The Effects of Component Duration on Multiple‐schedule Performance in Closed and Open Economies

Abstract

Pigeons responded on multiple variable-interval variable-interval schedules of reinforcement in an open and a closed economy. Equal duration components were increased in duration while the component rates of reinforcement were held constant, the component schedules were reversed, and component duration was decreased. In the open economy, daily sessions were limited to 1 hr, and subjects were maintained at 80% of their free-feeding weights through supplemental feeding when necessary in their home cages. In the closed economy, subjects were housed in their experimental chambers and no deprivation regimen was enforced. Relative response rate decreased as components were lengthened in the open economy, whereas in the closed economy relative rate increased as components were lengthened. Response proportions overmatched reinforcer proportions to a greater extent at long component durations in the closed economy, but there was no systematic effect of component duration on responding in the open economy.Key words: multiple schedules, component duration, closed economy, variable-interval schedules, overmatching, key peck, pigeonsIn multiple schedules, two or more independent schedules of reinforcement, each correlated with a different stimulus, are alternated at some specified interval. The alternation of schedules, or components, typically occurs independent of behavior. The distribution of responses between the successive components of a multiple variable-interval schedule is well described by a power function that relates the number of responses emitted per component and the number of reinforcers obtained per component. This function, called the generalized matching law (Baum, 1974b), takes the form:in which P represents the rate of responding emitted within a component, R represents the obtained rate of reinforcement within a component, and the subscripts denote the components of a multiple schedule. The fitted parameter a represents a constant preference for one of the alternatives that cannot be accounted for by the obtained distribution of reinforcers (bias), and the fitted exponent b represents a metric by which changes in behavior produced by a given change in reinforcement can be measured, and is said to capture the degree to which matching takes place on multiple schedules. Perfect matching is the case in which b is equal to 1.0, whereas values of b less than or greater than 1.0 represent instances in which undermatching or overmatching has been obtained. The absence of bias is indicated by values of a equal to 1.0. McSweeney, Farmer, Dougan, and Whipple (1986)

Responding on Concurrent‐chains Schedules in Open and Closed Economies

“…These variables covaried with changes in economic context and constitute a potential confound in any study that uses as an independent variable the type of nominal economy differences presented here. Similar problems associated with a withinsubjects comparison of the effects of closed versus open economies have been discussed with respect to successive schedule performance (LaFiette & Fantino, 1988). Because the terminal links of concurrent-chains schedules may be viewed as successively presented schedules (see Wardlaw & Davison, 1974), and the effects of these variables (those related to open and closed economies) on responding in successive schedules are well documented (Elliffe & Davison, 1985;Herrnstein & Loveland, 1974), there is the possibility that these factors might have interacted to produce the present results.…”

Closed‐economy Multiple‐schedule Performance: Effects of Deprivation and Session Duration

“…Such a situation is called a closed economy (Hursh, 1980(Hursh, , 1984, in contrast with the more commonly arranged open economy, in which subjects are maintained at a constant level of food deprivation by supplementary food given outside the experimental session. Elliffe LaFiette and Fantino (1988) have also reported overmatching in closed-economy multiple schedules, at least at component durations similar to those used by Elliffe and Davison. One variable known to affect sensitivity in multiple schedules is the level of deprivation of the scheduled reinforcer. Both Herrnstein and Loveland (1974) and Charman and Davison (1983), using pigeons responding for food reinforcers on open-economy multiple VI schedules, found that sensitivity approached strict matching (a = 1.0) as subjects were made less food deprived.…”

“…Elliffe and Davison (1985) reported consistent overmatching (a > 1.0) in an experiment in which pigeons received their total intake of food as reinforcers for responding on multiple schedules available 24 hr per day. Such a situation is called a closed economy (Hursh, 1980(Hursh, , 1984 Fantino (1988) have also reported overmatching in closed-economy multiple schedules, at least at component durations similar to those used by Elliffe and Davison. One variable known to affect sensitivity in multiple schedules is the level of deprivation of the scheduled reinforcer. Both Herrnstein and Loveland (1974) and Charman and Davison (1983), using pigeons responding for food reinforcers on open-economy multiple VI schedules, found that sensitivity approached strict matching (a = 1.0) as subjects were made less food deprived.…”

“…Both Herrnstein and Loveland (1974) and Charman and Davison (1983), using pigeons responding for food reinforcers on open-economy multiple VI schedules, found that sensitivity approached strict matching (a = 1.0) as subjects were made less food deprived. Both Elliffe and Davison's (1985) and LaFiette and Fantino's (1988) subjects were able to earn …”

“…Both Herrnstein and Loveland (1974) and Charman and Davison (1983), using pigeons responding for food reinforcers on open-economy multiple VI schedules, found that sensitivity approached strict matching (a = 1.0) as subjects were made less food deprived. Both Elliffe and Davison's (1985) and LaFiette and Fantino's (1988) subjects were able to earn Table 1 Sequence of conditions in all parts of the experiment, and deprivation level for each bird, expressed as a percentage of its free-feeding body weight. Conditions 49 to 55 (Part 4) are each listed more than once, because each bird experienced a different pair of VI schedules and arranged overall reinforcer rate.…”