Short-term and long-term test-retest reliability of memory, complexity, and randomness of EEG microstates sequence

Short-term and long-term test-retest reliability of memory, complexity, and randomness of EEG microstates sequence | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Short-term and long-term test-retest reliability of memory, complexity, and randomness of EEG microstates sequence Povilas Tarailis, Fiorenzo Artoni, Thomas Koenig, Christoph M. Michel, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5875634/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 09 Dec, 2025 Read the published version in Cognitive Neurodynamics → Version 1 posted 9 You are reading this latest preprint version Abstract EEG microstates sequence analysis gained a lot of attention in recent years and different sequence analysis methods have been applied to study microstates sequence randomness, complexity, speed, periodicity, and long-range memory. A few reliability studies reported somewhat consistent results of temporal parameters, yet no study so far addressed the within subject stability and reliability over time of different microstate sequence metrics. Here, we performed EEG microstate segmentation on data recorded from 60 healthy young adults and evaluated short-term (90 min), and long-term (30 days) reliability and agreement of EEG microstate sequence long-range memory as estimated with Hurst exponent, complexity as evaluated with two different Lempel-Ziv complexity algorithms, and its randomness as quantified with entropy and entropy rate. Our results showed mostly good short-term reliability across all 5 metrics (0.831 < ICC < 0.902), and moderate to good (0.651 < ICC < 0.793) long-term reliability. Reliability and agreement over time demonstrated in this work strongly suggests that microstates sequence dynamics is a stable trait of neural activity that can be utilised as a possible reliable neurophysiological biomarker. EEG microstates test-retest reliability sequence analysis complexity long-range memory randomness Figures Figure 1 Figure 2 Figure 3 Introduction With the increasing need and interest in potential easily accessible sensitive biomarkers of neuropsychiatric conditions, electroencephalography (EEG) has received extensive attention. One of the well-established method to analyse EEG in order to study large-scale brain cortical networks during resting state is EEG microstates (MS). The method uses multichannel approach to simultaneously quantify information from all channels; thus, the recorded oscillations are defined as the non-overlapping maps characterized by a unique spatial distribution (class) that allows to capture not only temporal but also spatial dynamics of the ongoing brain electrical activity (Khanna et al., 2015 ; Michel & Koenig, 2018 ; Tarailis et al., 2023 ). In MS analysis, the goal is to group EEG time samples into clusters using various algorithms (k-means, TAAHC, AAHC, PCA, ICA (Murray et al., 2008 ; Pascual-Marqui et al., 1995 ; Poulsen et al., 2018 ; von Wegner et al., 2018 ) so that the EEG samples that belong to the same class would have as similar topographical displays as possible. Resting state EEG MS approach gained a lot of attention in recent years as evidenced by increased number of clinical (de Bock et al., 2020 ; Deiber et al., 2023 ; Férat, Arns, et al., 2022 ; Hanoglu et al., 2022 ; Smailovic et al., 2019 ), task-related (Comsa et al., 2019 ; D’croz-Baron et al., 2021 ; Deolindo et al., 2021 ; Milz et al., 2016 ; Nazare & Tomescu, 2024 ; Zanesco et al., 2020 ), methodological (Férat, Seeber, et al., 2022; Haydock et al., 2025 ; Kalburgi et al., 2023 ; Koenig et al., 2023 ; Murphy et al., 2023 ), review and meta-analytical papers (Chivu et al., 2023 ; Das et al., 2022 ; Khanna et al., 2014 ; Michel & Koenig, 2018 ; Rieger et al., 2016 ; Schiller et al., 2023 ; Tarailis et al., 2023 ; Zanesco, 2023 ). Specifically, a handful of EEG MS studies suggested different microstate features as possible biomarkers for clinical and research settings (Chivu et al., 2023 ; Deiber et al., 2023 ; Rieger et al., 2016 ; Tarailis et al., 2023 ). With this increased interest in EEG microstate analysis, the question of within-subject reliability (stability of measurements taken from the same individuals across different points in time) is crucial. If extracted parameters are not reliable, conclusions drawn from them could cause misleading interpretations and even inaccurate diagnosis. In a classical assessment pipeline, EEG microstates are evaluated by its temporal parameters – duration (reflecting how long microstate class is present), occurrence (reflecting how often the microstate class is present), coverage (reflecting the proportion that microstate class takes in the recording). However, only eight studies up to date have addressed the question of test-retest reliability of EEG microstates with a focus on the temporal parameters (Antonova et al., 2022 ; Bagdasarov et al., 2024 ; Q. Guo et al., 2024 ; Khanna et al., 2014 ; Kleinert et al., 2023 ; J. Liu et al., 2020 ; Popov et al., 2023 ; Zhang et al., 2021 ) with somewhat consistent results. Khanna et al. ( 2014 ) reported high test-retest reliability (Cronbach's α of 0.7–0.9) of duration, occurrence, and coverage measures recorded approximately 48 hours apart, independent from clustering algorithm (k-means vs TAAHC) used and number of channels (30 vs 19 vs 8) included. Zhang et al. ( 2021 ) demonstrated high reliability of temporal parameters acquired with 91, 64 and 32 channels EEG. Reliability decreased using 19 and 8 channels EEG, suggesting that it might be sensitive to the low number of channels. Liu et al., ( 2020 ) showed that EEG recordings with a duration greater than 2 minutes display high short-term (23–25 hours) reliability (Intraclass correlation coefficient (ICC) > 0.6) for all temporal parameters of microstates. Similar results were reported by Guo et al., ( 2024 ) in healthy controls and Parkinson’s patients with high reliability of temporal parameters acquired from EEG longer than 3 minutes. Antonova et al. ( 2022 ) reported mixed short-term (approximately 5 to 10 minutes apart) reliability results for temporal parameters of microstate with ICC values ranging from as low as 0.3 to as high as 0.9. Popov et al., ( 2023 ) showed mixed short-term (7–9 days) reliability of duration, occurrence, coverage, global filed power (GFP) and global explained variance (GEV), outlining the decrease of reliability with subjects’ age, while Bagdasarov et al ( 2024 ) demonstrated high reliability in infants, interestingly showing that even microstate parameters extracted from 1 min. EEG show good to excellent reliability. Finally, Kleinert et al., ( 2023 ) reported good to excellent short-term (average interval between measures: 99 min) reliability and moderate to good long-term (average interval 63 days) reliability of temporal parameters obtained with different number of channels (64 and 30) and different clustering algorithms (k-means, AAHC). Overall, the abovementioned suggest that temporal parameters of microstates are reliable, i.e. stable across different points in time. However, the switch between different microstates in the EEG recording reflects important information. It was suggested that microstates represent short-lasting, rapidly switching global mental states, thus were nicknamed ‘atom of thoughts’ (Lehmann, 1990 ). In order to study transition patterns from one global state to the next one, in 1993, Wackermann (Wackermann et al., 1993 ) introduced microstate syntax method. The method is based on the first order Markov process, where the probability of occurrence of each microstate depends only on the one preceding it, i.e. a memoryless process. While many studies reported microstate syntax sensitivity to different task-related conditions (Bréchet et al., 2019 ; Jabès et al., 2021 ; Seitzman et al., 2017 ) personality traits (Du et al., 2022 ; P. Guo et al., 2020 ; Hu et al., 2021 ), levels of vigilance (Ke et al., 2021 ) and clinical conditions (Nishida et al., 2013 ; Vellante et al., 2020 ), the main limitation of this method is in its temporal scope. EEG microstate syntax considers only a single time step dynamics and assumes that transitions between microstates are stationary and the next state depends only on the one preceding it. It was suggested that quantification and evaluation of microstate sequences should extend beyond the first and even second-order Markovian transitions (Antonova et al., 2022 ; Artoni et al., 2023 ) in order to capture longer-range temporal dependencies and higher-order dynamics, thereby enabling a more comprehensive characterization of brain state transitions and their functional significance. Moreover, a few studies assessing the reliability of EEG microstate parameters reported that EEG syntax displays poor to moderate within-subjects consistency (Antonova et al., 2022 ; Kleinert et al., 2023 ). Contradictory to EEG MS syntax assumption, a handful of studies reported that MS sequences show non-Markovian, higher order temporal structure (Artoni et al., 2023 ; Gschwind et al., 2015 ; von Wegner et al., 2017 ), which is driven by the brain’s dominant frequency (Hermann et al., 2024 ; von Wegner et al., 2021 ; Wiemers et al., 2023 ). (Antonova et al., 2022 ; Bagdasarov et al., 2024 ; Q. Guo et al., 2024 ; Khanna et al., 2014 ; Kleinert et al., 2023 ; J. Liu et al., 2020 ; Popov et al., 2023 ; Zhang et al., 2021 )In our recent systematic review (Tarailis et al., 2023 ) we noted that due to advances in the field of data science, microstates sequence analysis gained a lot of attention and different sequence analysis methods have been applied to study microstate sequence randomness, speed, complexity, periodicity and long-range memory (Artoni et al., 2023 ; von Wegner et al., 2018 , 2023 ). Several sequence analysis methods are implemented in microstate analysis toolboxes for MATLAB (Tait & Zhang, 2022 ) and python (Férat, Scheltienne, et al., 2022 ; von Wegner & Laufs, 2018 ) and are freely available to researchers. Three studies already paid a great deal of attention to microstate sequences sensitivity to cluster algorithms, band pass filtering settings and electrode density. A work by (von Wegner et al., 2018 ) addressed the question of impact of the clustering algorithm selection (AAHC vs k-means vs k-mediods vs PCA vs ICA) on outcomes of sequence analysis methods (Hurst exponent, transition matrix mixing time, entropy, entropy rate and autoinformation function) and concluded that these parameters are not affected by the clustering algorithm. Another study by (Férat, Seeber, et al., 2022) showed that although the topographies are not affected, temporal order of microstate sequences is sensitive to the band-pass filtering settings (broadband vs narrowband). Zhang et al. ( 2021 ) reported stable entropy rate values for EEGs recorded using 91, 64, 32 and 19 channels, while this metric increased when 8 channels EEG was analysed. Yet, no study so far aimed to assess the test-retest within-subjects reliability of various sequence characteristics of EEG microstates. Thus, this work aims to evaluate how stable different outcomes reflecting short-range and long-range memory, complexity, sequence randomness and uncertainty of microstates across different points in time are. We performed EEG microstate segmentation on the publicly available EEG data recorded from 60 healthy young adults and evaluated short-term (after 90 min), and long-term (after 30 days) reliability on the outcomes. of the five microstate sequence analysis methods reflecting short-range and long-range memory, complexity, sequence randomness and uncertainty. Materials and methods Dataset Preprocessed EEG files were downloaded from public repository and are available at OpenNeuro (‘Dataset ds004148’; https://openneuro.org/datasets/ds004148/versions/1.0.1 ). EEG were obtained from 60 anonymized participants (M = 28, F = 32, mean age = 20.01 years and SD = ± 1.88, range between 18 and 28 years old). Detailed information about subjects and inclusion criteria are reported in the original work (Wang et al., 2022 ). Recording and preprocessing Detailed information about experimental design, data collection and preprocessing is described in (Wang et al., 2022 ). Briefly, EEG data was collected over three sessions, where session 1 (S1) and session 2 (S2) were held on the same day 90 minutes apart, and session 3 (S3) was scheduled 30 days later, matching the time of the day with the first recording (S1). Each session consisted of eyes open, eyes closed and three cognitive states, each lasting 5 minutes. In this study we used only eyes-closed EEG data. EEG data was recorded using 63 or 64 channels, two of these electrodes were used to record eye movements. FCz was used as the online reference, data was sampled at 500 Hz and impedance was kept below 5kΩ. To unify the channels from the different sizes of caps during EEG recordings, EEG was reconstructed to 61 channels, filtered between 0.3 and 45 Hz using Finite Impulse Response (FIR) filter. EEG recordings were visually inspected and problematic channels were rejected. Linear interpolation was used to reconstruct missing channels. EEG signal was segmented into 4-s epochs and then all bad epochs were manually screened for removal. Before the microstate segmentation, the reference was re-projected to average, data was downsampled to 250 Hz and bandpass filtered between 1 and 30 Hz using Butterworth filter of the second order. Microstate segmentation K-means clustering The microstate segmentation was performed using Cartool toolbox (Brunet et al., 2011 ). Cluster analysis was performed separately for each session and was performed in two stages. First, to speed up clustering and increase its reliability, individual EEGs were resampled into 28 random epochs of 7500 time frames. For each epoch, GFP was calculated as the spatial standard deviation and only topographical maps at GFP peaks were extracted and submitted to k-means clustering algorithm, ignoring their polarity. To identify the optimal number of microstate templates for each epoch, the number of clusters was set between 1 and 12 with 100 iterations for each number of clusters to maximize GEV. The spatial filter was applied to the topographical maps before cluster analysis to spatial smooth topographies and increase signal-to-noise ratio (Michel & Brunet, 2019 ). If topography failed to reach minimum correlation threshold with cluster’s centroid (< 0.5), it remained unassigned to any cluster. The optimal number of clusters for each epoch was determined by meta-criteria implemented in Cartool (see Cartool’s Reference Guide). At the second stage, 1680 sets (28 epochs x 60 subjects) of the most dominant topographies in each session of each epoch were concatenated and submitted to the second level k-means clustering, once again ignoring polarity and applying minimal spatial correlation threshold. Concatenated data was resampled into 100 random epochs of 1000 time frames. In this stage, the number of clusters ranged between 1 and 15 and segmentation was repeated 200 times for each number of clusters. The optimal number of group level clusters was also based on meta-criteria. Backfitting Based on meta-criteria, the optimal number of group level topographical maps for all three sessions was 5 (Fig. 2A). These maps were backfitted to the original individual EEGs using winner-takes-all approach, where spatial correlation was calculated between the group level maps and topography at each time frame of individual EEG. Polarity was ignored, and minimal correlation threshold was applied. Temporal smoothing (window size of 20 ms, weight smoothness factor of 10) was applied to ensure that the noise