Contrast and reallocation of extraneous reinforcers as a function of component duration and baseline rate of reinforcement.
Stretch component length or lower baseline payoff to shrink behavioral contrast and extra-reinforcer drift.
01Research in Context
What this study did
Sturmey (1995) worked with pigeons on two-key multiple VI-VI schedules. The team changed how long each schedule stayed on and how often food was given.
They watched what happened to key pecks and to extra food that could land on the second key. The goal was to see if time and rate changed behavioral contrast and reinforcer reallocation.
What they found
Longer components and lower base food rates made both contrast and reinforcer reallocation shrink. Shorter, richer components made the birds shift more pecks and more extra food toward the lean side.
In plain words, quick schedule switches pump up contrast; slow switches calm it down.
How this fits with other research
The finding lines up with Szatmari (1992), the earlier pigeon study that first showed extra food moves between components. The 1995 paper adds the rule: stretch the component or drop the base rate and the moving stops.
Pickering et al. (1985) looks like a contradiction. Those rat data said component duration did nothing to contrast. The gap is likely species: pigeons notice time changes; rats may not.
Reiss et al. (1982) also used pigeon multiple-VI schedules and saw shorter parts boost response rates, but only when paired with higher reinforcement. Sturmey (1995) ties that rate jump to the hidden flow of extra reinforcers, closing the loop between rate and reallocation.
Why it matters
If you run multiple schedules in practice, remember that short, rich components can spike contrast and side-behavior. When you want steady performance, lengthen the component or thin the baseline reinforcement first. This small timing tweak can save you from surprise bursts or drops in responding.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Lengthen each schedule component by 30 s and track if contrast spikes calm down.
02At a glance
03Original abstract
Four pigeons responded on multiple schedules arranged on a "main" key in a two-key experimental chamber. A constant schedule component was alternated with another component that was varied over conditions. On an extra response key, conjoint schedules of reinforcement that operated in both components were arranged concurrently with the multiple schedule on the main key. On the main key, changes in reinforcement rate in the varied component were inversely related to changes in response rates in the constant component (behavioral contrast). On the extra key, some reinforcers were reallocated between components, depending on the schedules in effect on the main key in the varied component. In the varied component, the obtained rates of reinforcement on the extra key were inversely related to main-key reinforcement rate. In the constant component, extra-key reinforcer rates were positively related to main-key reinforcer rates obtained in the varied component, and were not a function of response rates on the extra key. In two comparisons, the rate at which components alternated and the value of the main-key schedule in the constant component were varied. Consistent with earlier work, long components reduced the extent of contrast. Reductions in contrast as a function of component duration were accompanied by similar reductions in the extent of reinforcer reallocation on the extra key. In the second comparison, lowering the rate of reinforcement in the constant component increased the rate at which extra-key reinforcers were obtained, reduced the extent of reinforcer reallocation, and reduced contrast. Overall, the results are consistent with the suggestion that some contrast effects are due to the changes in extraneous reinforcement during the constant component, and that manipulations of component duration, and manipulations of the rate of reinforcement in the constant component, affect contrast because they influence the extent of extraneous reinforcer real-location.
Journal of the experimental analysis of behavior, 1995 · doi:10.1901/jeab.1995.63-203