Transfer of verbal discriminations based on differential reward magnitudes.

Publisher Website
Google Scholar
Abstract
A group of 40 human Ss was 1st given training in a modified verbal discrimination situation in which the 2 stimuli in each pair had different associated reward values, and then was tested on various new combinations of the training stimuli. On the basis of a final series of single-stimulus tests, gradients were constructed for the distributions of probabilities that Ss would tend to predict each of the possible reward values upon presentation of each of the stimuli singly. These gradients, together with a scanning model for choice responding, generated relatively accurate predictions of the proportions of choices to test combinations. A comparison of these data with those from a similar experiment conducted with monkeys by Meyer, LoPopolo, and Singh (see 40:9) tends to support a common interpretation of the operation of reward. (PsycINFO Database Record (c) 2006 APA, all rights reserved)

This publication has 0 references indexed in Scilit: