Covariation Bias in Blood-Injection-Injury Fear: The Application of Methodological Modifications Within the Illusory Correlation Paradigm

Previous studies have failed to support the prediction that covariation bias functions as a possible etiological and/or maintenance mechanism of blood-injection-injury (BII) fears. Methodological considerations may partially account for previous null findings. The current study extends previous studies by examining covariation bias in high (n = 32) and low (n = 30) BII fearful individuals using a systematically modified illusory correlation paradigm. Previous examinations have used multi-emotion stimuli (e.g., blood) or disproportionate stimulus–outcome emotion categories (e.g., Fear, Neutral, Neutral). Therefore, we incorporated fear-specific, disgust-specific, and neutral stimuli paired with ecologically valid, affectively-matched outcomes. Stimulus–outcome pairings were randomly paired and presented with equal frequency across trials. Participants provided estimations on the proportion of presented stimulus–expression pairings. Contrary to previous findings, our modified paradigm revealed a general covariation bias for affectively representative stimuli in high versus low BII fearful individuals. Further, the present study suggests a functional role for methodological variance in probability estimation. The current study’s limitations, in addition to theoretical and methodological implications are discussed.