Judging compound social categories: Compound familiarity and compatibility as determinants of processing mode

Three experiments tested the hypothesis that judgments about a group formed by two paired categories would rely on stored instances of individual category members (i.e., exemplars) in some cases, but not in others. Specifically, judgments of a relatively unfamiliar compound category (e.g., male elementary schoolteachers) were expected to rely on exemplars, whereas alternative sources of information, particularly abstract stereotypes, would be available for making judgments of a more familiar category (e.g., female elementary schoolteachers). Experiments 1 and 2 demonstrated support for these hypotheses. Experiment 3 ruled out the possibility that the differences in judgment strategy between the familiar and unfamiliar compound categories arose from the relative incompatibility of the two constituent categories (e.g., males and elementary schoolteachers), rather than familiarity. Implications for stereotype development and change are discussed.