The role of attention shifts in the categorization of continuous dimensioned stimuli

Results of human category learning experiments, using stimulus dimensions w ith binary values, have implicated a rapidly acting mechanism of attention shifts. Theories of categorization desire that stimuli with binary, discret e and continuous valued dimensions should all be treated similarly. Theoret ical analyses of attention shifting, however, have up to now only been deve loped for shifts between features, or shifts between entire dimensions, not shifts within dimensions. Here we present a model of how people learn to d iscriminate categories made up of stimuli with continuous-valued dimensions . The model uses rapid shifts in attention within stimulus dimensions to re duce errors during learning; the model generalizes J. K. Kruschke's (Psycho logical Review, 99, 22-44, 1992) ADIT model. In an experiment in category l earning, subjects were trained to discriminate four bivariate normal distri butions that are presented with differential base rates. The base-rate mani pulation produces several qualitative effects, for which the model accounts very well. With attention shifting turned off, the model fails to account for some aspects of the data, suggesting that attentions shifts are an impo rtant mechanism in the model.