Generalization of expectancy changes as a function of the nature of reinforcement.

Publisher Website
Google Scholar
Abstract
To test whether a change in behavior in S1 will generalize more to S2 the more S2 contains cues that have been associated with the same reinforcement as the cues in S1, 24 rats both hungry and thirsty throughout the expt. were divided into two major groups and each learned to turn right in a white T1-maze and left in a black T1-maze. For Group A the responses in the two situations were followed by different rewards (either water or food), while for Group B the same responses were followed by the same rewards. After training to a criterion of 18/20 correct choices, one response in one maze was extinguished to a criterion of 90-sec. latency. Then the other response was extinguished to the same criterion in the second T1-maze. The results in this limited laboratory situation offer statistical support for the hypothesis that the effect of the extinction in Maze 1 would generalize more to Maze 2 for the group which received the same rewards, i.e., for the group for which the cues in Maze 2 had been paired with the same reward as in Maze l.

This publication has 0 references indexed in Scilit: