Behavioral Contrast Effects Research Articles

The role of response-independent reinforcement in producing behavioral contrast effects in the rat

This experiment examined the role of response-independent reinforcement in producing behavioral contrast effects on a multiple schedule when rats were used as subjects. Retraction of the lever was correlated with one component (S2) of the multiple schedule so that there was no opportunity to make the lever-press response in S2. Three different schedules of response-independent reinforcement (VI 30 sec, VI 2 min, and Extinction) were presented during S2 in successive blocks of sessions. The schedule of reinforcement in S1 (VI 2 min) remained unchanged. Results indicated that response rate in S1 was inversely related to reinforcement frequency in S2. Reliable transient contrast effects were also reported.

RELATIVE FREQUENCY OF REINFORCEMENT AND RATE OF PUNISHED BEHAVIOR1

After training on a multiple variable-interval variable-interval schedule of reinforcement, each response in one component of the schedule was followed by a brief electric shock. When the rate of punished responses stabilized, the frequency of reinforcement in the other component was first decreased and then increased from the baseline frequency. The effects of these manipulations were consistent with reports of interactions in multiple schedules involving only unpunished behavior, i.e., the rate of punished responses increased with a higher relative frequency of reinforcement in the punishment component and decreased with a lower relative frequency of reinforcement in that component. The relevance of such findings to a further generality of behavioral contrast effects is discussed.

Effects of Percentage of Reinforcement upon Behavioral Contrast in the Rat

A large amount of data has been reported in recent years concerning the interaction of a given condition of reinforcement and previous or contemporary conditions of reinforcement. This interaction has been commonly referred to as “behavioral contrast”. In this paper, it was tested whether the contrast effect was obtained in the discrete trial situation in terms of running behavior and behavior variability.Of two experiments reported here, Exp. I was concerned with simultaneous contrast effect, while Exp. II with successive contrast effect. The apparatus used was one pair of T maze. Except a start box and goal boxes, one maze was painted white (S1) and the other black (S2). A start box and goal boxes were painted gray in both mazes.In Exp. I, Ss were randomly assigned to five groups of 10, each. The groups, defined in terms of the percentage of reinforcement administered in their positive and negative mazes, respectively, were; Group I (100%-100%), Group II (100%- 50%), Group III (100%-0%), Group IV (50%-50%), Group V (50%-0%). After exploration and pretraining, all Ss were given 21 successive free-choice trials per day in two T mazes for 10 days.In Exp. II, Ss were divided into four groups of 10, each. The groups were; Group I (100% control), Group II (100% downshift), Group III (30% control), Group IV (30% upshift). During first 5 days of training, Groups II and IV received identical treatment with Groups I, III, respectively. Subsequently in Group II, percentage of reinforcement of S2 were shifted from 100% to 30% during Days 6 to 15. On the contrary, Group IV received upshift of S2 from 30% to 100%.Three response measures were employed in both experiments, i. e., starting latency, running time and choice direction.The results were as follows : 1) In Exp. I, the result of starting latency indicated a significant evidence for the positive contrast effect. That is, the starting latency in the 100% maze was shortest for Group III, intermediate for Group II and longest for Group I (Fig. 1). Though there was statistically no difference among 3 groups in running time, the similar tendency was found.2) The choice behavior significantly differed among 5 groups. Although Groups I and IV tended to chose the same side with about equal frequency at the end of training, the Ss subjected to the differential conditioning (Groups II, III and V) preferred one side in 100%, maze and avoided that side in, the other maze (Fig. 2). This fact would suggest that some difference in percentage of reinforcement tends to fixate the choice behavior and these tendencies result from behavioral contrast effect.3) Unlike Exp. I, Exp. II showed no evidence of contrast effect. The successive shift of percentage reinforcement of S2 did not result in any change in response time (Fig. 3) and choice behavior. It would be shown from the present experiments that behavioral contrast effect was observed in the simultaneous shift of percentage of reinforcement, but not in the successive shift.

Behavioral contrast and frustration effect in multiple and mixed fixed-interval schedules in ther rat.

Behavioral contrast and frustration effect in multiple and mixed fixed-interval schedules in ther rat.

"Behavioral contrast and frustration effect in multiple and mixed fixed-interval schedules in the rat": Erratum.

"Behavioral contrast and frustration effect in multiple and mixed fixed-interval schedules in the rat": Erratum.

FIXED‐RATIO PUNISHMENT BY TIMEOUT OF CONCURRENT VARIABLE‐INTERVAL BEHAVIOR1

Pigeons' responding was maintained by two concurrently available variable-interval reinforcement schedules. A fixed-ratio punishment schedule of timeout periods from the concurrent reinforcement schedules was arranged for responding during one of the variable-interval schedules. The greater the probability of a timeout after a response on the punished variable-interval schedule (the smaller the fixed ratio that produced timeout), the greater the decline in the relative punished response rates. Relative reinforcement rates remained invariant when relative response rates declined. Both behavioral contrast and induction effects were observed on the unpunished variable-interval schedule as a function of timeout punishment of the other schedule.