Modelo Experimental de Corrupção: efeitos da magnitude e da probabilidade de punição em humanos

The study proposed an experimental model of corruption, to investigate the effect of magnitude and probability of punishment, in an experimental situation. The experiment simulated a typical corrupt act, in a game with university students and was divided into two phases, with one session each, also to assess the effect of exposure to the contingency of punishment. Participants should choose between two alternatives: one with reinforcers of smaller magnitude with no associated punishment, and another with reinforcers of greater magnitude, with punishment being manipulated in seven experimental conditions. Findings show that there was initially a greater reduction in undesirable behavior with the magnitude of the penalty but, after longer contact with the contingency and better discrimination of the probabilities, a growing effect of probability was noted. There was an order effect, with recovery of responses in groups in which punishment was applied in a decreasing way.