Noncontingent Condition Research Articles

Instrumental conditioning of the GSR through the positive and the negative reinforcer (author's transl)

2 experiments were designed to examine the effects of positive and negative reinforcers on the frequency of spontaneous GSRs. The first experiment in contingent condition consisted of 3 groups: C-PN (positive reinforcer-negative reinforcer), C-NP (negative-positive), NR (no reinforcer). The second experiment in non-contingent condition consisted of 2 groups: NC-PN (positive-negative), NC-NP (negative-positive). Each S in NC condition was matched with C condition based on operant level. The results were as follows. During the first reinforcement period, C-PN showed an increase as did NC-NP and C-NP showed a decrease as did NC-PN. During the second reinforcement period, C-PN showed a decrease as did NC-NP and C-NP showed an increase as did NC-PN. NR showed a continuous decrease.

Contingency of consequences and type of incentive in interdependent escape

Two experiments were conducted on the effects of contingency and type of incentive on collective incoordination in an interdependent escape situation. The purpose was to examine two alternative explanations for the incoordination phenomena: (1) a rational evaluation hypothesis, that incoordination stems from a “rational” perception of the interdependence which under certain circumstances could dictate highly individualistic behavior; (2) an emotional arousal hypothesis, that incoordination reflects response interference from arousal created by anticipation of the task outcome. The outcomes of the escape situation were presented to groups of five subjects as being either (a) contingent on each subject's escape, or (b) noncontingent—arbitrarily assigned to subjects regardless of their performance. In one experiment, a comparison was made within each contingency condition between shock and reward as task incentives. A “no outcome” condition served as a control treatment. It was reasoned that the rational evaluation hypothesis implies high incoordination only under contingent conditions (regardless of kind of incentive) whereas the emotional arousal hypothesis implies high inco-ordination under both contingent and noncontingent conditions (though with different patterns depending on the incentives involved). The results indicated that the noncontingent outcome conditon produced an amount of incoordination intermediate between the contingent and control conditions, and this was the case in both incentive conditions. Thus, each hypothesis received some support.

Probability Performance: Reinforcement Procedure and Number of Alternatives

With the number of response alternatives (two or four) and the type of informative feedback (contingent or noncontingent) manipulated in a probabilistic task, there were fewer choices of the high-probability alternative in the four-alternative contingent condition than in the two-alternative one, but the reverse was true in the two noncontingent conditions. The amount of information transmitted on each feedback trial may explain the results.

A Study of Operant Conditioning under Delayed Reinforcement in Early Infancy

MILLAR, W. STUART. A Study of Operant Conditioning under Delayed Reinforcement in Early Infancy. Monographs of the Society for Research in Child Development, 1972, 37 (2, Serial No. 147). The paucity of definitive data on operant conditioning under delayed reinforcement in the preverbal infant stimulated the present research. A series of four interrelated experiments was undertaken with 4-7-month-old infants in order to examine the temporal parameters involved in response-feedback contingency integration. An initial experiment demonstrated operant acquisition of an instrumental "hand-pulling" response in 4-8-month-old infants. Moreov;er, the data suggested an age difference in sensitivity to responsecontingent feedback. Experiments II, III, and IV were designed to examine the temporal span of integration in 6-7-month-old infants using delayed perceptual reinforcement within the free-operant framework. Experiment II demonstrated clear-cut response acquisition under immediately contingent feedback, limited acquisition under 1-second and 2-second delayed-contingent feedback but no acquisition under 3-second and noncontingent feedback. In Experiment III a more exhaustive design incorporated more stringent control procedures. Contingent and noncontingent conditions produced characteristic divergent effects. The effect of 1-second and 2-second delays was less clear-cut than in Experiment II. Experiment IV was designed primarily to clarify the discrepancy between Experiments II and III by providing extended exposure to the delayed-feedback scheduling. Under this condition, reliable acquisition occurred under 1-second and 2-second delayed feedback but not under the 3-second delay condition. Prior exposure to noncontingent feedback had a facilitating effect on subsequent response acquisition under 1-second and 2second delayed feedback but not under 3-second delayed feedback. The consistent effect achieved with contingent, noncontingent, delayed reinforcement, and, in particular, the suppression, recovery, and facilitation effects of noncontingent stimulation are discussed in relation to our understanding of how infants process contingent information. This content downloaded from 157.55.39.35 on Fri, 02 Sep 2016 05:30:35 UTC All use subject to http://about.jstor.org/terms

The role of reinforcement procedure in children's probability learning as a function of age and number of response alternatives

This study consisted of a factorial manipulation of type of reinforcement procedure (contingent-noncontingent), number of response alternatives (2–4) and age (4.5–5.5 years), using a probabilistic task. Subjects in the noncontingent condition were always informed of the correct response, even on trials on which S made an incorrect response, whereas Ss in the contingent condition were informed which response was correct only on trials on which S made the correct response. An interaction ( p < .01) between number of response alternatives and type of reinforcement procedure was found, such that the level of choice of the higher probability alternative in the four-alternative contingent group was significantly below the two-alternative, contingent, two-alternative noncontingent, and four-alternative noncontingent groups, which did not differ among themselves. The findings are related to those of earlier adult studies.

THE ROLE OF EXPERIMENTER ATTITUDE AND CONTINGENT REINFORCEMENT IN A VIGILANCE TASK.

Experimenter attitude toward one group of monitors was democratic (expansive and permissive) and experimenter attitude toward another group was autocratic (brief and brusque). Both groups monitored under three conditions of environmental stimulation (1) control (no stimulation) (2) radio contingent on detection performance, and (3) radio non-contingent on detection performance. The group treated in a democratic manner showed significantly higher levels of detection performance on all three conditions of environmental stimulation. The performance under radio contingent and radio noncontingent conditions was significantly higher than the control condition, and the radio condition contingent showed no performance decrement.