs and
PT
behaviors (Festinger, 1957; Heider, 1958; Newcomb, 1957). We strive to maintain and preserve
RI
harmony between our own attitudes and those espoused by important groups and important others within those groups (Cooper & Mackie, 1983; Festinger, 1954; Glasford, Dovidio, &
SC
Pratto, 2009; Glasford, Pratto, & Dovidio, 2008; Hogg & Smith, 2007; Tajfel & Turner, 1986).
NU
As decades of research in the cognitive dissonance tradition have taught us (Cooper, 2007; Harmon-Jones & Mills, 1999), when we perceive contradiction, we feel pressure to resolve that
MA
inconsistency through justification or rationalization, attempting to bring our changeable attitudes in line with our immutable past behavior.
TE
D
Norton, Monin, Cooper, and Hogg (2003) expanded the reach of dissonance theory to examine the impact of witnessing other people behaving in ways that contradict their attitudes
AC CE P
(see also, Monin, Norton, Cooper, & Hogg, 2004). They suggested that people who observe dissonant behavior by another member of a valued ingroup experience dissonance vicariously, and are motivated to change their own attitudes. Tied as we are to an actor by the strong bonds that hold an ingroup together, they predicted that people would experience dissonance vicariously and change their attitudes as a consequence. Simulating a classic induced compliance procedure (Festinger & Carlsmith, 1959), Norton et al. (2003) had participants witness a fellow group member agree to write an essay advocating a position they actually opposed. The results of several experiments showed that participants who were strongly identified with their ingroup changed their attitudes in the direction of the ingroup member‘s advocacy simply by observing the confederate‘s agreement to write a counterattitudinal speech. Personal vs. Vicarious Dissonance
ACCEPTED MANUSCRIPT 4
EGOCENTRISM IN VICARIOUS DISSONANCE A crucial question left open by the set of studies on vicarious dissonance is the process that motivates people to change their attitude. Results have shown that vicarious dissonance is
PT
sensitive to many of the moderators that affect personal cognitive dissonance. For example, the
RI
ingroup member whom people witness acting in a counterattitudinal fashion must have had the freedom to decline the request, the behavior must be counterattitudinal for the writer, and the
SC
there must be a plausible unwanted consequence pursuant to the behavior. Prior research has also
NU
shown that there is discomfort associated with vicarious dissonance: people who change their attitudes in the high dissonance condition report a feeling of greater discomfort.
MA
The current research is designed to shed light on the basis of people‘s motivation to change their own attitudes after observing a group member‘s behavior. One plausible view is that
D
people take the perspective of someone to whom they feel close (e.g., a member of their social
TE
group) and experience the discrepant situation from that person‘s perspective. The tendency of
AC CE P
group members to fuse their identities with those of other members of their group (Hogg & Smith, 2007) gives credence to this possibility. Adopting my fellow group member‘s perspective, I may experience what she experiences and change my attitudes to reduce the discomfort.
On the other hand, research has shown that adopting the perspective of others – even those to whom we feel close -- is not an easy task (Epley, Keysar, van Boven & Gilovich, 2004; Savitsky, Keysar, Epley, Carter & Swanson, 2011). People seem to begin their attempt at perspective taking by adopting an egocentric point of view, often making insufficient corrections from there. This prompts us to consider the possibility that the process that leads to attitude change in a group situation may be fundamentally an egocentric one. It is possible that it is more about ‗me‘ (the observer) and less about taking the perspective of the actor. That is, it is more
ACCEPTED MANUSCRIPT 5
EGOCENTRISM IN VICARIOUS DISSONANCE about how the actor would be affected by the group member‘s actions. Instead of viewing observers as changing their attitudes because of their fused identity with the group member who
PT
is in the throes of dissonance, the egocentric view is that vicarious dissonance feels
RI
uncomfortable because it is the way I think I would feel were I in the shoes of the actor. Prior research suggests that the egocentric view of vicarious dissonance is a viable approach. In one of
SC
Norton et al. (2003) studies, participants were asked about their discomfort after witnessing the
NU
counterattitudinal behavior of their fellow group member. They found that the magnitude of attitude change was not related to how they thought their fellow group member felt, but rather
MA
how they thought they would feel if they were in their partner‘s shoes. Rather than sharing their partner‘s discomfort, the participants seemed motivated by how they thought they would feel if
TE
D
they were in the essay writer‘s situation.
