(en) A large body of behavioral studies has evidenced attentional biases towards emotions in anxiety. Anxious people show enhanced detection, faster orienting and sustained attention towards threat stimuli, as for instance fearful or angry faces. This bias phenomenon has often been attributed to an hypervigilance to threat cues. However, an alternative explanation has proposed a main implication of disengagement process. Hence, controversies remain about the exact nature of the attentional processes implicated in the development of emotional biases. Due to their excellent temporal resolution, ERPs allow to explore the timing of emotional modulation and to identify which stages of cognitive processing (e.g. perceptive, attentional or executive) are affected and involved. As a consequence, we conducted six ERPs studies in order to answer to four questions: first, where is located the cognitive origin of emotional biases in anxiety, and what is the role of anxiety subtypes Second, do some emotions give rise to more pronounced processing biases? Third, what kind of attentional processes are mostly involved in the deployment of emotional biases? Fourth, does the addition of a depressive state modify the level of occurrence of emotional biases? Our thesis presents insights from six ERPs studies approaching social anxiety and trait anxiety through different paradigms, and shows how these points can be clarify. First, we observed that social anxiety is associated with enhanced perceptive processing of faces, indexed by increased P1/N1 waves. This initial orientation towards faces objects is followed by difficulties to disengage attentional resources from them, leading to reduced visual processing of subsequent targets to detect. In contrast, trait-anxiety did not modulated perceptive or attentional stages of cognitive processing but rather impacted on later stages related to response preparation (P300). Beside, the modulation of the emotional appraisal stage, indexed by the N300 component, was a common point between social anxiety and trait-anxiety. Lastly, the addition of an anxious state and depressive affects alters the strategic deployment of attentional resources towards emotional stimuli. Bias towards threat was not amplified in sub-clinical anxious states, but social anxiety was correlated with a perceptual hypervigilance towards all types of emotional expressions, positive or negative. To sum up, these studies suggest earlier modulations of emotional processing linked to social anxiety, while trait-anxiety influences later decisional stage through hyper-reactivity. Theoretical implications and clinical applications are raised, and different propositions concerning futures works are proposed.
Rossignol, M. (2009). Exploration électrophysiologique des biais attentionnels envers les expressions faciales émotionnelles dans l’anxiété. https://hdl.handle.net/2078.5/131323