SEQUENTIAL RESPONSE EFFECTS IN THE WHITE RAT DURING CONDITIONING AND EXTINCTION ON A DRL SCHEDULE.

FERRARO et al. (1965) watched rats press a lever under a DRL 60-s schedule. The rats had to wait at least one minute between presses to earn food.

The team tracked the exact time between responses. They wanted to see if one pause predicted the next pause.

Reinforced pauses clumped together. If a rat just earned food for waiting 60 s, the next pause was also likely to earn food.

When the schedule stopped paying off, responding died out fast. When food came back, the rats quickly returned to their old pause pattern.

REYNOLDS (1964) showed pigeons can space responses under DRL 20 s. P et al. moved from ‘can they wait?’ to ‘does the last wait predict the next wait?’

Mintz et al. (1966) added skin-potential wires to the same DRL 60-s setup a year later. They found bigger stress signals after non-reinforced responses, backing up the idea that reinforced IRTs feel different.

Richardson (1973) later proved DRL actually cuts overall response rate versus a matched VI schedule. P et al. gave the microscopic reason: rats bunch their successful long pauses together.

If you use DRL to slow a client’s behavior, know that reinforced waits may cluster. One successful long pause makes the next long pause more likely. Watch for these ‘hot streaks’ and don’t mistake them for accidental improvement. When you restart reinforcement after a break, the old pause pattern can bounce back quickly—so keep data collection running.