Interaction of knowledge-driven and data-driven processing in category learning

The present paper argues that category learning is both a data-driven and a knowledge-driven process. This is described in a generic model that distin guishes between categorical knowledge, conceptual knowledge, and implicit c ognitive theories. The model assumes that each of these knowledge aspects m ay affect the process of category learning by affecting the way similaritie s between objects are perceived. This central assumption of the model is te sted in two experiments. The first experiment shows that the presence or ab sence of prior categorical acid conceptual knowledge affects the psychologi cal stimulus space by changing the saliency of the stimulus dimensions. The second experiment uses these weights to predict the distribution of errors over the stimuli and the number of trials to criterion in category learnin g by other participants under the same knowledge conditions. We conclude th at prior categorical and conceptual knowledge affect category learning by m ediation of similarity perception, and discuss the implications of these re sults.