Bayesian Models of Psychological Inference: How to Predict Actions under Uncertainty Situations?

This study investigated if capacity for predicting actions under uncertainty conditions can be understood like a Bayesian inferential processes. Two experiments were carried out, each one with a different group of basic cycle undergraduate students. In both experiments it was presented a film that showed an agent getting into mace for eating meal. Participants had to predict which of four entrances would use the agent for getting inside of the mace. Results showed a high correlation, significant to a p < 0.01, between predictions of a Bayesian inferential model and the answers of the participants. Findings reveal that capacity for predicting actions involve estimations of posterior probability in accordance with the Bayesian model proposed.