Fixed-interval and fixed-ratio reinforcement schedules with human subjects.

Stoddard et al. (1988) asked adults to press buttons under two rules. Press fast on one button. Press slowly on the other. The slow button paid only if the press came inside a short window near the end of a fixed interval.

The team first let people count seconds out loud. Then they added a tiny window—just a few seconds—where a press would count. They wanted to see if the window would stop people from counting and start pressing more often.

When the short window appeared, counting stopped. People no longer waited until the last second. They pressed steadily through the whole interval.

The schedule, not the clock in their head, now controlled the presses. The brief limited-hold broke the self-timing trick and put behavior back under the timer.

Rapport et al. (1982) and Sanders (1969) ran similar FR-FI games but gave people all the time in the world. They saw the same fast-slow split, yet people still used silent counting. Stoddard et al. (1988) kept the same schedules and just clipped the response window—wiping out the count strategy.

Harzem et al. (1978) tried the opposite move: they paid people for long pauses. That also wrecked the pause, showing whichever contingency is tightest wins. Stoddard et al. (1988) confirms the rule from the other side—tightening the window beats self-timing.

Okouchi (2003) later showed that past schedules can color today’s FI rate. Stoddard et al. (1988) adds that a tiny tweak in the current schedule can override any history.

If your learner is ‘waiting out’ an interval prompt, don’t just talk about it—tighten the window. Add a brief limited-hold to the DRO, DRL, or FI. The learner must respond a little earlier each cycle. The clock-watching melts away and steady responding returns. One small timer change can save many session minutes.