Fear-relevant illusory correlations: what types of associations promote judgmental bias?

A. J. Tomarken, S. Mineka, and M. Cook (1989) found that high-fear individuals markedly overestimated the covariation between fear-relevant stimuli and aversive outcomes. The authors assessed what features of stimulus-outcome associations promote illusory correlations. In Experiment 1, participants with high snake fear exhibited significant covariation bias for slides of snakes and shocks, but not for slides of damaged electric outlets (DEOs) and shocks. In Experiment 2, individuals with high and low snake fear rated DEOs and shocks as belonging together better than snakes and shocks. However, the shapes of high-fear individuals' affective response profiles to snakes and shocks were more similar than their profile shapes involving other pairings. In addition, their affective responses to snakes and snake-shock profile similarity predicted snake-shock belongingness ratings. These results suggest the importance of emotional responses and emotional profile similarity in mediating illusory correlations involving fear-relevant stimuli.