Lee Kangjoo, Wang Yimeng, Cross Nathan E, Jegou Aude, Razavipour Fatemeh, Pomares Florence B, Perrault Aurore A, Nguyen Alex, Aydin Ümit, Uji Makoto, Abdallah Chifaou, Anticevic Alan, Frauscher Birgit, Benali Habib, Dang-Vu Thien Thanh, Grova Christophe
Department of Psychiatry, Yale University School of Medicine, New Haven, Connecticut, USA, 06510.
Multimodal Functional Imaging Lab, Biomedical Engineering Department, McGill University, Montréal, QC, Canada H3A 2B4.
bioRxiv. 2024 Jul 2:2024.06.28.601285. doi: 10.1101/2024.06.28.601285.
Decrease in cognitive performance after sleep deprivation followed by recovery after sleep suggests its key role, and especially non-rapid eye movement (NREM) sleep, in the maintenance of cognition. It remains unknown whether brain network reorganization in NREM sleep stages N2 and N3 can uniquely be mapped onto individual differences in cognitive performance after a recovery nap following sleep deprivation. Using resting state functional magnetic resonance imaging (fMRI), we quantified the integration and segregation of brain networks during NREM sleep stages N2 and N3 while participants took a 1-hour nap following 24-hour sleep deprivation, compared to well-rested wakefulness. Here, we advance a new analytic framework called the hierarchical segregation index (HSI) to quantify network segregation across spatial scales, from whole-brain to the voxel level, by identifying spatio-temporally overlapping large-scale networks and the corresponding voxel-to-region hierarchy. Our results show that network segregation increased in the default mode, dorsal attention and somatomotor networks during NREM sleep compared to wakefulness. Segregation within the visual, limbic, and executive control networks exhibited N2 versus N3 sleep-specific voxel-level patterns. More segregation during N3 was associated with worse recovery of working memory, executive attention, and psychomotor vigilance after the nap. The level of spatial resolution of network segregation varied among brain regions and was associated with the recovery of performance in distinct cognitive tasks. We demonstrated the sensitivity and reliability of voxel-level HSI to provide key insights into within-region variation, suggesting a mechanistic understanding of how NREM sleep replenishes cognition after sleep deprivation.
睡眠剥夺后认知能力下降，随后睡眠恢复，这表明睡眠尤其是非快速眼动（NREM）睡眠在维持认知方面起着关键作用。目前尚不清楚NREM睡眠N2和N3阶段的脑网络重组是否能独特地映射到睡眠剥夺后小睡恢复后的个体认知表现差异上。我们使用静息态功能磁共振成像（fMRI），在参与者24小时睡眠剥夺后进行1小时小睡期间，对NREM睡眠N2和N3阶段的脑网络整合和分离进行了量化，并与充分休息的清醒状态进行了比较。在这里，我们提出了一种新的分析框架，称为层次分离指数（HSI），通过识别时空重叠的大规模网络和相应的体素到区域层次结构，在从全脑到体素水平的空间尺度上量化网络分离。我们的结果表明，与清醒状态相比，NREM睡眠期间默认模式、背侧注意和躯体运动网络的网络分离增加。视觉、边缘和执行控制网络内的分离表现出N2与N3睡眠特定的体素水平模式。N3期间更多的分离与小睡后工作记忆、执行注意力和心理运动警觉性的较差恢复相关。网络分离的空间分辨率水平在不同脑区有所不同，并且与不同认知任务中表现的恢复相关。我们证明了体素水平HSI的敏感性和可靠性，为区域内变异提供了关键见解，这表明对NREM睡眠如何在睡眠剥夺后补充认知有了一种机制性理解。
