Continuous vs Partial Reinforcement

When reinforcement is arranged so that a target response is reinforced every single time it occurs, this is called continuous reinforcement; this schedule produces a high rate of responding while it lasts, but once reinforcement is withheld, response rates drop off very quickly and the response soon extinguishes, because withdrawal of the (previously constant) reinforcer is very easy for the organism to detect. When reinforcement is not given after every occurrence of the response — that is, only some responses are reinforced — this is called partial or intermittent reinforcement. Partial schedules, particularly ones based on a ratio of responses (reinforcing only after a certain number of responses have occurred), tend to produce very high rates of responding, since the organism must respond frequently to have any chance of being reinforced. Because reinforcement is not given every time under a partial schedule, it becomes difficult to tell exactly when a reinforcement has been discontinued completely and when it has merely been delayed — and this uncertainty is exactly why extinction of a response is markedly more difficult, and takes far longer, following partial reinforcement than following continuous reinforcement. This finding, that responses built up under pa …