DISTINGUISHING THEORIES OF REPRESENTATION - CRITIQUE OF ANDERSON ARGUMENTS CONCERNING MENTAL-IMAGERY

In a recent article, J. R. Anderson (see record 1979-22767-001) argued that behavioral data could not distinguish alternative theories of memory representations. The theorem Anderson proved does not support his conclusions. Five fallacies in his argument are criticized: (a) Behavioral data cannot distinguish alternative models. (b) An invertible function relates corresponding states of alternative representational theories. (c) Propositional models can account for data supporting image models and, thus, warrant equal credibility. (d) Propositional descriptions constitute parsimonious models. (e) Unless explicitly operationalized, image models cannot support theoretical or empirical research. Contrary to these beliefs, representational differences can qualitatively affect performance, and such important differences apparently characterize the analog class of imagery models. (33 ref) (PsycINFO Database Record (c) 2006 APA, all rights reserved).