during low GFP did not artificially interrupt the temporal segments of stable topography and small segments, less or equal to 5 time frames (20 ms), were rejected, split in half with the first half added to the preceding segment and the second half added to the proceeding segment. This resulted in time series of microstates, where every data frame was labelled as one of the possible microstates. Although temporal parameters and syntax are out of scope of this paper, additionally we calculated the mean GEV, duration, time coverage, occurrence of each microstate and first order Markovian syntax between each pair of microstates (Lehmann et al., 2005 ; Wackermann et al., 1993 ). Sequence analysis Hurst exponent Hurst exponent is a well-established method to assess the degree of temporal dependence. It describes the long-range dependency and provides information about the persistent or anti-persistent patterns within the time series (Gschwind et al., 2015 ; Van De Ville et al., 2010 ). Hurst exponent values theoretically ranges from 0 to 1, where values H < 0.5 suggest short-term correlations when recent states influence the immediate future states but without a sustained trend, values of H = 0.5 indicates more random, less predictable sequence behavior, and values H > 0.5 imply that the temporal structure of the sequence is not random and past states influence future states. Thus, a higher value of Hurst exponent indicates that sequence is more predictable, repetitive and indicates the degree of long-range temporal autocorrelations present between shorter and longer timescales (Díaz & Córdova, 2021 ). A handful of different methods exist to estimate Hurst exponent (Gschwind et al., 2015 ; von Wegner et al., 2018 ); here we used detrended fluctuation analysis (DFA).We modified publicly available code implemented in + microstate toolbox (Tait & Zhang, 2022 ). Individual symbolic microstates sequences were embedded into random walk by parting microstates into two classes and associating each class with a negative and a positive step, respectively (Van De Ville et al., 2010 ). 50 logarithmically spaced time scales (S) between 0.2 and 30 seconds were used. Random walk sequences were parcellated into number of epochs according to different time scales. Each epoch was detrended (F) by subtracting the local trend using least-squares fit (Jia et al., 2021 ; Ros et al., 2017 ). The Hurst parameter of the random walk corresponds to the slope between F and S on the log-log coordinates plane. Since from biological perspective there is no good basis how to partition microstates, we followed method explained in literature (Jia et al., 2021 ; von Wegner et al., 2023 ) for each subject, Hurst exponent was computed for every possible partition and then averaged across all partitions. Lempel-Ziv complexity Lempel-Ziv complexity (LZC) quantifies the complexity or compressibility of a sequence by identifying repeated subsequences required to explain the full sequence and does not rely on assumptions of Markovianity (Tait et al., 2020 ). To compute LZC we used publicly available code (Tait & Zhang, 2022 ).Microstate sequence with low LZC indicates regular, repetitive, small number of transitioning patterns within the sequence, whilst high LZC suggests irregular, complex microstates sequences. Individual microstate sequences were transformed into a “Jump” sequence by setting each microstate activity to the size of 1 time frame, which allowed focusing only on the temporal sequence of the microstates, independently of their individual durations. LZC starts with scanning the sequence from left to right and finding repeating patterns in the sequence. Complexity refers to the number of unique patterns in the entire sequence. Since LZC increases with the length of the sequence, to account for the length differences between individual sequences, we normalized LZC by the length of the “Jump” sequence. In order to test optimized version, the Microsynt approach to LZC calculation, we used the LZMA2 algorithm for lossless data compression with 64MB dictionary, BT4 MatchFinder, and BCJ2 Filter (Artoni et al., 2023 ) The LZC for each ‘Jump’ sequence was calculated via a sliding window approach (1000 microstates windows length, 80% window overlap) and averaged across windows. The sliding-window approach offers a solution to the issue of comparing complexity of sequences with different lengths. Rather than truncating all sequences to the length of shortest one, as in Tait et al. ( 2020 ), sequences with different lengths can be compared without loss of data as the window size remains the same (Artoni et al., 2022 ). For more details about method see (Artoni et al., 2023 ). Entropy-related measures Entropy measures uncertainty or randomness of the microstate sequence (von Wegner et al., 2018 ). The number of times a specific microstate occurs in the sequence gives the probability distribution of microstates occurrence(von Wegner et al., 2017 , 2018 ). The shape of this distribution can be characterized by its entropy (H): $$\:H\left(x\right)=-\sum\:_{k=1}^{\text{max}k}{p}_{k}\bullet\:{\text{log}}_{2}{p}_{k}\:\:\:\:\left(1\right)$$ where p is the probability of the microstate k occurrence in the sequence x. The sequence’s entropy is bounded between 0 – entire sequence is dominated by one single microstate, to log 2 ( max k ) – all microstates have the same occurrence probability. This is schematically represented in (Fig. 1 ). Unfortunately, entropy ignores the temporal structure of the microstate sequence, and a randomly shuffled sequence will result in exactly the same values. Entropy rate is an extension of Shannon’s entropy that can help to describe temporal structure of the sequence. It measures how much uncertainty or randomness per symbol in a sequence, given knowledge about the past states of the sequence (von Wegner et al., 2018 ). It measures the entropy (H) about the next symbol in a sequence ( \(\:{x}_{t+1})\) , given knowledge about past states \(\:{{x}_{t}}^{\left(n\right)}\) : $$\:{h}_{1}=H\left({x}_{t+1}\right|{{x}_{t}}^{\left(n\right)}\left)\:\:\:\:\:\right(2)$$ Since entropy rate becomes less reliable with longer memory ( n ), we limited memory to maximum 7 steps, as it is recommended in the literature (von Wegner et al., 2018 ; Wiemers et al., 2023 ; Zhang et al., 2021 ). For both entropy and entropy rate metrics, low values indicate less complex, more stable sequences, while higher values suggest more random processes. Statistical analysis To test short term (90 minutes) and long term (30 days) test-retest reliability of each extracted parameter, we calculated intraclass correlation coefficient (ICC) (Shrout & Fleiss, 1979 ) implemented in JASP statistical software (Version 0.17.3) (Love et al., 2019 ). ICC is a correlation coefficient that assesses the consistency between measures. ICC is similar to the Pearson’s correlation coefficient; it requires a linear relationship between the variables. The main difference is that ICC also takes into account differences in the means of the measures being considered (J. Liu et al., 2016 ). ICC values are bounded between 0 and 1, where, following the guidelines by (Koo & Li, 2016 ) based on 95% confidence interval (CI), values less than 0.5 indicate a poor, values between 0.5 and 0.74 indicate moderate, values between 0.75 and 0.89 indicate good, and values between 0.9 and 1 indicate excellent reliability. As a complementary analysis, we performed one-way ANOVA for each parameter separately (McKeown et al., 2023 ) to test if there are any significant differences between parameters obtained from S1, S2 and S3. In case of significant ANOVA result, post-hoc paired sample t-test was performed with a Bonferroni correction to control type-I error. Bland-Altman (BA) was used as a simple method to visualize and evaluate a bias between the mean differences, and to estimate 