Perspective of Bem's Self-Perception Theory
The traditional view is that attitudes determine behavior. However, Daryl Ben (1972) stood conventional logic on its head when he proposed that behavior often determines (or causes people to draw inferences about) their attitudes. Subsequent research on attribution has shown that sometimes people do infer their attitudes from their behavior.
According to Daryl Bem’s self-perception theory, people often infer their attitudes from their behavior. Thus, Bem holds that people are not nearly as aware of, nor as clear about, their own internal feelings and beliefs as they think they are. Bem’s self-perception principle states that
- Individuals come to “know” their own attitudes, emotions, and other internal states partially by inferring them from observations of their own overt behavior and/or the circumstances in which this behavior occurs. Thus, to the extent that internal cues are weak, ambiguous, or uninterpretable, the individual is functionally in the same position as an outside observer, an observer who must necessarily rely upon those same external cues to infer the individual’s inner state (1972, p.2)
From the perspective of self-perception theory (Bem, 1972), we get to know ourselves in much the same way we get to know others. Just as we might observe someone’s action (behavior) to make inferences about their desires, we may try to understand our attitudes by observing our own behavior. In Jamesian terms, self-perception theory posits that the perceiver self learns about internal attributes by observing external aspects of the perceived self.
In the beginning, Bem says, every child learns about his or her external environment through discrimination training by adults and other children around him or her. By verbal labeling and corroboration or correction, children learn to distinguish between dogs and cats, for instance, and between anger and happiness in other people.
In the same way, says Bem, through self-perception we also learn to label our own inner feelings of hunger, anger, anxiety, or liking. To use his favorite example, we decide that we like brown bread by observing the fact that we eat a lot of it. The other factor that is important in determining our self-attributions is whether the external circumstances constrain our behavior. If we realize that the only bread our mother serves us is brown bread, we will be less likely to conclude that our eating it indicates fondness for it.
In his first theoretical papers, Bem (1976) stated even more strongly the dramatic and unorthodox claim that people do not really know their attitudes and beliefs until they act, that they can infer these internal states only from their behavior, and that they even may be unable to remember any internal states that are discrepant from their behavior (Bem & McConnell, 1970).
Bem has also applied his theory to attitude change, positing that it occurs in reaction to self-observed behaviors combined with observation of external cues, which indicate whether or not the behavior is apt to be valid or truthful. For instance, if an actor in a TV commercial says “I like Busy Bakers’ brown bread,” but we know that he was paid to make the commercial, we may doubt whether that is his true attitude.
In a contrary example, if a person is subtly induced to say something opposed to his former opinion under conditions that suggest truthfulness, he is apt to decide that he really believes the statement that he has made.
As a real-life instance of this process, Bem (1970) has asserted that police station interrogation conditions constitute a truth-telling situation for most people, and that in such situations certain wily interrogation procedures can induce prisoners to make and to believe in false confessions about crimes, which they have not really committed. This is a particularly surprising and dramatic example of attitude change, and Bem has backed up his claims with clear-cut evidence from a laboratory experiment, which showed exactly this process at work. However, it should be noted that other investigators have had difficulty in replicating these results (Kiesler & Munson, 1975).
The challenging part of Bem’s theory is that other people can make the same conclusion about one’s attitude just by observing one’s behavior and by looking at the environmental constraints on it. People do not necessarily know their own attitudes better than some observer would.
We are used to thinking that we know ourselves best and that we know our own preferences better than other people do. But maybe not.
- Have you ever had a close friend say, “You really like that person, don’t you?” before you admitted it to yourself?
Bem’s theory best accounts for those circumstances where one does not know one’s own mind or attitude ahead of time.
In comparison with Kelley's covariation Opens in new window or Jones and Davis’ correspondent inference Opens in new window theories of attribution, Bem’s theory is a very simple one, and it proposes that people normally do very little cognitive work in making their attributions Opens in new window.
In its emphasis on the simple, shortcut nature of social cognition, Bem’s theory is consistent with the work on heuristics and biases. Thus it anticipated the subsequently developed cognitive miser view of social cognition—the conclusion that people’s attributions generally do not follow the normative models of rational inference processes, but rather take shortcuts that require as little time and effort in information processing as possible (Fiske & Taylor, 1984).