Ressurgência: estudos de variáveis controladoras

Abstract

The present study verified through three experiments with university students, how the influence of cost and response topography can control resurgence. Such phenomenon is characterized by the reappearance of behaviors that have been reinforced in the past and come back in later situations, even in the discontinuation of reinforcement. Software was developed that measured the frequency of different responses through the touches emitted on a sensitive computer screen. The phases consisted of Baseline (LB), Training (TR) and Extinction (EXT). In Experiment 1, the measures were responses at different costs distributed in links. In a link A the participant could issue a response, in B three and C five. The results showed that resurgence occurred due to the experimental control of reinforcement that was allocated according to the Training Phases (1 and 2), where in the operand of lower choice reinforcement was allocated. Extinct responses in Training Phase 1 were found to reappear in Extinction Phase (EXT 1). Thus, in Experiment 2 a new Training Phase (TR 3) was allocated after EXT 1 and an Extinction Phase 2 (EXT 2) after TR3. However, the measurements were verified in only one operand for the three links (A, B and C). The results made it possible to observe resurgence in both EXT 1 and EXT 2 phases, possibly due to history building (EXT 1) and recent training (EXT 2). Experiment 3 investigated whether topographically different responses could resurface. The phases were similar to those of Experiment 1 and the results found that only a single response resurfaced in Phase EXT, by participants 1, 2 and 4, thus raising a discussion regarding the criteria for a behavior to be considered resurgent or not. One of the considerations made endorsed the variability of behavior as a function of its reinforcement history, as well as the possibility of a single criterion for considering behavior as resurgent