Perspective Taking and Racial Bias 1
Running Head: PERSPECTIVE TAKING AND RACIAL BIAS
Perspective Taking Combats Automatic Expressions of Racial Bias
Andrew R. Todd
University of Cologne
Galen V. Bodenhausen, Jennifer A. Richeson, and Adam D. Galinsky
IN PRESS, Journal of Personality and Social Psychology
Andrew R. Todd
Universität zu Köln
Phone: +49 (0) 221 470 1212
Fax: +49 (0) 221 470 1216 0
Perspective Taking and Racial Bias 2
Five experiments investigated the hypothesis that perspective taking—actively contemplating
others’ psychological experiences—attenuates automatic expressions of racial bias. Across the
first three experiments, participants who adopted the perspective of a Black target in an initial
context subsequently exhibited more positive automatic interracial evaluations, with changes in
automatic evaluations mediating the effect of perspective taking on more deliberate interracial
evaluations. Furthermore, unlike other bias-reduction strategies, the interracial positivity
resulting from perspective taking was accompanied by increased salience of racial inequalities
(Experiment 3). Perspective taking also produced stronger approach-oriented action tendencies
toward Blacks (but not Whites; Experiment 4). A final experiment revealed that face-to-face
interactions with perspective takers were rated more positively by Black interaction partners than
were interactions with non-perspective takers, a relationship that was mediated by perspective
takers’ increased approach-oriented nonverbal behaviors (as rated by objective, third-party
observers). These findings indicate that perspective taking can combat automatic expressions of
racial biases without simultaneously decreasing sensitivity to ongoing racial disparities.
Perspective Taking and Racial Bias 3
The 20th century witnessed a dramatic shift in both the public espousal and legal
enforcement of the principle of racial equality. Indeed, survey data have revealed a substantial
decline in overt expressions of racial bias since the passage of civil rights legislation nearly 50
years ago (Schuman, Steeh, Bobo, & Krysan, 1997), prompting one researcher to claim that this
shift toward egalitarianism represents “[t]he single clearest trend in studies of racial attitudes”
(Bobo, 2001, p. 269). This collective attitudinal shift notwithstanding, the attainment of genuine
racial equality continues to be impeded by contemporary manifestations of bias—ones that are
qualitatively distinct from the “old-fashioned” racism that plagued previous generations but that
are equally capable of exerting pernicious effects. Because these biases are driven, in part, by
normal psychological processes that operate relatively automatically (Dovidio & Gaertner, 2004;
Macrae & Bodenhausen, 2000), designing strategies to combat them presents a formidable
The current research investigated the efficacy of perspective taking—the active
contemplation of others’ psychological experiences—as a strategy for counteracting automatic
expressions of racial bias. Although there is now a substantial literature attesting to the promise
of perspective taking for attenuating overt expressions of bias (Batson, Polycarpou, et al., 1997;
Dovidio et al., 2004; Galinsky & Ku, 2004; Shih, Wang, Bucher, & Stotzer, 2009; Vescio,
Sechrist, & Paolucci, 2003; Vorauer & Sasaki, 2009), little is currently known about whether
perspective taking likewise tempers the more indirect and automatic forms of racial bias that
pervade contemporary society. To fill this empirical gap, we conducted five experiments
examining the impact of perspective taking on several critical (but largely untested) intergroup
outcomes: automatic evaluations, approach–avoidance reactions, and behaviors displayed during
Perspective Taking and Racial Bias 4
Contemporary Racial Bias: Automatic Negativity and Behavioral Avoidance
The various forms that contemporary racial bias can take have been articulated in several
prominent theories (e.g., ambivalent racism, Katz & Hass, 1988; aversive racism, Dovidio &
Gaertner, 2004; modern racism, McConahay, 1986; symbolic racism, Sears & Henry, 2005).
Despite differences in their defining features and operating characteristics, these theories
generally posit that many Whites (and others) experience an inner conflict arising from
competing response tendencies toward Blacks. One set of tendencies is grounded in the
democratic principles of justice and equality and thus encourages nonbiased responses; the other
is based on an underlying, automatically activated negative affective reaction that encourages
discriminatory responses. Numerous studies have now shown that, despite a personal disavowal
of prejudice, individuals’ underlying interracial negativity often finds behavioral expression,
particularly in behaviors that are difficult to monitor and control (e.g., many nonverbal
behaviors; Dovidio & Gaertner, 2004).
Face-to-face interracial interactions provide one such context. Because the prospect of
interracial contact can be a source of anxiety and discomfort (Blascovich, Mendes, Hunter,
Lickel, & Kowai-Bell, 2001; Stephan & Stephan, 1985; Trawalter, Richeson, & Shelton, 2009),
many people try to avoid interracial contact whenever possible. Yet, oftentimes interracial
contact is unavoidable; in such cases, individuals’ underlying negativity may “leak out”
behaviorally. For instance, studies have shown that Whites who harbor negative automatic
reactions toward Blacks tend to display less nonverbal “friendliness”—fewer approach-oriented
(e.g., smiling, forward body leaning) and more avoidance-oriented (e.g., gaze aversion, increased
interpersonal distance) behaviors—during interracial interactions (e.g., Dovidio, Kawakami, &
Gaertner, 2002; Fazio, Jackson, Dunton, & Williams, 1995; McConnell & Leibold, 2001; see
Perspective Taking and Racial Bias 5
Greenwald, Poehlman, Uhlmann, & Banaji, 2009, for a meta-analytic review). Insofar as Black
interaction partners are able to detect underlying interracial negativity in Whites’ nonverbal
behaviors (see Dovidio et al., 2002; Richeson & Shelton, 2005), they, like Whites, may approach
future interracial interactions with a sense of reticence (Shelton, Dovidio, Hebl, & Richeson,
2009; Tropp, 2007). Importantly, this reticence can undermine attempts to establish the rapport
and trust that are critical to the development of positive intergroup relations.
Although social scientists have long been interested in unearthing effective strategies for
reducing intergroup bias, only recently has attention shifted to strategies targeting automatically
activated intergroup reactions. Despite a common assumption that automatic intergroup
reactions reflect highly robust mental representations that are rooted in long-term socialization
experiences (e.g., Rudman, 2004; Wilson, Lindsey, & Schooler, 2000), there is now considerable
evidence that automatic intergroup reactions are readily influenced by a variety of contextual and
psychological variables (e.g., Blair, Ma, & Lenton, 2001; Gawronski, Deutsch, Mbirkou, Seibt,
& Strack, 2008; Kawakami, Phills, Steele, & Dovidio, 2007; Olson & Fazio, 2006; Richeson &
Nussbaum, 2004; Turner & Crisp, 2010; see Gawronski & Sritharan, 2010, for a comprehensive
review). The current research sought to add to this accumulating body of research by exploring
the effects of one promising bias-reduction strategy—perspective taking—on automatically
activated expressions of racial bias.
Perspective Taking and the Attenuation of Contemporary Forms of Bias
The ability and propensity to consider others’ psychological perspectives is an invaluable
tool for inferring the contents of others’ minds and for predicting and explaining their actions.
Social theorists have long argued that a well-developed perspective-taking capacity is critical for
managing the complexities of social life (Higgins, 1981; Mead, 1934; Piaget, 1932; Smith,
Perspective Taking and Racial Bias 6
1759), with some viewing it as a critical antecedent to altruistic behavior (Batson, 1991) and to
the development of moral reasoning more generally (Selman, 1980). Its presence can promote
cooperation (Batson & Moran, 1999) and facilitate conflict resolution (Galinsky, Maddux, Gilin,
& White, 2008a). Perspective-taking deficiencies, in contrast, have been linked to severe social
dysfunction (as in the case of autism; Baron-Cohen, 1995) and to arrogant, inconsiderate, and
even aggressive styles of interpersonal responding (Richardson, Hammock, Smith, Gardner, &
Signo, 1994)—behaviors certain to add fuel to already fiery conflict situations.
Given the wide range of interpersonal benefits resulting from strategic perspective taking,
there is good reason to suspect that actively contemplating outgroup members’ psychological
perspectives could be an efficacious strategy for cultivating more positive intergroup relations.
Consistent with this supposition, there are now numerous studies attesting to the merits of
perspective taking as a strategy for reducing intergroup bias. Whereas some studies have linked
perspective taking to decreased activation and application of negative group stereotypes
(Galinsky & Moskowitz, 2000), others have shown that adopting the perspective of a particular
outgroup target leads to more positive evaluations of other individual members of the target’s
group (Shih et al., 2009) and of the target’s group as a whole (Batson, Polycarpou, et al., 1997;
Dovidio et al., 2004; Galinsky & Ku, 2004; Vescio et al., 2003; Vorauer & Sasaki, 2009).
Although these findings have greatly enhanced our understanding of the intergroup
consequences of perspective taking, this work has focused almost exclusively on overt forms of
bias (e.g., deliberate evaluations) to the exclusion of the more subtle forms of bias discussed
previously (see Galinsky & Moskowitz, 2000, for an exception). Indeed, we are not aware of
any published studies investigating the effects of perspective taking on automatic expressions of
racial bias. Very few studies, moreover, have examined the behavioral implications of
Perspective Taking and Racial Bias 7
intergroup perspective taking, and what little research there is has yielded mixed results (Blatt,
LeLacheur, Galinsky, Simmens, & Greenberg, in press; Vorauer, Martens, & Sasaki, 2009;
Vorauer & Sasaki, 2009).
Why might perspective taking engender more positive automatic interracial reactions?
