Variabilidade comportamental negativamente reforçada em humanos sob contingências de esquiva

Abstract

The behavioral variability has been investigated through procedures in which varied units produce reinforcement in one condition, while another condition maintains the same intermittency of reinforcement, but requires only the emission of the units, without the need for variation. The literature has a wide investigation of this phenomenon under positive reinforcement contingencies, but it’s scarce in the study under negative reinforcement contingencies. Among the few studies that investigated this phenomenon under negative reinforcement contingencies, no work has been identified that has used avoidance procedures with human participants. Therefore, this research aimed to investigate the acquisition and maintenance of behavioral variability under avoidance contingencies in human participants. To this end, 5 university students participated in a procedure that included three phases: CRF, VAR and ACO. In the first phase, units composed of four responses of pressing the Q and P keys of a computer avoided the loss of points in a continuous reinforcement schedule (CRF). In the VAR Phase, the loss of points was avoided by emitting a unit that differed from the last five units emitted (Lag 5 reinforcement schedule). In the ACO Phase, the intermittency of reinforcement was similar to that obtained in the VAR Phase and the variation was allowed, but was not required. The data of the proportion of reinforced attempts and latency of the reinforced units showed that there was learning and maintenance of the avoidance response throughout all the procedure. For all participants, there was an increase in the VAR Phase in the levels of variability ascertained by the U value, although the arbitrarily value defined in the area as indicative of behavioral variability (0,8) has not been reached in four cases. The analysis of the final half of the VAR phase indicated values between 0,78 and 0,86 for three participants. With the introduction of the ACO Phase, three participants (P1, P2 and P3) showed a decrease in this measure and two participants (P4 and P5) showed no change. The analysis of only the final half of the ACO Phase, however, indicated a drop in the U value for all participants. The control of the intermittency of reinforcement was effective in demonstrating slight increases of 1 to 4 percentage points in the ACO Phase compared to the VAR Phase. The analysis of the number of different units emitted and the distribution of the relative frequency of each behavioral unit also provided evidence of the increase in behavioral variability with the introduction of the variation requirement in the VAR Phase. These results suggest the operant control of behavioral variability under avoidance contingencies with human participants