A priori expectancy bias in patients with panic disorder

Several studies have indicated that phobic participants tend to overassociate fear-relevant stimuli and aversive outcomes, i.e, they show a covariation bias. Such a bias seems to be a powerful way to confirm danger expectations and enhance fear. Therefore, a covariation bias might be an important factor in the maintenance of fear.

To investigate a covariation bias in patients with panic disorder, we had 29 patients and 29 healthy control participants rate the a priori probabilities with which they would expect pictures of mushrooms, spiders, erotic scenes, and emergency situations to be paired with a tone, shock, or nothing.

This is the first study to show that patients with panic disorder specifically overestimate the association between panic-relevant stimuli and a following negative consequence. This distorted contingency expectancy represents a panic-specific covariation bias, since it was not observable for other stimuli-consequence combinations and only to a significantly lesser degree in control participants.

The underpinning hypothesis is that overestimation of threat plays a casual role in the origins and maintenance of anxiety. Thus anxiety may induce a covariation bias, which in turn may enhance the perceived threat, which in turn may intensify the anxiety etc. This reciprocal relationship between covariation bias and anxiety may have clinical implications for prediction and treatment in patients with panic disorder.