Analysis of the Partial Reinforcement Extinction Effect in Humans as a Function of Sequence of Reinforcement Schedules

Abstract

The present experiment was conducted to investigate the effects that changes in reinforcement schedule have on the persistence of learned responses. Human subjects learned two simple tasks concurrently, both tasks being under either continuous reinforcement (CRF) or partial reinforcement (PRF). After 50 acquisition trials on both tasks, half the subjects in each schedule condition received an additional 50 acquisition trials with the same reinforcement schedules on the two tasks. The remaining subjects received 50 additional trials with the opposite reinforcement schedule on both tasks. All subjects then received 50 extinction trials on both tasks. The extinction data were biphasic, consisting of a short-term phase of declining performance in which group differences were clear, followed by a long-term phase in which the groups converged to a uniform high level of persistence. A conventional partial reinforcement effect was observed during the short-term phase, but only in terms of the most recently experienced schedule, that is, the first schedule experienced appeared to have no effect upon persistence, contrary to some current theories of the partial reinforcement effect. The role of task demands in this type of research is discussed.