The present research seeks to clarify the mechanism underlying vicarious dissonance by
AC CE P
assessing two possible processes. We differentiate other-oriented perspective taking from egocentric simulation. In the first view, people experience the dissonance situation from the perspective of the actor, feeling or imagining what the actor feels. Adopting the tension state of cognitive dissonance on behalf of the actor, the observer reduces dissonance in the same way as the actor. In the second view, ingroup members cognitively simulate the dissonance situation. They imagine what they would experience if they were in the actor‘s situation. In the egocentric view, observers make the actor‘s situation their own and respond accordingly. We first report the results of two studies using the previously validated vicarious induced compliance paradigm and directly manipulate participants‘ perspective. In the third experiment, we report the first use of a new paradigm, adding converging evidence to the notion that people take an egocentric approach when observing a fellow group member engage in a dissonant act.
ACCEPTED MANUSCRIPT 6
EGOCENTRISM IN VICARIOUS DISSONANCE Study 1 Study 1 is designed to systematically investigate the role of perspective taking in
PT
vicarious dissonance. Participants observed a member of their ingroup engage in behavior that
RI
contradicted their attitude. In order to facilitate the shared connection between the participant and group member, some of our participants were instructed to view the situation from the
SC
perspective of the group member. By contrast, in order to facilitate egocentric simulation, we
NU
asked other participants to adopt their own perspective, imagining how they would feel if they were in the group member‘s situation. Although both processes are viable, our prediction—based
MA
on the clue provided by Norton et al.‘s (2003) discomfort findings—is that vicarious dissonance is facilitated when taking one‘s own egocentric perspective and diminished when trying to take
Method
AC CE P
Participants
TE
D
the perspective of the ingroup member.
Participants were recruited online via Amazon‘s Mechanical Turk service. Because no vicarious dissonance experiment has been reported using an on-line procedure, we chose to recruit at least 20 participants per condition to provide sufficient power. Our strategy was to leave mTurk open for 48 hours with the constraint that a minimum of 120 workers participate. 156 people volunteered for the experiment. Of these, 144 completed the entire survey. Three were ultimately excluded for failing to follow all of the instructions, resulting in a sample of N=141. Participants ranged in age from 18 to 74 (M = 33.9). 59.9% of the sample were women. Procedure Participants gave their consent to participate in an experiment comparing the writing styles of people in different countries. They believed that they were participating at the same
ACCEPTED MANUSCRIPT 7
EGOCENTRISM IN VICARIOUS DISSONANCE time as another American MTurk Worker (i.e., a member of their ingroup)—making their ingroup identity as an American salient through highlighting metacontrast (Brewer & Gardner,
PT
1996; Tajfel & Turner, 1979)—and were told that they had been assigned to read an essay
RI
written by that Worker. They then saw instructions indicating that their partner was either told (low choice condition) or asked (high choice condition) to write an essay in favor of government
SC
influence in the media. Specifically, in the high choice condition, the partner was asked, ―Are
NU
you willing to write the essay?‖ and the participant was permitted to decline to write. In the low choice condition, there was no permission to decline. Participants were also informed that the
MA
other Worker did not personally agree with the position they wrote about. While their partner was ostensibly writing the essay, participants in the experimental conditions were given a
D
perspective-taking priming manipulation. In the egocentric condition, participants were told,
TE
―While your partner is writing his/her essay, we would like you to take this moment to write a
AC CE P
few sentences about what you imagine you would be feeling if you were writing this essay. What would be going through your head?‖ In the other-oriented perspective taking conditions, they were asked to imagine what their partner is feeling while writing the essay. In a control condition, no instructions about perspective taking were given and participants went directly to the dependent measures.
After the priming manipulation (or control), participants filled out the study‘s dependent measures. Attitudes towards government influence in the media were assessed by asking participants, ―To what extent are you in favor of government influence in the media?‖, assessed on a 15 point scale. We also collected data on participants‘ perceptions of their own and their partner‘s representativeness and typicality, their identification with the ingroup (Americans), their liking and perceived similarity to their partner and their level of personal and vicarious
ACCEPTED MANUSCRIPT 8
EGOCENTRISM IN VICARIOUS DISSONANCE discomfort. In addition, we collected measures of their perceptions of the attitude of the average American and participants‘ perceptions of the importance of the issue. Participants were led to
PT
believe that they would read their partner‘s essay after hearing some information about it in the
RI
instructions, but after filling out the study‘s dependent measures the participants were instead
Results
NU
Manipulation check
SC
thanked and debriefed.
Our manipulation check indicated that our manipulation of choice in the study was
MA
successful, as participants in the high-choice conditions (M = 10.38, SD = 4.88) had significantly higher perceptions of their alleged partner‘s level of choice, F(1, 135) = 16.02, p < .001, η2p =
Attitude measures
TE
D
.10, than those in the low-choice conditions (M = 6.71, SD = 5.45).