95% limits of agreement (LOA) (Bland & Altman, 1986 ; Giavarina, 2015 ). Bias is calculated as the mean difference between two measurements (in this study – sessions) and LOA are estimated as ± 1.96 standard deviations of the mean difference (bias) (Bland & Altman, 1986 ). An ideal agreement between two measurements is zero and the smaller the range between LOA, the better the agreement is (Gerke, 2020 ; Myles & Cui, 2007 ). Since evaluated parameters have different ranges, instead of default approach, the difference for each subject was expressed as percentage of the observation represented on the X axis [(Differences between sessions / mean between sessions)*100] (Giavarina, 2015 ). A custom written MATLAB code was used to graphically plot the values between two sessions. Results Based on the meta-criteria, the optimal number of clusters for all 3 sessions was 5. The normalized values of each cluster evaluation parameter across all 3 sessions are presented in Fig. 2A. All five microstates’ maps for each session visually corresponded very well with topographies reported in the literature (Tarailis et al., 2023 ) and meta-microstates (Koenig et al., 2023 ). Topographical maps were labelled according to the updated labelling system (Tarailis et al., 2023 ) where topography with right frontal to left posterior configuration is labelled as MS A, topography with left frontal to right posterior – MS B, topography with frontal to occipital configuration – MS C, topography with fronto-central configuration – MS D and topography with posterior configuration – MS E. Topographies and shared spatial variance are presented in Fig. 2B and Fig. 2C. Figure 2. A) Assessment of extracted microstates with different criterion implemented in Cartool. The best solution for each criterion is colored green. B) Group level topographies for each session. C). Shared spatial variance between extracted topographies. Temporal parameters of duration and occurrence fell in line with 95% prediction intervals (Zanesco, 2023 ) while mean coverages of MS C were higher for all three sessions and average time coverages for MS D were lower for all three sessions than expected. Since the stability of spatiotemporal parameters and EEG syntax are out of scope of this study, descriptive statistics and ICC values for short-term and long-term reliability of spatiotemporal parameters and EEG microstates syntax are presented in supplementary material and will not be discussed here. Short-term and long-term reliability Descriptive statistics of Hurst exponent, nLZC, Microsynt, entropy and entropy rate for each session are summarized in the supplementary material (S. Table 1 ). Hurst exponent Mean values and standard deviations of the Hurst exponent for each session were as follows: 0.646 ± 0.040 for S1, 0.653 ± 0.039 for S2 and 0.643 ± 0.043 for S3. ANOVA did not reveal any significant differences between the sessions [F(2,177) = 0.89, p = 0.411]. ICC value for short-term reliability of Hurst exponent showed good consistency and bias was low. For long-term reliability, ICC value was in the moderate reliability range and the bias was once again close to zero, indicating high agreement. Short-term and long-term ICC values for reliability and agreement are summarized in Table 1 . Bias and LOA are visualized in Fig. 3 A. For the interpretation of BA plots refer to Statistical analysis section. Table 1 Intraclass correlation coefficient values with lower and upper 95% CI and bias with Limits of Agreement (LOA) for Hurst exponent, normalized Lempel-Ziv complexity (nLZC), Microsynt, Shannon‘s entropy and entropy rate. Short-term Long-term Reliability Point estimate Lower 95% CI Upper 95% CI Point estimate Lower 95% CI Upper 95% CI Hurst 0.843 0.744 0.905 0.670 0.504 0.789 nLZC 0.831 0.733 0.896 0.651 0.447 0.776 Microsynt 0.896 0.714 0.952 0.793 0.676 0.871 Entropy 0.902 0.841 0.940 0.750 0.613 0.843 Entropy rate 0.882 0.675 0.946 0.769 0.642 0.855 Agreement Bias Lower LOA Upper LOA Bias Lower LOA Upper LOA Hurst -0.994 -7.273 5.286 0.560 -9.254 10.374 nLZC 0.768 -4.024 5.459 1.270 -6.251 8.791 Microsynt 1.122 -2.191 4.435 0.541 -5.329 6.412 Entropy 1.724 -5.143 8.591 2.307 -10.343 14.959 Entropy rate 2.888 -5.479 11.254 1.629 -13.479 16.737 Normalized Lempel-Ziv complexity Mean values and standard deviations of the normalized LZ complexity for each session were as follows: 0.154 ± 0.006 for S2, 0.153 ± 0.007 for S2 and 0.152 ± 0.009 for S3. ANOVA did not show significant differences of nLZC between sessions [F(2,177) = 1.15, p = 0.320]. ICC value for short-term reliability of normalized LZC showed good consistency with high agreement (Table 1 ). For long-term reliability, ICC value was in the moderate reliability range and the agreement was high (low bias). Short-term and long-term ICC values for reliability and agreement are summarized in Table 1 . Bias and LOA for nLZC are visualized in Fig. 3 B. Microsynt Mean values and standard deviations of the sequence complexity as measured with Microsynt method for each session were as follows: 258 ± 10.76 for S1, 255.92 ± 11.34 for S2 and 257.46 ± 12.33 for S3. No significant differences were obtained between sessions for Microsynt [F(2,177) = 0.93, p = 0.397]. ICC value for short-term reliability indicated good consistency and high agreement. For long-term reliability, ICC value was in the moderate reliability range and agreement was close to zero (Table 1 ). Bias and LOA for complexity evaluated with Microsynt approach are visualized in Fig. 3 C. Entropy Mean values and standard deviations of entropy for each session were as follows: 2.097 ± 0.148 for S1, 2.063 ± 0.164 for S2 and 2.053 ± 0.189 for S3. Again. ANOVA did not reveal any differences of the measure between the sessions [F(2,177) = 1.13, p = 0.325]. ICC value for short-term reliability of entropy was just above excellent consistency threshold and the bias was close to zero (high agreement). For long-term reliability, ICC value dropped to the good reliability range, while agreement stayed high (Table 1 ). Entropy bias and LOA with 95% CI are visualized in Fig. 3 D. Entropy rate Mean values and standard deviations of the entropy rate for each session were as follows: 0.417 ± 0.040 for S1, 0.406 ± 0.042 for S2 and 0.411 ± 0.046 for S3. As indicated by ANOVA, no differences in entropy rates were observed entropy rate [F(2,177) = 1.1, p = 0.337]. ICC value for short-term reliability of entropy rate showed good consistency with high agreement. For long-term reliability, ICC value was in the good reliability range and agreement was high (Table 1 ). Bias and LOA for entropy rate are visualized in Fig. 3 E. Discussion The aim of EEG microstate analysis is to describe the ongoing EEG patterns by a small set of representative templates. In this case the EEG can be seen as a process with a finite set of discrete states which evolves in continuous time. With the increasing interest in microstate approach as a tool to quantify resting state EEG in both normal and pathological conditions, a number of papers addressing the methodological questions regarding the topographical stability of EEG microstates (Zanesco, 2020 ), algorithm selection (Khanna et al., 2014 ; von Wegner et al., 2018 ), potential bias in group level analysis (Murphy et al., 2023 ), the impact of band-pass filtering settings (Férat, Seeber, et al., 2022) and electrode density (Zhang et al., 2021 ), relationship with EEG spectral amplitude (Zulliger et al., 2022 ), non-uniform labelling system (Custo et al., 2017 ; Tarailis et al., 2021 , 2023 ) the impact of temporal smoothing (Hermann et al., 2024 ) and ICA (Artoni & Michel, 2025 ) were addressed. However, only eight studies (Antonova et al., 2022 ; Bagdasarov et al., 2024 ; Q. Guo et al., 2024 ; Khanna et al., 2014 ; Kleinert et al., 2023 ; J. Liu et al., 2020 ; Popov et al., 2023 ; Zhang et al., 2021 ) specifically focused