Page 258 - The Social Animal

240 The Social Animal

found yourself writing an essay excusing the brutality of the New
Haven police (as in Cohen’s experiment described earlier). Accord-
ing to Bem, you would dispassionately observe your own behavior,
shrug your shoulders and say, “Hmmm, because I wrote that essay
(for only 50 cents!), I guess I must believe what I wrote . . . or else
I wouldn’t have written it.” No dissonance, no discomfort, no self-
justification; merely self-perception.
Bem’s notion is elegant in its simplicity. If attitude change in this
kind of situation is simply a matter of cool self-perception, then we
do not need all this theorizing about discomfort, the self-concept,
self-justification, and the like.
It turns out that Bem is partly right. Self-perception does play a
role; but it seems to be operative only in those situations where a per-
son doesn’t have a clear, unambiguous belief to begin with. On the
other hand, where a person has a fairly clear initial belief (e.g., the
New Haven Police behaved badly; packing spools is a boring task; I
am a decent, sensible person), then discomfort and threats to the self-
concept do come into play. 72
How can I be sure that discomfort plays a major role in these dis-
sonant situations? Well, one reason is that people in these situations
73
say so. For example, Andrew Elliot and Patricia Devine found that
when people are put in a dissonance-arousing situation, they do in-
deed report feeling more agitated and more uncomfortable than peo-
ple in the control condition.
Participants reporting their own discomfort is convincing. In ad-
dition, there is independent behavioral evidence of discomfort. For
example, we know that discomfort is distracting. In a clever experi-
74
ment, Michael Pallak and Thane Pittman demonstrated that peo-
ple experiencing dissonance perform a complex task more poorly
than people not experiencing dissonance. The people experiencing
dissonance show the same decrement in performance as people in
other uncomfortable drive states like extreme hunger and thirst.
In addition, several investigators have shown some striking be-
havioral evidence for the motivating qualities of dissonance. In one
experiment, Mark Zanna and Joel Cooper 75 gave participants a
placebo pill. Some were told that the pill would arouse them and
make them feel tense. Others were told that the pill would make
them feel calm and relaxed. Participants in the control condition
were told that the pill would not affect them in any way. After in-
gesting the pill, each person was induced to write a counterattitudi-