Abstract
This study tested the application of the learning concepts of response hierarchies, and concept-mediated generalization to stereotypes, using political stereotypes as prototypes. The training phase of this study endowed a CVC with a political label of either “conservative” or “liberal.” A paired-associate, anticipation, concept-formation paradigm was used. The training stimuli were sets of six adjectives contained in the previously measured stereotypes. The sets were varied in terms of their hierarchical value or saliency. In a sample of 123 college students, training stimuli with high saliency were associated with greater learning on five dependent measures. When the same subjects were tested for generalization, they generalized all of the belief components comprising either a liberal or conservative stereotype to the nonsense syllable which had been used in earlier training. The original hierarchical arrangements of component beliefs were also present in the generalization data.
Get full access to this article
View all access options for this article.
