Response-Time Approach to Contrasting Models of Perceptual Classification

Abstract

Abstract : The long-term objective of this work is the development of general computational models of human perceptual classification and memory. An important goal is to develop and test models that explain the time course of classification and recognition decision making. The first specific goal involved the extension of Nosofsky and Palmeri's (1997) exemplar-based random-walk (EBRW) model of classification response times (RTs) to the domain of memory search. Several empirical studies demonstrated successful applications of the new theory in this domain. The second goal involved the development and testing of a new set of logical-rule models of classification RTs. Various cognitive architectures may underlie the application of logical rules in classification, including serial-, parallel-, and coactive-processing architectures. A highly diagnostic paradigm was developed that yielded sharply contrasting predictions from these alternatives. Several sets of experiments provided support for the logical-rules framework and allowed one to identify which processing architectures were used under alternative experimental conditions.