ON THE RELATIONSHIP AMONG IDENTIFICATION, SIMILARITY, AND CATEGORIZATION - REPLY

Nosofsky and Smith (1992) challenged the theoretical, methodological, and empirical results of Ashby and Lee (1991). This reply (a) shows that models derived from general recognition theory (GRT) can predict categorization from identification without incorporating selective attention-at least, in the data sets suggested by Nosofsky and Smith; (b) argues that the categorization processes postulated by GRT are extremely dissimilar to the exemplar-based similarity model proposed by Nosofsky (1986); (c) criticizes Nosofsky and Smith's post hoc reanalysis of Ashby and Lee's identification-confusion data; (d) answers questions raised by Nosofsky and Smith about the identification and similarity models tested by Ashby and Lee; and (e) argues that with the excellent fits reported by Ashby and Lee (i.e., 99.7% of variance accounted for), least squares and maximum likelihood model fitting procedures lead to similar conclusions.