Choice of a terminating over a non-terminating signal in free-operant avoidance.

Researchers gave four rats two free-operant avoidance schedules. Both schedules delivered a warning light before possible shock.

In one schedule, a single lever press turned the light off right away. In the other, the light stayed on for 30 extra seconds no matter how many times the rat pressed.

Every rat picked the schedule that let them shut the light off immediately. They never chose the 30-second delay option.

The chance to end the warning signal acted like a strong reward.

Hearst et al. (1970) looks like the opposite story. They added a one-minute warning that could not be avoided. The rats stopped pressing and got more shocks. The difference is simple: in the 1970 study the signal came with unavoidable shock, while in Bennett et al. (1973) the signal could be ended by the rat.

Liberman et al. (1973) and Badia et al. (1972) line up with the new finding. Rats picked signaled shock even when it was longer or stronger, showing that predictability beats intensity.

Glover et al. (1976) adds a twist: rats only prefer the signal when shocks are far apart. When shocks come every 45 seconds, the value of the safety signal drops.

Your prompts, timers, or warning cues are like those lights. If a client can turn them off with a correct response, the prompt itself becomes a reinforcer. If the prompt stays on, it may lose value or even suppress responding. Next time you use a visual or verbal cue, give the learner a clear way to end it immediately after the right response.