Frequency versus magnitude of reinforcement: New data with a different procedure.
Reinforcement frequency influences behavior more strongly than reinforcement magnitude — schedule design should prioritize rate over size.
01Research in Context
What this study did
The team worked with pigeons pecking a key for food. They asked: does it matter more how often food comes, or how big each bit is?
They used a new procedure to pull frequency and size apart. Then they watched which change moved the birds' pecking most.
What they found
Pecking shifted fast when food came more often. When the food pellet only got bigger, the birds barely budged.
Frequency won. Size hardly mattered. The old rule held even with the new setup.
How this fits with other research
Wilkie et al. (1981) and Bradshaw et al. (1978) saw the same in rats years earlier. Smaller drinks just meant you had to serve more sips to keep the work going. Todorov et al. (1984) now show the rule works in pigeons too.
Thomas (1974) looks like a clash. People did not follow the matching law that pigeons obey here. The gap is species, not science. Birds chase rates; humans chase other things.
Nevin (1967) and Davis et al. (1972) add lab proof that rate, not size, steers key pecks. The new data tighten the same thread.
Why it matters
For your client, schedule beats prize size. Give 30 tiny praises in ten minutes instead of three big cookies. You will see more correct responses without buying fancier treats. Check your data sheet for how often you deliver, not how much.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Count your current praise rate; double it for one session and graph the change.
02At a glance
03Original abstract
Two pigeons, with previous exposure to concurrent schedules, were submitted to 29 sessions of 8 hours each with concurrent variable-interval variable-interval schedules in which reinforcement parameters changed from session to session. In the first nine sessions reinforcement durations were equal in both schedules while reinforcement frequencies varied; in Sessions 10 through 18, both frequency and duration of reinforcement were varied; in Sessions 19 through 29, only reinforcement duration was varied. Results with this different procedure confirm previous findings that behavior is more sensitive to changes in reinforcement frequency than to reinforcement magnitude.
Journal of the experimental analysis of behavior, 1984 · doi:10.1901/jeab.1984.41-157