Research indicates that associative representations of many, if not most, social groups contain a
mixture of both positive and negative aspects (Fiske, Cuddy, Glick, & Xu, 2002). One
implication of this representational ambivalence is that factors that highlight the positive
associations should promote more positive automatic intergroup reactions (Gawronski &
Bodenhausen, 2006). To the extent that the cognitive elaboration stemming from perspective
taking calls to mind different (and more positive) group-based associative content than might
otherwise be considered, perspective taking holds the potential to engender more favorable
automatic interracial evaluations. Furthermore, if the positive mental representations activated in
the course of perspective taking elicit correspondingly more positive spontaneous behavior
during interracial encounters, then perspective taking also holds the potential to produce more
positive interracial contact experiences. Indeed, research indicates that when a given variable
influences the activation of mental associations, there are often corresponding downstream
effects on spontaneous forms of behavior (e.g., nonverbal behavior; see Gawronski & Sritharan,
Despite the benefits accrued from altering automatic interracial evaluations and
behaviors, strategies whose primary goal is to increase interracial harmony can have unintended
consequences that limit their utility. For instance, although focusing on intergroup
commonalities has long been argued to promote more positive intergroup evaluations, focusing
solely on commonalities can limit motivation for actual social change by desensitizing people to
Perspective Taking and Racial Bias 8
the persistence of interracial disparities (Dovidio, Gaertner, & Saguy, 2009; Saguy, Tausch,
Dovidio, & Pratto, 2009). If perspective taking, which has been shown to increase perceptions
of intergroup commonality (Galinsky, Ku, & Wang, 2005; Galinsky & Moskowitz, 2000;
Galinsky, Wang, & Ku, 2008b), is susceptible to this unintended side effect, its general value as
a strategy for navigating interracial contexts could be limited. However, providing evidence that
perspective taking can produce more positive automatic interracial reactions without shrouding
interracial disparities would indicate that the benefits of perspective taking do not come with
psychological strings attached.
Overview of the Current Research
The aim of the current research was to investigate the impact of perspective taking on
automatic interracial reactions and behaviors. As noted earlier, we define perspective taking
broadly as the active contemplation of others’ psychological experiences. In each of our
experiments, we manipulated perspective taking in an ostensibly unrelated context prior to the
administration of the dependent measures. Specifically, we introduced participants to a Black
male (either via video or a photograph) and instructed them to adopt his perspective as they
watched him in a video or as they wrote a brief essay about a day in his life. Because previous
research has found important psychological differences depending on how perspective taking is
manipulated (see Batson, 2009, for a review), we included two different manipulations of
perspective taking in Experiment 1. Whereas some participants tried to imagine the target’s
perspective (perspective-taking–other), others tried to imagine their own perspective as if they
were in the target’s situation (perspective-taking–self). In the remaining experiments, we
employed only the perspective-taking–other manipulation. For comparison purposes, we
introduced other participants to the same Black male and instructed them to adopt an objective
Perspective Taking and Racial Bias 9
focus, or we provided them with no additional instructions. Thus, the current research is perhaps
most aptly described as an investigation of the effects of a perspective-taking mindset1 on
automatic expressions of racial bias.
Our first two experiments assessed the influence of perspective taking on automatic
evaluations of Black Americans versus White Americans. Because prior research has
demonstrated that factors that produce more positive intergroup evaluations can have the
unintended consequence of obscuring intergroup inequalities (Dovidio et al., 2009; Saguy et al.,
2009), Experiment 3 investigated whether perspective taking is vulnerable to this unintended
side effect. Our final two experiments explored the behavioral implications of perspective
taking. Experiment 4 explored whether changes in automatic interracial reactions following
perspective taking are target-group-specific by assessing approach–avoidance action tendencies
separately for Black and White targets. Experiment 5 examined the impact of perspective taking
on behaviors displayed during an actual interracial interaction and on interaction partners’
subjective experiences of the interaction.
In general, we predicted that perspective taking would lead to more positive automatic
interracial evaluations and action tendencies. Based on the proposition that changes in mental
representations elicit corresponding changes in behavior (see Gawronski & Sritharan, 2010), we
further predicted that perspective-taking-induced changes in automatic interracial reactions
would lead to more positive interracial interactions.
Experiment 1: Automatic Interracial Evaluations
The primary purpose of Experiment 1 was to examine the impact of perspective taking on
automatic evaluations of Black Americans relative to White Americans. Participants watched a
video depicting a series of discriminatory acts directed toward a Black man versus a White man
Perspective Taking and Racial Bias 10
(Dovidio et al., 2004; Esses & Dovidio, 2002). As they watched the video, participants either
adopted the Black man’s perspective or they attempted to remain objective and detached. We
included two different perspective-taking conditions in this experiment. Some participants tried
to imagine the Black man’s thoughts, feelings, and experiences (perspective-taking–other
condition) as they watched the video; others tried to imagine their own thoughts, feelings, and
experiences as if they were in the Black man’s situation (perspective-taking–self condition).
Because both approaches have been used in past research (e.g., Batson, Early, & Salvarini, 1997;
Davis, Conklin, Smith, & Luce, 1996; Galinsky et al., 2008b) and, in some cases, have been
found to have different psychological consequences (Batson, 2009), we wanted to explore
whether the specific form of perspective taking would qualify our results. After watching the
video, participants completed a variant of the IAT that assesses automatic evaluations of Black
Americans relative to White Americans (i.e., personalized evaluative race Implicit Association
Test [IAT]; Olson & Fazio, 2004).
If adopting the perspective of a Black target activates different (and more positive)
group-based associative content than the negative content ordinarily activated when processing a
Black exemplar (Devine, 1989), then one could reasonably expect that perspective takers would
exhibit more positive automatic interracial evaluations than would non-perspective takers.
Alternatively, it is possible that adopting the perspective of a Black target simply heightens the
motivation to be—or at least to appear to be—unbiased (Plant & Devine, 1998) and that
perspective taking, despite having benefits for self-reported interracial evaluations (Dovidio et
al., 2004; Vescio et al., 2003), exerts little effect on automatic interracial evaluations.
Perspective Taking and Racial Bias 11
Participants and design. Fifty-one undergraduates (57% female; 67% White, 33%
Asian) received $8 for participating. They were randomly assigned to one of three experimental
conditions: perspective-taking–other vs. perspective-taking–self vs. objective focus.
Procedure and materials. On arriving to the laboratory, participants were greeted by an
experimenter and led to an individual cubicle where they were asked to perform several
ostensibly unrelated experimental tasks. All tasks were administered via computer.
Perspective-taking manipulation. First, as a part of a “documentary assessment” task,
participants watched a 5-min video clip depicting a Black man (Glen) and a White man (John)
engaging in a variety of everyday activities (Dovidio et al. 2004). Participants watched as the
two men received differential treatment while browsing in a department store, attempting to
purchase an automobile at a car dealership, and interacting with local police. It was clear from
the content of the video that Glen was treated unfairly because of his race.
Before watching the video clip, participants received one of three sets of instructions.
Participants in both perspective-taking conditions were asked to take Glen's (the Black man's)
perspective. Participants assigned to the perspective-taking–other condition received additional
instructions urging them to visualize clearly and vividly what Glen might be thinking, feeling,
and experiencing as he goes through the various activities depicted in the documentary.
Participants assigned to the perspective-taking–self condition, on the other hand, were asked to
imagine what they might be thinking, feeling, and experiencing if they were Glen, looking at the
world through his eyes and walking in his shoes as he goes through the various activities
depicted in the documentary. Finally, participants assigned to the objective-focus condition were
asked to remain objective and emotionally detached as they watched the video—to not let
Perspective Taking and Racial Bias 12
themselves get caught up in imagining what the men might be thinking, feeling, and
experiencing as the men go through the various activities depicted in the documentary.
Personalized evaluative race IAT. Next, as part of a “speeded categorization” task,
participants completed a personalized evaluative race IAT (Olson & Fazio, 2004), which
consisted of seven blocks of trials. In the first block (20 trials), participants assigned four facial
images of Black people (two male, two female) to the category African American (left-hand key)
and four facial images of White people (two male, two female) to the category European
American (right-hand key). In the second block (20 trials), participants assigned 10 normatively
positive words (e.g., honesty, love, vacation) and 10 normatively negative words (e.g., cancer,
failure, vomit) to the categories I Like (left-hand key) and I Dislike (right-hand key). The third
(20 trials) and fourth blocks (40 trials) consisted of a combination of the first two blocks.
Specifically, participants pressed the left-hand key whenever an image of a Black person or a
disliked word appeared and a right-hand key whenever an image of a White person or a liked
word appeared. In the fifth block (40 trials), the initial target-concept discrimination completed
in the first block was repeated but with the categorization keys switched. The sixth (20 trials)
and seventh blocks (40 trials) consisted of reversed versions of the third and fourth blocks (i.e.,
left-hand key for images of White people and disliked words, right-hand key for images of Black
people and liked words).
Before each block of trials, participants received brief instructions and were urged to
respond as quickly as possible. No error feedback was provided (see Olson & Fazio, 2004). An
inter-trial interval of 250 ms followed each response. We counterbalanced the order of the
experimental blocks across participants and randomized the order of the trials within each block
Perspective Taking and Racial Bias 13
for each participant. Preliminary analyses revealed no effects of block order; therefore, we
collapsed across this factor in the analyses reported below.
Manipulation check. Finally, participants completed three manipulation check items
assessing the orientation they adopted while watching the video: “To what extent did you try to
imagine what Glen might be thinking, feeling, and experiencing?” “To what extent did you try
imagine what you might be thinking, feeling, and experiencing if you were Glen?” and “To what
extent did you try to be objective and emotionally detached?” These ratings were made on 7-
point scales (0 = not at all, 6 = very much so).
Results and Discussion
Preliminary analyses in each experiment always included participant gender and
participant ethnicity. We retained these variables in the reported analyses as covariates when
significant effects emerged; otherwise, we collapsed the data across these variables.
Manipulation check. Inspection of the manipulation check items revealed that
participants in both the perspective-taking–other (M = 4.75, SD = .78) and perspective-taking–
self conditions (M = 4.95, SD = .91) reported imagining Glen’s thoughts, feelings, and
experiences more than did objective-focus participants (M = 3.56, SD = 1.75), ts ≥ 2.48, ps ≤
.022, ds ≥ 1.09, whereas the two perspective-taking conditions did not differ from each other (t <
1, p > .49, d = .24). Participants in the perspective-taking–other (M = 4.69, SD = 1.01) and
perspective-taking–self conditions (M = 4.63, SD = 1.26) also reported imagining what they
might be thinking, feeling, and experiencing if they were Glen more than objective-focus
participants did (M = 3.69, SD = 1.40), ts ≥ 2.25, ps ≤ .029, d ≥ .65, whereas the two perspective-
taking conditions did not differ from each other (t < 1, p > .89, d = .04). Finally, objective-focus
participants (M = 4.00, SD = 1.27) reported trying to be more objective and emotionally detached
Perspective Taking and Racial Bias 14
than did participants in the perspective-taking–other (M = 2.50, SD = 1.97) and perspective-
taking–self conditions (M = 2.63, SD = 1.57), ts ≥ 2.49, ps ≤ .029, d ≥ .72, who did not differ
from each other (t < 1, p > .81, d = .07). Overall, the effect of instruction set was significant for
all three items in separate one-way analyses of variance (ANOVAs), Fs ≥ 3.39, ps ≤ .042, ηp2 ≥
.12. Thus, it appears that our two sets of perspective-taking instructions had largely comparable
Automatic interracial evaluations. We computed IAT scores using the scoring
algorithm developed by Greenwald, Nosek, and Banaji (2003), with higher D-scores reflecting
an automatic preference for Whites over Blacks (i.e., pro-White bias). Because the two
perspective-taking conditions were virtually indistinguishable from each other on the
manipulation check items, we examined our hypotheses by conducting two planned contrasts
(Rosenthal, Rosnow, & Rubin, 2000): The first contrast compared the two perspective-taking
conditions to the objective-focus condition; the second contrast compared the two perspective-
taking conditions to each other (see Davis et al., 1996, for a similar analytical approach). We
also report the omnibus ANOVA.
