Effects of positive or negative reinforcement history and extinction arrangement on resurgence

Abstract

The recurrence of previously reduced behavior following successful differential reinforcement has been termed resurgence. Across studies, resurgence has occurred following either positive or negative reinforcement, but no direct within-experiment comparisons have been conducted. The primary purpose of the current experiments was to address this gap by comparing resurgence following a history of positive or negative reinforcement with human participants responding to earn or avoid losing money. Across experiments, we manipulated the arrangement of the resurgence test following negative reinforcement; money losses were discontinued in Experiment 1 and made inescapable during Experiment 2. Avoidance responding persisted after the discontinuation of monetary loss during Experiment 1, precluding comparisons of resurgence. Conversely, unavoidable monetary losses resulted in greater resurgence relative to discontinuation of monetary earnings during Experiment 2. The present experiments provide the first controlled within-subject comparisons of resurgence following positive and negative reinforcement and show that the form of extinction affects avoidance responding.