Resurgence: Response competition, stimulus control, and reinforcer control.
Keep the alternative response option visible after you stop reinforcing it—removing the cue alone won’t prevent relapse.
01Research in Context
What this study did
The team ran a lab resurgence test with three phases. First, participants earned tokens for pressing one button. Next, that button stopped paying off and a new button paid instead. Last, both buttons paid nothing.
In the final phase the researchers pulled away the second button for some participants. They wanted to know if the cue or the lost reward drove the old button-pressing to return.
What they found
Taking the alternative button away made the old response come back faster. Total relapse stayed the same. The lost reward, not the missing cue, controlled the final burst.
How this fits with other research
Bai et al. (2016) seems to disagree. They say stimulus-reinforcer links set during training rule later relapse. The two studies clash because Bai watched long persistence while A et al. watched speed of return; both can be true.
King et al. (2025) sweep up both papers in a new review. They side with A et al.: what the reinforcer used to do matters most when you pick relapse-prevention rewards.
Kimball et al. (2025) add another wrinkle. They show relapse gets worse when a changed room plus accidental reward team up. Their work extends A et al. by showing real-world relapse can stack several triggers at once.
Why it matters
When you fade DRA, leave the alternative item or task in view. Pulling the cue away will only hurry the old problem behavior back. Focus on thinning the reinforcer, not hiding the stimulus, and watch for any accidental payoff that could turbo-charge relapse.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →During DRA fade-out, leave the alternative material on the table while you thin the schedule; do not remove it.
02At a glance
03Original abstract
Resurgence is the relapse of a previously reinforced and then extinguished target response when extinguishing a more recently reinforced alternative response. We designed the present study to assess the contribution of stimulus-control and reinforcer-control processes in determining resurgence. In a modified resurgence procedure, we removed the alternative discriminative stimulus signaling alternative reinforcement when extinguishing the alternative response. This produced more abrupt resurgence of target responding than in a typical resurgence procedure maintaining the alternative discriminative stimulus when extinguishing the alternative response. The overall amount of resurgence did not differ. Importantly, a "renewal" control added and removed the alternative stimulus during extinction, identically as in the modified resurgence procedure. However, alternative responding was never reinforced, which produced no relapse of target responding. Therefore, the more abrupt resurgence with the modified procedure than with the typical procedure suggests removing the alternative stimulus reduced the competition between alternative and target responding. These findings revealed the importance of adding and removing alternative reinforcement in producing resurgence (reinforcer control) but little influence of simply adding and removing the alternative stimulus (stimulus control). These data suggest that clinicians should consider the long-term availability of the alternative response option when developing differential-reinforcement interventions.
Journal of the experimental analysis of behavior, 2014 · doi:10.1002/jeab.102