CATEGORY INVENTION IN UNSUPERVISED LEARNING

This research aimed to discriminate between 2 general approaches to un supervised category learning. one based on learning explicit correlati onal rules or associations within a stimulus domain (autocorrelation) and the other based on inventing separate categories to capture the co rrelational structure of the domain (category invention). An ''attribu te-listing'' paradigm was used to index unsupervised learning in 3 exp eriments. Each experiment manipulated the order in which instances fro m 2 different categories were presented and evaluated the effects of t his manipulation in terms of the 2 competing theoretical approaches to unsupervised learning. Strong evidence was found for the use by Ss of a discrete category invention process to learn the categories in thes e experiments. These results also suggest that attribute listing may b e a valuable method for future investigations of unsupervised category learning.