The influence of two variables upon the establishment of a secondary reinforcer for operant responses.

Abstract

The acquistion by an initially neutral stimulus of power to reinforce operant responses was studied as a function of its temporal relation to, and the number of times it was paired with, a primary reinforcing stimulus. In Expt. I, 6 groups of white rats were given 160 paired presentations of a continuous light and a pellet of food. The light durations prior to pellet delivery were 0, 0.5, 1, 2, 4, and 10 sec., respectively. A subsequent test was made of the relative effectiveness of the light in conditioning a new operant. The obtained function suggests a maximum effectiveness in the 0-2 sec. range. The light was significantly less effective as a secondary reinforcer for the 10 sec. group than for the other groups. In Expt. II, 5 exptl. groups were given 10, 20, 40, 80 and 120 regular reinforcements for bar-pressing. Each response produced a 3-sec. light, delivery of a food pellet occurring 1 sec. after light onset. A control group was also given 120 regular reinforcements, each reinforcement occurring 1 sec. after bar-pressing but without an accompanying light. Subsequently, the light was tested as a secondary reinforcer. The resulting function was a negatively accelerated increasing curve of low initial slope. The light was significantly more effective as a secondary reinforcer for the 120 reinforcement exptl. group than for the control group. Other exptl. groups were, however, not significantly better than the control group.