The Psychology of Bigotry 511with dissimilar beliefs. Although the link between attitude
similarity and interpersonal attraction had already been well
demonstrated by that point, Rokeach et al.’s provocative
contribution was to extent it to the domain of prejudice and to
argue that allforms of prejudice were essentially different
forms of belief prejudice. Thus, according to BCT, the racial
conflict between Blacks and Whites in the United States is
not due to race per se but rather to opposite or conflicting
stands on key issues such as affirmative action in em-
ployment and education. Likewise, the antipathies between
English and French in Canada are not due to ethnicity per se,
but rather to conflict over the issue of Quebec’s role, and the
place of the French language, within Canada. In other words,
racial and ethnic prejudice, as two examples, presumably
reflect belief prejudice.
BCT clearly suggests research in which belief is pitted
against group membership characteristics such as race or eth-
nicity. Rokeach et al. (1960), for example, had samples of
White university students from northern and southern parts of
the United States rate their desires to be friends with members
of pairs of stimulus persons whose races and beliefs, both
race-relevant and -irrelevant, were specified. For example,
Type R pairs varied in race but kept belief constant (e.g., a
White person who believes in God vs. a Black person who be-
lieves in God). Type B pairs kept race constant but varied be-
lief (a Black person who believes in God vs. a Black person
who is an atheist). Type RB pairs varied both race and belief
simultaneously. Differences in friendliness ratings for mem-
bers of a stimulus person pair were taken as reflecting
discrimination. A critical comparison suggested by BCT
involved a choice between an in-group member with dissimi-
lar beliefs versus an out-group member with beliefs similar to
one’s own. For this pair comparison, individuals’ preference
typically goes to the latter, consistent with BCT. Likewise,
Rokeach and Mezei (1966) showed that belief similarity ex-
cels race in predicting preferences for work partners among
employment applicants following actual interpersonal inter-
action and discussion between Black and White participants
with similar and dissimilar beliefs on an issue.
BCT remains as relevant a theory of prejudice in the
twenty-first century as it was in the latter half of the twentieth
century, largely due to the research over the past several
decades of Insko and his colleagues (e.g., Cox, Smith, &
Insko, 1996; Insko, Nacoste, & Moe, 1983) as well as recent
contributions by Biernat and her colleagues (Biernat, Vescio,
& Theno, 1996; Biernat, Vescio, Theno, & Crandall, 1996).
For example, Insko et al. (1983) reviewed the literature and
compared the strong version of BCT (when social pressure is
absent, only belief determines racial-ethnic discrimination)
to a weak version (when social pressure is absent, belief is
more important than race in determining discrimination or
prejudice). They concluded that the weak version of BCT was
clearly supported by the evidence, whereas the strong version
was more problematic (e.g., race effects in the form of in-
group favoritism occur even in the absence of social pressure).
Cox et al. (1996) reported results of three cross-sectional
surveys conducted over several decades of Black and White
teenagers sampled from a North Carolina school system who
had responded to stimulus persons varying in race and belief,
using a belief discrepancy manipulation in which dissimilar
beliefs were ones that respondents themselves had previously
attributed to the other race. For White respondents, race ef-
fects (i.e., preferring their own race to Blacks on social dis-
tance and other attitude measures) steadily declined across
three points in time from 1966 to 1993, as did perceived dis-
approval of interracial contacts and relationships. The effects
of belief similarity affected all of their dependent variables
and were constant across decades for White respondents. For
Black respondents, more complex findings were obtained:
Specifically, race effects (i.e., in-group preference) did not
decline between 1979 and 1993 (the only two time periods
including Black respondents), and belief similarity primarily
influenced same-race rather than interracial evaluations.
BCT has clear links to contemporary perspectives on im-
pression formation and prejudice. For example, Cox et al.
(1996) noted that BCT is very similar to Fiske and Neuberg’s
(1990) temporal-continuum model of impression formation.
In the latter model, a perceiver begins with categorical infor-
mation (viz., race, ethnicity, sex, age, etc.) about a person
but proceeds, if time permits and circumstances require, to
process individuating information (e.g., beliefs of the stimu-
lus person). Like Fiske and Neuberg’s model, BCT deals with
the issue of when individuating information (viz., beliefs and
values) about a stimulus person overcomes competing cate-
gorical information (viz., group membership) in the impres-
sions we form of others. Likewise, the importance that BCT
accords to perceived belief dissimilarity in eliciting prejudice
is shared today by terror management theory, a perspective
focusing on the psychological consequences of being aware
of, or sensitized to, one’s mortality (Solomon, Greenberg, &
Pyszczynski, 2000).
BCT has also been extended to the value domain. Schwartz
and Struch (1989) proposed that perceptions of value dissim-
ilarities between groups underlie intergroup antagonisms and
undercut feelings of shared humanity. Likewise, Biernat,
Vescio, Theno, and Crandall (1996) reported studies in which
group membership cues (race and sexual orientation, respec-
tively, in separate studies) were crossed with value violation
(e.g., a lazy vs. dependable worker in the race study or a good
vs. bad parental example in the sexual orientation study).