Relation of amount of primary reinforcement to discrimination and to secondary reinforcement strength.

Abstract

Thirty-six male rats of Wistar strain were divided at random into six equal groups and maintained on 22 hr. of food deprivation throughout the experiment. Each group was given light-dark discrimination training on the bar-press response for one of six sucrose solutions ranging from a 5% concentration to a 50% concentration. Reinforcement was administered on a 1-min. variable-interval schedule in the light, and reinforcement was withheld altogether in the dark. Following discrimination training, all animals were given further extinction in the dark until all had reached the same level of responding, At this point the light was reintroduced for 1 sec. following each response. Following this procedure, all animals were extinguished in the presence of the light until all had reached the same level of responding. The sound of the liquid feeder (dipper) was then reintroduced following each response. The rate-concentration function in the light was ostensibly linear between the 5% and the 32% sucrose concentration but dropped sharply between the 32% and the 50% concentration. By nonparametric one-way analysis of variance the function was significant (p < .005). The response rate in the dark was well below the rate in the light and was apparently unaffected by changes in the value of the reinforcing agent, rate differences between concentrations being statistically unreliable (p > .90). The rate-concentration function for the light as a secondary reinforcer appeared to be monotonic and negatively accelerated and to approach an asymptote between the 32 % and the 50% concentration. By nonparametric analysis of variance the function was significant (p < .01). The test for the dipper sound as a secondary reinforcer yielded a relationship to the sucrose concentration similar in shape to the function obtained with the light. Application of the same statistical test indicated significant over-all differences for this function also (p < .04). (PsycINFO Database Record (c) 2006 APA, all rights reserved)