AC CE P
To see whether our manipulations of choice and perspective taking had an effect on participants‘ attitudes towards government influence in the media, we conducted a 2 x 3 factorial ANOVA on the data. Our analyses revealed a significant interaction of choice and perspective taking, F(2, 135) = 3.37, p = .04 p2 = .06. To explore this interaction, we conducted analyses to explore the simple effect of choice within each perspective taking condition. In line with our hypotheses, there was no effect of choice in the no perspective-taking condition, p = .38, nor in the other perspective-taking condition, p = .87. When participants were instructed to take their own ―vicarious‖ perspective, however, participants in the high-choice condition (M = 4.21, SD = 2.88) exhibited significantly more positive attitudes towards government influence in the media, F(1, 62) = 7.05, p = .01, η2p = .16, than did participants in the low-choice condition (M = 2.14, SD = 2.01; see Figure 1).
ACCEPTED MANUSCRIPT 9
EGOCENTRISM IN VICARIOUS DISSONANCE
Affect measures
PT
We assessed participants‘ self-reported discomfort and their vicariously experienced
RI
discomfort. For participants‘ own discomfort, a 2 x 3 factorial ANOVA showed that there was not a significant effect for either perspective taking, p = .87, or choice, p = .84, on participants‘
SC
levels of discomfort, nor was there a significant interaction, p = .87. Similarly, for participants‘
NU
reported vicarious discomfort, neither perspective taking, p = .37, nor choice, p = .36, nor was the interaction significant, p = .29.
MA
Discussion
This experiment sought to investigate the importance of perspective taking in the process
TE
D
of vicarious dissonance. As predicted, the pattern of results followed the previously reported vicarious dissonance findings (Norton et al., 2003) only when participant were given instructions
AC CE P
designed to enhance an egocentric simulation of the fellow group member‘s situation. When given instructions to vicariously imagine writing the essay rather than to empathize with the essay-writer, participants in the high-choice condition expressed more favorable attitudes towards government influence in the media than participants in the low-choice condition. These canonical dissonance effects of choice on attitude change were not present, however, when participants were told to take their partner‘s perspective or received no instructions at all. One aspect of our study that merits further attention is that we found no differences across conditions in participants’ reported self or vicarious discomfort. This is consistent with prior research in both personal (Elliot &Devine, 1994) and vicarious dissonance (Norton et al, 2003) that did not find differences in discomfort when discomfort was measured after people had the opportunity to change their attitudes. When Norton et al
ACCEPTED MANUSCRIPT 10
EGOCENTRISM IN VICARIOUS DISSONANCE
(2003) systematically manipulated the order of measurement in the vicarious dissonance paradigm, they found that when discomfort was measured first, people reported
PT
differences in their level of discomfort but did not show attitude change. On the other
RI
hand, when discomfort was measured following the attitude measure, there was attitude
SC
change but no reported discomfort. In the Norton et al study, as in the current study, it is likely that when participants report their attitudes towards the issue before reporting their
NU
discomfort, the discomfort is reduced by the change in attitude.
In addition, the current study was not designed to address one of our key theoretical
MA
assumptions – that strong identification with the ingroup is important for vicarious dissonance to occur. Based on prior research that showed that strong group identification is necessary for
TE
D
vicarious dissonance, we elected to run Study 1 in a way that maximized identification with the in-group for all participants. We used participants‘ nationality (American citizens) as the
AC CE P
ingroup, adopting Brewer & Gardner‘s (1996) method of metacontrasts to keep identification high. We assessed the degree of identification by asking participants how strongly they identified with being an American citizen. There was very little variance in response to the identification question, and we were not in a position to show that the magnitude of identification facilitated vicarious dissonance in the egocentric conditions. In order to provide evidence for this key assumption, we conducted a second study for the purpose of showing that egocentric vicarious dissonance depends on strong in-group identification.
ACCEPTED MANUSCRIPT 11
EGOCENTRISM IN VICARIOUS DISSONANCE
PT
Study 2
RI
METHOD Participants and Procedure
SC
We recruited 84 new mTurk workers for the second study who ranged in age from 19 to
NU
66 (M= 36.8). 46.5% were female. As in the first experiment, the study was introduced as a study on the writing styles of people from different countries and participants believed their
MA
partner was another American mTurk worker. We precisely repeated the instructions of the egocentric conditions of Study 1, with half of the participants believing their partner chose – and
media.
