Evaluating resurgence procedures in a human operant laboratory.

Resurgence of previously extinguished behavior may occur when a recently reinforced alternative response is placed on extinction. Understanding the conditions that produce and reduce resurgence is important for both basic and applied researchers. Research on resurgence of human behavior may benefit from methods that facilitate comparison and replication of nonhuman animal studies. These studies often include an inactive control response to differentiate resurgence from extinction-induced variability. In contrast, human research typically does not. Sweeney and Shahan (2016) tested a brief, trial-based procedure that included an inactive control response with human participants, but they did not observe resurgence. The current study extended their methods by examining four different conditions in a free-operant task lasting <1h. Modifications across conditions included changing the number of response options available in each phase and how signals associated with each response were presented. Only one condition resulted in responding resembling resurgence. Our results suggest the utility of the inactive control response and the influence of contextual cues in human research should be investigated further.