Abstract
Anxiety disorders are characterized by biased implicit threat associations, which can be measured by indirect reaction time tasks. These tasks might provide a useful tool in the assessment of individual diagnoses and therapeutic changes. However, sufficient psychometric properties of the applied tasks are a prerequisite for these applications. Therefore, we comparatively investigated the reliability and validity of an Extrinsic Affective Simon Task (EAST), an Approach-Avoidance Task (AAT), and an Affective Priming Task (APT) by presenting the same tasks twice within 1 week. Data show retest reliabilities of around r = .42 for the EAST, r = .35 for the AAT, and r = .63 for the APT. Internal consistencies varied between .44 and .49 for the EAST, .66 and .70 for the AAT, and .53 and .76 for the APT. Validity correlations with self-report questionnaires ranged between r = .43 and r = .59, being lowest for the EAST and highest for the AAT. We argue that while these instruments might not be applicable to individual diagnostics yet, they are sufficiently reliable and valid to be used in the assessment of group differences.
