The partial reinforcement effect sustained through blocks of continuous reinforcement.

Abstract

Three groups of 14 rats each received 70 trials of random 40% reinforcement in a runway. Then different numbers of continuously reinforced trials were interpolated between the 40% reinforcement and extinction. Group P-O received none, Group P-25 received 25, and Group P-70 received 70 interpolated 100% reinforced trials. Two control groups (N = 9 each) received only continuous reinforcement during training: Group C-0 was extinguished with partial Group P-0; Group C-70 was given 70 additional reinforced trials and extinguished with Group P-70. There was no significant difference in resistance to extinction between the two continuously reinforced groups. Group P-70 resisted extinction significantly (P < .001) than the two continuously reinforced groups. Resistance to extinction was not significantly different between Groups P-0 and P-25, but of these groups both were slightly more resistant than Group P-70. These results cast doubt on the adequacy of the discrimination hypothesis that the partial reinforcement extinction effect results from difficulty in discriminating the transition from training to extinction.