on the reliability of temporal parameters. The knowledge on the reliability of parameters evaluate is crucial for further implementation of measures as potential personalized biomarkers. Notably, alongside spatiotemporal parameters, microstate sequence can be evaluated, as it is believed that the order and transition between different microstates reflect dynamics necessary for cognitive functions (Eqlimi et al., 2023 ; Haydock et al., 2025 ; Jia et al., 2021 ; Lehmann et al., 2005 ; Wackermann et al., 1993 ). It was suggested that EEG syntax is controlled by a probabilistic mechanism where transitions from one microstate to another microstate occur with unequal probabilities, and that the current microstate class and the internally or externally received information co-determine the next microstate occurrence (Lehmann et al., 2005 ; Wackermann et al., 1993 ). A handful of studies have reported altered transition probabilities in different clinical conditions (Nishida et al., 2013 ; Vellante et al., 2020 ) and during different experimental settings (Bréchet et al., 2019 ; Seitzman et al., 2017 ). Unfortunately, transition probabilities have never been included in the meta-analyses. Moreover, studies assessing test-retest reliability of transitions reported mixed results on reliability estimates ranging from poor to moderate (Antonova et al., 2022 ; Kleinert et al., 2023 ), which raises a question: – is microstate syntax reliable? Importantly, instead of focusing on the first or even second order Markovian transition probabilities, a handful of studies applied different theoretical information parameters to study various aspects of sequence characteristics - periodical activity, complexity, speed and randomness - demonstrating that it displays non-random higher order temporal structure. However, the reliability of outcomes was never assessed. In this study, using data-driven approach to extract the optimal number of microstates, we for the first time assessed the short-term (90 minutes) and long-term (30 days) test-retest reliability of different sequence characteristics - long-range memory as evaluated with Hurst exponent, complexity as computed with two different Lempel-Ziv algorithms, and sequence randomness and uncertainty measured with entropy and entropy rate - that can be utilized to define microstate sequence dynamics and reflect physiologically-relevant aspects. Sequence memory and complexity We showed that sequence memory of EEG microstates as estimated using Hurst exponent displays good short-term (ICC = 0.843) and moderate long-term (ICC = 0.670) reliability, indicating that it could be reliably used in different experimental settings. This was further substantiated by the absence of significant differences between sessions. Hurst exponent is a well-established method to assess the self-similarity within time series. The time scale-free dynamics indicates that the system is being maximally flexible for adapting to different kinds of upcoming stimuli (Chialvo, 2010 ; Ros et al., 2017 ) due to efficient and flexible information flow between multiple sources (Avramiea et al., 2022 ; Kello et al., 2010 ; Shew & Plenz, 2012 ). In terms of functionality, it may be related to the structural memory of the activation of neural network that constrains perception and behavior (Lewis et al., 2009 ; Linkenkaer-Hansen et al., 2001 , 2004 ). Also, it may define the dynamical regime of conscious resting-state activity (Demertzi et al., 2019 ; Tagliazucchi et al., 2013 ). The long-range memory of EEG microstate sequence was previously estimated with a promising results in (pre-)clinical states (Lassi et al., 2023 ; Tomescu et al., 2015 ), task-related conditions (Jia et al., 2021 ) and different sleep stages (von Wegner et al., 2023 ), all demonstrating affected long-range temporal dependencies. Since sequence complexity increases with the length of the sequence, some sort of normalization is required to control the difference of sequence lengths between the subjects. Here, we normalized LZC values by the length of “Jump” sequence. Similarly to Hurst exponent, it displayed good short-term (ICC = 0.831) and moderate long-term (ICC = 0.651) stability and no differences between sessions were observed. In the case of Microsynt method, on the contrary, no normalization for window length was necessary, and the sliding window approach implemented in Microsynt method resulted in good short-term (ICC = 0.896) and long-term reliability (ICC = 0.793) with no differences between sessions. Importantly, LZC appears to be sensitive to clinical conditions and states of consciousness: studies reported decreased LZC in Alzheimer’s patients (Lassi et al., 2023 ; Tait et al., 2020 ) and decreased LZC in non-REM sleep stages (von Wegner et al., 2023 ) and propofol-induced loss of consciousness (Artoni et al., 2022 ). Sequence randomness and uncertainty Many different entropy measures to quantify different aspects of randomness and predictability of time series (Delgado-Bonal & Marshak, 2019 ; Keshmiri, 2020 ) exist and can be potentially used in the context of EEG microstates sequence analysis. For example, few studies reported sample entropy changes in early course of psychosis (Murphy et al., 2020 ), obsessive compulsive disorder (Ren et al., 2024 ), propofol induced loss of consciousness (Z. Liu et al., 2022 ), yet no differences between genders was observed (Niu et al., 2024 ). Recently, changes of excess entropy from wakefulness to NREM sleep stages were demonstrated (von Wegner et al., 2023 ). Here, we focused on Shannon’s entropy and entropy rate. Entropy showed excellent short-term reliability (ICC = 0.902), while entropy rate showed good short-term reliability (ICC = 0.882). Both measures displayed good reliability for long-term stability (ICC = 0.750 and 0.769, respectively; no significant differences between sessions). Although entropy did not gain a lot of attention in the field of microstates so far, however, it was reported to be sensitive to levels of sleepiness (Wiemers et al., 2023 ). On the contrary, entropy rate of microstates sequence was shown to be affected by cognitive tasks (Jia et al., 2021 ), levels of sleepiness (von Wegner et al., 2023 ; Wiemers et al., 2023 ) and propofol induced loss of consciousness (Hermann et al., 2024 ; Zhang et al., 2021 ). Considerations and limitations EEG data by its nature contains a lot of unwanted noise. The signal-to-noise ratio decreases even more in the spontaneous data, where noise cannot be averaged out like with event-related potentials data. Even after rejection of artefacts and artefactual ICA components there is still some noise left, which could last for a few time frames. To reduce noise, especially at low GFP moments or during polarity inversions, post-processing of temporal smoothing and/or rejection of small segments are usually applied resulting in longer, more stable, less frequently changing segments. However, it alters the microstate sequence to a certain extent. Two recent studies showed that a common practice of backfitting template maps only on GFP peaks or using only sequences where all the duplicate states are removed (‘Jump’ sequences), can destroy EEG microstate sequence periodicity, decrease its complexity and increase its randomness (Hermann et al., 2024 ; von Wegner et al., 2023 ). Although in this study we used all time frames for backfitting (not only GFP peaks) we also applied temporal smoothing that could potentially affect certain parameters, such as entropy rate and LZC. On the other hand, unsmoothed sequence results in shorter mean duration of microstates (~ 20 ms) which goes against the interpretation of EEG microstates being related with different cognitive and physiological functions and brain states (Michel & Koenig, 2018 ; Tarailis et al., 2023 ). This results in the double-edged sword situation, where authors have