If perspective taking encourages less negative (more positive) automatic evaluations of
Black Americans relative to White Americans, then one would expect lower IAT scores in the
two perspective-taking conditions than in the objective-focus condition. The critical contrast
testing our primary hypothesis revealed that participants in both the perspective-taking–other (M
= .32, SD = .59) and perspective-taking–self conditions (M = .43, SD = .41) exhibited
significantly weaker pro-White bias than did objective-focus participants (M = .80, SD = .37),
t(48) = 3.06, p = .004, d = .88, whereas the two perspective-taking conditions did not differ from
Perspective Taking and Racial Bias 15
each other, t(48) = 1.07, p = .47, d = .21. Overall, the effect of instruction set was significant in a
one-way ANOVA, F(2, 48) = 4.84, p = .01, ηp2 = .17.
These findings provide initial support for our contention that adopting the perspective of
a Black target in one context can engender more favorable automatically activated interracial
evaluations in a subsequent context. Furthermore, we found no differences between the two
perspective-taking conditions, a pattern that was confirmed by the results of the manipulation
check, which indicated that participants did not distinguish between the two perspective-taking
instruction sets. Although some previous research has demonstrated important emotional,
cognitive, motivational, and neurophysiological differences when comparing these two
perspective-taking conditions (Batson, 2009), numerous other studies have observed null effects
(Davis et al., 1996, Experiment 1; Davis et al., 2004, Experiment 2; Finlay & Stephan, 2000;
Galinsky et al., 2008b, Experiment 2a). Batson (2009) has argued that null effects are especially
likely when participants have very limited information about the target whose perspective they
are asked to adopt, though this was not necessarily the case in Experiment 1 or in Finlay and
Stephan’s (2000) study. Nevertheless, because participants in our remaining experiments
received very little information about the perspective-taking target, we dropped the perspective-
taking–self condition from these experiments.
Experiment 2: Automatic Interracial Evaluations Redux
The goal of Experiment 2 was to replicate and extend the results from Experiment 1
using a different induction of perspective taking. Instead of watching a video depicting racial
discrimination, participants received a photograph of a young Black male and wrote an essay
about a day in his life (Galinsky & Moskowitz, 2000; Macrae, Bodenhausen, Milne, & Jetten,
1994). In this way, and unlike Experiment 1, participants were unconstrained in the context in
Perspective Taking and Racial Bias 16
which they chose to imagine the target and the manner in which they described him. As they
wrote their essays, participants either imagined the target person’s thoughts, feeling, and
experiences or they remained objective and detached. Afterwards, participants again completed
a personalized evaluative race IAT (Olson & Fazio, 2004).
Participants and design. Thirty-eight undergraduates (79% female; 58% White, 32%
Asian, 10% Latino[a]) received either partial course credit or $7 for participating. They were
randomly assigned to one of two experimental conditions: perspective taking vs. objective focus.
Procedure and materials. On arrival at the lab, participants were led to an individual
cubicle where they were asked to perform several ostensibly unrelated experimental tasks. All
tasks were administered via computer.
Perspective-taking manipulation. First, as part of a linguistic task investigating “how
people construct life event details from visual information,” participants wrote a short narrative
essay about a randomly-selected person whom they had never met. To emphasize the seemingly
random selection of the target, we presented participants with 8 different numbered boxes, each
of which ostensibly corresponded to a specific individual. After clicking on one of the boxes, all
participants saw a photograph of the same target person (a young Black man) along with
instructions to spend about 5 min writing about a day in his life. Participants in the perspective-
taking condition received additional instructions that were similar to the perspective-taking–
other instructions from Experiment 1. Participants in the objective-focus condition received
additional instructions that were modeled after those from Experiment 1.
Personalized evaluative race IAT. Next, participants completed a personalized
evaluative race IAT (Olson & Fazio, 2004) that was similar to the one used in Experiment 1,
Perspective Taking and Racial Bias 17
with the following exceptions. First, we changed the number of trials in each block: The first,
second, third, and sixth blocks contained 24 (rather than 20) trials, and the fourth, fifth, and
seventh blocks contained 48 (rather than 40) trials. Second, instead of using normatively
positive and negative words, we included 12 entities (e.g., coffee, football, tequila) shown in
previous research to have no clear normative evaluation but a large degree of variability in
personal evaluation (see Olson & Fazio, 2004). Finally, because there were no effects of block
order in Experiment 1 and because our primary interest was to examine relative differences in
associative evaluations as a function of instruction set rather than the absolute magnitude of
associations per se, we did not counterbalance the order of the critical trial blocks.2
Results and Discussion
As in Experiment 1, we computed IAT scores using the Greenwald et al. (2003) scoring
algorithm, with higher D-scores reflecting an automatic preference for Whites over Blacks (i.e.,
pro-White bias). Once again, if perspective taking encourages less negative (more positive)
automatic evaluations of Black Americans versus White Americans, then we should observe a
less pronounced IAT effect among perspective takers than objective-focus participants. As
expected, perspective takers (M = .01, SD = .52) exhibited significantly weaker pro-White bias
than did objective-focus participants (M = .49, SD = .70), t(36) = 2.39, p = .02, d = .78.
These findings provide additional support for our contention that perspective taking can
attenuate the automatic interracial negativity that characterizes contemporary racial bias. Taken
together, the results of Experiments 1 and 2 are consistent with previous research documenting
decreases in self-reported prejudicial attitudes following perspective taking (Batson, Polycarpou,
et al., 1997; Dovidio et al., 2004; Galinsky & Ku, 2004; Vescio et al., 2003; Vorauer & Sasaki,
Perspective Taking and Racial Bias 18
2009) and suggest that the effects of perspective taking on deliberate interracial evaluations are
not fully explained by an increased motivation to appear unbiased to oneself or others.
Experiment 3: Automatic and Deliberate Interracial Evaluations and Perceptions of
Experiment 3 sought to replicate these findings using a different measure of automatic
interracial evaluations. Another goal of Experiment 3 was to determine what other types of race-
related associations are automatically activated following perspective taking. Galinsky and
Moskowitz (2000) discovered that one mechanism through which perspective taking reduces
intergroup bias is by increasing perceptions of commonality between the self and the target of
perspective taking (and other members of the target’s group). However, studies have shown that
heightened perceptions of intergroup commonality, although beneficial for reducing intergroup
prejudice, can inadvertently cause perceivers to overlook and underestimate intergroup
inequalities (Dovidio et al., 2009; Saguy et al., 2009). Thus, even though perspective taking
appears to be an effective strategy for reducing automatic prejudice, it might simultaneously
reduce acknowledgment of intergroup inequalities, which would raise concerns about its general
utility as a strategy for improving intergroup relations and motivating social change.
To examine these issues, we first had participants complete the narrative essay task from
Experiment 2. As before, some participants received perspective-taking instructions, whereas
others wrote their essays without any additional instructions. Including this control condition
allowed us to determine whether the results obtained in the first two experiments reflect the
benefits of perspective taking or the detriments of an objective focus.
After writing their essays, participants completed two IATs, one of which was
conceptually similar to the personalized evaluative race IAT used in Experiments 1 and 2—it
Perspective Taking and Racial Bias 19
assessed automatic evaluations of Black Americans relative to White Americans (standard
evaluative race IAT; Greenwald et al., 1998). Because there are controversies surrounding
which IAT is optimal for assessing automatic evaluations (e.g., Gawronski, Peters, & LeBel,
2008; Han, Czellar, Olson, & Fazio, 2010; Han, Olson, & Fazio, 2006; Nosek & Hansen, 2008a,
2008b; Olson & Fazio, 2004; Olson, Fazio, & Han, 2009), Experiment 3 utilized the standard
IAT to extend the findings from Experiments 1 and 2. Although we expected the effects of
perspective taking on the two IAT variants to be comparable, it is ultimately an empirical
The second IAT was designed to capture automatic tendencies to perceive interracial
inequalities. More specifically, this latter IAT assessed automatically activated associations
between Black Americans (vs. White Americans) and oppression-related (vs. privilege-related)
concepts (racial oppression IAT; Uhlmann, Brescoll, & Paluck, 2006). If perspective taking
produces more favorable automatic interracial evaluations without shrouding the existence of
racial disparities, then perspective takers should show a weaker association between Black
Americans and negative concepts, coupled with a stronger association between Black Americans
and oppression-related concepts.
Finally, we assessed deliberate intergroup evaluations using a set of feeling thermometer
items. Participants reported their feelings of warmth versus coldness toward each of several
different racial/ethnic groups (including Blacks and Whites).
Participants and design. Fifty-six undergraduates (54% female; 71% White, 21%
Asian, 4% Latino[a], 4% mixed or other races/ethnicities) received $8 for participating. They
were randomly assigned to one of two experimental conditions: perspective taking vs. control.
Perspective Taking and Racial Bias 20
Procedure and materials. On arrival at the lab, participants were led to an individual
cubicle and were asked to perform several ostensibly unrelated experimental tasks. All tasks
were administered via computer.
Perspective-taking manipulation. First, participants composed a brief essay about a day
in the life of a photographed Black male as in Experiment 2, following one of two sets of
instructions. Participants in the perspective-taking condition received the same perspective-
taking instructions from Experiment 2, whereas participants in the control condition simply
wrote about the person in the photograph without any additional instructions.
