Attitudes and Information Processing 313compatible with their highly heritable attitudes. Tesser (1993;
Tesser & Crelia, 1994) tested this idea in several ingenious
ways. In all of his studies, attitudes that had been shown by
Eaves et al. (1989) to have either high or low heritability co-
efficients were studied. In one study, individuals were found
to provide answers more quickly for high than for low heri-
tability attitudes. In another study, individuals were found to
be less affected by conformity pressure when reporting high
than when reporting low heritability attitudes. In a third
study, interpersonal similarity on high heritability attitudes
was shown to affect liking for others more than did similarity
on low heritability attitudes. Finally, in two studies, individu-
als found agreement feedback more reinforcing when the
agreement occurred for highly heritable attitudes than when
it occurred for less heritable attitudes. These findings suggest
that attitude strength is positively correlated with attitude
heritability (see also Olson et al., 2001).
ATTITUDES AND INFORMATION PROCESSING
One of the fundamental functions of attitudes, as discussed
earlier, is the object-appraisal function, which refers to the
capacity of attitudes to facilitate both the identification of
objects and the rapid appraisal of the objects’ implications
for the self. This function underscores that attitudes influ-
ence how objects are perceived and how information about
those objects is processed. In this section we review research
on the effects of attitudes on information processing. The
theme of this section is selectivity—attitudes tend to facilitate
the processing of information that is consistent with them and
to inhibit the processing of inconsistent information.
Selective Attention
Festinger (1957) proposed in hisdissonance theorythat peo-
ple want to believe that their decisions and attitudes are cor-
rect. Whereas individuals attend in an unbiased way to
information prior to making decisions or forming attitudes,
Festinger argued that after attitudes are formed, they moti-
vate people to pay attention to consistent information and
avoid inconsistent information. Early tests of thisselective
exposure hypothesisyielded little support (see Freedman &
Sears, 1965), but researchers gradually identified boundary
conditions for the effect (see Frey, 1986). For example, the
utility, novelty, and salience of consistent versus inconsis-
tent information must be controlled so that the effects of
attitudinal consistency can be tested clearly. Researchers
have documented selective attention in the laboratory (e.g.,
Frey & Rosch, 1984) and in field settings (e.g., Sweeney &
Gruber, 1984), and there is evidence that individuals with
repressing-avoidance defensive styles may exhibit selective
attention to consistent information more than do individ-
uals with ruminative-approach defensive styles (Olson &
Zanna, 1979).
There is also some evidence of a broader form of selective
attention, which relates to the existence of strong attitudes
per se. Specifically, Roskos-Ewoldsen and Fazio (1992)
showed that objects toward which individuals have highly
accessible attitudes (whether positive or negative) are more
likely to attract attention than are objects toward which indi-
viduals have less accessible attitudes. Presumably, this selec-
tivity effect is not motivated by a desire to believe one’s
attitudes to be correct, but rather by the functional value of
quickly attending to objects that personal experience has
shown to be potentially rewarding or punishing.Selective PerceptionMany researchers have shown that attitudes influence the per-
ception or interpretation of attitude-relevant information, with
the effect generally of interpreting information as more sup-
portive of one’s attitudes than is actually the case. For ex-
ample, Vidmar and Rokeach (1974) found that viewers’
perceptions of the television showAll in the Familywere re-
lated to their racial attitudes: Low-prejudice viewers saw the
bigoted character of Archie Bunker as the principal target of
humor and sarcasm in the show, whereas high-prejudice view-
ers saw Archie sympathetically and considered his liberal son-
in-law Mike to be the principal target of humor and sarcasm.
Similarly, Lord, Ross, and Lepper (1979) found that individ-
uals’ attitudes toward capital punishment predicted their as-
sessments of the quality of two alleged scientific studies, one
supporting and one questioning the deterrence value of the
death penalty: Participants evaluated the study that apparently
supported their own view more favorably than they evaluated
the study that apparently disconfirmed their view. Houston
and Fazio (1989) replicated this study and showed that the bi-
asing effect of attitudes on the interpretation of information
was significant only when the attitudes were highly accessible
(see also Fazio & Williams, 1986; Schuette & Fazio, 1995).
In another domain, Vallone et al. (1985) found that indi-
viduals’ evaluations of the media coverage of an event were
biased by their relevant attitudes (see also Giner-Sorolla &
Chaiken, 1994).
If there is a general bias to perceive the world as consistent
with one’s attitudes, then existing attitudes might reduce the
ability of perceivers to detect that the attitude object has
changed. Indeed, Fazio, Ledbetter, and Towles-Schwen (2000)
have documented such an effect and related it to attitude