Effects of Ratio Reinforcement Schedules on Discrimination Performance by Japanese Monkeys

Abstract

In Experiment 1, Japanese monkeys were trained on three conditional positiondiscrimination problems with colors as the conditional cues. Within each session, each problem was presented for two blocks of ten reinforcements; correct responses were reinforced under continuous-reinforcement, fixed-ratio 5, and variable-ratio 5 schedules, each assigned to one of the three problems. The this effect, previous results have not been consistent. For example, Ferster (1960) found that pigeons' matching-to-sample accuracy (with no intertrial interval) was near chance under a continuous-reinforcement (CRF) schedule but became higher as the fixed-ratio size increased to 20. Nevin et al. (1963) obtained the opposite results. In their experiment, matching-to-sample accuracy (with a 1-s intertrial interval) was highest with CRF. In order to resolve this inconsistency, Thomas (1979) tested pigeons' matching performance on FR schedules with and without an intertrial interval. Without an intertrial interval, he obtained results similar to Ferster's (1960). With intertrial intervals (larger than 5 s), however, accuracy was fairly constant for ratios from 1 (CRF) to 20, and these results were more like but not exactly consistent with those of Nevin et al. (1963). Similarly, accuracy was almost constant for ratios ranging from 3 to 10 in Davidson and Osborne's (1974)

Response sequence learning as a function of primary versus conditioned reinforcement

“…When every third correct response led to food reinforcement (SLight, 3-response sequence), accuracy was higher than when every seventh correct response (SLight, 7-response sequence) or every response (SFood condition) led to primary reinforcement (Figure 1). This result is similar to matching-to-sample tasks where intermediate reinforcement ratios also produced the highest response accuracy (e.g., Ferster, 1960;Fujita, 1985;Thomas, 1979).…”

“…However, variables such as fixed-ratio length, intertrial interval (IT! ), and level of training appear to interact with the conditioned versus primary reinforcement dimension in determining response accuracy on complex discrimination tasks (e.g., Fujita, 1985;Nevin, Cumming, & Berryman, 1963;Thomas, 1979). The present research varied both number of responses per sequence and minimum required interresponse interval in asking whether primary reinforcers are generally more effective in training sequences of responses or whether, as in matching-to-sample and conditional discrimination learning, there are parameters under which conditioned reinforcing feedback yields higher levels of accuracy.…”

Do bantams (Gallus gallus domesticus) amodally complete partly occluded lines? An analysis of line classification performance.

“…criminations (Ferster, 1958;Fujita, 1985) and is also useful for increasing the number of test trials inserted among baseline trials.…”

Perception of the standard and the reversed Müller-Lyer figures in pigeons (Columba livia) and humans (Homo sapiens).

“…Correct responses not followed by food were secondarily reinforced by a brief flash of the hopper light. Such intermittent reinforcement has been suggested as a way to raise the accuracy of conditional discriminations (Ferster, 1958;Fujita, 1985) and is also useful to increase the number of test trials inserted among baseline trials.…”