TE
D
the other half believing their partner was told – to write in favor of government influence in the
AC CE P
Attitudes were assessed with the same 15-point scale used in Study 1. Participants were then asked about their level of group identification. Instead of asking about their identification as Americans, we asked participants about how much they identified with the specific group, ―mTurk workers.‖ Following the dependent measures, participants were thanked and debriefed. Results and Discussion
Consistent with the assumption underlying vicarious cognitive dissonance, level of identification had a significant effect on reported attitudes in the high choice condition. When participants learned that their fellow mTurk worker agreed to make a speech advocating increased influence of the government in media, those who identified strongly with their group (i.e., were above the median) came to agree with that position (M = 4.62) more than those who were below the median on identification (M = 1.18) (F (1, 82) = 4.32, p <.02, p2= .019. The
ACCEPTED MANUSCRIPT 12
EGOCENTRISM IN VICARIOUS DISSONANCE
interaction between choice and level of identity was also significant (F(1, 82) = 9.93, p <.01, p2 = .108.
PT
Figure 2 depicts the attitude scores regressed on choice and level of identification
RI
(standardized). The choice x identification interaction was significant, b=.5472, t(82) = 3.21, p
SC
= .002. As predicted, higher levels of identification led to higher attitude scores for participants in the high choice condition (b= .51; t (41) = 3.77; p <.01). We had no a priori prediction about
NU
the effect of identification in the low choice conditions. The data show a slight reversal of the effect of identification on attitude change in the low choice condition, but that effect was not
MA
significant (p<.40).
Overall, the results of the first two studies offer strong support for the notion that
TE
D
vicarious dissonance is a function of egocentric identification. The first study showed that people change their attitudes in the direction advocated by a fellow group member when they
AC CE P
adopt an egocentric position, imagining how they would feel if they were in their partner‘s position. As predicted, attitudes became more positive to the partner‘s counterattitudinal statement if the partner chose, rather than was told, to write the position. Study 2 was designed to clarify an ambiguity in the first study. In that study, we had asked participants about their identification with their nationality and found very little variance in the degree of identification. In the second study, we asked about participants‘ identification with a more specific and less nationalistic identity, asking about their identification with the group ―mTurk workers.‖ We found that the effect for choice in the egocentric condition depended on people feeling highly identified with that in-group. Study 3
ACCEPTED MANUSCRIPT 13
EGOCENTRISM IN VICARIOUS DISSONANCE
In our third experiment, we approached the egocentric model of vicarious dissonance in a completely novel way. We adopted the classic free choice paradigm of dissonance (Brehm,
PT
1956) in which participants witnessed a fellow group member make a choice between two items.
RI
In a typical study that arouses personal cognitive dissonance (e.g., Kitayama, Snibbe, Markus, & Suzuki, 2004; Harmon-Jones, Harmon-Jones, Fearn, Sigelman, & Johnson, 2008), participants
SC
are asked to rate a number of objects and are then asked to make a choice between two of the
NU
items. Because choosing one object over another necessarily entails accepting the unwanted elements of the chosen object and losing the desirable elements of the unchosen alternative,
MA
those consequences create dissonance. The dissonance is resolved by spreading the attractiveness of the alternatives following the choice; the more difficult the choice (i.e., the closer the
TE
D
attractiveness prior to the choice), the greater the dissonance. The extant literature has not used free choice as a method for studying vicarious
AC CE P
dissonance. One purpose of the current study is to demonstrate that people, after observing an ingroup member make a difficult choice, also spread the choice alternatives. Second, the free choice instantiation of dissonance allows us converging methodology to examine the egocentric simulation vs. other-oriented perspective taking process that underlies the effect. In this study, people observed a fellow group member make a choice that was either easy or difficult for that group member. Orthogonal to that manipulation, we systematically manipulated whether the choice would have been difficult or easy for the observer. The question is: Whose choice difficulty drives the spreading of the alternatives? We hypothesized that participants‘ attitudes would respond egocentrically to the choice difficulty—by exhibiting spreading of the alternatives when the observed choice would have been difficult for them. By contrast, we expected that participants would be unaffected by the fellow group member‘s actual difficulty in
ACCEPTED MANUSCRIPT 14
EGOCENTRISM IN VICARIOUS DISSONANCE making their choice. We predicted that it is the way the choice difficulty affects the perceiver
(the egocentric perspective)—not the fellow group member—that drives the vicarious dissonance
PT
effect.
RI
Method Participants
SC
Based on prior laboratory research on vicarious cognitive dissonance (Norton et
NU
al<,2003; Monin et al, 2004), we sought 60 participants for Study 2. 70 undergraduate students were recruited for an in-lab study and compensated either with $8 or partial course credit. Ten
MA
participants were eliminated from the analyses due to suspicion. This left us with a final valid sample of N = 60. 65% of the sample was women; the mean age was 19.5 years.
