Unwanted Elements of Norms

Unwanted Elements of Norms

The elicitation of a norm has been described here as a process of parallel activation of multiple representations that are recruited by a stimulus or by a category. We assume that the process of recruitment is rapid, automatic, and essentially immune to voluntary control after its initiation. In particular, it is not possible for the individual to sift through activated elements and discard irrelevant or misleading ones. This limitation on the voluntary control of norms helps explain two well-documented phenomena of social judgment: the perseverance of discredited beliefs and the correspondence bias in person perception.

Imagine a discussion of a Canadian athlete, in which someone who is unfamiliar with metric measures reads from a sheet: “Brian weighs 102 kg. That’s 280 lbs, I think. No, it’s actually about 220 lbs.” Does the speaker’s initial error affect listeners’ subsequent responses to questions about Brian’s size and strength? The literature on perseverance of discredited beliefs (C. A. Anderson, 1983; Fleming & Arrowood, 1979; Ross & Anderson, 1982; Ross & Lepper, 1980; Schul & Bernstein, 1985) suggests that it does. The message of this literature is that traces of an induced belief persist even when its evidential basis has been discredited. The discarded message is not erased from memory, and the norm elicited by a subsequent question about Brian’s weight could therefore contain the original message as well as its correction. Thus, a listener might “know” immediately after the message that Brian’s true weight is 220 lbs, and this value would presumably retain an availability advantage, but the category norm associated with Brian’s weight would still be biased toward the erroneous value of 280 lbs. Judgments that depend indirectly on the activation of the norm would be biased as well.

Some aspects of the phenomenon known as the correspondence bias (Jones, 1979) or the fundamental attribution error (Ross, 1977) could be explained in similar terms. Many studies have shown that people make unwarranted inferences concerning personal traits and attitudes from observations of behavior that are in fact entirely constrained by the situation. Subjects in one famous series of experiments in this tradition observed an individual who was explicitly instructed to write an essay or to read aloud a speech advocating an unpopular position (Jones, 1979; Jones & Harris, 1967; Jones & McGillis, 1976). In response to subsequent questions, observers commonly attributed to target persons an attitude consistent with the position that these persons had been constrained to advocate.

In these experiments, as in studies of discredited beliefs, a behavioral observation is accompanied by information that challenges its validity. As in a theatrical performance, the actor in the experiments conducted by Jones (1979; see also Jones & Harris, 1967; Jones & McGillis, 1976) engages in a behavior (e.g., advocating the regime of Fidel Castro) that does not have its usual significance because of the special demands of the situation. We propose that in both the theater and these experiments, two traces are laid down: (1) a literal memory of the person expressing pro-Castro sentiments, and (2) a memory of the behavior in the context of the situational constraints. Both memories are elements of the set that is evoked by further observations of the actor’s political opinions or by a question concerning those opinions. As in the preceding example, this norm is biased even for an observer who “knows” that the actor’s behavior is constrained, and therefore uninformative. Quite simply, pro-Castro behaviors are more normal for the actor than for a random stranger. Belief perseverance, generalizations from atypical examples, and failures of discounting all illustrate the same principle: Any observation of behavior – even if it is discounted or discredited – increases the normality of subsequent recurrences of compatible behaviors.