Comparing generalization by humans and adaptive networks

One goal of study reported was to examine how people generalize in a simple deterministic categorization task in which each pattern is characterized in terms of known binary features. While we expected certain similarities to emerge across human learners we anticipated that the particular generalizations might be subject to considerable individual differences. To test this idea, we used an experimental paradigm that would permit us to to observe individual subjects during the learning of a categorization task on a set of training patterns and then allow us examine the types of categorizations they made on a set of novel test patterns. Later we compared human generalizations to those of a small adaptive network. We have demonstrated that subjects who learn the same pattern categorization may abstract different principles and therefore show large individual differences in their generalization behavior. Adaptive networks with the minimum number of hidden units exhibit a similar behavior but generalize differently.