Response control with titration of punishment.

The researcher put pigeons and rats on a special shock schedule. Each peck or lever press raised shock intensity a little. If the animal waited, the shock dropped. The animal, not the experimenter, set the pain level.

The study also tested steady shock that never changed. The goal was to see if the animals would keep a steady response rate to hold shock at one intensity.

The birds and rats settled into a narrow band of responses. This kept shock intensity almost flat. They acted like living thermostats for pain.

When shock stayed on at a fixed level, response rates only rose if the shock was continuous. Pulsed shock did not push rates up.

Dallemagne et al. (1970) ran the same titration setup two years earlier with rats. Their animals also held shock low and stopped responding when shock was turned off. Rachlin (1972) widens the picture by adding pigeons and comparing steady versus pulsed shock.

Mosk et al. (1984) later showed that, in signaled avoidance, more shock past a minimum does not help and can even hurt. Rachlin (1972) lines up: organisms aim for one intensity, not "as low as possible."

Zimmerman (1969) found that fixed-interval shock can maintain positively accelerated responding. That looks like a contradiction—shock as reinforcer versus shock as punisher. The difference is schedule: fixed-interval arranges shock after time, titration arranges shock after each response. Timing, not the shock itself, decides the effect.

When you thin consequences, remember that clients may adjust their behavior to keep feedback at a familiar level. If you fade punishment too fast, they might speed up responses to bring the intensity back. Use titration logic in your thinning steps: let the learner’s behavior guide the level, and hold the schedule rules steady.