The goal of this research is to investigate the perceptual and cognitive processes used in the categorization of complex stimuli. There are two broadly-defined explanations of categorization decisions: decisions may be based on abstract rules, or they may be based on recall of particular examples from memory. These explanations are implemented as formal models (e.g., Nosofsky's GCM Model and Ashby's GRT model). A series of probability-learning experiments will help determine which of the two explanations provide a better account across various stimulus conditions. Preliminary results indicate that people use memory exemplars when there are a few to-be-classified stimuli, but they use rules when there are several to-be-classified stimuli. The working hypothesis is that people are flexible: using either rule-based or exemplar-based strategies depending on the cognitive demands of the task. Exemplar-based categorization is used when the cognitive demands of the task are relatively light while rule-based categorization is used when the cognitive demands of the task are relatively heavy. The research will allow for a fuller understanding of the nature of categorization processes which will aid our understanding of human decision-making and have important implications for education.