TE
D
Design
In this experiment, participants witnessed a partner (actually a confederate) make a
AC CE P
choice between two candies. The perceived difficulty of this choice (easy or hard), however, was manipulated for the participant and the partner independently. This created a 2 (participant choice difficulty: easy, hard) by 2 (partner choice difficulty: easy, hard) design. Procedure
Consistent with prior free choice studies, participants were asked to rate a number of items, two of which were to be featured in a choice later in the study. Before taking part in the experiment, participants completed an online pre-measure. In this pre-measure survey, participants were asked to rate the desirability of fourteen candies using a 1-9 Likert scale with anchors assessing the desirability of the items. Upon arrival, the experimenter explained that the study was, ostensibly, a replication of a previous study originally run at a different, rival university. Because vicarious dissonance relies
ACCEPTED MANUSCRIPT 15
EGOCENTRISM IN VICARIOUS DISSONANCE
on high levels of group identification, this explanation referencing a neighboring, rival university served to elicit identification of the participant with their university identity (Brewer & Gardner,
PT
1996), ensuring a near-uniform level of identification across all participants in the study. The experimenter further explained that the study, originally run in dyads, was split up due to room
RI
constraints. Participants were then told that one participant, the ‗chooser,‘ would be video
SC
recorded during the experiment and this video would be shown to their partner, the ‗judger,‘ at a
NU
later time. In actuality, all participants were assigned to the ‗judger‘ role. The experimenter loaded a video on the computer. In the video, participants watched a
MA
gender-matched ‗partner‘ seated in the same room go through a free choice experiment. The partner in the video was first asked to rate all items (candies) before being asked to make a
D
choice between two of them. When asked to make their choice, a box was placed between the
TE
partner and the camera, clearly labeled ‗A‘ and ‗B.‘ Because it was important to make the videos
AC CE P
identical for all participants, the actual candy within the box was not seen. Matching the labeled boxes in the video to a key previously provided to the participant allowed participants to know what choice of candies was available to their partner without actually seeing the items themselves. The perceived difficulty of the choice for the participant and the perceived difficulty of the choice for the partner were manipulated. In order to manipulate the perception that their partner was making an easy or a difficult choice, we manipulated the amount of deliberation time that the partner exhibited before selecting one of the candies in the video (cf. Liberman & Förster, 2006). For an easy choice, the partner made his/her decision instantly, whereas they hesitated for 10 seconds before deciding to indicate a difficult choice. To create choices that the participant perceived to be easy, participants believed their partner was choosing between one candy that the participant had rated high and one candy that the participant had rated low on the
ACCEPTED MANUSCRIPT 16
EGOCENTRISM IN VICARIOUS DISSONANCE pretest, with a minimum separation of three points on the scale. In the hard choice condition, participants believed their partner was choosing between two candies that the participant had
PT
rated highly on the pretest and that also had scores that were identical or separated by only one
RI
point on the scale. This operationalization of deliberation time as a proxy for explicit choice difficulty has the additional benefit of implying that the partner holds similar attitudes towards
SC
the two alternatives when the decision takes a long time, echoing our manipulation of choice
NU
difficulty for the actual participant. Finally, when it came time for the partner to select a candy in the video, the partner always chose the candy that the participant rated more positively1.
MA
Following the video, participants completed the post-measure that asked participants to rate the desirability of two candies on the same 1-9 Likert scale used in the pre-measure. In
TE
D
Brehm‘s (1956) study and subsequent replications, dissonance is measured using the degree to which attitudes towards the chosen and unchosen alternatives grow more polarized following the
AC CE P
choice—that is, spreading of alternatives. Our dependent measure of attitude change, therefore, was the difference in attitude change (i.e., spreading of alternatives) for the chosen and unchosen candies. This measure was created in two steps. First, two difference scores were calculated by subtracting the pre-measure attitude score from the post-measure attitude score for both the chosen and unchosen items. Finally, the spreading of the two choice alternatives was computed by subtracting attitude change towards the unchosen candy from attitude change towards the chosen candy. In addition to this attitude change measure, participants were asked to rate the perceived degree of difficulty their partner demonstrated when making the choice as a manipulation check. Upon completion of the questionnaire, participants were debriefed and compensated. Results
ACCEPTED MANUSCRIPT 17
EGOCENTRISM IN VICARIOUS DISSONANCE Manipulation check
To assess whether our manipulation of the partner‘s choice difficulty was successful, we
PT
conducted a 2 (participant choice difficulty) by 2 (partner choice difficulty) ANOVA on
RI
participants‘ perceptions of their partner‘s difficulty making the choice in the video. As expected, participants who witnessed their partner deliberate on the choice longer (M = 3.97, SD
SC
= 1.51) believed that their partner‘s choice was more difficult than participants whose partner
NU
made the choice instantly (M = 1.39, SD = 0.57), F(1, 56) = 70.16, p < .001, η2p = .56. Spreading of alternatives
MA
We hypothesized that participants who witnessed their partner choose between two items rated highly by the participant would change their attitudes about those items more than
TE
D
participants who witnessed their partner choose between a highly-rated and lowly-rated item, regardless of the amount of difficulty their partners appeared to have in making the choice. To
AC CE P
investigate this, we conducted a 2 (participant choice difficulty) by 2 (partner choice difficulty) ANOVA on our measure of spreading of alternatives. For this measure, a higher score translates to an increasing distance between participants‘ ratings for the chosen and unchosen items after witnessing their partner‘s choice (relative to pretest ratings). In line with our predictions, we found a main effect of the participant‘s own choice difficulty, but not that of their partner, on spreading of alternatives. In other words, participants who witnessed their partner make a choice between two items that the participant had previously rated highly (M =0.88, SD = 2.49) exhibited greater spreading in their ratings of the items relative to participants who witnessed their partner choose between one high- and one low-ranked item (M = -0.77, SD = 2.29), F(1, 56) = 7.20, p = .01, η2p = .11 (see Figure 3). In contrast, the effect of the partner‘s choice difficulty, and the interaction, did not reach statistical significance, ps > .14.
