Abstract
60 Ss were trained on an instrumental response for money. A CR training schedule was employed with half of the group and the remainder received PR. 10 Ss from each training condition subsequently received one of 3 types of instructions designed to manipulate the informative property of blanks. One instruction treatment defined blank (nonreward) outcomes for a response as a reliable “r” predictor of future nonreward, another identified blanks as an equivocal “e” predictor. The third instruction treatment served as a control, supplying no comment about blanks. The r treatment was predicted to decrease instrumental responses, e should increase responding and the control treatment should yield intermediate performance. Support for this hypothesis was obtained when behaviour was examined under intermittent punishment (50% electric shock and 50% blank outcomes) and under extinction (100% blanks). Some theoretical and practical implications of these findings were discussed.
Get full access to this article
View all access options for this article.