Jirsa V K, Sporns O, Breakspear M, Deco G, McIntosh A R
Theoretical Neuroscience Group, CNRS UMR 6233, Université de la Méditerranée, Institute of Movement Sciences, Faculté des Sciences du Sport, Marseille, France.
Arch Ital Biol. 2010 Sep;148(3):189-205.
Neurocomputational models of large-scale brain dynamics utilizing realistic connectivity matrices have advanced our understanding of the operational network principles in the brain. In particular, spontaneous or resting state activity has been studied on various scales of spatial and temporal organization including those that relate to physiological, encephalographic and hemodynamic data. In this article we focus on the brain from the perspective of a dynamic network and discuss the role of its network constituents in shaping brain dynamics. These constituents include the brain's structural connectivity, the population dynamics of its network nodes and the time delays involved in signal transmission. In addition, no discussion of brain dynamics would be complete without considering noise and stochastic effects. In fact, there is mounting evidence that the interaction between noise and dynamics plays an important functional role in shaping key brain processes. In particular, we discuss a unifying theoretical framework that explains how structured spatio-temporal resting state patterns emerge from noise driven explorations of unstable or stable oscillatory states. Embracing this perspective, we explore the consequences of network manipulations to understand some of the brain's dysfunctions, as well as network effects that offer new insights into routes towards therapy, recovery and brain repair. These collective insights will be at the core of a new computational environment, the Virtual Brain, which will allow flexible incorporation of empirical data constraining the brain models to integrate, unify and predict network responses to incipient pathological processes.
利用逼真的连接矩阵的大规模脑动力学神经计算模型，加深了我们对大脑中操作网络原理的理解。特别是，已经在包括与生理、脑电图和血液动力学数据相关的各种空间和时间组织尺度上研究了自发或静息状态活动。在本文中，我们从动态网络的角度关注大脑，并讨论其网络组成部分在塑造脑动力学中的作用。这些组成部分包括大脑的结构连接性、其网络节点的群体动力学以及信号传输中涉及的时间延迟。此外，不考虑噪声和随机效应，对脑动力学的讨论就不完整。事实上，越来越多的证据表明，噪声与动力学之间的相互作用在塑造关键脑过程中起着重要的功能作用。特别是，我们讨论了一个统一的理论框架，该框架解释了结构化的时空静息状态模式是如何从噪声驱动的对不稳定或稳定振荡状态的探索中出现的。从这个角度出发，我们探索网络操纵的后果，以理解大脑的一些功能障碍，以及为治疗、恢复和脑修复途径提供新见解的网络效应。这些共同的见解将成为一个新的计算环境——虚拟大脑的核心，它将允许灵活纳入限制脑模型的经验数据，以整合、统一和预测对初始病理过程的网络反应。
