Response-reinforcer independence and the economic continuum: A preliminary analysis

Three pigeons were exposed to 1-hr and 4-hr sessions during which they earned food under a fixed-ratio 50 schedule of reinforcement while obtaining additional food according to either a variable-interval or a variable-time schedule. Postsession food was provided after the 1-hr sessions. The frequency of the variable-interval and variable-time food presentations was varied under the two session durations. The various combinations of within-session earned and unearned food, as well as the postsession food, defined conditions on the open-to-closed economy continuum. Key pecks tended to increase as the frequency of either variable-interval or variable-time food decreased. An economic-continuum analysis based on an independence quotient as a measure of response-reinforcer independence is presented to account for the effects.     

Pigeons' wait-time responses to transitions in interfood-interval duration: Another look at cyclic schedule performance

Recent developments reveal that animals can rapidly learn about intervals of time. We studied the nature of this fast-acting process in two experiments. In Experiment 1 pigeons were exposed to a modified fixed-time schedule, in which the time between food rewards (interfood interval) changed at an unpredictable point in each session, either decreasing from 15 to 5 s (step-down) or increasing from 15 to 45 s (step-up). The birds were able to track under both conditions by producing postreinforcement wait times proportional to the preceding interfood-interval duration. However, the time course of responding differed: Tracking was apparently more gradual in the step-up condition. Experiment 2 studied the effect of having both kinds of transitions within the same session by exposing pigeons to a repeating (cyclic) sequence of the interfood-interval values used in Experiment 1. Pigeons detected changes in the input sequence of interfood intervals, but only for a few sessions—discrimination worsened with further training. The dynamic effects we observed do not support a linear waiting process of time discrimination, but instead point to a timing mechanism based on the frequency and recency of prior interfood intervals and not the preceding interfood interval alone.     

Operant conditioning of behavioral variability using a percentile reinforcement schedule.

The present investigation developed and tested a new percentile reinforcement schedule suited to study pattern variability, whose main feature was the relative dissociation it provided between the variability requirement defining criterional responses and overall probability of reinforcement. In a discrete-trials procedure, pigeons produced patterns of four pecks on two response keys. If the pattern emitted on the current trial differed from the N preceding patterns, reinforcement was delivered with probability mu. The schedule continuously adjusted the criterion N such that the probability of a criterional response, estimated from the subject's recent behavior, was always constant. In these circumstances, the criterion corresponded to an invariant percentile in the distribution of recent responses. Using a between-subjects design, Experiment 1 manipulated the variability requirement--the percentile--while keeping overall reinforcement probability constant. The degree of variability varied directly with the requirement. In addition, an inverse relationship existed between the requirement and within-group variance. Experiment 2 manipulated probability of reinforcement while maintaining the variability requirement constant. No consistent relationship was found between variability and reinforcement probability. A tentative hypothesis was advanced ascribing the operant conditioning of behavioral variability to a process of probability-dependent selection.     

The roles of stimulus control and reinforcement frequency in modulating the behavioral effects of d-amphetamine in the rat.

The behavioral effects of d-amphetamine have been shown to be modulated by stimulus control, with less impairment of performance occurring when control is great. When the fixed-consecutive-number schedule is used (on which at least a specified consecutive number of responses must be made on one operandum before a single response on another will produce a reinforcer), response rate tends to be invariant but reinforcement frequency is not. This study asks whether the differences in reinforcement frequency that usually accompany changes in stimulus control could themselves be responsible for the performance differences. Two versions of the fixed-consecutive-number schedule of reinforcement were combined into a multiple schedule within which stimulus control was varied but differences in reinforcement frequency were minimized by omitting some reinforcer deliveries during the component that usually had the higher reinforcement frequency. In one component, a compound discriminative stimulus was added with the eighth consecutive response on the first lever; a single response on the second lever was then reinforced. In the other component, no such stimulus was presented. With no added stimulus, large decreases occurred in the number of runs satisfying the minimum requirement for reinforcement at doses of drug that produced only minimal changes when an added stimulus controlled behavior. Thus, increased stimulus control diminishes the behavioral changes produced by d-amphetamine even when the possible contribution by baseline reinforcement rate is minimized.     

Extent of food restriction affects probability but not delay-based decision-making

Rodent studies on decision-making often use food rewards and food-restrict subjects in order to motivate performance. However, food restriction has widespread effects on brain and behavior, which depend on factors including extent of restriction and feeding schedule. These factors are well recognized for their effects on motivation, but may also cause effects on decision-making independent of motivation. We examined how the degree of weight-based food restriction in rats influenced decision-making on the probability and delay discounting tasks. Additionally, we examined how the method of food restriction (consistent amount vs. time constrained feeding schedule) influenced decision-making. Our results showed that the degree of weight-based food restriction significantly altered probability, but not delay discounting, and that these effects were not entirely explainable by differences in motivation. Additionally, the method of food restriction did not significantly influence discounting when animals were within the same range of weight-based restriction. Together, our findings suggest that the degree of food restriction may modulate the neural circuitry responsible for selective aspects of decision-making related to probability. Further, these data support the need for tight control and reporting of weight and feeding in studies relying on food restriction, and suggest that the effects of food restriction may be broader than previously considered.     

Generating variable and random schedules of reinforcement using Microsoft Excel macros

Variable reinforcement schedules are used to arrange the availability of reinforcement following varying response ratios or intervals of time. Random reinforcement schedules are subtypes of variable reinforcement schedules that can be used to arrange the availability of reinforcement at a constant probability across number of responses or time. Generating schedule values for variable and random reinforcement schedules can be difficult. The present article describes the steps necessary to write macros in Microsoft Excel that will generate variable-ratio, variable-interval, variable-time, random-ratio, random-interval, and random-time reinforcement schedule values.