Biased Attentional Processing of Positive Stimuli in Social Anxiety Disorder: An Eye Movement Study

Abstract

Despite the established relationship between social anxiety and attentional bias towards threat, a growing base of evidence suggests that social anxiety is additionally maintained by a deficit in the attentional processing of positive information. However, it remains unclear which component of attention is implicated in this deficit. Using eye movement-based measures and a novel attentional cuing methodology, the present study sought to investigate the presence of anxiety-linked bias in attentional engagement with, attentional disengagement from, and total fixation time to, socially relevant emotional stimuli in individuals diagnosed with social anxiety disorder, relative to non-socially anxious controls. Socially anxious individuals were found to exhibit faster attentional disengagement from positive stimuli, and reduced total fixation time to all emotional stimuli, relative to controls. Additionally for socially anxious individuals, lower total fixation times to positive stimuli were associated with higher levels of state anxiety. No differential pattern of engagement was evident between groups. We conclude that social anxiety is maintained in part by the aberrant processing of positive social stimuli.

Keywords:

• social anxiety
• attentional bias
• eye tracking
• gaze
• emotion

Acknowledgements

The authors would like to acknowledge the assistance of Dr Robert Battisti for statistical advice, Ms Amanda Kenyon for assistance with recruitment and assessment of clinical participants, and Mr Benjamin Grafton and Mr Daniel Rudaizky for assistance with control participant recruitment and experiment setup. This project was partially supported by an ARC Linkage project grant LP110200562 to Associate Professor Guastella.

Notes

^1. The engagement–disengagement terminology was guided by previous literature (e.g., Koster et al., 2006; Mogg et al., 2008). However, facilitated engagement has also previously been referred to as attentional vigilance, and delayed disengagement as attentional dwelling.

^2. It has been suggested that 88 is the optimal cut-off point for the full form SPAI (Peters, 2000). As the maximum score is 192 for the SPAI and 64 for the SPAI-23, the cut-off for the SPAI-23 was set to the closest integer to 88 divided by 3, which results in 29.

^3. The following sets of EM analyses were additionally repeated without the inclusion of covariates. The key anxiety-linked EM findings remained significant at the .05 level. In addition, control reimbursement method did not influence EM findings.