Applying the Implicit Association Test to Measure Intolerance of Uncertainty

Intolerance of Uncertainty (IU) is a key trans-diagnostic personality construct strongly associated with anxiety symptoms. Traditionally, IU is measured through self-report measures that are prone to bias effects due to impression management concerns and introspective difficulties. Moreover, self-report scales are not able to intercept the automatic associations that are assumed to be main determinants of several spontaneous responses (e.g., emotional reactions). In order to overcome these limitations, the Implicit Association Test (IAT) was applied to measure IU, with a particular focus on reliability and criterion validity issues. The IU-IAT and the Intolerance of Uncertainty Inventory (IUI) were administered to an undergraduate student sample (54 females and 10 males) with a mean age of 23 years (SD = 1.7). Successively, participants were asked to provide an individually chosen uncertain event from their own lives that may occur in the future and were requested to identify a number of potential negative consequences of it. Participants' responses in terms of cognitive thoughts (i.e., cognitive appraisal) and worry reactions toward these events were assessed using the two subscales of the Worry and Intolerance of Uncertainty Beliefs Questionnaire. The IU-IAT showed an adequate level of internal consistency and a not significant correlation with the IUI. A path analysis model, accounting for 35% of event-related worry, revealed that IUI had a significant indirect effect on the dependent variable through event-related IU thoughts. By contrast, as expected, IU-IAT predicted event-related worry independently from IU thoughts. In accordance with dual models of social cognition, these findings suggest that IU can influence event-related worry through two different processing pathways (automatic vs. deliberative), supporting the criterion and construct validity of the IU-IAT. The potential role of the IU-IAT for clinical applications was discussed.