Page 183 - The Social Animal

Social Cognition 165

ior could be seen as either hostile or gentle, and rated Donald’s level
of hostility. Consistent with our earlier discussion of contrast effects,
we would expect the different puzzles to influence judgments about
Donald. Compared with Hitler and Manson, almost everyone looks
gentle—including Donald; compared with the Pope and Santa
Claus, almost everyone appears hostile—including Donald. This is
exactly what Herr found. Those subjects primed with the extremely
hostile persons rated Donald as less hostile than those who received
the gentle primes.*
But Herr’s experiment didn’t stop there. Next, the subjects
played a bargaining game with a person whom they thought was
Donald. In this game, participants were required to choose between
one of two strategies—competing or cooperating. Herr found that
when subjects expected to play against a hostile Donald, they played
in a highly competitive manner; when they expected a gentle Don-
ald, they played with far more cooperation. Interestingly, the subjects
who were naively playing the role of Donald also perceived this com-
petitiveness; they rated their opponent’s level of hostility in a man-
ner consistent with the way he or she played the game. In sum, a
relatively subtle context had influenced attitudes and expectations
that, in turn, affected behavior and subsequently affected the next
round of perceptions.
Carol Dweck and her colleagues have demonstrated the behav-
ioral consequences of people’s more enduring beliefs. According to
Dweck, children develop implicit theories about the permanence of

*The reader should note the crucial difference between this experiment and one
by Higgins et al., discussed earlier in this chapter. In the Higgins experiment, the
researchers were priming a category—negativity. This influenced observers to see
subsequent ambiguous stimuli (like Donald) more negatively—because that is what
people are primed to look for. In the Herr experiment, the researchers were prim-
ing exemplars of hostility (like Hitler). Here, a contrast effect occurs: Compared to
extremely hostile people, an ambiguous person (like Donald) comes off looking like
a teddy bear. Summary: What then can we conclude from the considerable research
on attitudes and behavior? First and foremost, the collective research on attitudes
and behavior underscores a principle we will see quite often in this book: Subtle sit-
uational variables are often strong determinants of our behavior. Second, most
people tend to overlook the importance of the situation in explaining behavior, pre-
ferring instead to explain other people’s actions in terms of assumptions about their
personalities and attitudes. In other words, most of us assume that people’s attitudes
do indeed forecast their behavior, and then we overapply this belief in interpreting
the behavior of others. We see attitude-behavior relationships even when they may
not exist in reality.