Evaluative race IAT. Next, participants completed two separate IATs (order was
counterbalanced across participants). The standard evaluative race IAT (Greenwald et al., 1998)
assessed the degree to which participants automatically associate Blacks and Whites with
positivity versus negativity. This IAT was nearly identical to the personalized evaluative race
IAT employed in Experiments 1 and 2, except that we changed the categories I Like and I
Dislike to Good and Bad, respectively. The stimuli consisted of the same 8 facial images of
Black and White men and women and the same 10 positive and 10 negative words used in
Racial oppression IAT. The racial oppression IAT (Uhlmann et al., 2006) assessed
automatic associations of Blacks and Whites with oppression versus privilege. In this IAT,
participants assigned Black and White facial images and oppression-related (e.g., victimized,
mistreated, exploited) and privilege-related words (e.g., advantaged, dominant, powerful) to the
categories African American, European American, Oppressed, and Privileged, respectively. In
all other respects, the IATs were identical both to each other and to the IATs used previously.
Perspective Taking and Racial Bias 21
Feeling thermometers. Lastly, participants completed items assessing the degree of
warmth versus coldness they felt toward four different racial/ethnic groups: Whites, Blacks,
Asians, and Latino(a)s. Participants were asked to focus on their feelings toward each group and
to provide their ratings on 7-point scales (0 = very cold, 6 = very warm).
Results and Discussion
Automatically activated associations. Once again, we computed IAT scores using the
Greenwald et al. (2003) scoring algorithm. For the evaluative race IAT, higher D-scores reflect
an automatic preference for Whites over Blacks (i.e., pro-White bias); for the racial oppression
IAT, higher D-scores reflect stronger Black–oppressed (White–privileged) associations.
If perspective taking encourages less negative (more positive) automatic evaluations of
Black Americans versus White Americans, then we should observe lower scores on the
evaluative race IAT among perspective takers than control participants. Similarly, if perspective
taking heightens (implicit) recognition of racial inequality, then we should observe higher scores
on the racial oppression IAT among perspective takers than controls. As expected and displayed
in Figure 1, perspective takers exhibited weaker pro-White bias on the evaluative race IAT, t(54)
= 2.01, p < .05, d = .55, and stronger Black–oppressed (White–privileged) associations on the
racial oppression IAT, t(54) = 2.03, p < .05, d = .55, than did control participants.
Deliberate intergroup evaluations. If perspective taking engenders more positive
deliberate evaluations of Black Americans, then feelings of warmth toward Blacks as a group
should be greater for perspective takers than control participants. Furthermore, if the effects of
perspective taking are target-group-specific, then feelings of warmth toward the other groups
should not differ as a function of instruction set. As expected, perspective takers (M = 4.52, SD
= 1.24) reported stronger feelings of warmth toward Blacks as a group than did control
Perspective Taking and Racial Bias 22
participants (M = 3.88, SD = 1.09), t(54) = 2.02, p < .05, d = .55, whereas feelings of warmth
toward Whites, Latino(a)s, and Asians did not differ for perspective takers and control
participants (|t|s < 1.44, ps > .15, |d|s < .43).
Mediation analyses. We next conducted several mediation analyses to examine the
underlying relationship between perspective taking and changes in automatic and deliberate
interracial evaluations. According to the associative–propositional evaluation (APE) model
(Gawronski & Bodenhausen, 2006), several different meditational patterns could be expected.
First, perspective taking could exert a direct effect on automatic interracial evaluations, which
then exert an indirect effect on deliberate evaluations of Blacks. Second, perspective taking
could exert a direct effect on deliberate evaluations, which could exert an indirect effect on
automatic interracial evaluations. Third, perspective taking could exert both direct and indirect
effects on both automatic and deliberate evaluations (see Gawronski & Bodenhausen, 2006, for a
more detailed discussion of these meditational patterns).
Using the bootstrapping procedures advocated by Shrout and Bolger (2002) and the SPSS
macros created by Preacher and Hayes (2008), we first assessed whether the effect of instruction
set (0 = control, 1 = perspective taking) on deliberate evaluations (i.e., feeling thermometer
ratings) of Blacks was mediated by automatic interracial evaluations (i.e., evaluative race IAT
scores). As displayed in Figure 2, results revealed a significant direct effect of automatic
evaluations on deliberate evaluations (t = -1.93, p = .06). When controlling for this effect, the
effect of instruction set on deliberate evaluations was no longer significant (t = 1.48, p = .14).
We also observed a 95% confidence interval around the indirect effect of automatic evaluations
ranging from .0138 to .4727, indicating significant mediation (p < .05).
Perspective Taking and Racial Bias 23
Next, we examined the reverse meditational pattern—that deliberate evaluations of
Blacks mediated the relationship between perspective taking and automatic interracial
evaluations. This analysis yielded no evidence for mediation; the 95% confidence interval
around the indirect effect included zero (-.1390 to .0087).
Finally, we tested whether automatic perceptions of intergroup inequality (i.e., racial
oppression IAT scores) mediated the relationship between perspective taking and deliberate
evaluations of Blacks. Results failed to reveal a direct effect of the mediating variable on the
outcome variable (t = 1.28, p > .22).
These findings are notable for several reasons. First, we replicated the findings from
Experiments 1 and 2 using a different measure of automatic evaluations. Second, we replicated
previous research showing that perspective taking leads to more positive deliberate evaluations
of African Americans as a group (e.g., Dovidio et al., 2004; Vescio et al., 2003). Third, we
demonstrated that changes in automatic evaluations mediated changes in deliberate evaluations.
Taken together, the results of Experiment 3 suggest that perspective taking might create a
complex web of automatic interracial associations. On the one hand, perspective taking
strengthened automatic associations between Blacks (Whites) and general positivity (negativity).
On the other hand, it strengthened automatic associations between Blacks (Whites) and concepts
related to oppression and disadvantage (power and privilege), which are clearly negative in
valence. Arguably, the most important question is which of these seemingly inconsistent
associations exerts a more pronounced effect on behavior. One way to explore this question is to
examine the effect of perspective taking on basic approach–avoidance action tendencies.
Experiment 4: Approach–Avoidance Action Tendencies
Perspective Taking and Racial Bias 24
One of the most basic functions of attitudes is to provide an orienting framework for
interactions with the social environment, with favorable evaluations leading to engagement with
a stimulus and negative ones to disengagement (e.g., Eaton, Majka, & Visser, 2008).
Accordingly, extensive research has confirmed that approach-related motor responses (e.g.,
pulling an object toward oneself) are facilitated when people have a positive evaluation of a
particular entity, whereas avoidance-related responses (e.g., pushing an object away from
oneself) are facilitated when people harbor a negative evaluation of that entity (e.g., Chen &
Bargh, 1999; Duckworth, Bargh, Garcia, & Chaiken, 2002; Eder & Rothermund, 2008; for a
review, see Neumann, Förster, & Strack, 2003). The goal of Experiment 4 was to determine
whether perspective taking might affect basic interracial approach–avoidance tendencies.
We investigated this possibility by having participants complete the same narrative essay
task (with perspective-taking versus control instructions) used in Experiment 3, after which they
completed a motor task that involved moving a gaming joystick either toward (approach) or
away from (avoidance) themselves in response to facial images of Black people, White people,
and inanimate objects (i.e., pieces of furniture). Because participants were required to respond to
faces of both races using the same motor response (and to pieces of furniture using the opposite
motor response) within the same block of trials, this task allowed us to assess approach–
avoidance reactions to Blacks and Whites separately. After completing the joystick task,
participants were asked to help a different research assistant with a separate, unrelated task in a
different room in the lab. In preparation for this task, participants were instructed to set up two
chairs (one for themselves, the other for the research assistant), with the distance between the
chairs serving as a second measure of automatic approach–avoidance reactions (e.g., Kawakami
et al., 2007; Macrae et al., 1994). Depending on condition, participants were informed that the
Perspective Taking and Racial Bias 25
research assistant’s name was either “Jake” (a stereotypically White name) or “Tyrone” (a
stereotypically Black name), which again allowed us to assess approach–avoidance reactions to
Blacks and Whites separately.
Participants and design. Seventy-one undergraduates (58% female; 39% White, 54%
Asian, 3% mixed or other races/ethnicities) received $8 for participating. They were randomly
assigned to one of four conditions in a 2 (instruction set: perspective taking vs. control) × 2 (race
of research assistant interviewer: White vs. Black) between-participants design.
Procedure and materials. Participants arrived to the lab individually and were greeted
by an experimenter who led them to an individual cubicle where they were asked to perform
several ostensibly unrelated experimental tasks. The first task was same the narrative essay task
used in Experiments 2 and 3. Participants wrote a narrative essay about a young Black male who
appeared in a photograph, following either perspective-taking or control instructions.
Approach–avoidance joystick task. Next, as part of “motor task” investigating “how
quickly people can make different motor movements in response to different stimuli,”
participants responded to different images either by pulling the joystick toward themselves or
pushing it away from themselves. The images consisted of the same 8 facial images of Blacks
and Whites from Experiments 1–3 as well as 8 images of different pieces of furniture (e.g., table,
chair, sofa), each of which appeared one-by-one in the middle of the screen.
The task consisted of two experimental blocks of 80 trials each, both of which were
preceded by blocks of practice trials (16 preceding the first experimental block, 32 preceding the
second block; see Nosek et al., 2007). In one block of experimental trials (approach faces/avoid
furniture), participants were asked to gently pull the joystick toward themselves as quickly as
Perspective Taking and Racial Bias 26
possible if the image is a face and to gently push the joystick away from themselves as quickly
as possible if the image is a piece of furniture. In the other block of experimental trials (avoid
faces/approach furniture), participants received the opposite instructions—to push the joystick if
the image is a face and to pull the joystick if the image is a piece of furniture. Critically, there
was no mention of race or the terms approach and avoid at any point during the task.
Each trial began with a fixation cross presented for 1000 ms, followed by a single image
(either a face or a piece of furniture) that remained on the screen until participants moved the
joystick in the appropriate direction. Incorrect responses were accompanied by a red “X”, which
appeared in the center of the screen until participants made the correct movement. An inter-trial
interval of 250 ms followed both correct and incorrect responses. We randomized the order of
trials within each trial block for each participant, and we counterbalanced the order of the trial
blocks across participants. Preliminary analyses indicated that block order did not moderate the
results; thus, we collapsed across this variable in the analyses reported below.
