Dynamics of gamma responses to negative emotional visual stimuli: from local activity to functional connectivity.

Many electroencephalographic (EEG) studies on the cortical dynamics induced by unpleasant picture view- ing demonstrated the modulation of event-related potentials (ERPs) components as a function of valence and the increase of gamma band responses to emotional stimuli; while only a few studies investigated phase synchronization phenomena such as inter-trial or between regions phase locking of gamma responses to emotional stimulation. The aim of this study was to provide a complete description of the cortical dynam- ics induced by unpleasant and neutral pictures viewing, from the ERP averages to gamma rhythm modula- tion, and its phase synchronization. Gamma rhythm modulation was estimated by the event-related synchronization (ERS) approach, and phase synchrony between trials and between cortical regions was studied by extending the phase-locking statistics (PLS) approach. Consistent with previous literature, an increase in P300 and late positive potential and an increase in gamma activity during viewing of unpleasant pictures as compared to neutral ones were found. No inter-trial syn- chronization was evoked by the stimuli, whereas widespread phase locking between sites was identi␣ed. In particular, differences in gamma synchronization between unpleasant and neutral stimuli were found. Speci␣cally, at early (0–250 ms) lags from stimulus onset, in the 38–45 Hz gamma interval, stronger inter- site synchronizations for the unpleasant stimuli, even though quite widespread across the scalp, mainly in- volved the interhemispheric synchronization between temporal and frontal regions. In contrast, in the 30–37 Hz gamma interval, stronger synchronizations for the responses to neutral trials were found in the 500–750 time interval, mainly involving the temporo-parietal regions. These ␣ndings suggest that the full elaboration of unpleasant stimuli requires a tight interhemispheric com- munication between temporal and frontal regions that is realized by means of phase synchronization at about 40 Hz. In addition, in contrast with the idea of a broadband modulation of high-frequency activity by cognitive/emotional stimuli, the present ␣ndings i.e. stronger BRS responses to either emotional or neutral trials at speci␣c frequency and time range, indicate that speci␣c intervals of gamma activity could be each pri- marily involved in a speci␣c aspect of stimulus processing