EFFECTS OF REINFORCEMENT SCHEDULING ON SIMULTANEOUS DISCRIMINATION PERFORMANCE¹

Abstract

Pigeons were trained on a discrete‐trials, simultaneous discrimination procedure, with confusable stimuli such that asymptotic performance was about 85% correct. Trials were terminated if no response occurred within 2 sec of stimulus onset, so that probability of responding was free to vary. The schedule of reinforcement for correct responses was varied, with the following results: (1) there was no relation between frequency of reinforcement and accuracy of responding. (2) In extinction, the probability of responding fell to low levels, but accuracy remained roughly constant. (3) When reinforcement was available after a fixed number of trials or after a fixed number of correct responses, the probability of responding increased with successive trials after reinforcement, but accuracy was generally constant. (4) When every fifth correct response was reinforced, accuracy decreased immediately after reinforcement if the birds were required to respond on every trial.

Reinforced Variability in Animals and People: Implications for Adaptive Action.

“…Accuracy increases as well (e.g., when learning to discriminate between stimuli or match one stimulus to another; Nelson, 1978;Nevin, 1967). It is as if the closer the reinforcer, the more vigorous and accurate is performance.…”

STIMULUS CONTROL WITH FIXED‐RATIO REINFORCEMENT1

“…Although many data show that a reinforcement for each response to a particular stimulus establishes differential responding, there is less information about the effects of schedules that require several responses for each reinforcement. Fixed-ratio (FR) schedules of intermittent reinforcement appear to be effective: Nevin (1967) obtained differential responding to different brightnesses with a schedule involving five responses per reinforcement (FR 5), several experiments (Dews, 1963;Ferster, 1960;Mintz, Mourer, and Weinberg, 1966;Nevin, Cumming, and Berryman, 1963) obtained accurate matchingto-sample with schedules ranging as high as FR 95, and Rohles (1961) produced responses to the odd member of a set of stimuli with ratios up to FR 19.…”

“…Nevin's (1967) matching-to-sample or oddity procedures in which the correct stimulus at any time depended on the sample, thereby correlating a different stimulus with reinforcement at different times. The purpose of the present experiment was to study the effects of fixed-ratio schedules on a free-operant simultaneous discrimination among three different stimuli.…”

“…This finding was like the present one in that responses to the negative stimuli usually occurred in the first quarter of the ratio and then declined. Nevin's (1967) data, showing that errors occurred early in fixed ratios only when a response was required to advance the ratio, raises the question of whether the relation is between pauses and errors or between the temporal remoteness of reinforcement and errors. If the discrete trial data may be generalized to the free responding situation, Nevin's experiment suggests that the temporal relation may be the important one.…”

The pattern of responding after extensive extinction

“…Although the most obvious effect of the prolonged omission of reinforcement is a considerable decrease in response rate, it has been clear since Pavlov (1927), and from the major textbooks of the 1940s through the 1970s, that extinction does not eliminate the original learning (Hilgard & Marquis, 1940;Hull, 1943;Kimble, 1961;Mackintosh, 1974). After extinction, there is considerable evidence for the retention of information that was acquired during original acquisition (see, e.g., Nevin, 1967, and many others). Further evidence from transfer designs have supported and extended this conclusion (Delamater, 1996;Rescorla, 1996Rescorla, , 2001.…”