We examined the relative contribution of specific components of multidimensional anxiety to cognitive biases in the processing of threat-related information in three experiments. Attentional bias was assessed by the emotional Stroop word color-naming task, interpretative bias by an on-line inference processing task, and explicit memory bias by sensitivity (d') and response criterion (β) from word-recognition scores. Multiple regression analyses revealed, first, that phobic anxiety and evaluative anxiety predicted selective attention to physical- and ego-threat information, respectively; cognitive anxiety predicted selective attention to both types of threat. Second, phobic anxiety predicted inhibition of inferences related to physically threatening outcomes of ambiguous situations. And, third, evaluative anxiety predicted a response bias, rather than a genuine memory bias, in the reporting of presented and nonpresented ego-threat information. Other anxiety components, such as motor and physiological anxiety, or interpersonal and daily-routines anxiety made no specific contribution to any cognitive bias. Multidimensional anxiety measures are useful for detecting content-specificity effects in cognitive biases.
The broad applicability of memory bias and its coexistence with the planning fallacy: Reply to Griffin and Buehler (2005).