Manipulation of dangerousness judgements to fear-relevant stimuli: effects on a priori UCS expectancy and a posteriori covariation assessment.

Using a covariation assessment procedure, the present study pre-experimentally manipulated subjects' assessments of the dangerousness of fear-relevant stimuli. The results demonstrated that this manipulation subsequently influenced both a priori expectancy that the stimulus will be followed by an aversive consequence (UCS), and inflated an a posteriori covariation bias indicating that subjects judged the stimulus to be followed more frequently by the aversive UCS than by a non-aversive stimulus. These findings support the view that judgments about such characteristics as the dangerousness of a stimulus causally contribute to a UCS expectancy bias and an a posteriori covariation bias, and thus contribute to the formation of selective associations.