The removal and restoration of stimulus control.

The team first taught a simple circle-versus-ellipse discrimination.

Then they let the control weaken so the learner mixed up the two shapes.

Finally they ran two short retraining rounds to see if the original, crisp choice would come back.

After only two targeted lessons the learner again picked the correct shape almost every time.

The old, accurate pattern returned without starting the whole program over.

REYNOLDS (1964) saw the same speed in a DRL study: two well-timed pellets brought the old 20-s pause pattern right back.

Schroeder et al. (2014) stretched the idea to stimulus-equivalence. They showed that untested relations stay quiet after extinction unless you first check them—evidence that control can be revived only under the right conditions.

Jackson et al. (2026) flip the coin. Instead of restoring lost control, they keep it from slipping in the first place by fading the therapy room back in after DRA. Renewal dropped from 94% to 50%.

If a client starts confusing SDs after a break, you do not need to rebuild the whole discrimination chain. Run a quick review session with clear reinforcement for the correct choice. Two or three good trials can bring the old topography back, saving you hours of reteaching and sparing the learner needless frustration.