• Citer

BOLD response to deviant face detection informed by P300 event-related potential parameters: A simultaneous ERP-fMRI study.

par Campanella, Salvatore ;Bourguignon, Mathieu ;Peigneux, Philippe ;Metens, Thierry ;Nouali, Mustapha ;Goldman, Serge ;Verbanck, Paul ;De Tiege, Xavier
Référence NeuroImage, 71, page (92-103)
Publication Publié, 2013-05

Article révisé par les pairs

• Accès en ligne
• Détails
• Contenu
• Statistiques

Résumé :

INTRODUCTION: Faces are multi-dimensional stimuli conveying parallel information about identity and emotion. Although event-related potential (ERP) studies have disclosed a P300 component in oddball responses to both deviant identity and emotional target faces, it is hypothesized that partially different neural processes should subtend emotion vs. identity within the core network of face processing. In the present study, we used simultaneous ERP-fMRI recordings and ERP-informed analysis of functional magnetic resonance imaging (fMRI) data to evidence the specific neural networks underlying P300 generation in response to different deviant emotional vs. identity faces. METHOD: 18 participants were scanned during a visual oddball task in which they had to detect 3 types of deviant faces representing a change in emotion-fear or happiness-or in identity, within a series of frequent neutral ones. Amplitude and latency parameters of the P300 component, recorded for each type of deviant faces, were used to constrain fMRI analyses. RESULTS: Analysis of fMRI data informed by single-trial parameters of the P300 component disclosed specific activation patterns for fearful, happy and identity deviant faces. For fearful faces, P300 amplitudes were associated with BOLD changes in the left fusiform gyrus whereas latencies were linked to left superior orbito-frontal and right fusiform activations. P300 amplitude modulations for happy deviant faces involved the left posterior cingulate gyrus and right parahippocampal regions whereas P300 latencies related to the right insula and left caudate regions. Finally, identity deviant faces were associated with widespread activities involving cortical and subcortical regions when P300 amplitudes were considered, and P300 latencies were associated with activity in right hippocampal/parahippocampal regions. DISCUSSION: Our results suggest the existence of differential cerebral functional processes involved in the responses to deviant face stimuli, depending on the quality of the deviance (fear vs. happiness vs. identity).