ABSTRACT: Whereas many theories have been proposed that parse mind into individual components, they do not emphasize the interaction between these various elements. We describe two human categorization experiments designed to address the three issues relevant to hybrid rule- and exemplar-based systems: the necessity of rules, the necessity of exemplar memory, and the interaction between these two sub-systems in learning and in classification performance. We account for the participants' classifications using an updated version of the hybrid rule and exemplar model described by Kruschke and Erickson (1994). This hybrid model consists of a rule module, an exemplar module, and a gating mechanism. This gating mechanism controls the influence of each module in decision-making and the extent of learning in each module. We also show that neither of these two sub-systems, acting alone, can adequately account for human behavior. This is significant inasmuch as the exemplar sub-system is a full implementation of ALCOVE (Kruschke, 1992), which has performed well in a variety of categorization tasks.