Secondary Reinforcement and Frustration

Abstract

It was pointed out that conventional tests for secondary reinforcement are also theoretically optimal conditions for the production of aversive frustration. Hence, positive demonstrations of secondary reinforcement create a paradox for frustration theory. A resolution of this paradox is offered in terms of the chain of behavior cued off by the secondary reinforcing stimulus. The analysis implies that secondary reward effects will be maintained longer the longer is the chain of behavior cued off by stimuli presented following the to-be-learned response. To test this prediction, rats were first trained to run an alley for food reward. Subsequently, they learned to press a lever to get out of the start box; Ss in the Short-chain condition exited directly into the goal box, whereas Ss in the Long-chain condition ran through the alley to the goal box. Sixty tests were given with food reward omitted from the goal box. As predicted, the bar pressing performance of the Long-chain Ss was consistently superior to that of the Short-chain Ss.