ACCEPTED MANUSCRIPT 18
EGOCENTRISM IN VICARIOUS DISSONANCE
PT
[Insert Figure 2]
RI
In order to deconstruct whether this effect stemmed from attitude change towards the chosen vs. unchosen item, we conducted two other 2x2 ANOVAs on the attitude change scores
SC
for both the chosen and unchosen candies. While scores for the chosen items were not affected
NU
by either of our choice difficulty variables, we did observe a significant main effect of participant choice difficulty for the unchosen alternative, F(1, 56) = 16.47, p < .001, ηp2 = .23, such that
MA
participants for whom the partner‘s choice was difficult (M = -1.12, SD = 1.86) derogated the partner‘s unchosen item more than participants for whom the partner‘s choice was easy (M =
TE
D
0.80, SD = 1.90). As expected, there was no main effect for the partner‘s choice difficulty or for the interaction between participant and partner difficulty.
AC CE P
Discussion
Adapting Brehm‘s free choice paradigm to Norton et al.‘s (2003) vicarious dissonance paradigm, the present study orthogonally manipulated the difficulty of a choice for both an actor (a confederate) and an observer (the participant). Participants watched an ostensible partner make a choice between two items ideographically tailored to the participants‘ preferences. We predicted that, after witnessing another ingroup member make a choice between two items, participants would change their attitudes in an egocentric fashion—in other words, that they would be sensitive only to the amount of difficulty the choice would have presented for themselves, and not for the observed ingroup other. Our data support these predictions. It is worth pointing out the result that participants‘ spreading of alternatives appears to be mainly driven by the derogation of the unchosen alternative rather than any increase in the
ACCEPTED MANUSCRIPT 19
EGOCENTRISM IN VICARIOUS DISSONANCE
desirability of the chosen alternative. This is consistent with results of prior researh using the free choice paradigm In particular, Shultz, Léveillé, and Lepper (1999) found that participants in a
PT
free choice paradigm reduced their dissonance by derogating the unchosen alternative when the
RI
alternatives were highly desirable, but reduced their dissonance by increasing desirability of the chosen alternative when the alternatives were initially undesirable.
SC
General Discussion
NU
Vicarious dissonance opens a new window into a classic psychological process. The experience of cognitive dissonance has typically been seen as a state of personal discomfort that
MA
prompts justification or rationalization through attitude change, but recent research has demonstrated that dissonance can be aroused vicariously through witnessing inconsistency on the
TE
D
part of a fellow ingroup member. Previous research to date, however, has not conclusively addressed whether the inconsistent behavior of the observed other is arousing because people are
AC CE P
sensitive to the consistency of other ingroup members (cf. McKimmie, Terry, & Hogg, 2009) and feel the other‘s discomfort or because people are uncomfortable imagining themselves behaving inconsistently in the same situation. In two experiments, one of which that constitutes the first use of a vicarious free choice paradigm, we provide evidence for the latter. Our results suggest that people process two crucial variables for dissonance arousal—choice and consequences—egocentrically. When watching another ingroup member make a choice between two items, participants only showed spreading of alternatives—i.e., attitude change—when those items would have constituted a difficult choice for them, and not when the ingroup member found the decision visibly difficult themselves. In other words, participants only processed information about consequences when those consequences mattered from their perspective. Similarly, when participants were either instructed to take the perspective of the ingroup other,
ACCEPTED MANUSCRIPT 20
EGOCENTRISM IN VICARIOUS DISSONANCE told to imagine their own ‗vicarious‘ perspective, or given no perspective-taking instructions, participants only showed sensitivity to information about the other‘s level of choice when they
PT
egocentrically imagined themselves in the other‘s shoes. Taken together with previous evidence
RI
for the role of vicarious discomfort in vicarious dissonance, these findings point to the conclusion that vicarious dissonance is fundamentally an egocentric process.
