Effects of nonrandom intermittent reinforcement schedules in human-eyelid conditioning.

Abstract

Random, single alternation, double alternation, and 2 nonrandom (1 biased toward runs and the other biased toward alternation) reinforcement schedules were employed in human classical eyelid conditioning. The major findings were that with the same alternation pattern some Ss responded according to the pattern and others did not and that the random and nonrandom probabilistic schedules yielded data generally consistent with a linear operator model, but that parameters depended upon experience with the specific schedule. It was concluded that the failure of parameters to remain invariant over early training necessitates more extensive training to achieve stability and that more attention will have to be paid to individual-S response characteristics, since the same reinforcement schedules did result in lawfully different behavior in different Ss.A characteristic common to all "strength" models of learning (e.g., Bush & Hosteller, 1951;Hull, 1943;Spence, 1956) is that the tendency to make a response increases following a reinforcement and decreases following a nonreinforcement. Mean response strength resulting from an intermittent reinforcement schedule then represents the averaged effects of increments and decrements in response strengths, with terminal performance reflecting the reinforcement ratio, the incremental effect of a reinforcement (6), the decremental effect of a nonreinforcement (0'), and the limits of performance, A and A', resulting from, respectively, an

Conditional probability of reinforcement and sequential behavior in human conditioning with intermittent reinforcement schedules

“…The sequential behavior observed in earlier studies (Prokasy et al, 1967;Higgins et al, 1968) is consistent with the hypothesis that response probability increases with successive reinforcements and decreases with successive nonreinforcements. These outcomes are precisely wh at would be expected from an incremental-decremental model (e.g., Bush & Mosteller, 1951;Spence, 1936).…”

“…In all eases, P(R I R) = P(R IR). METHOD The method of transducing and reeording eyelid responses is reported elsewhere (Prokasy et al, 1967). The es was a 600-msee, lOOO-Hz tone, the interstimulus interval was 600 msec, and the ues was a 50-msee puff of nitrogen of suffieient intensity to support a 150-mm column of mereury.…”

Sequential effects in differential human eyelid conditioning

“…The basic apparatus for transducing and recording has been reported by Prokasy, Carlton, & Higgins (1967). CS+ and CS-were changes in brightness of a .75 in.…”

Information processing and the decremental effect of intermittent reinforcement schedules in human conditioning

“…Two facts are wholly consistent with the incrementaldecremental account: Usually, an intermittent reinforce· ment schedule results in a lower average level of performance than that observed with a 100% reinforcement schedule (see, e.g., Gormezano & Moore, 1969, p. 166) and, in the human literature, there is clear evidence of an average decremental effect following nonreinforced trials & Higgins, 1967;Prokasy, Higgins, & Carlton, 1968). It is to be noted, however, that not in all instances (Gormezano & Moore, 1969, p. 166) is there an overall reduced level of performance.…”

“…The first is that performance is affected by the nature of the reinforcement schedule beyond the trial-specific consequences of US presence vs. absence. For example, Prokasy, Carlton, and Higgins (1967) have shown that a subset of human subjects exposed to a doublealternation schedule tend to respond on US trials and not respond on US trials, thus reflecting a discrimination of the double-alternation sequence. In addition, if an intermittent reinforcement schedule is biased toward runs of reinforced trials or runs of nonreinforced trials, response likelihood, respectively, increases and decreases more rapidly than if the run bias were not present Prokasy, Carlton, & Higgins, 1967;Prokasy & Kumpfer, 1969).…”