Comparison of reinforcement and test trials in paired-associate learning.

Abstract

In 2 experiments, 192 undergraduates learned pairs under 8 conditions varied systematically with respect to densities of reinforcement (R) trials and unreinforced test (T) trials, while the total trials (time) were held constant. Although the forgetting prevention effects were shared by both R and T trials, distinctive differences existed between the 2 types of trials, each having a unique function of its own. Conditioning effects were observed only during Rs. At the same time, R and T trials manifested intriguing interactions. Learning was enhanced by the existence of unreinforced Ts which did not produce conditioning by themselves and was retarded by the existence of Rs which acted as sole conditioning agents. However, there was an optimal combination of Rs and Ts which enabled S to use a given total time most efficiently. (PsycINFO Database Record (c) 2006 APA, all rights reserved)