Abstracts

Rationale: Event-related augmentation of high-gamma activity on electrocorticography (ECoG) has been suggested to be a summary measure of cortical activation with excellent temporal resolution. Here, we generated a four-dimensional map of the visual cortex, by using visually-elicited high-gamma amplitudes averaged across a large number of patients.   Methods: Patients with seizure onset zones involving the occipital cortex were excluded; thus, the present study included 63 patients with drug-resistant focal epilepsy. During extraoperative ECoG recording in a darkened room, a series of 50 full-field stroboscopic flash stimuli were given to each patient with a frequency of 0.5 Hz, at a distance of 30 cm from the closed eyes. At each electrode site, we measured the amplitude change at high-gamma band (70-110 Hz), in steps of 10 Hz and 5 ms, relative to the mean amplitude in the baseline period between 200 and 100 ms prior to stimulus onset. Event-related potentials (ERPs) were also measured by averaging voltage measures time-locked to stimulus onset. The percent changes in high-gamma amplitude as well as ERPs were sequentially animated on the FreeSurfer standard surface image as a function of time. Results: High gamma augmentation was confined to the anterior portions of bilateral lingual and cuneus gyri at 30 ms following stimulus onset and involved the entire portions of those gyri at 80 ms. High-gamma augmentation was modest in the lateral occipital and fusiform gyri. High-gamma augmentation in the lingual and cuneus gyri was subsequently replaced by prolonged high-gamma attenuation lasting from 200 to 600 ms following stimulus onset. Such high-gamma attenuation was particularly prominent in the posterior portions of bilateral lingual and cuneus gyri as well as the lateral occipital gyri close to the occipital poles. The initial negative ERP peak (N1) within 100 ms coincided with high-gamma augmentation, whereas the subsequent positive (P1) and negative (N2) peaks were rather associated with high-gamma attenuation.  Conclusions: Full-field flash stimuli elicited brief neuronal activation initially in the lower-order visual areas for the peripheral field and subsequently in those for the central field, whereas activation of the higher-order visual areas was modest. Prolonged suppression subsequently took place in the lower-order visual areas particularly in those for the central field as well as in the immediately surrounding higher-order visual areas. In contrast to event-related high-gamma amplitudes demonstrating alternate cortical activation and deactivation in the visual pathways, the amplitude measures on ERP peaks per se failed to infer the direction of neural modulation at given moments.  Funding: Grant support: R01NS064033 to E. Asano.