Seating distance task. Upon completing the “motor task,” participants were informed by
the computer that the experiment had ended. As participants exited their cubicle, the
experimenter casually asked participants if they would remain in the lab for another 5 min to help
a different research assistant with an unrelated task. The experimenter explained that this other
research assistant (either “Jake” or “Tyrone,” depending on condition) would soon be starting a
new experiment in which he would be interviewing students about their experiences in college. It
was further explained that Jake/Tyrone needed to practice his interviewing skills before he could
begin the new experiment. All participants agreed to the interview, at which point the
experimenter led them to an adjacent room in the lab.
Perspective Taking and Racial Bias 27
On entering the room, the experimenter (pointing to a stack of chairs in the corner of the
room) instructed participants to remove two chairs from the stack, to set them up across from
each other, and to have a seat. The experimenter then left the room, presumably to get
Jake/Tyrone. A few seconds later, the experimenter returned, informed participants that the
experiment had ended, and asked participants to complete a final questionnaire in their original
cubicle. Contained within this questionnaire were two manipulation check items asking
participants to recall the name of the interviewer and to guess the race/ethnicity of the
interviewer. As participants completed the questionnaire, the experimenter measured (to the
nearest quarter-inch) the distance between the two chairs. During debriefing, no participant
voiced suspicions that the interview was related to the other experimental tasks.
Results and Discussion
Approach–avoidance reactions. After eliminating incorrect responses (3%) and
response latencies < 300 ms and > 1500 ms (< 1%), we subjected the remaining latencies to a
log-transformation prior to analysis (Chen & Bargh, 1999). For interpretive ease, descriptive
statistics are reported in milliseconds. Mean response latencies and standard deviations for the
different conditions appear in Table 1.
If perspective taking elicits more favorable interracial approach–avoidance reactions, then
perspective takers should be faster to approach Black targets and slower to avoid Black targets
relative to control participants. Furthermore, if the effects of perspective taking are specific to
Black targets, then approach–avoidance reactions to White targets and inanimate objects should
not differ as a function of instruction set. A 2 (instruction set: perspective taking vs. control) × 2
(movement: approach vs. avoidance) × 3 (target: Black people vs. White people vs. inanimate
objects) mixed ANOVA, with repeated measures on the last two factors, revealed the critical
Perspective Taking and Racial Bias 28
three-way interaction, F(2, 138) = 3.62, p = .03, ηp2 = .05. This analysis also yielded main effects
for movement, F(1, 69) = 27.96, p < .001, ηp2 = .29, and target, F(2, 138) = 12.27, p < .001, ηp2 =
.14, indicating that approach reactions were faster than avoidance reactions and that reactions to
the facial images were faster than reactions to furniture, respectively.
To specify the critical three-way interaction in terms of the current hypotheses, we
calculated indices of approach-oriented action tendencies by subtracting participants’ approach
latencies from their avoidance latencies for each of the targets. As expected and displayed in
Figure 3, perspective takers exhibited stronger approach reactions for Black targets than did
control participants, t(69) = 2.78, p < .01, d = .67, whereas approach reactions for the White
targets and inanimate objects did not differ between perspective takers and control participants
(|t|s < 1, ps > .37, |d|s < .22).
Seating distance. Due to time constraints in several experimental sessions, four
participants were unable to complete the seating distance task. We also excluded data from two
other participants for failing to correctly identify the intended race of the interviewer, leaving 65
participants for the analyses reported below.
If perspective taking engenders stronger approach-oriented action tendencies toward
Black Americans, then seating distances from the ostensibly Black interviewer (“Tyrone”) should
be closer for perspective takers than for control participants. Moreover, if the effects of
perspective taking are specific to Black targets, seating distances from the ostensibly White
interviewer (“Jake”) should not differ as a function of instruction set. As expected and displayed
in Figure 4, perspective takers sat closer to the Black interviewer than did control participants,
t(32) = 2.15, p = .04, d = .76, whereas seating distances from the White interviewer did not differ
for perspective takers and control participants (t < 1, p > .41, d = .30). Furthermore, whereas
Perspective Taking and Racial Bias 29
control participants sat closer to the White interviewer than to the Black interviewer, t(31) = 2.00,
p = .05, d = .72, perspective takers sat non-significantly closer to the Black interviewer than to
the White interviewer, t(30) = 1.12, p = .27, d = .41. This pattern of means produced a
significant instruction set by interviewer race interaction, F(1, 61) = 4.86, p = .03, ηp2 = .07.
Employing two different measures of approach–avoidance action tendencies, the results
of Experiment 4 indicate that adopting the perspective of a Black male target in one context
strengthened automatic approach reactions toward Blacks as a group and encouraged stronger
behavioral approach tendencies toward a different Black target in a subsequent context. These
findings are consistent with our prediction—derived from the attitudes literature (Neumann et al.,
2003)—that the changes in automatic interracial evaluations found in Experiments 1–3 should
translate to stronger approach reactions and weaker avoidance reactions to Black Americans.
Furthermore, the effects of perspective taking were specific to Black targets (see also Batson,
Polycarpou, et al., 1997; Shih et al., 2009; Vescio et al., 2003), suggesting that the results of the
first three experiments are likely to reflect increased positivity toward Black Americans rather
than increased negativity toward White Americans.
Thus far, we have primarily examined the impact of perspective taking on intrapersonal
dependent variables, albeit ones that should theoretically have noteworthy interpersonal
consequences. Although showing marked differences on these outcomes is valuable in its own
right, a critical litmus test of any bias-reduction strategy is whether it produces positive changes
in actual behavior during encounters with members of the targeted group.
Experiment 5: Interracial Contact Experiences
The goal of our final experiment was to more directly investigate the interpersonal
consequences of perspective taking. More specifically, we examined its effects on behaviors
Perspective Taking and Racial Bias 30
displayed during an interracial interaction. Participants first completed the narrative essay task
used in the previous experiments; some participants received perspective-taking instructions,
others received objective-focus instructions, and remaining participants received no additional
instructions (i.e., no-instruction control condition). Inclusion of these two control conditions
afforded an assessment of how participants behave both in the absence of perspective taking as
well as under default circumstances. Afterwards, rather than completing a response-latency
measure of automatic interracial reactions as in the first four experiments, participants instead
engaged in a brief, unexpected interracial interaction with one of two Black female
experimenters, who later provided her perceptions of the positivity of the interaction. These
interactions were videotaped and later coded by trained judges, who rated participants’ nonverbal
behaviors along several dimensions indicative of approach versus avoidance (e.g., smiling, eye
contact, body posture; Andersen, 1985).
Based on the results of the first four experiments, one could reasonably predict that
positive changes in automatic evaluations and approach–avoidance reactions might provide the
impetus for more positive spontaneous behavior during intergroup interactions (see Gawronski &
Sritharan, 2010). In contrast to this view, Vorauer and colleagues (2009) have argued that
perspective taking, despite its bias-reducing potential in non-interaction contexts, can disrupt
displays of positive behavior when enacted during an actual interaction. Specifically, Vorauer et
al. (2009, Experiment 4) found that Aboriginal Canadian students reported less positive affect
after face-to-face interactions with White Canadian students who had (versus had not) actively
taken their perspective during the interaction. Interestingly, this was only true for interactions
with lower-prejudiced White participants; perspective taking produced non-significantly positive
effects for interactions with higher-prejudiced participants. Thus, contrary to the current
Perspective Taking and Racial Bias 31
hypotheses, Vorauer et al.’s (2009) work suggests that perspective taking may, at best, have
small effects during actual interracial interactions and might even backfire, resulting in more
negative contact experiences.
Participants and design. Forty-nine undergraduates (67% female; 77% White, 17%
Asian, 6% Latino[a]) received partial course credit for their participation. They were randomly
assigned to one of three experimental conditions: perspective taking vs. objective focus vs. no-
Procedure and materials. Participants arrived to the lab individually for a study
investigating “the dynamics of interpersonal interactions.” On arrival, participants were greeted
by a White male experimenter (blind to experimental condition) and led to an individual room in
the laboratory where they learned that they would be completing several unrelated tasks. These
included a linguistic task that involved writing a short essay, which was to be followed by a brief
discussion with another student on a “to-be-determined” topic. Participants also learned that
after the discussion with the other student they would answer a few questions prepared by their
introductory psychology instructors. Before leaving the room, the White experimenter informed
participants that another experimenter would assist them during the remainder of the
experimental session. At this point, participants had received no information regarding the race
of the other experimenter. During the linguistic task, participants composed a narrative essay
about a photographed Black male target, following either perspective-taking, objective-focus, or
After completing their essays, participants were greeted by one of two Black female
experimenters (both blind to experimental condition and hypotheses) and led to another room in
Perspective Taking and Racial Bias 32
the laboratory for the interaction (see McConnell & Leibold, 2001, for a similar procedure). This
room was equipped with a table and two chairs. In addition (and unbeknownst to participants), a
hidden video camera was positioned to record participants as they interacted with the
experimenter. Upon entering the room, the Black experimenter informed participants that the
other student with whom they would have their discussion had arrived late and was still
completing the first task. In actuality, there was no other student. The Black experimenter then
informed participants that, so as not to waste their time, they could answer the questions
prepared by the introductory psychology instructors as they were waiting for the other student to
finish the linguistic task. After assuring participants that their responses would remain
confidential, the Black experimenter asked a series of mundane questions about their
introductory psychology course (e.g., “What are your favorite and least favorite aspects about the
class?”), pausing between each question and pretending to record participants’ responses on a
notepad. The interactions lasted for approximately 3 min.
After the interaction, the Black experimenter excused herself to “go check on the other
participant.” In actuality, she reported her subjective experiences of the interaction (see Black
experimenter ratings below). At this point, the original experimenter returned, informed
participants that the experiment had ended, and requested permission to analyze their videotapes.
Black experimenter ratings. Immediately following the interaction, the Black
experimenters completed items assessing (a) their perceptions of participants’ behaviors (i.e.,
friendly, pleasant, likeable, engaged, relaxed, cold, curt, tense, uncomfortable; α = .94; see
Dovidio et al., 2002) and (b) their subjective enjoyment of the interaction (i.e., enjoyable,
awkward, comfortable; α = .80; see Apfelbaum & Sommers, 2009). All ratings were made on 9-
point scales (0 = not at all, 8 = very much so) and were reverse-coded where applicable.