to choose: to apply temporal smoothing to reduce the noise, obtain temporal parameters that are similar to normative (Zanesco, 2023 ), but affect some aspects of sequence characteristics, or do not apply temporal smoothing, preserve certain aspects of sequence metrics, yet also preserve unwanted noise that results in short duration and high occurrence rate. The optimal approach remains an open question (Haydock et al., 2025 ). The temporal parameters provide certain information about specific microstate which has its own functional role and underlying sources, and changes of those parameters are more straightforward to interpret. In contrast, sequence analysis methods are not microstate specific. As a result, the functional roles of different parameters become more difficult to interpret as they reflect the overall network dynamics. Nevertheless, EEG microstate sequence analysis could be a new frontier in the field. Conclusion We showed that microstate sequence parameters of Hurst exponent, (Microsynt) Lempel-Ziv complexity, entropy and entropy rate display moderate and good short-term and long-term reliability with no differences between sessions, outperforming reliability estimates of transition probabilities that range from poor to moderate. Inter-class correlation values reported here strongly suggest that dynamics microstates sequence represents stable traits of neural activity and could serve as a possible neurophysiological biomarker. Thus, further application of these and other sequence analysis methods in both clinical and experimental studies is encouraged. Declarations Author Disclosure statement No competing financial interests exist. Funding No funding was used in support of this research. Author Contribution PT: conceptualization, formal analysis, methodology software, visualization, writing original draft, reviewing and editing. FA: formal analysis, methodology, reviewing and editing. TK: reviewing and editing. CMM: supervision, reviewing and editing. IG-B: conceptualization, supervision, reviewing and editing. Acknowledgement We would like to thank the original authors who gathered and made available the de-identified data on which this manuscript is based. We also thank Frederic von Wegner for his help computing entropy rate. Data Availability Data is provided within the manuscript. References Antonova, E., Holding, M., Suen, H. C., Sumich, A., Maex, R., & Nehaniv, C. (2022). EEG microstates: Functional significance and short-term test-retest reliability. Neuroimage: Reports , 2 (2), 100089. https://doi.org/10.1016/J.YNIRP.2022.100089 Artoni, F., Maillard, J., Britz, J., Brunet, D., Lysakowski, C., Tramèr, M. R., & Michel, C. M. (2023). Microsynt: Exploring the syntax of EEG microstates. NeuroImage , 277 , 120196. https://doi.org/10.1016/J.NEUROIMAGE.2023.120196 Artoni, F., Maillard, J., Britz, J., Seeber, M., Lysakowski, C., Bréchet, L., Tramèr, M. R., & Michel, C. M. (2022). EEG microstate dynamics indicate a U-shaped path to propofol-induced loss of consciousness. NeuroImage , 256 , 119156. https://doi.org/10.1016/J.NEUROIMAGE.2022.119156 Artoni, F., & Michel, C. M. (2025). How does Independent Component Analysis Preprocessing Affect EEG Microstates? Brain Topography , 38 (2), 26. https://doi.org/10.1007/S10548-024-01098-4 Avramiea, A. E., Masood, A., Mansvelder, H. D., & Linkenkaer-Hansen, K. (2022). Long-Range Amplitude Coupling Is Optimized for Brain Networks That Function at Criticality. Journal of Neuroscience , 42 (11), 2221–2233. https://doi.org/10.1523/JNEUROSCI.1095-21.2022 Bagdasarov, A., Brunet, D., Michel, C. M., & Gaffrey, M. S. (2024). Microstate Analysis of Continuous Infant EEG: Tutorial and Reliability. Brain Topography 2024 , 1–18. https://doi.org/10.1007/S10548-024-01043-5 Bland, J. M., & Altman, D. G. (1986). Statistical methods for assessing agreement between two methods of clinical measurement. Lancet , 1 (8476), 307–310. https://doi.org/doi.org/10.1016/S0140-6736(86)90837-8 Bréchet, L., Brunet, D., Birot, G., Gruetter, R., Michel, C. M., & Jorge, J. (2019). Capturing the spatiotemporal dynamics of self-generated, task-initiated thoughts with EEG and fMRI. NeuroImage , 194 , 82–92. https://doi.org/10.1016/j.neuroimage.2019.03.029 Brunet, D., Murray, M. M., & Michel, C. M. (2011). Spatiotemporal Analysis of Multichannel EEG: CARTOOL. Computational Intelligence and Neuroscience , 2011 , 15. https://doi.org/10.1155/2011/813870 Chialvo, D. R. (2010). Emergent complex neural dynamics. Nature Physics , 6 (10), 744–750. https://doi.org/10.1038/nphys1803 Chivu, A., Pascal, S. A., Damborská, A., & Tomescu, M. I. (2023). EEG Microstates in Mood and Anxiety Disorders: A Meta-analysis. Brain Topography . https://doi.org/10.1007/S10548-023-00999-0 Comsa, I. M., Bekinschtein, T. A., & Srivas Chennu, ·. (2019). Transient Topographical Dynamics of the Electroencephalogram Predict Brain Connectivity and Behavioural Responsiveness During Drowsiness. Brain Topography , 32 , 315–331. https://doi.org/10.1007/s10548-018-0689-9 Custo, A., Van De Ville, D., Wells, W. M., Tomescu, M. I., Brunet, D., & Michel, C. M. (2017). Electroencephalographic Resting-State Networks: Source Localization of Microstates. Brain Connectivity , 7 (10), 671–682. https://doi.org/10.1089/brain.2016.0476 Das, S., Zomorrodi, R., Enticott, P. G., Kirkovski, M., Blumberger, D. M., Rajji, T. K., & Desarkar, P. (2022). Resting state electroencephalography microstates in autism spectrum disorder: A mini-review. Frontiers in Psychiatry , 13 , 988939. https://doi.org/10.3389/FPSYT.2022.988939/BIBTEX D’croz-Baron, D. F., Bréchet, · Lucie, Baker, M., & Karp, T. (2021). Auditory and Visual Tasks Influence the Temporal Dynamics of EEG Microstates During Post-encoding Rest. Brain Topography , 34 , 19–28. https://doi.org/10.1007/s10548-020-00802-4 de Bock, R., Mackintosh, A. J., Maier, F., Borgwardt, S., Riecher-Rössler, A., & Andreou, C. (2020). EEG microstates as biomarker for psychosis in ultra-high-risk patients. Translational Psychiatry , 10 (1), 300. https://doi.org/10.1038/S41398-020-00963-7 Deiber, M. P., Piguet, C., Berchio, C., Michel, C. M., Perroud, N., & Ros, T. (2023). Resting-State EEG Microstates and Power Spectrum in Borderline Personality Disorder: A High-Density EEG Study. Brain Topography . https://doi.org/10.1007/s10548-023-01005-3 Delgado-Bonal, A., & Marshak, A. (2019). Approximate Entropy and Sample Entropy: A Comprehensive Tutorial. Entropy , 21 (6). https://doi.org/10.3390/E21060541 Demertzi, A., Tagliazucchi, E., Dehaene, S., Deco, G., Barttfeld, P., Raimondo, F., Martial, C., Fernández-Espejo, D., Rohaut, B., Voss, H. U., Schiff, N. D., Owen, A. M., Laureys, S., Naccache, L., & Sitt, J. D. (2019). Human consciousness is supported by dynamic complex patterns of brain signal coordination. Science Advances , 5 (2). https://doi.org/10.1126/SCIADV.AAT7603/SUPPL_FILE/AAT7603_SM.PDF Deolindo, C. S., Ribeiro, M. W., de Aratanha, M. A. A., Scarpari, J. R. S., Forster, C. H. Q., da Silva, R. G. A., Machado, B. S., Amaro Junior, E., König, T., & Kozasa, E. H. (2021). Microstates in complex and dynamical environments: Unraveling situational awareness in critical helicopter landing maneuvers. Human Brain Mapping , 42 (10), 3168–3181. https://doi.org/10.1002/hbm.25426 Díaz, H. A. M., & Córdova, F. (2021). On the meaning of Hurst entropy applied to EEG data series. Procedia Computer Science , 199 , 1385–1392. https://doi.org/10.1016/j.procs.2022.01.175 Du, M., Peng, Y., Li, Y., Zhu, Y., Yang, S., Li, J., Zou, F., Wang, Y., Wu, X., Zhang, Y., & Zhang, M. (2022). Effect of trait anxiety on cognitive flexibility: Evidence from event-related potentials and resting-state EEG. Biological Psychology , 170 , 108319. https://doi.org/10.1016/J.BIOPSYCHO.2022.108319 Eqlimi, E., Bockstael, A., Schönwiesner, M., Talsma, D., & Botteldooren, D. (2023). Time course of EEG complexity reflects attentional