What Just Changed?

 

Behavior can improve for days or even weeks and then suddenly return after something changes.

A new teacher.

A different room.

A new task.

A shift in routine.

That does not necessarily mean the original intervention failed.

A 2026 study published in the Journal of Applied Behavior Analysis examined 98 clinical patients and 749 context changes during treatment for challenging behavior. Researchers looked at changes involving the person implementing treatment, the setting, and the task.

They identified 189 instances of behavioral renewal following a context change.

The return of behavior was often fast.

51.3% of renewal episodes appeared during the very first session after the change.

The average onset was just 1.72 sessions.

That alone is important, but the persistence data may be even more useful.

Almost half of the renewal episodes disappeared after one session. However:

32.3% persisted for at least three sessions

and

12.2% persisted for at least five sessions.

Interestingly, the type of context change—whether it involved a new person, setting, or task—did not meaningfully predict how quickly the behavior returned or how long it lasted.

What did matter was the size of the initial return. Larger increases in behavior were associated with more persistent renewal.

This has a practical implication for anyone observing behavior over time.

Imagine a student whose challenging behavior decreases for three weeks. Then a new teacher takes over, and the behavior suddenly increases.

It may be tempting to conclude that the original understanding of the behavior was wrong.

But that is not necessarily the case.

The new findings reinforce an important principle in behavior support:

Context matters.

Behavior does not occur in a vacuum. Changes in people, places, tasks, routines, expectations, and other environmental conditions can affect whether previously reduced behavior reappears.

That means behavioral data are more meaningful when they include information about what was different at the time.

A sudden increase after a major context change should not automatically be treated as proof that everything previously learned was incorrect.

Instead, it may be a signal to observe carefully, gather additional data, and see whether the change is temporary or persistent.

For parents, educators, trainees, and behavior-support professionals, this is an important reminder:

When behavior changes, ask what else changed too.

That question may explain far more than the behavior alone.

Source: Journal of Applied Behavior Analysis, 2026.