Perspective Taking and Racial Bias 33
Nonverbal video coding. Several weeks after data collection, we extracted a 30-sec
video clip from the middle of each interaction (Ambady, Bernieri, & Richeson, 2000). We then
edited each clip so that only the participant was visible, thereby concealing the interracial
composition of the interaction. Video clips for two participants were lost due to a recording
malfunction, and an additional participant requested that his videotape be erased. Therefore,
analyses are based on the 46 remaining participants. Three female coders (two Black, one White)
who were blind to experimental condition and hypotheses viewed the videos without sound and
rated for the presence of approach-oriented action tendencies, which were operationalized here in
terms of several spontaneous nonverbal behaviors reflecting an approach orientation (i.e.,
smiling, eye contact, leaning toward [vs. away from] experimenter, fidgeting; see Andersen,
1985). All items were rated on 9-point scales (reverse-coded where applicable), with endpoints
tailored to the particular behavior (e.g., 0 = not smiling at all, 8 = smiling a lot). The judges’
ratings were sufficiently reliable (intraclass r = .68) and, thus, their ratings were combined (α =
.82), with higher scores reflecting more approach-oriented nonverbal behaviors.
Results and Discussion
Preliminary analyses revealed main effects of participant gender on the Black
experimenters’ ratings, indicating more positive ratings for female than male participants; thus,
we retained participant gender as a covariate in those analyses. Additional analyses revealed no
differences as a function of the particular Black experimenter (#1 vs. #2) or participant ethnicity,
so we collapsed across these variables.
Because previous research has observed null effects between objective-focus and control
conditions (e.g., Dovidio et al., 2004), we examined our hypotheses by conducting a planned
contrast (Rosenthal et al., 2000) comparing the perspective-taking condition to both the
Perspective Taking and Racial Bias 34
objective-focus and no-instruction control conditions (see Dovidio et al., 2004, for a similar
analytical approach). We also report the results of all simple comparisons and the omnibus
analyses of (co)variance.
If perspective taking encourages more positive behaviors, then the Black experimenters
should report more positive perceptions of participants’ behaviors and more positive subjective
experiences of the interaction itself following interactions with perspective takers than following
interactions with objective-focus and control participants. Furthermore, we expected that
perspective takers would exhibit more positive automatic reactions (operationalized here as
objective, third-party observers’ ratings of participants’ approach-oriented nonverbal behaviors)
than would objective-focus and no-instruction control participants. Means and standard
deviations appear in Table 2.
Black experimenters’ perceptions of participants’ behaviors. As predicted, the
critical contrast comparing the positivity of perspective takers’ behaviors to that of non-
perspective takers was reliable, t(44) = 2.41, p = .02, d = .73. Additional comparisons revealed
that behavior ratings for perspective takers were more positive than were ratings for objective-
focus participants, t(44) = 2.24, p = .03, d = .68, and marginally more positive than were ratings
for control participants, t(44) = 1.91, p = .06, d = .58. Behavior ratings for control and objective-
focus participants did not differ (t < 1, p > .70, d = .11). Overall, the effect of instruction set was
marginally significant in a one-way ANOVA, F(2, 44) = 2.93, p = .06, ηp2 = .12.
Black experimenters’ subjective experiences. Also as expected, Black experimenters
reported more positive subjective experiences following interactions with perspective takers than
following interactions with non-perspective takers, t(44) = 2.60, p = .01, d = .78. Additional
comparisons indicated that interactions with perspective takers were rated more positively than
Perspective Taking and Racial Bias 35
were interactions with either objective-focus, t(44) = 2.26, p = .03, d = .68, or control
participants, t(44) = 2.23, p = .03, d = .67. Ratings for control and objective-focus participants
did not differ (t < 1, p > .98, d = .02). A one-way ANOVA testing the omnibus effect of
instruction set on the Black experimenters’ subjective experiences was significant, F(2, 44) =
2.93, p = .04, ηp2 = .13.
Approach-oriented nonverbal behaviors. Consistent with predictions, perspective
takers displayed more approach-oriented nonverbal behaviors than did non-perspective takers,
t(43) = 2.39, p = .02, d = .73. Additional comparisons revealed that perspective takers displayed
more approach-oriented behaviors than did either objective-focus, t(43) = 2.10, p = .04, d = .64,
or control participants, t(43) = 2.01, p = .05, d = .61. Ratings for control and objective-focus
participants did not differ (t < 1, p > .87, d = .16). A one-way ANOVA testing the omnibus
effect of instruction set on approach-oriented nonverbal behaviors was marginally significant,
F(2, 43) = 2.86, p = .07, ηp2 = .12.
Mediation analysis. We next conducted a mediation analysis testing whether the effect
of instruction set on the Black experimenters’ ratings was mediated by objective, third-party
observers’ ratings of approach-oriented nonverbal behaviors. Because the two sets of ratings
provided by the Black experimenters were highly positively correlated (r = .87, p < .001), we
combined them to form a positivity of interaction composite. To dichotomize the independent
variable, we tested the effect of the contrast comparing perspective takers to the combination of
objective-focus and control participants (see Dovidio et al, 2004). As shown in Figure 5, results
revealed a significant direct effect of approach-oriented behaviors on positivity of interactions (t
= 2.45, p = .02). When controlling for this effect, the previously significant effect of instruction
set on positivity of interactions (t = 2.36, p = .02) was no longer significant (t = 1.50, p = .14).
Perspective Taking and Racial Bias 36
We also observed a 95% confidence interval around the indirect effect of approach-oriented
behaviors ranging from .0271 to .2386, indicating significant mediation (p < .05).
These results indicate that face-to-face interactions with perspective takers were viewed
more positively by Black experimenters than were interactions with objective-focus and control
participants. Furthermore, the increase in approach-oriented nonverbal behaviors among
perspective takers (as assessed by objective, third-party observers) mediated the relationship
between perspective taking and the positivity experienced by the Black experimenters. These
findings generally complement those obtained in the first four experiments by demonstrating that
the salutary effects of perspective taking extend to interpersonal processes and outcomes.
These findings are entirely consistent with what might be predicted from the results of
Experiments 1–4 and with theorizing by Galinsky et al. (2005), who assert that one of the
primary functions of perspective taking is to facilitate both the creation and maintenance of
social bonds. Although other research has observed negative intergroup behaviors following
perspective taking (Vorauer et al., 2009; Vorauer & Sasaki, 2009), our procedures differed from
those of Vorauer and colleagues in several notable ways that could potentially account for the
divergence in findings. We elaborate on these procedural differences in the General Discussion.
The current research investigated the viability of perspective taking as a strategy for
attenuating automatic expressions of racial bias. Results obtained across five experiments, using
two different perspective-taking manipulations, two different comparison conditions, and a
combination of self-report, latency-based, and behavioral dependent measures, consistently
documented the merits of perspective taking for generating more favorable automatic interracial
evaluations, approach–avoidance tendencies, and interpersonal behaviors. Specifically, we
Perspective Taking and Racial Bias 37
provided the first empirical demonstration that perspective taking—regardless of the specific
form (perspective taking–self vs. perspective taking–other)—can positively alter automatic
intergroup evaluations. We also demonstrated that, unlike other bias-reduction strategies
(Dovidio et al., 2009; Saguy et al., 2009), perspective taking does not blind perceivers to the
realities of interracial disparities. Not only did it weaken associations between Blacks and
general negativity (Whites and general positivity), but it also strengthened associations between
Blacks and oppression-related concepts (Whites and privilege-related concepts). Furthermore,
perspective taking elicited parallel effects on automatic and deliberate interracial evaluations,
with changes in the former mediating changes in the latter (Gawronski & Bodenhausen, 2006).
The current research also revealed that the benefits of perspective taking were evident
both in participants’ action tendencies and their actual behavior. First, we showed that
perspective taking strengthened automatic approach-oriented action tendencies toward Blacks
(but not Whites) as a group and toward a specific Black (but not White) person. Whereas the
IATs employed in Experiments 1–3 were unable to distinguish between positive reactions to
Black targets and negative reactions to White targets (or vice versa), the results of Experiment 4
suggest that the effects of perspective taking are indeed target-group-specific (Batson,
Polycarpou, et al., 1997; Shih et al., 2009; Vescio et al., 2003). Second, we found that
perspective taking led to more favorable interracial interactions, according to Black interaction
partners and to objective, third-party observers. Furthermore, the perspective-taking-induced
changes in approach-oriented nonverbal behaviors (as rated by the observers) mediated changes
in the experiences reported by Black interaction partners. Collectively, these findings indicate
that perspective taking can be a viable strategy for engendering more positive automatic
evaluations, approach–avoidance reactions, and spontaneous behaviors in intergroup contexts.
Perspective Taking and Racial Bias 38
Mechanism(s) Underlying the Current Findings
We demonstrated that perspective taking can enrich interracial contact experiences by
increasing the positivity of automatic reactions, specifically by strengthening approach-oriented
action tendencies. But how exactly does perspective taking produce shifts in automatic
intergroup reactions? According to the APE model (Gawronski & Bodenhausen, 2006), strategic
attempts to alter automatic reactions can produce changes either (a) by temporarily modifying
the activation pattern of preexisting associations or (b) by reshaping the underlying structure of
associations in memory. Although determining how perspective taking alters automatic
interracial evaluations was not the focus of the current research, we believe that it could produce
changes through either route described by the APE model.
First, because representations of African Americans contain both positive and negative
aspects (e.g., Fiske et al., 2002; Katz & Hass, 1988), it is possible that perspective taking alters
automatic interracial reactions by activating more positive group-based associative content than
the predominantly negative content that is ordinarily activated upon encountering a Black
individual (Devine, 1989). Consistent with this view, Galinsky and Moskowitz (2000) showed
that accessibility for negative group-based stereotypes can be attenuated by perspective taking.
Second, it is possible that perspective taking alters automatic evaluations by increasing
perceptions of psychological connectedness between the self and the target of perspective taking
(e.g., Davis et al., 1996; Galinsky et al., 2005, 2008b; Galinsky & Moskowitz, 2000). The logic
underlying this associative self-anchoring account is that once an associative link between the
self and an object of evaluation is established, to the extent that people espouse positive
automatic evaluations of themselves—as most people do (e.g., Bosson, Swann, & Pennebaker,
2000)—these positive self-evaluations should transfer to the object of evaluation (Gawronski,
Perspective Taking and Racial Bias 39
Bodenhausen, & Becker, 2007; Prestwich, Perugini, Hurling, & Richetin, 2010; Zhang & Chan,
2009). Consistent with this reasoning, Galinsky and Ku (2004) found that only people with
higher levels of self-esteem evince more positive self-reported intergroup attitudes following
perspective taking. Future research could examine whether positive automatic self-evaluations
moderate the effects of perspective taking on automatic intergroup evaluations.