engagement during listening to speech in noise. European Journal of Neuroscience . https://doi.org/10.1111/ejn.16159 Férat, V., Arns, M., Deiber, M. P., Hasler, R., Perroud, N., Michel, C. M., & Ros, T. (2022). Electroencephalographic Microstates as Novel Functional Biomarkers for Adult Attention-Deficit/Hyperactivity Disorder. Biological Psychiatry: Cognitive Neuroscience and Neuroimaging , 7 (8), 814–823. https://doi.org/10.1016/J.BPSC.2021.11.006 Férat, V., Scheltienne, M., Brunet, D., Ros, T., & Michel, C. (2022). Pycrostates: a Python library to study EEG microstates. Journal of Open Source Software , 7 (78), 4564. https://doi.org/10.21105/joss.04564 Férat, V., Seeber, | Martin, Michel, C. M., & Ros, T. (2022). Beyond broadband: Towards a spectral decomposition of electroencephalography microstates. Human Brain Mapping . https://doi.org/10.1002/HBM.25834 Gerke, O. (2020). Reporting standards for a bland-altman agreement analysis: A review of methodological reviews. In Diagnostics (Vol. 10, Issue 5). MDPI. https://doi.org/10.3390/diagnostics10050334 Giavarina, D. (2015). Understanding Bland Altman analysis. Biochemia Medica , 25 (2), 141–151. https://doi.org/10.11613/BM.2015.015 Gschwind, M., Michel, C. M., & Van De Ville, D. (2015). Long-range dependencies make the difference-Comment on “A stochastic model for EEG microstate sequence analysis.” In NeuroImage (Vol. 117, pp. 449–455). Academic Press Inc. https://doi.org/10.1016/j.neuroimage.2015.05.062 Guo, P., Cui, J., Wang, Y., Zou, F., Wu, X., & Zhang, M. (2020). Spontaneous microstates related to effects of low socioeconomic status on neuroticism. Scientific Reports 2020 10:1 , 10 (1), 1–8. https://doi.org/10.1038/s41598-020-72590-7 Guo, Q., Liu, S., Wang, L., Feng, K., & Yang, S. (2024). Analysis of microstate features for Parkinson’s disease based on reliability validation. Journal of Neuroscience Methods , 406 , 110115. https://doi.org/10.1016/J.JNEUMETH.2024.110115 Hanoglu, L., Toplutas, E., Saricaoglu, M., Velioglu, H. A., Yildiz, S., & Yulug, B. (2022). Therapeutic Role of Repetitive Transcranial Magnetic Stimulation in Alzheimer’s and Parkinson’s Disease: Electroencephalography Microstate Correlates. Frontiers in Neuroscience , 16 . https://doi.org/10.3389/FNINS.2022.798558 Haydock, D., Kadir, S., Leech, R., Nehaniv, C. L., & Antonova, E. (2025). EEG Microstate Syntax Analysis: A Review of Methodological Challenges and Advances. NeuroImage , 121090. https://doi.org/10.1016/J.NEUROIMAGE.2025.121090 Hermann, G., Tödt, I., Tagliazucchi, E., Todtenhaupt, I. K., Laufs, H., & von Wegner, F. (2024). Propofol Reversibly Attenuates Short-Range Microstate Ordering and 20 Hz Microstate Oscillations. Brain Topography . https://doi.org/10.1007/s10548-023-01023-1 Hu, N., Long, Q., Li, Q., Hu, X., Li, Y., Zhang, S., Chen, A., Huo, R., Liu, J., & Wang, X. (2021). The modulation of salience and central executive networks by acute stress in healthy males: An EEG microstates study. International Journal of Psychophysiology , 169 , 63–70. https://doi.org/10.1016/J.IJPSYCHO.2021.09.001 Jabès, A., Klencklen, G., Ruggeri, P., Michel, C. M., Banta Lavenex, P., & Lavenex, P. (2021). Resting‐State EEG Microstates Parallel Age‐Related Differences in Allocentric Spatial Working Memory Performance. Brain Topography . https://doi.org/10.1007/s10548-021-00835-3 Jia, W., von Wegner, F., Zhao, M., & Zeng, Y. (2021). Network oscillations imply the highest cognitive workload and lowest cognitive control during idea generation in open-ended creation tasks. Scientific Reports 2021 11:1 , 11 (1), 1–23. https://doi.org/10.1038/s41598-021-03577-1 Kalburgi, S. N., Aryan, D., Nash, K., Schiller, B., & Koenig, T. (2023). MICROSTATELAB: The EEGLAB toolbox for resting-state microstate analysis . https://doi.org/10.21203/rs.3.rs-3097311/v1 Ke, M., Li, J., & Wang, L. (2021). Alteration in Resting-State EEG Microstates Following 24 Hours of Total Sleep Deprivation in Healthy Young Male Subjects. Frontiers in Human Neuroscience , 15 . https://doi.org/10.3389/FNHUM.2021.636252 Kello, C. T., Brown, G. D. A., Ferrer-i-Cancho, R., Holden, J. G., Linkenkaer-Hansen, K., Rhodes, T., & Van Orden, G. C. (2010). Scaling laws in cognitive sciences. Trends in Cognitive Sciences , 14 (5), 223–232. https://doi.org/10.1016/j.tics.2010.02.005 Keshmiri, S. (2020). Entropy and the Brain: An Overview. Entropy (Basel, Switzerland) , 22 (9). https://doi.org/10.3390/E22090917 Khanna, A., Pascual-Leone, A., & Farzan, F. (2014). Reliability of resting-state microstate features in electroencephalography. PLoS ONE , 9 (12), e114163. https://doi.org/10.1371/journal.pone.0114163 Khanna, A., Pascual-Leone, A., Michel, C. M., & Farzan, F. (2015). Microstates in resting-state EEG: Current status and future directions. Neuroscience and Biobehavioral Reviews , 49 , 105–113. https://doi.org/10.1016/j.neubiorev.2014.12.010 Kleinert, T., Koenig, T., Nash, K., & Wascher, · Edmund. (2023). On the Reliability of the EEG Microstate Approach. Brain Topography 2023 , 1 , 1–16. https://doi.org/10.1007/S10548-023-00982-9 Koenig, T., Diezig, S., Kalburgi, S. N., Antonova, E., Artoni, F., Brechet, L., Britz, J., Croce, P., Custo, A., Damborská, A., Deolindo, C., Heinrichs, M., Kleinert, T., Liang, Z., Murphy, M. M., Nash, K., Nehaniv, C., Schiller, B., Smailovic, U., … Michel, C. M. (2023). EEG-Meta-Microstates: Towards a More Objective Use of Resting-State EEG Microstate Findings Across Studies. Brain Topography 2023 , 1–14. https://doi.org/10.1007/S10548-023-00993-6 Koo, T. K., & Li, M. Y. (2016). A Guideline of Selecting and Reporting Intraclass Correlation Coefficients for Reliability Research. Journal of Chiropractic Medicine , 15 (2), 155–163. https://doi.org/10.1016/j.jcm.2016.02.012 Lassi, M., Fabbiani, C., Mazzeo, S., Burali, R., Vergani, A. A., Giacomucci, G., Moschini, V., Morinelli, C., Emiliani, F., Scarpino, M., Bagnoli, S., Ingannato, A., Nacmias, B., Padiglioni, S., Micera, S., Sorbi, S., Grippo, A., Bessi, V., & Mazzoni, A. (2023). Degradation of EEG microstates patterns in subjective cognitive decline and mild cognitive impairment: Early biomarkers along the Alzheimer’s Disease continuum? NeuroImage: Clinical , 38 , 103407. https://doi.org/10.1016/J.NICL.2023.103407 Lehmann, D. (1990). Brain Electric Microstates and Cognition: The Atoms of Thought. In Machinery of the Mind (pp. 209–224). Birkhäuser Boston. https://doi.org/10.1007/978-1-4757-1083-0_10 Lehmann, D., Faber, P. L., Galderisi, S., Herrmann, W. M., Kinoshita, T., Koukkou, M., Mucci, A., Pascual-Marqui, R. D., Saito, N., Wackermann, J., Winterer, G., & Koenig, T. (2005). EEG microstate duration and syntax in acute, medication-naïve, first-episode schizophrenia: A multi-center study. Psychiatry Research - Neuroimaging , 138 (2), 141–156. https://doi.org/10.1016/j.pscychresns.2004.05.007 Lewis, C. M., Baldassarre, A., Committeri, G., Romani, G. L., & Corbetta, M. (2009). Learning sculpts the spontaneous activity of the resting human brain. Proceedings of the National Academy of Sciences of the United States of America , 106 (41), 17558–17563. https://doi.org/10.1073/PNAS.0902455106/SUPPL_FILE/0902455106SI.PDF Linkenkaer-Hansen, K., Nikouline, V. V., Palva, J. M., & Ilmoniemi, R. J. (2001). Long-Range Temporal Correlations and Scaling Behavior in Human Brain Oscillations. The Journal of Neuroscience , 21 (4), 1370. https://doi.org/10.1523/JNEUROSCI.21-04-01370.2001 Linkenkaer-Hansen, K., Nikulin, V. V., Palva, J. M., Kaila, K., & Ilmoniemi, R. J. (2004). Stimulus-induced change in long-range temporal correlations and scaling behaviour of sensorimotor oscillations. European Journal of Neuroscience , 19 (1), 203–218. https://doi.org/10.1111/J.1460-9568.2004.03116.X Liu, J., Tang, W., Chen, G., Lu, Y., Feng, C., & Tu, X. M. (2016). Correlation and agreement: overview and clarification of competing concepts and measures. Shanghai Archives of