Nicholas G Cicero^1,3,4 (), Michaela Klimova², Sam Ling^1,2, Laura Lewis^3,4,5,6; ¹Graduate Program in Neuroscience, Boston University, ²Department of Psychological and Brain Sciences, Boston University, ³Department of Biomedical Engineering, Boston University, ⁴Institute for Medical Engineering and Sciences, Massachusetts Institute of Technology, ⁵Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, ⁶Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital

While asleep, our brains withdraw from our external environment and presumably suppress the processing of external sensory information. Although sensory disconnection is necessary for our brain to transition into a sleep state, the mechanisms regulating sensory gating remain debated. While the eyelid is the first barrier to visual input while we sleep, with closed eyes, high luminance inputs still reach the lateral geniculate nucleus (LGN) and primary visual cortex (V1). Here, we investigated how arousal state impacts visual processing using simultaneous EEG-fMRI in humans resting with their eyes closed (n=15), across sleep and wakefulness. During an eyes-closed resting-state scan, we presented a temporal contrast modulated flickering stimulus while subjects naturally transitioned in and out of sleep. Subjects were also instructed to perform a self-paced behavioral task to track behavioral responsivity. We first found that while awake with closed eyes, flickering stimuli induced responses in LGN and in V1, whereas extrastriate cortical responses were suppressed. During non-rapid eye movement (NREM) sleep, the visual-evoked response pattern substantially changed: visually-evoked responses were largely preserved in the LGN, whereas we observed substantial stimulus-locked suppression in visual cortex. Specifically, the magnitude of stimulus-locked visuocortical suppression scaled with the stimulus intensity, wherein higher intensity stimuli induced even larger negative visually evoked responses. To further investigate when visuocortical deactivation occurred as subjects transitioned from wakefulness to sleep, we also separated trials by behavioral state and found that this visuocortical suppression occurred only when subjects were behaviorally unresponsive. During the fully unresponsive state, the EEG steady-state visual evoked potential was also significantly attenuated, and low frequency delta power, a signature of sleep depth, was greatest. Given that the LGN showed preserved visual- evoked activation across arousal states, these results suggest that cortical inhibitory circuits are a key mechanism by which visual cortex disengages from environmental stimuli during sleep.