Maintaining responding during stimulus generalization testing in extinction.
Finish each session with one big reinforcer after a long wait to keep stimulus control strong during later extinction probes.
01Research in Context
What this study did
Crossman et al. (1973) worked with lab rats in a small chamber.
They wanted to see if the last reinforcer of the day could protect stimulus control during later extinction tests.
One group ended every session with a single large pellet after a long, stretching wait.
The other group got many tiny pellets on a short, random schedule.
Next day both groups faced generalization probes with no food.
What they found
The rats that got the big final pellet kept pressing the lever longer.
Their generalization curve stayed tall and smooth across repeated probes.
The short-schedule rats quit fast and their curve flattened.
One big reinforcer at the end acted like a shield against extinction.
How this fits with other research
Horner (1971) showed that non-contingent food makes behavior drop faster.
That sounds opposite, but D spread food all through the session, not just at the end.
Timing matters: early free food weakens control, while one final earned food protects it.
Terrace (1974) tracked ‘active non-responding’—rats freezing to avoid frustration.
K’s terminal reinforcer kept the rats working, so they never reached that freeze point.
Together the papers say: arrange the last reinforcer, not the first, to keep behavior alive.
Why it matters
When you probe stimulus generalization, clients often stop responding after the first unreinforced trial.
End the session with one large, delayed reinforcer the learner has to work for.
This simple shift keeps the skill strong across future extinction tests and makes your data cleaner.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →After the last trial of your probe session, set a 2-min VI schedule and deliver a double-size edible for the final response.
02At a glance
03Original abstract
Resistance to extinction and generalization gradients were studied following training with a long-adjusting-interval schedule. One large reinforcer occurred at the end of each daily training session. Sessions varied in length from 20 sec to 42.66 min, but were usually the latter. Repeated generalization tests were subsequently conducted for these subjects and subjects trained with a more conventional short-random-interval schedule. The long-adjusting-interval schedule produced generalization gradients that were not qualitatively different from those produced by the conventional procedure. However, the advantages of the long-adjusting-interval schedule are: (1) greater resistance to extinction both within and across generalization tests and (2) more stable gradient slopes within and across tests.
Journal of the experimental analysis of behavior, 1973 · doi:10.1901/jeab.1973.19-199