Prospects for Perspective Taking in Interaction Contexts
The results of Experiment 5 indicate that perspective taking may offer a valuable tool for
curbing behavioral displays of bias. Although these results are entirely consistent with what
might be predicted based on the results of the first four experiments, they are at odds with other
research showing that perspective taking can have negative behavioral consequences during
intergroup interactions (Vorauer et al., 2009; Vorauer & Sasaki, 2009). How then can we
reconcile this divergent pattern of findings? As noted earlier, there were a number of procedural
differences across the two research programs.
First, Vorauer et al.’s (2009) participants learned very early on that they would be
interacting with an outgroup member—in fact, they were explicitly told that the experiment was
investigating “first meeting situations involving members of different ethnic groups” (p. 821).
Participants in the current research, in contrast, received no indication that they would have an
interracial interaction prior to its occurrence, leaving little time for participants to dwell on race-
based evaluative concerns before the interaction itself (cf. Vorauer, Main, & O’Connell, 1998).
Second, whereas Vorauer and colleagues (2009) participants’ were explicitly instructed
to engage in perspective taking (or not) during the interaction itself, participants in the current
research engaged in perspective taking prior to the interaction (ostensibly as part of another
study) and thus were never explicitly instructed to adopt their interaction partner’s perspective.
Perspective Taking and Racial Bias 40
These procedural differences suggest the possibility that approaching an intergroup interaction
with a perspective-taking mindset may foster positive outcomes, whereas actively taking the
perspective of the outgroup individual with whom one is interacting may have negative
consequences. Future research could examine this possibility by varying whose perspective is
taken and when. A related possibility is that adopting the perspective of an interaction partner
during the interaction itself may impose additional attentional demands on participants (see also
Malle & Pearce, 2001)—ones that may be absent for non-perspective takers or participants
induced with a perspective-taking mindset. These attentional demands could interfere with
otherwise egalitarian thought processes among lower-prejudiced individuals, thereby producing
negative behavioral outcomes. Future research could examine this possibility by including a
condition wherein participants’ attentional resources are taxed during the interaction via a
secondary task that disallows the opportunity to focus on potential evaluation.
Third, our interaction scenario was much more constrained than the one utilized by
Vorauer and colleagues. In the current research, participants answered a series of innocuous
questions about their experiences in introductory psychology for approximately 3 min. In
contrast, participants in the Vorauer et al. (2009) study had a much lengthier interaction
(approximately 15 min), during which they discussed a number of different topics (e.g., career
goals, relationships with family members). These procedural differences raise the possibility that
perspective taking may be more likely to yield positive behavioral effects when interactions are
relatively brief and/or when discussion topics are relatively innocuous. In line with this
reasoning, research has shown that behaviors displayed during interracial interactions can vary
substantially depending on both the duration of the interaction and the nature of the discussion
topic(s) (Shelton & Richeson, 2006; Trawalter & Richeson, 2008; Trawalter et al., 2009). Future
Perspective Taking and Racial Bias 41
research could examine these possibilities by varying the length of the interaction, the topic(s) of
discussion, or both.
In sum, there are a number of procedural differences between the current research and
that of Vorauer and colleagues (Vorauer et al., 2009; Vorauer & Sasaki, 2009), any one (or
combination) of which may have accounted for the observed differences. Based on these
considerations, it seems that the most pressing question for future research is not whether
perspective taking yields positive or negative behavioral effects in intergroup contexts, but rather
under what circumstances and for whom positive versus negative behaviors can be expected.
Although the blatant racism of earlier eras has declined dramatically in recent decades,
contemporary forms of bias continue to thwart the realization of genuine racial equality. The
current research provided converging evidence for the utility of perspective taking as a strategy for
combating automatic expressions of racial bias and for facilitating more favorable interracial
contact experiences. Although this research most directly addresses the efficacy of intergroup
perspective taking under relatively controlled lab conditions, we believe it has important
implications outside the laboratory as well. Specifically, we hope these experiments lay an
empirical and theoretical foundation upon which effective intergroup relations programs (i.e., in
classrooms, workshops, etc.) may be based. Future research should continue to explore the
implications of intergroup perspective taking by determining the contexts in which its benefits can
most effectively be harnessed.
Perspective Taking and Racial Bias 42
Ambady, N., Bernieri, F., & Richeson, J. A. (2000). Towards a histology of social behavior:
Judgmental accuracy from thin slices of behavior. In M. Zanna (Ed.), Advances in
Experimental Social Psychology (Vol. 32, pp. 201-271). San Diego, CA: Academic Press.
Apfelbaum, E. P., & Sommers, S. R. (2009). Liberating effects of losing executive control:
When regulatory strategies turn maladaptive. Psychological Science, 20, 139-143.
Andersen, P. A. (1985). Nonverbal immediacy in interpersonal communication. In A. W.
Siegman & S. Feldstein (Eds.), Multichannel integration of nonverbal behavior (pp. 1-36).
Hillsdale, NJ: Erlbaum.
Baron-Cohen, S. (1995). Mindblindness: An essay on autism and theory of mind. Cambridge,
MA: MIT Press.
Batson, C. D. (1991) The altruism question: Toward a social-psychological answer. Hillsdale,
Batson, C. D. (2009). Two forms of perspective taking: Imagining how another feels and
imagining how you would feel. In K. D. Markman, W. M. Klein, & J. A. Suhr (Eds.),
Handbook of imagination and mental simulation (pp. 267-279). New York: Psychology
Batson, C. D., Early, S., & Salvarini, G. (1997). Perspective taking: Imagining how another feels
versus imagining how you would feel. Personality and Social Psychology Bulletin, 23, 751-
Batson, C. D., & Moran, T. (1999). Empathy-induced altruism in a prisoner’s dilemma.
European Journal of Social Psychology, 29, 909–924.
Perspective Taking and Racial Bias 43
Batson, C. D., Polycarpou, M. P., Harmon-Jones, E., Imhoff, H. J., Mitchener, E. C., Bednar, L.
L., Klein, T. R., & Highberger, L. (1997). Empathy and attitudes: Can feeling for a member of
a stigmatized group improve feelings toward the group? Journal of Personality and Social
Psychology, 72, 105-118.
Blascovich, J., Mendes, W. B., Hunter, S. B., Lickel, B., & Kowai-Bell, N. (2001). Perceiver
threat in social interactions with stigmatized others. Journal of Personality and Social
Psychology, 80, 253–267.
Blatt, B., LeLacheur, S. F., Galinsky, A. D., Simmens, S. J., & Greenberg, L. (in press).
Perspective-taking: A method to increase satisfaction in medical encounters. Academic
Bobo, L. (2001). Racial attitudes and relations at the close of the twentieth century. In N. J.
Smelser, W. J. Wilson, & F.M. Mitchell (Eds.), Racial trends and their consequences (Vol.
1, pp. 264–301). Washington, DC: National Academy Press.
Bosson, J. K., Swann, W. B., Jr., & Pennebaker, J. (2000). Stalking the perfect measure of
implicit self-esteem: The blind men and the elephant revisited? Journal of Personality and
Social Psychology, 79, 631-643.
Chen, M., & Bargh, J. A. (1999). Consequences of automatic evaluation: Immediate behavioral
predispositions to approach or avoid the stimulus. Personality and Social Psychology Bulletin,
Davis, M. H., Conklin, L., Smith, A., & Luce, C. (1996). Effect of perspective taking on the
cognitive representation of persons: A merging of self and other. Journal of Personality and
Social Psychology, 70, 713-726.
Perspective Taking and Racial Bias 44
Davis, M. H., Soderlund, T., Cole, J., Gadol, E., Kute, M., Myers, M., & Weihing, J. (2004).
Cognitions associated with attempts to empathize: how do we imagine the perspective of
another? Personality and Social Psychology Bulletin, 30, 1625-1635.
Devine, P. G. (1989). Stereotypes and prejudice: Their automatic and controlled
components. Journal of Personality and Social Psychology, 56, 5-18.
Dovidio, J. F., Gaertner, S. L., & Saguy, T. (2009). Commonality and the complexity of “we”:
Social attitudes and social change. Personality and Social Psychology Review, 13, 3-20.
Dovidio, J. F., & Gaertner, S. L. (2004). Aversive racism. In M. P. Zanna (Ed.), Advances in
experimental social psychology (Vol. 36, pp. 1-52). San Diego, CA: Academic Press.
Dovidio, J. F., Kawakami, K., & Gaertner, S. L. (2002). Implicit and explicit prejudice and
interracial interaction. Journal of Personality and Social Psychology, 82, 62–68.
Dovidio, J. F., ten Vergert, M., Stewart, T. L., Gaertner, S. L., Johnson, J. D., Esses, V. M., Riek,
B. M., & Pearson, A. R. (2004). Perspective and prejudice: Antecedents and mediating
mechanisms. Personality and Social Psychology Bulletin, 29, 1537-1549.
Duckworth, K. L., Bargh, J. A., Garcia, M., & Chaiken, S. (2002). The automatic evaluation of
novel stimuli. Psychological Science, 6, 515-519.
Eaton, A. A., Majka, E. A., & Visser, P. S. (2008). Emerging perspectives on the structure and
function of attitude strength. European Review of Social Psychology, 19, 165-201.
Eder, A. B., & Rothermund, K. (2008). When do motor behaviors (mis)match affective stimuli?
An evaluative coding view of approach and avoidance reactions. Journal of Experimental
Psychology: General, 137, 262-281.
Esses, V. M., & Dovidio, J. F. (2002). The role of emotions in determining willingness to engage
in intergroup contact. Personality and Social Psychology Bulletin, 28, 1202-1214.
Perspective Taking and Racial Bias 45
Fazio, R. H., Jackson, J. R., Dunton, B. C., & Williams, C. J. (1995). Variability in automatic
activation as an unobtrusive measure of racial attitudes: A bona fide pipeline? Journal of
Personality and Social Psychology, 69, 1013-1027.
