The extinction burst: Effects of alternative reinforcement magnitude
A large alternative reinforcer during extinction can prevent the burst.
01Research in Context
What this study did
Shahan et al. (2025) worked with rats that pressed a lever for food. The team then stopped the food and gave the rats nothing, one pellet, or six pellets on a nearby tray.
They counted how many rats showed an extinction burst and how big the burst was.
What they found
The bigger the alternative snack, the smaller the burst. Six pellets often wiped it out.
One pellet helped a little. No pellets let the burst rage on.
How this fits with other research
Shahan et al. (2025) also showed that rich histories make bigger bursts. The two papers are mirror images: one changes history, the other changes what you offer during extinction.
Lambert et al. (2024) saw the same pattern in adults with disabilities. High-value reinforcers before extinction worsened bursts; the rat study shows high-value alternatives during extinction tame them.
Hatton et al. (1999) found bursts or aggression in about half of self-injury cases when extinction stood alone. The new data give a fix: pair extinction with a generous alternative reward.
Lerman et al. (1995) pegged burst prevalence at 24% across 113 datasets. The six-pellet condition in the rat lab drove it toward zero, showing one way to beat that baseline risk.
Why it matters
You can shrink or skip the extinction burst by giving a big, immediate reinforcer for another response the moment you start extinction. Try it next time you fade reinforcement for stereotypy or attention-maintained behavior: deliver a handful of highly preferred edibles or 3-4 minutes of a favorite game on an FR 1 schedule for a competing response while you withhold the old reinforcer.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Start extinction with a generous competing item in the other hand—think 5-6 bites, not 1.
02At a glance
03Original abstract
Discontinuing reinforcement for an operant behavior sometimes produces a transient increase in responding (i.e., an extinction burst). A matching-law based quantitative model posits that the extinction burst results from the elimination of competition from reinforcement-related behavior with the onset of extinction. Consistent with this suggestion, retrospective case reviews in applied settings suggest reduced prevalence of the extinction burst when alternative reinforcement is available, but there has been no relevant prospective research. The present experiment used rats and methods approximating conditions in applied settings to examine the effects of alternative reinforcement magnitude on the extinction burst. Lever pressing in baseline produced a one-pellet reinforcer on a variable-interval 1.5-s schedule before a within-session transition to extinction that included no alternative reinforcement, one pellet, or six pellets of alternative reinforcement. A robust extinction burst was obtained when no alternative reinforcement was available, but the prevalence and magnitude of the extinction burst was reduced with an alternative reinforcer available, more so with the larger alternative reinforcer. The data were well described by the model. Similar prospective studies directly examining related effects in basic and applied settings could strengthen the empirical basis of current practice and further evaluate the utility of the present theoretical approach.
Journal of the Experimental Analysis of Behavior, 2025 · doi:10.1002/jeab.70045