Signal functions in delayed reinforcement.
A short signal during any delay keeps learners responding until reinforcement arrives.
01Research in Context
What this study did
The team worked with pigeons in a small lab chamber. Birds pecked a colored key for grain.
Sometimes grain came right away. Sometimes it came after a 5-second wait.
In half the tests a short blackout lit up during the wait. In the other half the wait stayed quiet. The scientists counted key pecks in each set-up.
What they found
Pigeons kept pecking hard when a blackout signaled the wait. They slowed down when no cue marked the delay.
The signal acted like a tiny promise: "Keep going, food is coming." Without that promise, the birds acted as if the link between peck and food had snapped.
How this fits with other research
Grosch et al. (1981) saw the same thing three years earlier. Their birds actually picked the signaled delay over the quiet one when given a choice.
Cullinan et al. (2001) later added a twist. They showed that smart, different signals protect preference for shorter waits, while vague signals weaken it.
Kuroda et al. (2014) stretched the idea into discrimination tasks. Full signals kept accuracy high; partial or no signals let it drop. Together the four studies draw a clear line: a clear cue during any wait keeps behavior strong.
Why it matters
You can borrow the blackout trick today. When reinforcement must be delayed, give a brief signal—a timer beep, a card flip, a short phrase. The learner hears or sees the cue and keeps working. One simple signal preserves response strength and cuts frustration.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Add a 2-second visual cue—like dimming the lights or turning on a tiny LED—during the wait before tokens or praise.
02At a glance
03Original abstract
Three experiments were conducted with pigeons to examine the role of the signal in delay-of-reinforcement procedures. In the first, a blackout accompanying a period of nonreinforcement increased key-peck response rates maintained by immediate reinforcement. The effects of dissociating the blackout from the delay interval were examined in the second experiment. In three conditions, blackouts and unsignaled delays were negatively correlated or occurred randomly with respect to one another. A signaled delay and an unsignaled delay that omitted the blackouts were studied in two other conditions. All delay-of-reinforcement conditions generally produced response rates lower than those produced by immediate reinforcement. Signaled delays maintained higher response rates than did any of the various unsignaled-delay conditions, with or without dissociated blackouts. The effects of these latter conditions did not differ systematically from one another. The final experiment showed that response rates varied as a function of the frequency with which a blackout accompanied delay intervals. By eliminating a number of methodological difficulties present in previous delay-of-reinforcement experiments, these results suggest the importance of the signal in maintaining responding during delay-of-reinforcement procedures and, conversely, the importance of the delay interval in decreasing responding.
Journal of the experimental analysis of behavior, 1984 · doi:10.1901/jeab.1984.42-239