A Bayesian Case-Controls Exploration of the Malleability of Attentional Bias for Threat in Social Phobia

Abstract

Recent studies have demonstrated that training individuals with a social phobia to attend to non-threat is related to short-term (Amir et al., 2008) as well as long-term decrease in anxiety (e.g., Schmidt et al., 2009). However, to date, no study has examined the adaptation of this attention bias modification procedure into a single-case design for social phobia. Using an attention training procedure based on a modified dot-probe task, the present single-case study examined whether such procedure enabled a client with a social phobia to reduce attentional biases and to transfer this rehabilitation on anxiety response to a speech performance, social anxiety symptoms severity, and diagnosis of social phobia. Using a Bayesian probabilistic approach, case’s performances were compared to a normative sample before, after the treatment, at a two-month follow-up, at a six-month follow-up, as well as one year after the treatment. The results suggested the efficacy of the rehabilitation program on the attentional bias for threat cues and the generalization of these beneficial effects to decrease in anxiety symptoms during the two- and six-month follow-up period. A significant decrease in both subjective and behavioral anxiety during speech performance was also observed. However, a set back of the attentional bias for threat as well as symptoms of social anxiety was observed at the one-year follow up.