Fear of negative evaluation and attentional bias for facial expressions: an event-related study.

Numerous studies have shown an exacerbation of attentional bias towards threat in anxiety states. However, the cognitive mechanisms responsible for these attentional biases remain largely unknown. Further, the authors outline the need to consider the nature of the attentional processes in operation (hypervigilance, avoidance, or disengagement). We adapted a dot-probe paradigm to record behavioral and electrophysiological responses in 26 participants reporting high or low fear of evaluation, a major component of social anxiety. Pairs of faces including a neutral and an emotional face (displaying anger, fear, disgust, or happiness) were presented during 200 ms and then replaced by a neutral target to discriminate. Results show that anxious participants were characterized by an increased P1 in response to pairs of faces, irrespective of the emotional expression included in the pair. They also showed an increased P2 in response to angry-neutral pairs selectively. Finally, in anxious participants, the P1 response to targets was enhanced when replacing emotional faces, whereas non-anxious subjects showed no difference between the two conditions. These results indicate an early hypervigilance to face stimuli in social anxiety, coupled with difficulty in disengaging from threat and sustained attention to emotional stimuli. They are discussed within the framework of current models of anxiety and psychopathology.

Electrophysiological correlates of enhanced perceptual processes and attentional capture by emotional faces in social anxiety.

Behavioural studies have used spatial cueing designs extensively to investigate emotional biases in individuals exhibiting clinical and sub-clinical anxiety. However, the neural processes underlying the generation of these biases remain largely unknown. In this study, people who scored unusually high or low on scales of social anxiety performed a spatial cueing task. They were asked to discriminate the orientation of arrows appearing at the location previously occupied by a lateralised cue (consisting of a face displaying an emotional or a neutral expression) or at the empty location. The results showed that the perceptual encoding of faces, indexed by P1, and mobilisation of attentional resources, reflected in P2 on occipital locations, were modulated by social anxiety. These modulations were directly linked to the social anxiety level but not to trait anxiety. By contrast, later cognitive stages and behavioural performances were not modulated by social anxiety, supporting the theory of dissociation between efficiency and effectiveness in anxiety.