Commonalities and Distinctions in Featural Stimulus Representations

This paper evaluates four featural models of stim-ulus similarity using data collected for a set of 16 nations. Algorithms are developed for Þnding stim-ulus representations, and the important issue of balancing data-Þt against model complexity is ad-dressed by using the Geometric Complexity Crite-rion. Although the data clearly incorporate both common and distinctive features, Tverskys (1977) Contrast Model seems unable to express these reg-ularities in an appropriate manner. However, we show that a new version of the Contrast Model that treats each feature as either being common or dis-tinctive is better able to capture the essential as-pects of the similarity judgments