Extinction-induced aggression.
Stopping reinforcement can spark aggression even in calm learners—plan for it and use the burst to teach new skills.
01Research in Context
What this study did
Researchers worked with pigeons in a small lab box.
Each bird had learned to peck a key for food.
Then they stopped giving food for pecking.
They watched what the birds did next.
What they found
Every time food stopped, the birds attacked.
They pecked hard at other birds or at fake birds in the box.
The more food they used to get, the harder they attacked.
This happened even though the birds had never fought before.
How this fits with other research
Harrison et al. (1975) ran the same test and got the same result.
They also showed that giving less food during training makes later attacks weaker.
Corfield-Sumner et al. (1977) added a twist.
They found that DRO works as well as plain extinction to stop reinforced aggression.
But neither method stops the extra aggression that pops up right after food.
Cengher et al. (2020) flipped the problem into a solution.
They used the burst of new behaviors that comes with extinction to teach kids with autism better ways to ask for things.
Matson et al. (2011) showed the same idea works with vocal words.
When they stopped giving treats for sign language, the kids started talking more.
Why it matters
When you stop giving reinforcement, expect a storm.
The behavior may get worse before it gets better.
Watch for hitting, yelling, or other new problems.
Have a plan ready.
You can use the burst of energy to shape better skills.
Reinforce any small good behavior that shows up during the storm.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Before your next extinction session, set a timer for 5 minutes and count any new aggressive or novel responses—then immediately reinforce calm or appropriate alternatives.
02At a glance
03Original abstract
Pigeons were conditioned to peck a response key under a procedure that alternated periods of food reinforcement with periods of extinction. The pigeons attacked a nearby pigeon at the onset of extinction. Some also attacked a stuffed model of a pigeon. The duration of attack was an inverse function of the time since the last food reinforcement and a direct function of the number of reinforcements. The pigeons attacked after the last food delivery whether or not the conditioned pecking response was required and whether or not the extinction period was signaled. The food had to be eaten; the mere sight and sound of food being delivered did not produce attack. Prior satiation reduced attack. The phenomenon was not attributable to a past history of competition between pigeons since socially deprived pigeons also attacked. Superstitious reinforcement of attack was not found to be a factor. The results indicated that the transition from food reinforcement to extinction was an aversive event that produced aggression.
Journal of the experimental analysis of behavior, 1966 · doi:10.1901/jeab.1966.9-191