Pigeons respond to produce periods in which rewards are independent of responding.

Pigeons pecked a key to turn on a green light. While the light stayed on, grain dropped every few seconds no matter what they did.

Each peck started a timer. The timer set how long the green "on" period lasted. The birds could not make food come faster by pecking more.

Short on periods made the birds peck like crazy. Long on periods made them slow down.

Other settings, like how often the grain came, barely changed the peck rate. The length of the light period was the main driver.

STEBBINROSS et al. (1962) saw the same birds peck faster when food came more often. Neuringer (1973) shows the opposite: peck rate drops when the light stays on longer, even though food rate stays the same. The two studies test different things. C et al. changed how often food arrived. A kept food timing fixed and only changed how long the bird’s own peck kept the light on.

Kelleher (1966) proved a brief light can act like a mini-reward. Neuringer (1973) uses that idea: the green light itself seems to feed the pecking, so shorter flashes make the birds work harder to get the next flash.

Buskist et al. (1988) later found that a half-second signal keeps pigeons pecking through short delays. Neuringer (1973) is the mirror image: the bird’s own response sets the signal length, and shorter signals boost rate.

If you use token boards, timers, or any brief cue, remember that the cue length can drive response rate more than the actual reward rate. Try shaving the cue to one or two seconds and watch responding climb. This gives you a free lever: you can keep the same reinforcement plan but get more work by tightening the access window to the signal.