Reinstatement and a resurgence‐like effect in an odor‐signaled multiple schedule in rats
Rich reinforcement history in one signaled context leaks resurgence into other contexts when extinction hits.
01Research in Context
What this study did
Mason's team worked with 12 rats in a three-part schedule. Each part had its own smell cue and its own lever. One part gave lots of pellets. One gave fewer. One gave none. After training, they stopped pellets in the rich part only. They watched if pressing jumped in the other two parts. Later they gave a single free pellet to see if pressing came back.
They ran the whole thing twice to check the pattern.
What they found
When the rich-smell lever stopped paying off, the rats started pressing the other levers more. It looked like resurgence, but it happened without extinction on those levers. The biggest rebound showed up where pellets had been richest and most recent. One free pellet later brought the top pressing back strongest in that same rich spot.
How this fits with other research
Greer et al. (2024) saw the same thinning-risk rule in kids with destructive behavior. Big early drops in reinforcement caused the biggest resurgence. Mason's rats prove the rule holds even when the drop is in a different smell context.
Arroyo Antúnez et al. (2026) used mice and found large alternative rewards knock behavior out fast but rebound harder. Mason's rich-pellet part fits that story: more reinforcement history, more later relapse.
Fisher et al. (2019) showed higher baseline rates during FCT bred more resurgence in children. Mason's rats mirror that: the component with the richest past produced the strongest return.
Why it matters
If you run multiple schedules or DRA in separate settings, do not assume extinction in one place keeps the others safe. Thoroughly thin or extinguish across all relevant cues. Check each context for resurgence, especially where reinforcement was richest. One quick probe with a single reward can show you where relapse will hit hardest.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Probe each signaled context for resurgence after you start extinction in the richest one.
02At a glance
03Original abstract
This study examined determinants of relapse in rats within a multiple schedule signaled by three odor stimuli (S1, S2, S3). In both experiments, reinforcement was programmed for all components in Phase A, extinction in S1 was programmed in Phase B, extinction in both S1 and S2 in Phase C, and extinction in all components in Phase D. In Experiment 1, an increase of S1 responding was observed in Phase C when extinction was introduced in S2. This resurgence‐like effect was not observed in Phase D when extinction was present in all three components. Experiment 2 tested whether the mitigation of relapse observed in Phase D was due to the absence of reinforcement in any component, but the results were equivocal. In Phase E, noncontingent food was delivered in the intercomponent interval or at the termination of S1. Reinstatement was generally highest in S3, which was associated with the greatest overall reinforcement amount and recency. When total reinforcement amount was equated across S1 and S2, more reinstatement was observed in the more recently reinforced component. Results demonstrate the utility of using a multiple schedule with odors to examine relapse in rats.
Journal of the Experimental Analysis of Behavior, 2026 · doi:10.1002/jeab.70079