Extinction of Sidman avoidance behavior.
Turn off the aversive event completely to end avoidance fast—warning cues alone won’t keep it going.
01Research in Context
What this study did
Thomas (1968) worked with rats in a Sidman avoidance box.
Thes animals could press a lever every 15 seconds to stop a mild shock.
The team then turned the shock off completely to see how fast avoidance would stop.
They also tested if a warning light would keep the rats pressing even when no shock came.
What they found
All rats quit pressing the lever in under four hours once the shock was gone.
The warning light slowed the drop a little, but the rats still stopped.
The light alone did not keep the avoidance going.
How this fits with other research
Kaufman (1965) mapped the normal pattern of lever pressing before any extinction.
That baseline helped Thomas (1968) know what to expect when the shock disappeared.
McConnell et al. (2020) later used the same idea with autistic young men at the dentist.
They removed escape breaks for screaming and saw quick, large drops in disruption.
Greer et al. (2024) showed that if you cut reinforcement too fast after extinction, problem behavior can pop back up.
This warns you to fade slowly after you first remove the aversive event.
Why it matters
When you stop an escape behavior, remove the aversive event 100 percent.
Warning lights, countdown timers, or staff cues will not keep the behavior alive on their own.
Fade any new reinforcement slowly to avoid resurgence later.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Identify the exact aversive event your client escapes from and block it fully during the next session.
02At a glance
03Original abstract
Extinction of Sidman avoidance behavior by eliminating the noxious stimulus was studied in Sprague-Dawley rats with bar-pressing as the response. Each of three subjects was trained and extinguished on each of the following schedules in a different order: nondiscriminated, response-shock interval = 20 sec, shock-shock interval = 5 sec; nondiscriminated, response-shock interval = 40 sec, shock-shock interval = 5 sec; discriminated, response-white noise interval = 15 sec, noise-shock interval = 5 sec, shock-shock interval = 5 sec. Less than one 4-hr session was required for extinction for all procedures. When a warning stimulus was present, resistance to extinction increased. Subjects did not, however, respond to avoid the signal. Only small differences in extinction were found after training on different schedules with no warning signal.
Journal of the experimental analysis of behavior, 1968 · doi:10.1901/jeab.1968.11-153