Person stereotypes and memory for people.

Abstract

Snyder and Uranowitz have proposed a memory-priming mechanism by which information about a person that is normally unavailable in episodic memory is made available by the activation of a person stereotype that subsumes that information. In our two experiments subjects read a biography of Betty K, who was later labeled as either a heterosexual or a lesbian before the subjects took a recognition memory test. A signal-detection model was used to assess the effects of labeling on response bias as well as on the amount of information available in memory. Neither experiment produced any improved recognition memory for biographic information due to activation of a sexual stereotype. Both experiments found a response bias (guessing), however, acting in the direction of the label the subject received. It is concluded that at this time no clear empirical support exists for the memory-priming mechanism proposed by Snyder and Uranowitz. In an important and influential article, Snyder and Uranowitz (1978) proposed a person-memory retrieval mechanism based on the operation of person stereotypes. According to their hypothesis previously presented information stored in memory but not available in a recognition task may become available after a person stereotype is activated. It is not necessary that this person stereotype play a role in encoding the information. So long as the information presented is relevant to the stereotype, the learner need not think of this relation at the time of presentation. This hypothesis thus goes against the encoding-specificity principle (Tulving & Thomson, 1973). Although Snyder and Uranowitz used person stereotypes as the activated structures to influence memory re