Effect of differential quantity of reward on acquisition and performance of a maze habit.

Abstract

Forty albino and hooded rats were trained to differentiate between a one- and a five-pellet reward placed at the ends of opposite arms of a single-choice T maze. After the subject (S) reached criterion or had 65 trials, it was placed in one of two differentially rewarded groups and given 20 trials, one per day, in a Stone 14-unit multiple-T maze. Both groups were on a 22-hr. hunger drive. Two and one-half months following the 20th trial, Ss were placed on a 22-hr. thirst drive, familiarized with drinking in a goal box, and then were given five relearning trials in the multiple-T maze. All Ss remained in the goal box until they had finished drinking. No significant differences between mean error scores were found in the original learning situation. By trial 20, there were significant differences between the two groups in terms of mean running time, favoring the group receiving the greater reward. No significant differences were found in either mean errors or mean time scores of the two groups during the five relearning trials. These data do not support the hypothesis that quantity of reward affects the acquisition of a habit per se, but suggest that the quantity of reward influences action potential or utilization of the habit.