SC
Support for an egocentric model has strong implications for the experience of vicarious
NU
dissonance. Perspective taking has in the past been implicated in vicarious experience (Goldstein & Cialdini, 2007; Eisenberg et al., 1991; Smith, 2006), yet a clear linkage to vicarious
MA
dissonance has not been clearly established in previous literature. Consistent with the work of Epley et al (2004; Savitsky et al, 2011), the present experiments suggest that simulation, rather
TE
D
than true perspective taking, is the driving force behind vicarious dissonance. Vicarious dissonance may additionally have untapped potential as a tool for changing
AC CE P
attitudes for the better. In cases where people possess positive attitudes towards destructive behaviors, such as smoking or drug abuse, or negative attitudes towards positive behaviors, such as flossing, leveraging the behavior of an ingroup member could be a productive way to change people‘s attitudes when other more direct methods (e.g. persuasion) are ineffective. Vicarious free choice adds a new dimension in broadening the types of attitudes targetable through vicarious dissonance and the ease with which they can be changed. One might leverage vicarious free choice to get children to choose healthier brands of snacks or beverages over other attractive but unhealthy ones, for instance, where merely witnessing another ingroup member choose a healthier brand on its own might not be aversive or inconsistent enough to arouse dissonance. The knowledge that vicarious dissonance functions egocentrically—i.e., that choice and consequences involved in the dissonant act seem to matter only for the observer—is essential for
ACCEPTED MANUSCRIPT 21
EGOCENTRISM IN VICARIOUS DISSONANCE designing interventions that can be tailored to any situation. With a thorough understanding of
the mechanisms by which vicarious dissonance functions, researchers have yet one more tool to
AC CE P
TE
D
MA
NU
SC
RI
PT
effect attitude change beyond classic means such as persuasion and social influence.
ACCEPTED MANUSCRIPT 22
EGOCENTRISM IN VICARIOUS DISSONANCE Acknowledgments
AC CE P
TE
D
MA
NU
SC
RI
PT
We appreciate the help of Ms. Lauren Feldman who assisted in the data collection.
ACCEPTED MANUSCRIPT 23
EGOCENTRISM IN VICARIOUS DISSONANCE References Brehm, Jack W. ―Postdecision changes in the desirability of alternatives.‖ The Journal of
PT
Abnormal and Social Psychology 52.3 (1956): 384. Brewer, Marilynn B., and Wendi Gardner. ―Who is this ―We‖? Levels of collective identity and
RI
self representations.‖ Journal of Personality and Social Psychology 71.1 (1996): 83.
SC
Cooper, J. (2007). Cognitive dissonance: 50 years of a classic theory. London: Sage.
NU
Cooper, J., & Mackie, D. (1983). Cognitive dissonance in an intergroup context. Journal of Personality and Social Psychology, 44(3), 536-544.
MA
Eisenberg, N., Fabes, R. A., Schaller, M., Miller, P., Carlo, G., Poulin, R., & Shell, R. (1991). Personality and socialization correlates of vicarious emotional responding. Journal of
TE
D
Personality and Social Psychology, 61(3), 459. Elliot, A. J., & Devine, P. G. (1994). On the motivational nature of cognitive dissonance:
AC CE P
Dissonance as psychological discomfort. Journal of Personality and Social Psychology, 67(3), 382.
Epley, N., Keysar, B., Van Boven, L., & Gilovich, T. (2004). Perspective taking as egocentric anchoring and adjustment. Journal of Personality and Social Psychology, 87(3), 327. Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7(2), 117-140. Festinger, L. (1957). A theory of cognitive dissonance. Evanston, IL: Row, Peterson. Festinger, L., & Carlsmith, J. M. (1959). Cognitive consequences of forced compliance. The Journal of Abnormal and Social Psychology, 58(2), 203. Glasford, G. E., Dovidio, J. F., & Pratto, F. (2009). I continue to feel so good about us: In-group identification and the use of social identity-enhancing strategies to reduce intragroup dissonance. Personality and Social Psychology Bulletin, 35(4), 415-427.
ACCEPTED MANUSCRIPT 24
EGOCENTRISM IN VICARIOUS DISSONANCE Glasford, G. E., Pratto, F., & Dovidio, J. F. (2008). Intragroup dissonance: Responses to in-
group violation of personal values. Journal of Experimental Social Psychology, 44, 1057-
PT
1064.
