El efecto de la frecuencia de juicio y de la experiencia previa no contingente en el aprendizaje causal

Résumé

This work aimed to study the interaction between judgment frequency and a previous non-contingent experience upon the causal relationships detection and attribution in a subsequent contingent task. Experiment 1 showed an impaired detection after a non-contingent experience independently of the judgment frequency during non-contingency, revealing the reliability of this previously found effect. Experiment 2 showed firstly, a clear judgment of frequency effect upon causal attribution in the contingent phase; secondly, the results also showed that a non-contingent experience altered the influence of each type of trials upon causal attribution in the subsequent contingent task. These results raise doubts about the ability of single mechanism models, be it associative or statistical, to explain human causal attribution and learning. However, the recently proposed belief revision model was able to fully explain these results, as showed buy the simulation of this model based on the predictions and parameters derived from previous research.