Interresponse-time analysis of stimulus control in multiple schedules.

Haring (1985) watched pigeons work on two different color keys. Each color meant a different pay rate. The team sorted every peck by how long the bird waited since the last peck.

They wanted to know if the pause length changed how well the color could steer the next peck.

Pauses longer than one second followed the color rule best. Short, rapid pecks mostly followed body cues, not the key color.

In plain words: slow, thoughtful responses listen to outside signals. Quick, burst responses run on autopilot.

Rilling et al. (1969) first showed that timing rules alone do not break stimulus control. G’s data add that timing rules actually strengthen control—if the pause is long enough.

Tanno et al. (2009) later proved that animals can treat the reinforced pause itself as a signal. Together, the three studies form a line: timing matters, long pauses boost outside control, and those pauses can even become their own cue.

Shimp (1967) trained birds to peck fast and found short IRTs easy to reinforce. G found short IRTs hard to steer with outside cues. The papers seem opposite, but they ask different questions. P asks “Can speed be rewarded?” G asks “Does speed listen to outside signals?” Both answers can be true.

When you use DRL or any timing-based program, check the pause length before you assume the learner “knows” the cue. If responses come in bursts, add a brief wait prompt or model to shift control from body cues to your external signal.