The economics of daily consumption controlling food- and water-reinforced responding.
Daily limits change either response speed or time split, depending on whether the reinforcers can substitute for each other.
01Research in Context
What this study did
Pigeons worked on two keys at once. Each key gave food or water on its own timer.
The birds faced a daily limit. Once they earned all food or water for the day, that key shut off.
Researchers watched how limits changed what the birds did.
What they found
When food and water could replace each other, limits changed how birds split their time.
When food and water could not replace each other, limits changed how fast birds pecked.
The type of reinforcer pair mattered more than the limit itself.
How this fits with other research
McGonigle et al. (1982) ran the same setup and saw the same pattern. This gives us more trust in the finding.
Catania et al. (1974) used cocaine doses instead of food and water. Both studies show matching holds across very different commodities.
Hall (2005) later showed that earning rates can break the matching law. This builds on Killeen (1978) by naming the exact step where limits bite.
Davison et al. (1989) found no effect of overall feedback. This seems to clash with Killeen (1978), but the key difference is whether limits act on each key alone or on the whole session.
Why it matters
When you set daily limits on reinforcers, first ask if the client sees the items as swappable. If yes, expect shifts in time use. If no, expect changes in work speed. This tells you where to look first when behavior drifts under rationing.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →List each reinforcer you ration. Mark which ones the client could swap. Watch if limits change speed or choice next week.
02At a glance
03Original abstract
In the first experiment, two rhesus monkeys earned their entire ration of food and water during daily sessions with no provisions to ensure constant daily intakes. Two variable-interval schedules of food presentations were concurrent with one variable-interval schedule of water presentations; the maximum rate of food presentations arranged by one food schedule was varied. As the rate of food presentations was increased, the absolute level of responding on the two food schedules combined decreased, while responding on the water schedule increased. The preference for the variable food schedule compared to the other food schedule approximately matched the proportion of reinforcers obtained from it. The preference for the variable food schedule compared to the water schedule did not match, but greatly decreased, as the proportion of reinforcers from the food schedule increased. When Experiment I was replicated, with provisions to ensure constant daily intakes of food and water (Experiment II), the absolute response rates under the two food schedules combined and under the water schedule no longer changed with increases in the rate of food during the sessions. On the other hand, choice between the two food schedules remained proportional to the distribution of obtained food pellets. These results were interpreted as indicating that behavior to obtain nonsubstitutable commodities, such as food and water, is strongly controlled by the economic conditions of daily consumption, while choice between substitutable commodities is independent of these factors.
Journal of the experimental analysis of behavior, 1978 · doi:10.1901/jeab.1978.29-475