Changing attentional bias: Can it effect self-reported anxiety?

Studies conducted over the last decade have demonstrated the occurrence of attentional biases towards anxiety-relevant verbal information in people with anxiety disorders and those with high trait anxiety. Studies investigating the existence and significance of attentional bias in anxiety disorders have shown attentional bias to be a reliable correlate of anxiety that diminishes when anxiety is successfully treated (e.g., Lavy et nl., 1993). However, despite speculation that attentional bias may predispose the development of anxiety, or contribute to maintaining anxiety, only one empirical study has demonstrated that it is possible to alter attentional bias, and that this has an impact on subsequent state anxiety (MacLeod, 1995). Such demonstrations are crucial in determining the causal status of attentional bias in anxiety disorders. The present study reports a methodology for investigating the causal status of attentional bias in anxiety. Forty-four non-clinical volunteers were randomly allocated to one of two training conditions aimed at inducing an attentional bias either towards or away from spider-related verbal information. Procedures targeting attentional bias did not influence self-reported anxiety in this study, nor did changes in responding on the dot probe task generalise to a second commonly used measure of attentional bias, the Stroop task. Explanations of these findings with regard to the mediational role of attentional biases in anxiety are considered.