THE RELATIVE SENSITIVITIES OF SAME-DIFFERENT AND IDENTIFICATION JUDGMENT MODELS TO PERCEPTUAL DEPENDENCE

Probabilistic models of same-different and identification judgments ar e compared (within each paradigm) with regard to their sensitivity to perceptual dependence or the degree to which the underlying psychologi cal dimensions are correlated. Three same-different judgment models ar e compared. One is a step function or decision bound model and the oth er two are probabilistic variants of a similarity model proposed by Sh epard. Three types of identification models are compared: decision bou nd models, a probabilistic multidimensional scaling model, and probabi listic models based on the Shepard-Luce choice rule. The decision boun d models were found to be most sensitive to perceptual dependence, esp ecially when there is considerable distributional overlap. The same-di fferent model based on the city-block metric and an exponential decay similarity function, and the corresponding identification model were f ound to be particularly insensitive to perceptual dependence. These re sults suggest that if a Shepard-type similarity function accurately de scribes behavior, then under typical experimental conditions it should be difficult to see the effects of perceptual dependence. This result provides strong support for a perceptual independence assumption when using these models. These theoretical results may also play an import ant role in studying different decision rules employed at different st ages of identification training.