Abstinence as Choice: Exploring Voluntary Abstinence from Alcohol Self-Administration Using the Resurgence-as-Choice Framework
Resurgence can return even when the problem behavior is only displaced, not extinguished—plan for this in alcohol treatment.
01Research in Context
What this study did
Craig et al. (2024) looked at rat data on voluntary alcohol abstinence. The rats could press a lever for booze or for food. No one removed the booze. The animals simply chose to skip it when food paid better.
The authors asked, "Can the Resurgence-as-Choice model explain this pattern?" They mapped each drink skip as a choice, not as behavior that had been extinguished.
What they found
The model fit. Even without extinction, the old booze pressing came back when the food payoff shrank. The team showed that resurgence can happen even when the target behavior is only displaced, not wiped out.
They warn the model still needs more checks before we trust the numbers in the clinic.
How this fits with other research
Greer et al. (2019) first wrote the Resurgence-as-Choice equations. Craig et al. (2024) simply widen the lens to voluntary abstinence. Same math, new story.
Greer et al. (2024) later tested kids with destructive behavior. They found that sharp early drops in alternative reinforcement spark the biggest comeback. Craig's rat data line up: when the food lever quits paying well, booze pressing surges.
Shahan et al. (2020) showed the same rebound in pigeons when reinforcer rate fell. Craig's work extends the idea to alcohol use without using extinction at all.
Why it matters
You now have a choice-based way to think about relapse. If a client skips alcohol because gym tokens pay better, expect a slip the day the tokens stop. Build slow thinning of alternative rewards into the discharge plan. Track the value of the new behavior, not just the old one.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Plot the value of the client's alternative behavior on a graph and thin the payoff in small steps, not chunks.
02At a glance
03Original abstract
Resurgence is an increase in the rate of a previously suppressed behavior that occurs when an alternative source of reinforcement is made worse in some way. The Resurgence as Choice model offers a quantitative approach to understanding resurgence that may provide important insights into the variables that affect this form of relapse in the natural environment. Bringing this model to bear on relapse following reinforcement-based interventions for alcohol and other substance use disorders, however, may not be straightforward. Laboratory work on which the Resurgence as Choice model is based has almost exclusively focused on resurgence following extinction of target behavior, but abstinence from alcohol during intervention is often voluntary: Patients may drink alcohol and forfeit therapeutic reinforcers at any time. In this article, we first will review recent data from our group that demonstrate a method for studying resurgence following voluntary abstinence from alcohol seeking in rats. In a previous experiment, we reduced rats’ alcohol-maintained lever pressing to low levels without placing it on extinction by arranging nondrug differential reinforcement of other behavior. Further, when we suspended nondrug reinforcement, resurgence of lever pressing occurred. Next, we will explore methods for modeling these outcomes using the Resurgence-as-Choice framework. We conclude that the data under consideration may not be sufficient to discriminate between candidate models of resurgence following voluntary abstinence and point to areas for future empirical and theoretical development. This work may provide a stronger bridge between preclinical and conceptual work on resurgence and clinical treatments for alcohol use disorder.
Perspectives on Behavior Science, 2024 · doi:10.1007/s40614-024-00405-5