Abstracts

Rationale: Humans spontaneously perform saccades to look around during wakefulness. It has been hypothesized that the cortical processes of vision are transiently suppressed during a saccade and reactivated immediately following saccade offset. Here, we determined whether human occipital regions were indeed suppressed during saccades and subsequently reactivated, as measured by gamma activity at 80-150 Hz. We also determined whether occipital gamma activity was likewise modulated by saccadic eye movements without external visual stimuli during rapid eye movement (REM) sleep. Methods: We studied nine patients (age: 5-17 years) with focal epilepsy undergoing extraoperative electrocorticography (ECoG). All patients had normal intelligence and visual acuity. Using time-frequency analysis, we determined the spatiotemporal dynamics of cortical gamma activity relative to the onset and offset of spontaneous saccades during wakefulness as well as REM sleep. Results: In total, we found 725 spontaneous saccade events during wakefulness and 243 during REM sleep. No significant gamma-modulation was found prior to saccade onset. During wakefulness, significant gamma-attenuation involved 43 occipital sites during saccades, 12 of which subsequently showed augmentation; an additional 47 sites exhibited only post-saccade augmentation. The polar/lateral occipital lobe showed more extensive gamma-attenuation, while the medial/inferior occipital lobe showed more extensive gamma-augmentation (p<0.001 on Fisher exact probability test). We also found that a larger saccade duration was associated with a larger gamma-attenuation in all occipital area (p<0.001 on one-way repeated measures ANOVA). During REM sleep, the temporal profile of gamma-modulation was similar to that during wakefulness, but with smaller magnitude. Significant gamma-attenuation during REM sleep involved 6 of the same sites that attenuated during wakefulness; no sites exhibited significant gamma-augmentation during REM sleep. Conclusions: Portions of the human occipital cortex were suppressed immediately after saccade onset and throughout saccades, with subsequent reactivation following saccade offset, as measured by the amplitude of gamma activity at high temporal resolution. Such a transient suppression of occipital function may contribute to the maintenance of visual stability before and after saccades. Even without external visual stimuli, portions of occipital regions were modulated by saccades during REM sleep. This suggests that saccades themselves or the signals initiating saccades may induce occipital suppression regardless of visual input. The present study failed to find gamma-modulation preceding the onset of saccades.