Primacy in causal strength judgments: The effect of initial evidence for generative versus inhibitory relationships

The order in which people receive information has a substantial effect on s ubsequent judgment and inference. Our focus is on the order of covariation evidence in causal learning. The first experiment shows that the initial pr esentation of evidence suggesting a generative causal relationship (the joi nt presence or joint absence of cause and effect) leads to higher judged ca usal strength than does the initial presentation of evidence suggesting an inhibitory relationship (the presence of cause or effect in the absence of the other). Additional studies show that this primacy effect is unlikely to be due to fatigue or to an insufficient number of learning trials. These r esults are not readily explained by current contingency-based or associativ e theories of causal induction.