Occasion setting without feature-positive discrimination training

Abstract

In four experiments rats received training in which a brief conditioned stimulus, x, occurred during presentations of an extended feature stimulus, A; a second brief stimulus, y, occurred during presentations of a second feature, B. In Experiment 1 both x and y were paired with food and, in a subsequent test, they elicited more conditioned responding when they were presented in the feature with which they had been trained than when they were presented in the alternative feature. The results of Experiments 2 and 3 suggested that this effect was not the result of a return of unconditioned responding to x and y when they were presented outside the training feature. Experiment 4 employed a blocking test to demonstrate that neither was the effect caused by a restoration of associability enhancing the rate of extinction to x and y outside the training feature. These results suggest that this training procedure gave the features occasion-setting properties. The implications of this possibility for the various accounts of occasion setting are discussed.