Miller Stephanie R, Yu Shan, Pajevic Sinisa, Plenz Dietmar
Section on Critical Brain Dynamics, National Institute of Mental Health, Bethesda, MD, USA.
Section on Quantitative Imaging and Tissue Sciences, National Institute of Child Health and Development, NIH, Bethesda, MD, USA.
Netw Neurosci. 2021 Jun 3;5(2):505-526. doi: 10.1162/netn_a_00188. eCollection 2021.
Ongoing neuronal activity in the brain establishes functional networks that reflect normal and pathological brain function. Most estimates of these functional networks suffer from low spatiotemporal resolution and indirect measures of neuronal population activity, limiting the accuracy and reliability in their reconstruction over time. Here, we studied the stability of neuronal avalanche dynamics and corresponding reconstructed functional networks in the adult brain. Using chronically implanted high-density microelectrode arrays, the local field potential (LFP) of resting-state activity was recorded in prefrontal and premotor cortex of awake nonhuman primates. Avalanche dynamics revealed stable scaling exhibiting an inverted parabolic profile and collapse exponent of 2 in line with a critical branching process over many days and weeks. Functional networks were based on a Bayesian-derived estimator and demonstrated stable integrative properties characterized by nontrivial high neighborhood overlap between strongly connected nodes and robustness to weak-link pruning. Entropy-based mixing analysis revealed significant changes in strong link weights over weeks. The long-term stability in avalanche scaling and integrative network organization in the face of individual link weight changes should support the development of noninvasive biomarkers to characterize normal and abnormal brain states in the adult brain.
大脑中持续的神经元活动建立了反映正常和病理脑功能的功能网络。对这些功能网络的大多数估计都存在时空分辨率低和神经元群体活动间接测量的问题，限制了它们随时间重建的准确性和可靠性。在这里，我们研究了成年大脑中神经元雪崩动力学和相应重建功能网络的稳定性。使用长期植入的高密度微电极阵列，在清醒的非人类灵长类动物的前额叶和运动前皮层记录静息态活动的局部场电位（LFP）。雪崩动力学揭示了稳定的标度，呈现出倒抛物线轮廓，崩塌指数为2，这与许多天和几周内的临界分支过程一致。功能网络基于贝叶斯推导的估计器，并表现出稳定的整合特性，其特征是强连接节点之间具有非平凡的高邻域重叠，并且对弱连接修剪具有鲁棒性。基于熵的混合分析揭示了数周内强连接权重的显著变化。面对单个连接权重变化时，雪崩标度和整合网络组织的长期稳定性应支持开发非侵入性生物标志物，以表征成年大脑中的正常和异常脑状态。
