Effects of unsignaled delay of reinforcement on preference and resistance to change.
A silent 3-second wait kills reinforcement power, so deliver fast or clearly mark the delay.
01Research in Context
What this study did
Four pigeons pecked keys for food. The team added a 3-second delay between the peck and the grain. No light or sound told the birds the delay was coming.
They measured how often the birds pecked and how hard they worked when the delay showed up.
What they found
Three out of four birds almost stopped pecking. Their response rate dropped and they gave up faster when the session got tough.
The delay also changed how they pecked: lighter, slower hits.
How this fits with other research
Koegel et al. (1992) seems to disagree. They gave 3-second delayed praise to babies and crying went down while happy sounds went up. The twist: babies value parent voices different than pigeons value grain. Same 3-second wait, opposite result.
Cullinan et al. (2001) stretches the idea further. Kids with ADHD learned to wait a whole day for bigger rewards. They used fading and kept reward size high, showing delay damage can be trained away.
Laugeson et al. (2014) ranks delay last. When building response chains, good quality and big amount beat fast delivery. Delay hurts, but not as much as cheap or tiny rewards.
Why it matters
If you hand out tokens, snacks, or praise more than a second after the target behavior, you may be wasting it. With humans, signal the wait ("Great job, here's your sticker after I finish counting") or keep the delay under one second. When you must wait, pair it with a big or high-quality reinforcer and teach the learner to watch for it.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Count your current delay; if it wanders past 1 second, add a signal word or reduce it.
02At a glance
03Original abstract
In Phase 1, pigeons were trained on a concurrent chain in which a 3‐s unsignaled delay of reinforcement was imposed on responding in a terminal link in some conditions. Preference for that terminal link was always reduced in comparison with conditions in which there was no delay, substantially so for 3 of the 4 pigeons. In Phase 2, pigeons responded in a two‐component multiple schedule. The scheduled rates of reinforcement were equal, but a 3‐s unsignaled delay was imposed in one component. Resistance of responding to prefeeding and extinction was reduced in the delay component for the same 3 subjects for which the data had shown strong effects of delay on preference. Systematic observation revealed differences in response topography. In the delay component, subjects oriented more closely to the key and responses were less forceful compared with the no‐delay component. Our results give further evidence that preference and resistance to change covary within subjects. However, they challenge the premise that the critical determiners of preference (i.e., terminal‐link value) and resistance to change (behavioral mass) may be quantified purely in terms of stimulus—reinforcer relations.
Journal of the experimental analysis of behavior, 1998 · doi:10.1901/jeab.1998.69-247