Synthetic variable-interval schedules of reinforcement.

Pigeons pecked a key for food. The feeder only worked if the bird waited a set amount of time between pecks.

Ten different wait-time windows were tested. Each window was a tiny slice of the variable-interval schedule.

The goal was to see if reinforcing only certain wait times would shape when the birds pecked.

Reinforcement pulled the birds’ interresponse times toward the target window. Longer waits showed the clearest change.

A straight-line and a reciprocal curve described the data. Timing, not just rate, came under schedule control.

Misak et al. (2011) later saw the same IRT effect under concurrent VI-VI. Longer waits boosted preference for the richer schedule. That study conceptually replicates the 1973 finding in a choice setup.

Frame et al. (1984) extended the idea by adding colored lights that told the bird how long the current interval would last. Pauses lengthened when the cue signaled a long wait, showing external stimuli can sharpen the same temporal control.

McSweeney et al. (1993) pushed further, proving the post-reinforcement pause tracks the prior interfood interval. Together these papers form a line: from pure IRT class, to signaled duration, to memory-based pause rules.

You now know that schedules can sculpt the micro-timing of each response, not just the overall rate. When you shape waiting in a client—before a mand, between bites, or during transitions—think in IRT windows. Start with a broad window, then narrow it. Pair a brief signal if extra control is needed. The same principle that stretched pigeon pauses can stretch a child’s wait time without extra tokens.