Connectionist models of conditioning: A tutorial

Models containing networks of neuron-like units have become increasingly prominent in the study of both cognitive psychology and artificial intelligence. This article describes the basic features of connectionist models and provides an illustrative application to compound-stimulus effects in respondent conditioning. Connectionist models designed specifically for operant conditioning are not yet widely available, but some current learning algorithms for machine learning indicate that such models are feasible. Conversely, designers for machine learning appear to have recognized the value of behavioral principles in producing adaptive behavior in their creations.     

Control of responding during stimuli that precede transitions in reinforcement frequency

Pigeons' responses were reinforced on a variant of a mixed variable-interval extinction schedule of reinforcement in which the transition to the higher reinforcement rate was signaled by a trace stimulus projected on the response key prior to the onset of the component correlated with food delivery. In the first of two experiments, the duration of the trace stimulus preceding the component correlated with food delivery was varied from 1.5 to 50.0 s and in the second experiment, the reinforcement frequency in the same component was varied from 10 to 60 reinforcers per hour. Pigeons pecked at the trace stimulus preceding the onset of the component correlated with food delivery even though responding was not reinforced in its presence and only one of the changes in reinforcement rate (i.e., from extinction to reinforcement) was signaled. The rate of pecking during the trace stimulus was a function of its duration but not of the reinforcement frequency in the following component. Higher rates generally occurred at the shorter trace-stimulus durations. Component responding following the offset of the trace stimulus was under discriminative control of the trace stimulus whether or not responding occurred in the presence of the trace stimulus.     

Testing a stochastic foraging model in an operant simulation: Agreement with qualitative but not quantitative predictions

An operant simulation of foraging through baited and empty patches was studied with 4 pigeons. On a three-key panel, side keys were designated as patches, and successive opportunities to complete 16 fixed-ratio 10 schedules on side keys were defined as encounters with feeders. In a random half of the patches in any session, some of the fixed-ratio 10 schedules yielded reinforcement (baited feeders) and the other schedules yielded nonreinforcement (empty feeders). In the other half of the patches, all feeders were empty. Pigeons could travel between patches at any time by completing a fixed-ratio schedule on the center key. An optimal foraging model was tested in Experiments 1 and 2 by varying center-key travel time and number of baited feeders in baited patches. The ordinal predictions that number of feeders visited in empty patches would increase with travel time and decrease as number of baited feeders increased were supported, but pigeons visited far more feeders in empty patches than the optimal number predicted by the model to maximize energy/time. In Experiment 3, evidence was found to suggest that the number of empty feeders encountered before the first baited feeder in baited patches is an important factor controlling leaving empty patches.     

Effects of differences between stimuli, responses, and reinforcer rates on conditional discrimination performance

In a discrete-trial conditional discrimination procedure, 4 pigeons obtained food reinforcers by pecking a key with a short latency on trials signaled by one stimulus and by pecking the same key with a long latency on trials signaled by a second stimulus. The physical difference between the two stimuli and the temporal separation between the latency values required for reinforcement were varied factorially over four sets of conditions, and the ratio of reinforcer rates for short and long latencies was varied within each set of conditions. Stimulus discrimination varied directly with both stimulus and response differences and was unaffected by the reinforcer ratio. Sensitivity to reinforcement, estimated by generalized-matching-law fits to the data within each set of conditions, varied directly with the response difference but inversely with the stimulus difference arranged between sets of conditions. Because variations in stimulus differences, response differences, and reinforcer differences did not have equivalent effects, these findings question the functional equivalence of the three terms of the discriminated operant: antecedent stimuli, behavior, and consequences.     

Comparing Locomotion With Lever-press Travel In An Operant Simulation Of Foraging

An operant model of foraging was studied. Rats searched for food by pressing on the left lever, the patch, which provided one, two, or eight reinforcers before extinction (i.e., zero reinforcers). Obtaining each reinforcer lowered the probability of receiving another reinforcer, simulating patch depletion. Rats traveled to another patch by pressing the right lever, which restored reinforcer availability to the left lever. Travel requirement changed by varying the probability of reset for presses on the right lever; in one condition, additional locomotion was required. That is, rats ran 260 cm from the left to the right lever, made one response on the right lever, and ran back to a fresh patch on the left lever. Another condition added three hurdles to the 260-cm path. The lever-pressing and simple locomotion conditions generated equivalent travel times. Adding the hurdles produced longer times in patches than did the lever-pressing and simple locomotion requirements. The results contradict some models of optimal foraging but are in keeping with McNair's (1982) optimal giving-up time model and add to the growing body of evidence that different environments may produce different foraging strategies.     

Social reinforcement of operant behavior in rats: a methodological note.

An apparatus was developed to study social reinforcement in the rat. Four Long-Evans female rats were trained to press a lever via shaping, with the reinforcer being access to a castrated male rat. Responding under a fixed-ratio schedule and in extinction was also observed. Social access was found to be an effective reinforcer. When social reinforcement was compared with food reinforcement under similar conditions of deprivation and reinforcer duration, no significant differences were observed.     

A revised list of Spanish translations of operant terminology

The experimental analysis of behavior has grown enormously in Latin America within the last decade. This paper offers an updating revision of suggested Spanish translations of operant expressions.     

Stimulus control of respondent and operant key pecking: A single key procedure

Pigeons' responses to a uniformly illuminated response key were either reinforced on a variable-interval one-minute schedule of reinforcement or extinguished for one-minute periods. When 1.5 second signals were presented at the beginning of each component, so as to differentially predict reinforcement, the pigeons pecked at the signals, at rates higher than rates during the remainder of the component. When the brief signals were not differentially predictive of reinforcement, pecking in their presence decreased to near zero levels. Similar results were obtained with signals based upon colors and upon line orientations. Changes in rates of (unreinforced) pecking occurred during the signal whether pigeons responded differentially during the remainder of the component or not. Experiment II demonstrated that the presence of the signal correlated with extinction was not necessary for pecking to develop at the signal which preceded the component in which responding was intermittently reinforced. The experiments demonstrated a clear dissociation of respondent control from operant control of a response. In addition, operant behavior was shown to be relatively insensitive to differing rates of reinforcement, as compared to the sensitivity of respondent behavior to differing rates of reinforcement produced by the very same operant behavior.