1 Graduate Program in Neuroscience, Center for Neurotechnology, University of Washington, Seattle, Washington.
2 Integrated Brain Imaging Center, Department of Radiology, University of Washington, Seattle, Washington.
Brain Connect. 2019 Jul;9(6):488-499. doi: 10.1089/brain.2018.0596. Epub 2019 May 21.
Prior studies using functional magnetic resonance imaging, electroencephalography, and magnetoencephalography have observed both structured patterns in resting-state functional connectivity and spontaneous longitudinal variation in connectivity patterns independent of a task. In this first study using electrocorticography (ECoG), we characterized spontaneous, intersession variation in resting-state functional connectivity not linked to a task. We evaluated pairwise connectivity between electrodes using three measures (phase locking value [PLV], amplitude correlation, and coherence) for six canonical frequency bands, capturing different characteristics of time-evolving signals. We grouped electrodes into 10 functional regions and used intraclass correlation (ICC) to estimate pairwise longitudinal stability. We found that stronger PLV (PLV ≥0.4) in theta through gamma bands and strong correlation in all bands ('s ≥0.6) are linked to substantial stability (ICC ≥0.6), but that stability does not imply strong phase locking or amplitude correlation. There was no notable link between strong coherence and high ICC. All within-region PLVs are markedly stable across frequencies. In addition, we highlight interaction patterns across several regions: parahippocampal/entorhinal cortex is characterized by stable, weak functional connectivity except self-connections. Dorsolateral prefrontal cortex connectivity is weak and unstable, except self-connections. Inferior parietal lobule has little stability despite narrow connectivity bounds. We confirm prior studies linking functional connectivity strength and intersession variability, extending into higher frequencies than other modalities, with greater spatial specificity than scalp electrophysiology. We suggest further studies quantitatively compare ECoG to other modalities and/or use these findings as a baseline to capture functional connectivity and dynamics linked to perturbations with a task or disease state.
先前的研究使用功能磁共振成像、脑电图和脑磁图观察到静息状态功能连接的结构化模式，以及与任务无关的连接模式的自发纵向变化。在这项首次使用皮层电图（ECoG）的研究中，我们描述了与任务无关的静息状态功能连接的自发、会话间变化。我们使用三个度量（锁相值[PLV]、振幅相关和相干性）评估电极之间的成对连接，涵盖了时间演变信号的不同特征。我们将电极分为 10 个功能区域，并使用组内相关系数（ICC）估计成对的纵向稳定性。我们发现，θ到γ频段的较强 PLV（PLV≥0.4）和所有频段的强相关性（s≥0.6）与实质性稳定性（ICC≥0.6）相关，但稳定性并不意味着强相位锁定或振幅相关性。强相干性与高 ICC 之间没有明显的联系。所有频段内的 PLV 都在频率间表现出明显的稳定性。此外，我们强调了几个区域之间的交互模式：海马旁/内嗅皮层的特征是除自连接外稳定的弱功能连接。背外侧前额叶皮层的连接较弱且不稳定，除自连接外。下顶叶尽管连接范围狭窄，但稳定性较差。我们证实了先前的研究将功能连接强度和会话间变异性联系起来的研究，扩展到比其他模态更高的频率，具有比头皮电生理更高的空间特异性。我们建议进一步的研究定量比较 ECoG 与其他模态，或使用这些发现作为基线，以捕捉与任务或疾病状态相关的功能连接和动力学变化。
