Efeitos da liberação de reforçadores durante o intervalo entre tentativas sobre desempenho no emparelhamento ao modelo com atraso em sapajus spp.

Abstract

Studies with pigeons have shown that releasing reinforcers during the Intertrial Interval (ITI) on simple and conditional discrimination tasks can produce a drop on the stimulus control. In the case of conditional discrimination, the literature suggests that the conditions that deteriorate the performance involve the reinforcer’s release at the end or during all the ITI, in other words, when the reinforcer is released next to the beginning of a new trial. This study had the objective to investigate the effects of releasing of reinforcers during the ITI on the performance on the identity delayed-matching-to-sample task (DMTS) with three capuchin monkeys (Sapajus spp.). Each trial was initiated by presenting a sample stimulus; after the emission of five touches to the sample, it was initiated a delay and, at the end of the delay, comparison stimuli were presented so the choice response would occur. Correct responses were followed by the release of 190 mg banana’s pellet and by the ITI and incorrect responses were followed just by the ITI. The sessions were composed by trials with 1 s, 5 s, and 20 s delay and 30 s ITI. In order to study the effects of free reinforcer releasing , a 190 mg pellet was released also during the ITI, according to five conditions, which sequence was random and differed between subjetcs: Early – free reinforcer released 5 s after the beginning of the ITI; Middle – release 15 s after the beginning of the ITI; Late – release 25 s after the beginning of the ITI; All – release 5 s, 15 s and 25 s after the beginning of the ITI; and, None – there was no reinforcement release during the ITI. These conditions were applied in two procedures. On the first, named Maintenance, just one set of stimuli was used on all conditions of manipulation of reinforcement; on the second, named Acquisition, each ix condition presented a specific stimulus set. No effect of different conditions of free reinforcer releasing was observed on the general performance of the subjects. However, on the Acquisition procedure, it was verified a decay on the performance on the trials with maximum delay, 20 s, though the conditions of releasing reinforcement that affected the performance at these trials were different between the subjects. The result of one of the three subjects was similar to what was found on the literature, the deterioration on the performance on the trials with the longest delay, 20 s, occurred on the conditions in which the reinforcement was released at the end or during all the ITI. The other participants showed decay on the conditions in which there was no reinforcement during the ITI or when the release occurred at the beginning or end of the ITI. We suggest, for future researches, to test the same performances under the same conditions with new sets of stimuli seeking intra-subject replication and later to use trials with delays longer than the 20 s, according to the training history in different delays of each subject.