Body weight and response acquisition with delayed reinforcement.
New lever presses can be created with unsignaled 30-s delayed food, but leaner body weight and prior pairing sessions make success more likely.
01Research in Context
What this study did
Coe et al. (1997) tested whether rats learn a new lever press when food comes 30 seconds later. No click, no light, no magazine training. Just press now, eat later.
They varied body weight. Some rats worked at 70% free-feed weight, others at 80%. Half got magazine training first, half did not.
What they found
Rats did learn to press, even with the 30-s delay. But results were messy. At 70% weight, some rats pressed without any magazine training. At 80% weight, only rats with prior magazine training caught on.
Body weight and training history decided who succeeded.
How this fits with other research
Sutphin et al. (1998) used the same delay setup but added a second 'cancel' lever. Acquisition still happened at delays up to 32 s, showing the 1997 finding holds when the contingency is clearer.
Farrant et al. (1998) ran a parametric follow-up the next year. They showed that adding a 12-s delay flattens the usual rate differences between rich and lean schedules, building directly on the 1997 discovery.
Keely et al. (2007) kept the same unsignaled 30-s delay and asked how rats track the contingency. They found rats pressed more on the lever that started the delay chain, confirming that learning really is under contingency control, not just superstition.
Why it matters
For BCBAs, the takeaway is simple: delayed reinforcement can teach new skills, but satiation and learning history matter. If a child isn't responding, check motivation (think 'body weight') and look at whether earlier pairing sessions happened. You can run a teaching trial with a 20- or 30-second delay, but first ensure the reinforcer is strong and the contingency has been paired at least once.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Before your next session, run two pairing trials: deliver the preferred item right after the target response, then switch to a 20-s delay for the remaining trials.
02At a glance
03Original abstract
The relation between body weight and responding established with unsignaled delayed reinforcement was investigated. In three experiments, naive rats were deprived to either 70%, 80%, or 90% of ad libitum weight and were then exposed to tandem variable-interval 15-s differential-reinforcement-of-other-behavior 30-s schedules. The tandem schedule defined a resetting unsignaled delay-of-reinforcement procedure. In the first experiment, speed of magazine training, acquisition of lever pressing, and final rate of lever pressing were related to body weight. In the next experiment, lever pressing was established and maintained in rats that were magazine trained at 70% of ad libitum weight but that were then exposed to the delay procedure at 90% of ad libitum weight. Responding did not change consistently either across or within subjects in subsequent conditions in which body weight was manipulated. In the final experiment, lever pressing was established and maintained with delayed reinforcement in the absence of magazine training for each of 2 rats at 70% and for 1 of 2 rats at 90% of ad libitum weight. The results further illuminate the conditions under which responding can be established in the absence of training and when such responses are reinforced only following an unsignaled delay period.
Journal of the experimental analysis of behavior, 1997 · doi:10.1901/jeab.1997.67-131