Cues paired with alternative reinforcement mitigate resurgence in humans

Abstract

Resurgence is the return of a previously reinforced and extinguished operant response with the worsening of more recent conditions, including extinction of a more recently reinforced alternative response. Given the relevance of resurgence to relapse of clinically relevant behavior, this preclinical experiment evaluated a method for mitigating resurgence by arranging a cue paired with alternative reinforcement in adult humans. After reinforcing target button pressing with points during Phase 1, Phase 2 arranged extinction of target responding and reinforced alternative responding with point delivery and a paired cue. Phase 3 provided a within-subject evaluation of the presence and absence of the cue between two groups either with continued or removed alternative reinforcement. Target responding was greater during Phase 3 in the absence (i.e., resurgence) than presence of alternative reinforcement. Continued presentation of the cue decreased target responding with the removal but not when continuing alternative reinforcement. Alternative response rates did not differ, however, depending on the presence or absence of cues, which is inconsistent with interpretations based on conditioned reinforcement and generalization. These findings are consistent with previous studies with rats, pigeons, and children, though moderate effect sizes suggest future research is needed to identify variables that could enhance the effectiveness of such cues to mitigate resurgence clinically.