RI
Harmon-Jones, E., & Mills, J. (eds) (1999). Cognitive dissonance: Progress on a pivotal theory in social psychology, Washington, DC: APA.
SC
Harmon-Jones, E., Harmon-Jones, C., Fearn, M., Sigelman, J. D., & Johnson, P. (2008). Left
NU
frontal cortical activation and spreading of alternatives: tests of the action-based model of dissonance. Journal of Personality and Social Psychology, 94(1), 1-15.
MA
Heider, F. (1958). The psychology of interpersonal relations. NY: Wiley. Hogg, M. A., & Smith, J. R. (2007). Attitudes in social context: A social identity perspective.
TE
D
European Journal of Social Psychology, 18(1), 89-131. Shultz, T. R., Léveillé, E., & Lepper, M. R. (1999). Free choice and cognitive dissonance
AC CE P
revisited: Choosing ―lesser evils‖ versus ―greater goods.‖ Personality and Social Psychology Bulletin, 25(1), 40-48. Liberman, N., & Förster, J. (2006). Inferences from decision difficulty. Journal of Experimental Social Psychology, 42(3), 290-301. McKimmie, B. M., Terry, D. J., & Hogg, M. A. (2009). Dissonance reduction in the context of group membership: The role of metaconsistency. Group Dynamics: Theory, Research, and Practice, 13(2), 103-119. Monin, B., Norton, M. I., Cooper, J., & Hogg, M. A. (2004). Reacting to an assumed situation vs. conforming to an assumed reaction: The role of perceived speaker attitude in vicarious dissonance. Group Processes & Intergroup Relations, 7(3), 207-220.
ACCEPTED MANUSCRIPT 25
EGOCENTRISM IN VICARIOUS DISSONANCE Newcomb, T. M., & American Council on Public Affairs. (1957). Personality and Social Change. Dryden.
PT
Norton, M. I., Monin, B., Cooper, J., & Hogg, M. A. (2003). Vicarious dissonance: Attitude
RI
change from the inconsistency of others. Journal of Personality and Social Psychology, 85(1), 47-62.
SC
Savitsky, K., Keysar, B., Epley, N., Carter, T. & Swanson, A. (2011). The closeness-
NU
communication bias: Increased egocentrism among friends versus strangers. Journal of Experimental Social Psychology, 47, 269-273.
MA
Smith, A. (2006). Cognitive empathy and emotional empathy in human behavior and evolution. Psychological Record, 56(1).
TE
D
Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. The Social Psychology of Intergroup Relations, 33(47), 74.
AC CE P
Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behavior. Psychology of Intergroup Relations, 7-24.
ACCEPTED MANUSCRIPT 26
EGOCENTRISM IN VICARIOUS DISSONANCE Footnotes
AC CE P
TE
D
MA
NU
SC
RI
PT
1. In the case where alternatives in the participant hard choice condition were tied in terms of desirability, the actor‘s ostensible choice was randomly determined.
ACCEPTED MANUSCRIPT 27
EGOCENTRISM IN VICARIOUS DISSONANCE
Figure Captions
PT
Figure 1. Attitude towards government influence in the media in Study 1 as a function of
RI
perspective taking instructions.
SC
Figure 2. Attitudes toward government influence for egocentrically primed participants as a
NU
function of choice and level of in-group identification.
MA
Figure 3. Post-decision spread of alternatives as a function of observer and actor choice
AC CE P
TE
D
difficulty.
ACCEPTED MANUSCRIPT 28
PT
EGOCENTRISM IN VICARIOUS DISSONANCE
Low Choice
6
RI
High Choice
SC NU
4
MA
3
1 No PT
TE
D
2
Other PT Egocentric PT Perspective Taking Condition
AC CE P
Attitude
5
Figure 1
ACCEPTED MANUSCRIPT 29
AC CE P
TE
D
MA
NU
SC
RI
PT
EGOCENTRISM IN VICARIOUS DISSONANCE
Figure 2
ACCEPTED MANUSCRIPT 30
EGOCENTRISM IN VICARIOUS DISSONANCE
Easy Choice
Difficult Choice
PT
1
RI
0.5
SC
0 -0.5 -1
NU
Spreading of Alternatives
1.5
MA
-1.5
AC CE P
TE
D
Actor Condition
Figure 3
Observer Condition
ACCEPTED MANUSCRIPT 31
EGOCENTRISM IN VICARIOUS DISSONANCE Highlights
AC CE P
TE
D
MA
NU
SC
RI
PT
Tested an egocentric model for the underlying process of vicarious dissonance. For the first time, showed effects of vicarious dissonance in both free choice and induced compliance. Vicarious dissonance was facilitated when participants adopted an egocentric perspective. Results show that vicarious dissonance is “about me” rather than about a group member.