Fazio, R. H., & Olson, M. A. (2003). Implicit measures in social cognition research: Their
meaning and use. Annual Review of Psychology, 54, 297-327.
Finlay, K. A., & Stephan, W. G. (2000). Reducing prejudice: The effects of empathy on
intergroup attitudes. Journal of Applied Social Psychology, 30, 1720-1737.
Fiske, S. T., Cuddy, A. J. C., Glick, P., & Xu, J. (2002). A model of (often mixed) stereotype
content: Competence and warmth respectively follow from perceived status and
competence. Journal of Personality and Social Psychology, 82, 878-902.
Galinsky, A. D., & Ku, G. (2004). The effects of perspective-taking on prejudice: The moderating
role of self-evaluation. Personality and Social Psychology Bulletin, 30, 594-604.
Galinsky, A. D., Ku, G., & Wang, C. S. (2005). Perspective-taking: Fostering social bonds and
facilitating social coordination. Group Processes and Intergroup Relations, 8, 109-125.
Galinsky, A. D., Maddux, W. W., Gilin, D., & White, J. B. (2008a). Why it pays to get inside the
head of your opponent: The differential effects of perspective-taking and empathy in
strategic interactions. Psychological Science, 19, 378-384.
Galinsky, A. D., Magee, J. C., Inesi, M. E., & Gruenfeld, D. H. (2006). Power and perspectives
not taken. Psychological Science, 17, 1068-1074.
Galinsky, A. D., & Moskowitz, G. B. (2000). Perspective-taking: Decreasing stereotype
expression, stereotype accessibility, and ingroup favoritism. Journal of Personality and
Social Psychology, 78, 708-724.
Perspective Taking and Racial Bias 46
Galinsky, A. D., Wang. C. S., & Ku, G. (2008b). Perspective-takers behave more stereotypically.
Journal of Personality and Social Psychology, 95, 404-419
Gawronski, B., & Bodenhausen, G. V. (2006). Associative and propositional processes in
evaluation: An integrative review of implicit and explicit attitude change. Psychological
Bulletin, 132, 692–731.
Gawronski, B., Deutsch, R., & Banse, R. (in press). Response interference tasks as indirect
measures of automatic associations. To appear in K. C. Klauer, C. Stahl, & A. Voss (Eds.),
Cognitive methods in social psychology. New York: Guilford Press.
Gawronski, B., Deutsch, R., Mbirkou, S., Seibt, B., & Strack, F. (2008). When “just say no” is
not enough: Affirmation versus negation training and the reduction of automatic stereotype
activation. Journal of Experimental Social Psychology, 44, 370-377.
Gawronski, B., & Payne, B. K. (Eds.) (2010). Handbook of implicit social cognition. New York:
Gawronski, B., Peters, K. R., Brochu, P. M., & Strack, F. (2008). Understanding the relations
between different forms of racial prejudice: A cognitive consistency perspective. Personality
and Social Psychology Bulletin, 34, 648-665.
Gawronski, B., Peters, K. R., & LeBel, E. P. (2008). What makes mental associations personal or
extra-personal? Conceptual issues in the methodological debate about implicit attitude
measures. Social and Personality Psychology Compass, 2, 1002-1023.
Gawronski, B., & Srirathan, R. (2010). Formation, change, and contextualization of mental
associations: Determinants and principles of variations in implicit measures. In B. Gawronski
& B. K. Payne (Eds.), Handbook of implicit social cognition: Measurement, theory, and
applications (pp. 216-240). New York: Guilford Press.
Perspective Taking and Racial Bias 47
Gollwitzer, P. M. (1990). Action phases and mind-sets. In E. T. Higgins & R. M. Sorrentino
(Eds.), The handbook of motivation and cognition: Foundations of social behavior (Vol. 2,
pp. 53-92). New York: Guilford Press.
Greenwald, A. G., McGhee, D. E., & Schwartz, J. K. L. (1998). Measuring individual
differences in implicit cognition: The Implicit Association Test. Journal of Personality and
Social Psychology, 74, 1464–1480.
Greenwald, A. G., Nosek, B. A., & Banaji, M. R. (2003). Understanding and using the Implicit
Association Test: I. An improved scoring algorithm. Journal of Personality and Social
Psychology, 85, 197-216.
Greenwald, A.G., Poehlman, T.A., Uhlmann, E.L., & Banaji, M.R. (2009). Understanding and
using the Implicit Association Test: III. Meta-analysis of predictive validity. Journal of
Personality and Social Psychology, 97, 17-41.
Han, H. A., Olson, M. A., & Fazio, R. H. (2006). The influence of experimentally-created
extrapersonal associations on the Implicit Association Test. Journal of Experimental
SocialPsychology, 42, 259-272.
Han, H. A., Czellar, S., Olson, M. A., & Fazio, R. H. (2010). Malleability of attitudes or
malleability of the IAT? Journal of Experimental Social Psychology, 46, 286-298.
Higgins, E.T. (1981). Role taking and social judgment: Alternative developmental perspectives
and processes. In J. H. Flavell & L. Ross (Eds.), Social cognitive development: Frontiers and
possible futures. (pp. 119-153). Cambridge, England: Cambridge University Press.
Katz, I., & Hass, R. G. (1988). Racial ambivalence and value conflict: Correlational and priming
studies of dual cognitive structures. Journal of Personality and Social Psychology, 55, 893–
Perspective Taking and Racial Bias 48
Kawakami, K., Phills, C. E., Steele, J. R., & Dovidio, J. F. (2007). (Close) distance makes the
heart grow fonder: Improving implicit racial attitudes and interracial interactions through
approach behaviors, Journal of Personality and Social Psychology, 92, 957–971.
Külpe, O. (1904). Versuche ttber Abstraktion [Experiments on abstraction]. Bericht über den 1.
Kongrefl fur Experimentelle Psychologic, 1, 56-68.
Macrae, C. N., & Bodenhausen, G. V. (2000). Social cognition: Thinking categorically about
others. Annual Review of Psychology, 51, 93–120.
Macrae, C. N., Bodenhausen, G. V., Milne, A. B., & Jetten, J. (1994). Out of mind but back in
sight: Stereotypes on the rebound. Journal of Personality and Social Psychology, 67, 808-
Malle, B. F., & Pearce, G. E. (2001). Attention to behavioral events during social interaction:
Two actor-observer gaps and three attempts to close them. Journal of Personality and Social
Psychology, 81, 278-294.
McConahay, J. B. (1986). Modern racism, ambivalence, and the Modern Racism Scale. In J. F.
Dovidio & S. L. Gaertner (Eds.), Prejudice, discrimination, and racism (pp. 91-125).
Orlando, FL: Academic Press.
McConnell, A. R., & Leibold, J. M. (2001). Relations between the Implicit Association Test,
explicit racial attitudes, and discriminatory behavior. Journal of Experimental Social
Psychology, 37, 435–442.
Mead, G. H. (1934). Mind, self, and society. Chicago: University of Chicago Press.
Neumann, R., Förster, J., & Strack, F. (2003). Motor compatibility: The bidirectional link
between behavior and evaluation. In J. Musch & K. C. Klauer (Eds.), The psychology of
Perspective Taking and Racial Bias 49
evaluation: Affective processes in cognition and emotion (pp. 7-49). Mahwah, NJ: Lawrence
Nosek, B. A., & Hansen, J. J. (2008a). The associations in our heads belong to us: Searching for
attitudes and knowledge in implicit evaluation. Cognition and Emotion, 22, 553-594.
Nosek, B. A., & Hansen, J. J. (2008b). Personalizing the Implicit Association Test increases
explicit evaluation of the target concepts. European Journal of Psychological Assessment,
Olson, M. A., & Fazio, R. H. (2004). Reducing the influence of extra-personal associations on
the Implicit Association Test: Personalizing the IAT. Journal of Personality and Social
Psychology, 86, 653-667.
Olson, M. A., & Fazio, R. H. (2006). Reducing automatically-activated racial prejudice through
implicit evaluative conditioning. Personality and Social Psychology Bulletin, 32, 421-433.
Olson, M. A., Fazio, R. H., & Han, H. A. (2009). Conceptualizing personal and
extrapersonal associations. Social Psychology and Personality Compass, 3, 152-170.
Payne, B. K., Cheng, C. M., Govorun, O., & Stewart, B. (2005). An inkblot for attitudes: Affect
misattribution as implicit measurement. Journal of Personality and Social Psychology, 89,
Piaget, J. (1932). The moral judgment of the child. London: Kegan Paul, Trench, Trubner and
Plant, E. A., & Devine, P. G. (1998). Internal and external motivation to respond without
prejudice. Journal of Personality and Social Psychology, 75, 811–832.
Perspective Taking and Racial Bias 50 Download full-text
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and
comparing indirect effects in multiple mediator models. Behavior Research Methods, 40,
Prestwich, A., Perugini, M., Hurling, R., & Richetin, J. (2010). Using the self to change implicit
attitudes. European Journal of Social Psychology, 40, 61-71.
Purdie-Vaughns, V., & Eibach, R. P. (2008). Intersectional invisibility: The distinctive
advantages and disadvantages of multiple subordinate-group identities. Sex Roles, 59, 377-
Richardson, D. R., Hammock, G. S., Smith, S. M., Gardner, W., & Signo, M. (1994). Empathy
as a cognitive inhibitor of interpersonal aggression. Aggressive Behavior, 20, 275-289.
Richeson, J. A., & Nussbaum, R. J. (2004). The impact of multiculturalism versus color-
blindness on racial bias. Journal of Experimental Social Psychology, 40, 417–423.
Richeson, J. A., & Shelton, J. N. (2005). Thin slices of racial bias. Journal of Nonverbal
Behavior, 29, 75-86.
Richeson, J. A., & Trawalter, S. (2005). Why do interracial interactions impair executive
function? A resource depletion account. Journal of Personality and Social Psychology, 88,
Rosenthal, R., Rosnow, R. L., & Rubin, D. B. (2000). Contrasts and effect sizes in behavioral
research: A correlational approach. Cambridge University Press.
Rudman, L. A. (2004). Sources of implicit attitudes. Current Directions in Psychological
Science, 13, 80-83.
Saguy, T., Tausch, N., Dovidio, J. F., & Pratto, F. (2009). The irony of harmony: Intergroup
contact can produce false expectations for equality. Psychological Science, 29, 114-121.