Antonova Elena, Holding Martin, Suen Ho Chak, Sumich Alex, Maex Reinoud, Nehaniv Chrystopher
Division of Psychology, Department of Life Sciences, College of Health, Medicine and Life Sciences & Centre for Cognitive Neuroscience, Brunel University London, Kingston Lane, Uxbridge, UB8 3PH, United Kingdom.
Department of Psychology, Institute of Psychiatry Psychology & Neuroscience, King's College London, De Crespigny Park, London, SE5 8AF, United Kingdom.
Neuroimage Rep. 2022 Mar 12;2(2):100089. doi: 10.1016/j.ynirp.2022.100089. eCollection 2022 Jun.
EEG microstates, reflecting discrete topographical organization of the EEG signal power, may have clinical relevance; however, their functional significance and test-retest reliability remain unclear. To investigate the functional significance of the canonical EEG microstate classes and their pairwise transitions, and to establish their within-session test-retest reliability, we recorded 36-channel EEGs in 20 healthy volunteers during three eyes-closed conditions: mind-wandering, verbalization (silently repeating the word 'square' every 2 s), and visualization (visualizing a square every 2 s). Each condition lasted 3 min and the sequence of three conditions was repeated four times (two runs of two sequence repetitions). The participants' alertness and their sense of effort during the experiment were rated using visual-analogue scales. The EEG data were 2-20 Hz bandpass-filtered and analysed into the four canonical microstate classes: A, B, C, and D. EEG microstate classes C and D were persistently more dominant than classes A and B in all conditions. Of the first-order microstate parameters, average microstate duration was most reliable. The duration of class D microstate was longer during mind-wandering (106.8 ms) than verbalization (102.2 ms) or visualization (99.8 ms), with a concomitantly higher coverage (36.4% vs. 34.7% and 35.2%), but otherwise there was no clear association of the four microstate classes to particular mental states. The test-retest reliability was higher at the beginning of each run, together with higher average alpha power and subjective ratings of alertness. Only the transitions between classes C and D (from C to D in particular) were significantly higher than what would be expected from the respective microstates' occurrences. The transition probabilities, however, did not distinguish between conditions, and their test-retest reliability was overall lower than that of the first-order parameters such as duration and coverage. Further studies are needed to establish the functional significance of the canonical EEG microstate classes. This might be more fruitfully achieved by looking at their complex syntax beyond pairwise transitions. To ensure greater test-retest reliability of microstate parameters, study designs should allow for shorter experimental runs with regular breaks, particularly when using EEG microstates as clinical biomarkers.
脑电图微状态反映了脑电图信号功率的离散地形组织，可能具有临床相关性；然而，它们的功能意义和重测可靠性仍不清楚。为了研究典型脑电图微状态类别及其两两转换的功能意义，并确定它们在同一时间段内的重测可靠性，我们在20名健康志愿者闭眼的三种状态下记录了36导脑电图：走神、言语化（每2秒默念单词“正方形”）和视觉化（每2秒想象一个正方形）。每种状态持续3分钟，三种状态的顺序重复四次（两轮两次顺序重复）。使用视觉模拟量表对参与者在实验过程中的警觉性和努力感进行评分。脑电图数据经过2 - 20赫兹带通滤波，并分析为四个典型微状态类别：A、B、C和D。在所有状态下，脑电图微状态类别C和D始终比类别A和B更占主导。在一阶微状态参数中，平均微状态持续时间最可靠。在走神期间，D类微状态的持续时间（106.8毫秒）比言语化（102.2毫秒）或视觉化（99.8毫秒）更长，同时覆盖率更高（分别为36.4%、34.7%和35.2%），但除此之外，这四个微状态类别与特定心理状态没有明显关联。在每次运行开始时重测可靠性更高，同时平均α波功率和警觉性主观评分也更高。只有C类和D类之间的转换（特别是从C到D）显著高于各自微状态出现频率所预期的值。然而，转换概率无法区分不同状态，并且它们的重测可靠性总体低于持续时间和覆盖率等一阶参数。需要进一步研究来确定典型脑电图微状态类别的功能意义。通过研究它们除两两转换之外的复杂句法可能会更有成效地实现这一点。为确保微状态参数有更高的重测可靠性，研究设计应允许进行更短的实验运行并定期休息，特别是在将脑电图微状态用作临床生物标志物时。
