On the distinction between open and closed economies.
Reinforcement strength hinges on the work-to-reward ratio, not on whether extra rewards are available outside the task.
01Research in Context
What this study did
Timberlake et al. (1987) worked with pigeons that pecked a key for food.
They tested two setups. In an open economy the birds could eat extra food after sessions. In a closed economy all food came only from pecking.
The team varied how many rewards each minute of pecking produced. They watched how fast the birds pecked at each reward density.
What they found
Pecking speed followed a bitonic curve under both setups. Birds pecked fastest at middle reward densities and slower at very low or very high densities.
The open versus closed label did not matter. Only the ratio of work to payoff controlled the birds’ response rate.
How this fits with other research
Aznar et al. (2005) extends the idea to people. Two adults with developmental disabilities worked on a computer task. When free snacks were removed (closed economy) their work rate rose, matching the pigeon data.
Schlundt et al. (1999) also used pigeons in an open economy. They saw a smooth hyperbolic climb in pecking as reward density grew. The 1987 study adds the peak-and-drop shape, showing the curve bends when density gets too rich.
Together the papers say the same rule works across species: take away free reinforcers and performance climbs until rewards get so dense that extra food loses value.
Why it matters
You can drop the open-versus-closed debate. Focus on the work-to-pay ratio instead. If a client has free access to a reinforcer, thin that access before sessions. Watch for the sweet spot where the learner works briskly but does not satiate. Then fade the density slowly to keep responding strong.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Remove or delay free access to the client’s top snack for 30 minutes before the session, then run a short progressive-ratio program and note the highest ratio completed.
02At a glance
03Original abstract
Open and closed economies have been assumed to produce opposite relations between responding and the programmed density of reward (the amount of reward divided by its cost). Experimental procedures that are treated as open economies typically dissociate responding and total reward by providing supplemental income outside the experimental session; procedures construed as closed economies do not. In an open economy responding is assumed to be directly related to reward density, whereas in a closed economy responding is assumed to be inversely related to reward density. In contrast to this predicted correlation between response-reward relations and type of economy, behavior regulation theory predicts both direct and inverse relations in both open and closed economies. Specifically, responding should be a bitonic function of reward density regardless of the type of economy and is dependent only on the ratio of the schedule terms rather than on their absolute size. These predictions were tested by four experiments in which pigeons' key pecking produced food on fixed-ratio and variable-interval schedules over a range of reward magnitudes and under several open- and closed-economy procedures. The results better supported the behavior regulation view by showing a general bitonic function between key pecking and food density in all conditions. In most cases, the absolute size of the schedule requirement and the magnitude of reward had no effect; equal ratios of these terms produced approximately equal responding.
Journal of the experimental analysis of behavior, 1987 · doi:10.1901/jeab.1987.48-35