Visual orientation in children's discrimination learning under constant, variable, and covariable delay of reinforcement

Abstract

Sixty-four first-grade children learned a two-choice, simultaneous discrimination under one of four conditions of delayed reinforcement: (a) constant 2-sec delay; (b) constant 16-sec delay; (c) 2- and 30-sec variable delay; and (d) 2- and 30-sec covariable delay. The response measures consisted of acquisition data, visual orienting responses (eye movements), starting speeds, and response speeds. The results indicated that (a) constant 16-sec delay produced poorer discrimination performance than constant 2-sec delay; (b) constant 16-sec delay resulted in more visual orienting than constant 2-sec delay; (c) discrimination performance under covariable delay was superior to performance under constant 16-sec delay; and (d) differences in starting speeds between learners and nonlearners were not correlated with differences in the number of orienting responses. Alternative interpretations of delayed reinforcement effects were discussed.