Brain and Vision Research Laboratory, Department of Biomedical Engineering, Boston University Boston, MA, USA.
Front Neuroinform. 2013 Feb 15;7:3. doi: 10.3389/fninf.2013.00003. eCollection 2013.
Human perception, cognition, and action are supported by a complex network of interconnected brain regions. There is an increasing interest in measuring and characterizing these networks as a function of time and frequency, and inter-areal phase locking is often used to reveal these networks. This measure assesses the consistency of phase angles between the electrophysiological activity in two areas at a specific time and frequency. Non-invasively, the signals from which phase locking is computed can be measured with magnetoencephalography (MEG) and electroencephalography (EEG). However, due to the lack of spatial specificity of reconstructed source signals in MEG and EEG, inter-areal phase locking may be confounded by false positives resulting from crosstalk. Traditional phase locking estimates assume that no phase locking exists when the distribution of phase angles is uniform. However, this conjecture is not true when crosstalk is present. We propose a novel method to improve the reliability of the phase-locking measure by sampling phase angles from a baseline, such as from a prestimulus period or from resting-state data, and by contrasting this distribution against one observed during the time period of interest.
人类的感知、认知和行动是由相互关联的大脑区域组成的复杂网络支持的。人们越来越感兴趣的是测量和描述这些网络作为时间和频率的函数，而区域间相位锁定通常用于揭示这些网络。该测量方法评估了在特定时间和频率下两个区域的脑电活动相位角的一致性。非侵入性地，可以使用脑磁图（MEG）和脑电图（EEG）测量计算相位锁定的信号。然而，由于 MEG 和 EEG 中重建源信号的空间特异性缺乏，区域间相位锁定可能会因串扰而导致假阳性。传统的相位锁定估计假设当相位角分布均匀时不存在相位锁定。然而，当存在串扰时，这种假设并不成立。我们提出了一种新的方法，通过从基线（例如，在刺激前期间或静息状态数据期间）采样相位角，并将该分布与在感兴趣的时间段内观察到的分布进行对比，来提高相位锁定测量的可靠性。
