Punishment of an extinguishing shock-avoidance response by time-out from positive reinforcement.

Nigro (1966) tested whether a 30-second time-out from food would stop an avoidance response that was already dying out. The subject pressed a lever to avoid mild shock. Once the shock stopped, the response kept going even though it no longer helped. Researchers then gave a 30-second break from food each time the subject pressed. They also tried giving the same break at random times to see if timing mattered.

The study used a single-case lab design. It compared contingent time-out (only after the lever press) with noncontingent time-out (given no matter what).

When the 30-second food break followed the lever press, the response stopped fast. Latencies (time before pressing) grew and responses dropped to zero. When the same break came at random, little changed. The result shows timing beats mere exposure.

Fantino (1967) ran a near-copy the next year and got the same punch line: contingent time-out beats noncontingent. The pair forms a clean replication set.

Green et al. (1975) flips the story. In their shock-maintained task, removing the chance for a post-response time-out killed the behavior. That means time-out can act as a reward, not a penalty, when shock still runs the show. The clash is only skin-deep: R punished an already-extinguishing avoidance response, while L reinforced shock-based responding. Same tool, opposite jobs.

Later work stretched the idea in two ways. Leander et al. (1972) tried 1-, 15-, and 30-minute time-outs with children and found 15 min works as well as 30, showing the 30-second dose is only the starting point. Bell (1999) showed that once time-out maintains a habit, that habit dies hard—even after the original avoidance is gone.

For BCBAs, the core lesson is contingency. Time-out only suppresses behavior when it is a clear, immediate consequence of that behavior. Random breaks, or breaks given for other reasons, will not do the job. Check your timing first, then adjust length if needed.