Extinction burst vs spontaneous recovery

An extinction burst is a temporary increase in the behavior soon after extinction begins, while spontaneous recovery is the behavior suddenly reappearing later, after it had already decreased. Both are expected side effects of extinction. The deciding clue is timing: early spike means burst; return after a quiet period means spontaneous recovery.

What is an extinction burst?

An extinction burst is a short-lived rise in the frequency, duration or intensity of a behavior right after the reinforcer that maintained it stops coming. A learner whose whining used to get the tablet may whine louder and longer for the first few sessions of extinction. It often shows the reinforcer has been successfully withheld.

Bursts can also include new forms of behavior, such as adding crying or aggression. That is why supervisors plan for safety before starting extinction.

What is spontaneous recovery?

Spontaneous recovery is the reappearance of a behavior that had decreased under extinction, usually after a break such as a weekend or holiday. The returning behavior is typically weaker and briefer than before. If it still produces no reinforcement, it tends to fade again; if it is reinforced, it can return to its old level.

Extinction burst and spontaneous recovery side by side

Extinction burstSpontaneous recovery
When it happensEarly in extinctionAfter behavior had decreased
What it looks likeMore, longer or more intense behaviorBehavior suddenly returns
Typical triggerReinforcer first withheldTime away, such as after a weekend
Usual sizeOften above baselineUsually weaker than baseline
Correct responseStay consistent with extinctionStay consistent with extinction

How do you tell them apart on the exam?

Locate the behavior on a timeline. If extinction has just started and the behavior jumps, it is an extinction burst. If the behavior had already dropped to low levels, perhaps to zero, and then reappears, especially after a break, it is spontaneous recovery. If a different, previously reinforced behavior returns instead, that points to resurgence.

  • Just started extinction, behavior worse: burst.
  • Was low or gone, now back: spontaneous recovery.
  • An old, different behavior returns when a new one stops paying off: resurgence.

Worked scenarios, answered: burst or recovery

Scenario 1. On day one of planned ignoring for calling out, a learner calls out 25 times instead of the usual 10. Extinction burst.

Scenario 2. Calling out drops to zero over two weeks. After spring break, the learner calls out four times on Monday, then stops. Spontaneous recovery.

Scenario 3. A learner hits the vending machine harder and harder after it stops giving snacks. Extinction burst.

Scenario 4. A learner taught to request breaks stops getting breaks when the plan is thinned, and old head-banging returns. That is resurgence, not spontaneous recovery.

What are the common traps with these two?

The most common trap is reading a burst as proof the plan is failing and giving in. Reinforcing the behavior during a burst can teach the learner that more intense behavior works. Another is mistaking spontaneous recovery for relapse that needs a new plan. A third is confusing spontaneous recovery with resurgence, which involves a different, older behavior.

  • Delivering the reinforcer to end a burst.
  • Expecting extinction to work without any increase.
  • Not recording behavior after breaks, so recovery looks like a mystery.

When would you see each in a session?

Expect a burst in the first sessions after your supervisor starts an extinction procedure, and expect possible spontaneous recovery after weekends, holidays or illness. In both cases, follow the plan consistently, keep everyone safe, and take accurate data. If behavior becomes dangerous or reaches a level the plan did not foresee, follow crisis steps and contact your supervisor.

Caregivers are often the ones who see recovery at home. Passing on their reports objectively helps your supervisor tell a normal side effect from a real setback.

Practice questions

Question 1 · Behavior reduction

On the first day attention is withheld for Gabe's door slamming, as his plan specifies, slamming jumps from about 6 times a session to 15. What is this most likely?

Question 2 · Behavior reduction

After three weeks of extinction, Tess's table banging dropped to zero. On the Monday after a long weekend, she bangs the table a few times, even though the behavior still produces no reinforcement. What is this called?

Question 3 · Behavior reduction

Early in an extinction procedure, a learner's problem behavior gets temporarily worse but stays safe to manage. What should the technician do?

Question 4 · Behavior reduction

Felix's screaming for attention was replaced by handing staff an "excuse me" card. On a short-staffed day, no one responds to the card for a long stretch, and Felix begins screaming again. What is this most likely?

Question 5 · Behavior reduction

After a team stops giving attention for whining, the whining briefly becomes louder and more frequent. What is the most likely explanation?

Questions people ask

Does every extinction procedure cause a burst?

Not always, but bursts are common enough that plans should prepare for them.

Is spontaneous recovery a sign extinction failed?

No. It is an expected pattern. Continued consistency usually lets the behavior fade again.

How ready are you?

Ten questions, about 3 minutes, no account. Get your readiness score and your weakest domains.

Check my readiness