THE ROLE OF BETWEEN- AND WITHIN-SETTING IRRELEVANT DIMENSIONS IN CHILDREN'S SIMULTANEOUS DISCRIMINATION LEARNING

Abstract

120 children received training on simultaneous problems differing as to the number (0, 1, or 2) of irrelevant dimensions that varied between settings and within settings to determine their effects on choice behavior. The results indicated that the concurrent presentation (within setting) of the values on an irrelevant dimension disrupted performance on the simultaneous problem relative to the successive presentation (between setting) of these values. Performance was higher on problems having 1 irrelevant dimension than on problems with 2 irrelevant dimensions, both within settings and between settings; however, neither difference was reliable. The purpose of the experiment was to determine the effects of the number of irrelevant dimensions that varied within settings and between settings on children's choice behavior in simultaneous discrimination problems. The hypothesis of stimulus interaction (Spiker 1963), an extension of Hull-Spence theory specifying a rule for combining the generalized and conditioned habits and inhibitory tendencies of stimulus components in multidimensional stimuli, predicts that a discrimination problem having an irrelevant dimension varying either between settings or within settings is more difficult to solve than the standard simultaneous problem which has no irrelevant dimensions (other than position).