Efeitos da História Comportamental e de Instruções sobre a Aquisição e a Resistência à Extinção em um Esquema Múltiplo FR DRL / The Effects of Behavioral History and Instructions upon the Acquisition and Resistance to Extinction with a Multiple FR DRL Schedule

RESUMOPara verificar os efeitos da história comportamental e de instruções sobre a aquisição e a resistência à extinção, 15 universitários foram distribuídos em três grupos. Na Fase 1, o Grupo 1 foi exposto a aumentos graduais nos valores dos componentes até um múltiplo FR 60 DRL 20 s; os grupos 2 e 3 foram expostos diretamente a esse esquema, mas o Grupo 3 recebeu instruções. Na Fase 2, o FR foi ajustado e um múltiplo FR n DRL 20 s foi mantido até a estabilidade. Na Fase 3 estava em vigor um múltiplo Extinção Extinção. As instruções, em comparação com a exposição gradual ou direta aos esquemas, facilitaram a aquisição de taxas diferenciadas de respostas, mas aumentaram a resistência à extinção.

ABSTRACTTo investigate the effects of the behavioral history and instructions on acquisition and resistance to extinction with a multiple schedule, 15 students were distributed into three groups. In Phase 1, Group 1 was exposed to gradual increases in the values of the components until a multiple FR 60 DRL 20 s; groups 2 and 3 were directly exposed to that multiple schedule, but Group 3 received instructions. In Phase 2, the FR was adjusted and a multiple FR n DRL 20 s was maintained until stability. In Phase 3, a multiple Extinction schedule was in effect. Instructions, as compared to direct or gradual exposure to the schedules, facilitated acquisition of differentiated response rates, but increased resistance to extinction.

Adquisición y dinámica de preferencias en programas múltiples de reforzamiento / Acquisition and Preference Dynamic in Multiple Schedules of Reinforcement

Ocho palomas fueron entrenadas en programas múltiples de reforzamiento Razón Variable-Razón Variable (mult RV-RV) con cambios rápidos e imprevistos en las distribuciones de refuerzo en ambos componentes del programa múltiple. El objetivo principal fue evaluar cómo se ajustan las tasas de respuestas a cambios abruptos y no señalados en las condiciones de reforzamiento en distintos períodos y en particular determinar si la dinámica del ajuste del comportamiento es dirigida por la razón o por la diferencia en las probabilidades en las tasas de reforzamiento obtenido en dos componentes de un programa múltiple. Los principales hallazgos fueron que cuando las diferencias entre dos programas de reforzamiento (pobre y rico) son constantes, el desarrollo de la preferencia por una de las respuestas del programa múltiple fue más rápido cuando la razón de las probabilidades de reforzamiento fue mayor (5 a 1), lo cual es congruente con los resultados obtenidos en programas de reforzamiento concurrentes. Sin embargo, cuando la razón se mantuvo constante, la tasa de adquisición no fue más rápida cuando la diferencia entre la probabilidad de reforzamiento fue mayor, hallazgo distinto al reportado en experimentos con programas concurrentes. Los resultados resaltan la importancia de la discriminabilidad en probabilidades de reforzamiento entre la fase de entrenamiento y de transición.

Eight pigeons were trained on multiple variable ratio-variable ratio schedules of reinforcement (mult VR-VR) with rapid and unexpected changes in reinforcement distribution within both components of the multiple schedule. The main objective of the study was to assess the adjustment of response rates to abrupt and unsignaled changes in the conditions of reinforcement in different periods, particularly whether the dynamics of behavioral change is guided by the ratio or by the difference between the probabilities of reinforcement obtained in the two components of the multiple schedule. The main findings were that when the differences between two schedules of reinforcement (lean and rich) are constant, the development of preference for one of the responses in the multiple schedule was faster when the ratio of the probability of reinforcement was higher (5 to 1), which is consistent with the results obtained with concurrent schedules of reinforcement. However, different from the results obtained using concurrent schedules, when the ratio of probabilities remained constant the rate of acquisition was not faster when the difference between the probabilities of reinforcement was higher. The results highlight the importance of the discriminability of reinforcement probability between training and transition phases.