Aprendizagem observacional, formação e expansão de classes de equivalência.

Abstract

Observational learning of conditional discrimination has been described in the literature, with results thus far yielding few conclusions. The present study investigated the conditions under which observational learning of conditional discrimination lead to the formation and expansion of equivalence classes in different populations. Served as participants in Experiments 1, 2 and 3, college students; in Experiment 4, preschool children and, in Experiment 5, children and adolescents with special education needs. Experiment 1 systematically replicated an earlier experiment carried out with mentally retarded adults with increased control over stimuli presentation and the demonstrators responses. The experimental stimuli were familiar pictures, and instructions for the participant to observe were introduced. First, BA relations were taught, followed by modeling of CA relations, and by tests of baseline and emergent relations. Finally, modeling of two new relations involving stimulus sets D and E and tests for expanded classes were introduced. All six participants learned the conditional discriminations, and showed equivalence class formation and only one of them did not show class expansion. Experiment 2 had an experimental design analogous to Experiment 1, but all the conditional relations were taught through modeling. Three of four participants learned the conditional discriminations and showed formation and expansion of classes. Experiment 3 extended the Experiment 1 results to three classes of abstract stimuli, all six participants learned the conditional discrimination and 5 of them showed formation and expansion of equivalence classes. In Experiment 4, eight preschool children were submitted to a similar procedure, but with familiar pictures were used as stimuli and tokens that could be exchanged for a gift at the end of the session, regardless of their performance. Six of eight participants learned relations BA and CA and showed equivalence class formation and two of them showed class expansion. In Experiment 5, children and adolescents with special needs were taught BA and CA relations only through modeling. The delay between the demonstrator s behavior and the participants response was manipulated as well as the reinforcement system, and seemed to be important for three out of six participants. The remaining three participants did not learn the relations even after protracted training. In general, observational learning occurred for all populations investigated. Further investigation, however, is necessary to clarify the role of the observational response, differential reinforcement and the delay between demonstration and response, the demonstrator as a conditioned reinforcer, and other reinforcement sources for the observer s behavior.