Discrimination of variable schedules is controlled by interresponse times proximal to reinforcement.

Takayuki and team worked with rats to see how the animals tell one reinforcement schedule from another. They compared variable-ratio (VR) schedules with tandem schedules that mixed variable-interval, differential-reinforcement-of-low-rate, and differential-reinforcement-of-high-rate parts. The rats had to press a lever to earn food.

The researchers tracked every lever press and the tiny pauses between presses, called interresponse times. They asked: do rats look at the whole pattern of presses, or just the last few pauses right before the food arrives?

Rats could tell VR from the other schedules only when the last few reinforced pauses were very different. If the pause pattern just before food stayed the same, the rats acted as if the schedules were identical.

The study backs a copyist timing model: animals copy, or repeat, the pause that just worked. Long-term rate did not guide choice; only the most recent pause mattered.

Galbicka et al. (1981) showed that clear signals tied to each schedule help animals match response rates to payoff rates. Tanno et al. (2012) zoom in tighter, showing that even without extra signals, the last few pauses alone can do the job.

Smith et al. (1975) found that the number of responses in a fixed interval can itself become a cue. The 2012 study widens that idea: not just response count, but the tiny pauses between responses, can act as the cue.

Castañe et al. (1993) saw that any intermittent schedule makes rats defecate, pointing to the time between any two food deliveries. Takayuki refines this: only the pauses that end in food, not the overall wait, control what the rat does next.

When you shape a new skill or thin a schedule, think about the pause that just earned reinforcement. If you want the learner to tell two schedules apart, make the last response pattern noticeably different. Do not trust that overall rate will guide discrimination; tighten or loosen the required pause right before the reinforcer. Try inserting a brief DRL or DRH just before payoff to sharpen the difference.