Belief bias and the extinction of induced fear.

Some people show slower extinction of UCS expectancies than other people. Little is known about what predicts such delayed extinction. Extinction requires that people deduce the logical implication of corrective experiences challenging the previously learned CS-UCS contingency. "A strong habitual tendency to confirm beliefs" may therefore be a powerful mechanism immunising against refutation of UCS expectancies. This study investigated whether individual differences in such a belief confirming tendency (a process called "belief bias") may help in explaining individual differences in extinction. We tested whether relatively strong belief bias predicts delayed extinction of experimentally induced UCS expectancies. In a differential aversive conditioning paradigm, we used UCS-irrelevant (Experiment 1) and UCS-relevant (Experiment 2) pictorial stimuli as CS⁺ and CS⁻, and electrical stimulation as UCS. Belief bias indeed predicted delayed extinction of UCS expectancies when the CS⁺ was UCS-relevant (as is typically the case for phobic stimuli, Experiment 2). The study provides preliminary evidence that enhanced belief bias may indeed play a role in the persistence of UCS expectancies, and can thereby contribute to the development and persistence of anxiety disorders. The results also point to the relevance of reasoning tendencies in the search for predictors of delayed extinction of UCS expectancies.