Department of Psychology, Columbia University New York, NY, USA.
Front Integr Neurosci. 2012 Jul 19;6:44. doi: 10.3389/fnint.2012.00044. eCollection 2012.
Mounting experimental and theoretical results indicate that neural systems are poised near a critical state. In human subjects, however, most evidence comes from functional MRI studies, an indirect measurement of neuronal activity with poor temporal resolution. Electrocorticography (ECoG) provides a unique window into human brain activity: each electrode records, with high temporal resolution, the activity resulting from the sum of the local field potentials of ∼10(5) neurons. We show that the human brain ECoG recordings display features of self-regulated dynamical criticality: dynamical modes of activation drift around the critical stability threshold, moving in and out of the unstable region and equilibrating the global dynamical state at a very fast time scale. Moreover, the analysis also reveals differences between the resting state and a motor task, associated with increased stability of a fraction of the dynamical modes.
越来越多的实验和理论研究表明，神经网络处于一种临界状态。然而，在人类研究中，大多数证据来自功能磁共振成像研究，这是一种对神经元活动的间接测量，时间分辨率较差。脑电描记术（ECoG）为人类大脑活动提供了一个独特的窗口：每个电极以高时间分辨率记录来自约 10^5 个神经元的局部场电位的总和的活动。我们表明，人类大脑 ECoG 记录显示出自调节动力学临界的特征：激活的动力学模式在临界稳定性阈值周围漂移，在不稳定区域内外移动，并在非常快的时间尺度上使全局动力学状态达到平衡。此外，分析还揭示了静息状态和运动任务之间的差异，这与一部分动力学模式的稳定性增加有关。
