A Theory of the Extinction Burst
Extinction bursts are a traffic jam of activities, not tantrums — give the behavior another lane.
01Research in Context
What this study did
Shahan (2022) wrote a theory paper. No kids, no rats, no data sheets.
He asked: why do we see an extinction burst — a brief jump in the very behavior we just stopped reinforcing?
Instead of blaming frustration, he framed the burst as a traffic jam. Remove one reinforcer and other activities race to fill the open lane.
What they found
The model says the burst is normal competition, not emotional fallout.
When the old reinforcer vanishes, responses linked to other reinforcers gain strength for a moment.
Keep those other activities easy to reach and the burst shrinks.
How this fits with other research
Stancliffe et al. (2007) saw aggression during extinction and called it spontaneous recovery. Shahan keeps the spotlight on competition, not resurgence labels.
Laugeson et al. (2014) showed that yanking the alternative cue hastens relapse. That lab data lines up with Shahan’s advice: leave the alternative, lose the burst.
Craig et al. (2016) found high-rate alternative reinforcement fuels resurgence, a result that momentum theory missed. Shahan’s traffic-lane view swallows the same finding without bending rules.
Bai et al. (2016) warned that extra alternative reinforcement can make target behavior tougher to extinguish. The papers seem to clash — more alternatives look bad in Bai, good in Shahan. The gap is timing: Bai measured long-run resistance; Shahan explains the first-minute spike.
Why it matters
Next time you run extinction, watch for a quick flare. That jump is not the child “testing you.” It is the moment other reinforcers pull response strength.
Keep preferred toys, songs, or tasks within reach during the first extinction minutes. You are not bribing; you are giving the behavior system somewhere else to go.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Place two highly preferred alternatives right next to the client before you start extinction — keep them there for the first five minutes.
02At a glance
03Original abstract
A preliminary theory of a temporary increase in the rate of an operant response with the transition to extinction (i.e., the extinction burst) is proposed. The theory assumes reinforcers are events permitting access to some valuable activity, and that such activity can compete for allocation with the target response under some conditions (e.g., very high reinforcement rates). With the transition to extinction, elimination of this competition for allocation can produce an increase in the the target response, but the increase is transient because the value of the target response decreases with exposure to extinction. The theory provides a way to understand why the extinction burst is not ubiquitous, seems more common following very small ratio schedules, occurs for a short period of time following the transition to extinction, and may be eliminated with the availability of alternative reinforcement. It appears to provide a reasonable starting point for a theory of the extinction burst that does not necessarily require inclusion of invigorating effects of frustration, and it is closely aligned with Resurgence as Choice theory. Additional research on factors modulating reinforcement-related activities and how they affect the extinction burst could help to further evaluate the theory.
Perspectives on Behavior Science, 2022 · doi:10.1007/s40614-022-00340-3