Psychiatry , 28 (2), 115–120. https://doi.org/10.11919/j.issn.1002-0829.216045 Liu, J., Xu, J., Zou, G., He, Y., Zou, Q., & Gao, J. H. (2020). Reliability and Individual Specificity of EEG Microstate Characteristics. Brain Topography , 33 (4), 438–449. https://doi.org/10.1007/s10548-020-00777-2 Liu, Z., Si, L., Xu, W., Zhang, K., Wang, Q., Chen, B., & Wang, G. (2022). Characteristics of EEG Microstate Sequences During Propofol-Induced Alterations of Brain Consciousness States. IEEE Transactions on Neural Systems and Rehabilitation Engineering : A Publication of the IEEE Engineering in Medicine and Biology Society , 30 , 1631–1641. https://doi.org/10.1109/TNSRE.2022.3182705 Love, J., Selker, R., Marsman, M., Jamil, T., Dropmann, D., Verhagen, J., Ly, A., Gronau, Q. F., Šmíra, M., Epskamp, S., Matzke, D., Wild, A., Knight, P., Rouder, J. N., Morey, R. D., & Wagenmakers, E. J. (2019). JASP: Graphical statistical software for common statistical designs. Journal of Statistical Software , 88 (1). https://doi.org/10.18637/jss.v088.i02 McKeown, D. J., Finley, A. J., Kelley, N. J., Cavanagh, J. F., Keage, H. A. D., Baumann, O., Schinazi, V. R., Moustafa, A. A., & Angus, D. J. (2023). Test-retest reliability of spectral parameterization by 1/f characterization using SpecParam. Cerebral Cortex (New York, NY) , 34 (1), bhad482. https://doi.org/10.1093/CERCOR/BHAD482 Michel, C. M., & Brunet, D. (2019). EEG source imaging: A practical review of the analysis steps. Frontiers in Neurology , 10 (APR), 325. https://doi.org/10.3389/fneur.2019.00325 Michel, C. M., & Koenig, T. (2018). EEG microstates as a tool for studying the temporal dynamics of whole-brain neuronal networks: A review. NeuroImage , 180 , 577–593. https://doi.org/10.1016/j.neuroimage.2017.11.062 Milz, P., Faber, P. L., Lehmann, D., Koenig, T., Kochi, K., & Pascual-Marqui, R. D. (2016). The functional significance of EEG microstates-Associations with modalities of thinking. NeuroImage , 125 , 643–656. https://doi.org/10.1016/j.neuroimage.2015.08.023 Murphy, M., Stickgold, R., & Öngür, D. (2020). Electroencephalogram Microstate Abnormalities in Early-Course Psychosis. Biological Psychiatry: Cognitive Neuroscience and Neuroimaging , 5 (1), 35–44. https://doi.org/10.1016/J.BPSC.2019.07.006 Murphy, M., Wang, J., Jiang, C., Wang, L. A., Kozhemiako, N., Wang, Y., Wang, J., Jiang, C., Gai, G., Zou, K., Wang, Z., Yu, X., Wang, G., Tan, S., Murphy, M., Hall, M. H., Zhu, W., Zhou, Z., Shen, L., … Purcell, S. M. (2023). A Potential Source of Bias in Group-Level EEG Microstate Analysis. Brain Topography . https://doi.org/10.1007/s10548-023-00992-7 Murray, M. M., Brunet, D., & Michel, C. M. (2008). Topographic ERP analyses: A step-by-step tutorial review. In Brain Topography (Vol. 20, Issue 4, pp. 249–264). https://doi.org/10.1007/s10548-008-0054-5 Myles, P. S., & Cui, J. (2007). Using the Bland-Altman method to measure agreement with repeated measures. In British Journal of Anaesthesia (Vol. 99, Issue 3, pp. 309–311). Oxford University Press. https://doi.org/10.1093/bja/aem214 Nazare, K., & Tomescu, M. I. (2024). Valence-specific EEG microstate modulations during self-generated affective states. Frontiers in Psychology , 15 , 1300416. https://doi.org/10.3389/FPSYG.2024.1300416 Nishida, K., Morishima, Y., Yoshimura, M., Isotani, T., Irisawa, S., Jann, K., Dierks, T., Strik, W., Kinoshita, T., & Koenig, T. (2013). EEG microstates associated with salience and frontoparietal networks in frontotemporal dementia, schizophrenia and Alzheimer’s disease . https://doi.org/10.1016/j.clinph.2013.01.005 Niu, Y., Chen, X., Chen, Y., Yao, Z., Chen, X., Liu, Z., Meng, X., Liu, Y., Zhao, Z., & Fan, H. (2024). A gender recognition method based on EEG microstates. Computers in Biology and Medicine , 173 . https://doi.org/10.1016/J.COMPBIOMED.2024.108366 Pascual-Marqui, R. D., Michel, C. M., & Lehmann, D. (1995). Segmentation of Brain Electrical Activity into Microstates; Model Estimation and Validation. IEEE Transactions on Biomedical Engineering , 42 (7), 658–665. https://doi.org/10.1109/10.391164 Popov, T., Tröndle, M., Baranczuk-Turska, Z., Pfeiffer, C., Haufe, S., & Langer, N. (2023). Test–retest reliability of resting-state EEG in young and older adults. Psychophysiology , 60 (7), e14268. https://doi.org/10.1111/PSYP.14268 Poulsen, A. T., Pedroni, A., Langer, N., & Hansen, L. K. (2018). Microstate EEGlab toolbox: An introductory guide. BioRxiv , 1–30. https://doi.org/10.1101/289850 Ren, H., Ran, X., Qiu, M., Lv, S., Wang, J., Wang, C., Xu, Y., Gao, Z., Ren, W., Zhou, X., Mu, J., Yu, Y., & Zhao, Z. (2024). Abnormal nonlinear features of EEG microstate sequence in obsessive–compulsive disorder. BMC Psychiatry , 24 (1), 881. https://doi.org/10.1186/S12888-024-06334-6 Rieger, K., Hernandez, D., Baenninger, L., & Koenig, A. (2016). 15 Years of Microstate Research in Schizophrenia – Where Are We? A Meta-Analysis. A Meta-Analysis. Front. Psychiatry , 7 , 22. https://doi.org/10.3389/fpsyt.2016.00022 Ros, T., Frewen, P., Théberge, J., Michela, A., Kluetsch, R., Mueller, A., Candrian, G., Jetly, R., Vuilleumier, P., & Lanius, R. A. (2017). Neurofeedback Tunes Scale-Free Dynamics in Spontaneous Brain Activity. Cerebral Cortex , 27 (10), 4911–4922. https://doi.org/10.1093/CERCOR/BHW285 Schiller, B., Sperl, M. F. J., Kleinert, T., Nash, K., & Gianotti, L. R. R. (2023). EEG Microstates in Social and Affective Neuroscience. Brain Topography 2023 , 1 , 1–17. https://doi.org/10.1007/S10548-023-00987-4 Seitzman, B. A., Abell, M., Bartley, S. C., Erickson, M. A., Bolbecker, A. R., & Hetrick, W. P. (2017). Cognitive manipulation of brain electric microstates. NeuroImage , 146 , 533–543. https://doi.org/10.1016/j.neuroimage.2016.10.002 Shew, W. L., & Plenz, D. (2012). The Functional Benefits of Criticality in the Cortex. Http://Dx.Doi.Org/10.1177/1073858412445487 , 19 (1), 88–100. https://doi.org/10.1177/1073858412445487 Shrout, P. E., & Fleiss, J. L. (1979). Intraclass correlations: uses in assessing rater reliability. Psychological Bulletin , 86 (2), 420–428. https://doi.org/10.1037//0033-2909.86.2.420 Smailovic, U., Koenig, T., Laukka, E. J., Kalpouzos, G., Andersson, T., Winblad, B., & Jelic, V. (2019). EEG time signature in Alzheimer´s disease: Functional brain networks falling apart. NeuroImage: Clinical , 24 . https://doi.org/10.1016/j.nicl.2019.102046 Tagliazucchi, E., Von Wegner, F., Morzelewski, A., Brodbeck, V., Jahnke, K., & Laufs, H. (2013). Breakdown of long-range temporal dependence in default mode and attention networks during deep sleep. Proceedings of the National Academy of Sciences of the United States of America , 110 (38), 15419–15424. https://doi.org/10.1073/pnas.1312848110 Tait, L., Tamagnini, F., Stothart, G., Barvas, E., Monaldini, C., Frusciante, R., Volpini, M., Guttmann, S., Coulthard, E., Brown, J. T., Kazanina, N., & Goodfellow, M. (2020). EEG microstate complexity for aiding early diagnosis of Alzheimer’s disease. Scientific Reports 2020 10:1 , 10 (1), 1–10. https://doi.org/10.1038/s41598-020-74790-7 Tait, L., & Zhang, J. (2022). +microstate: A MATLAB toolbox for brain microstate analysis in sensor and cortical EEG/MEG. NeuroImage , 258 , 119346. https://doi.org/10.1016/J.NEUROIMAGE.2022.119346 Takarae, Y., Zanesco, A., Keehn, B., Chukoskie, L., Müller, R. A., & Townsend, J. (2022). EEG microstates suggest atypical resting-state network activity in high-functioning children and adolescents with autism spectrum development. Developmental Science . https://doi.org/10.1111/desc.13231 Tarailis, P., Koenig, T., Michel, C. M., & Griškova-Bulanova, I. (2023). The Functional Aspects of Resting EEG Microstates: A Systematic Review. Brain Topography . https://doi.org/10.1007/S10548-023-00958-9 Tarailis, P., Šimkutė, D., Koenig, T., & Griškova-Bulanova, I. (2021). Relationship between Spatiotemporal Dynamics of the Brain at Rest and Self-Reported Spontaneous Thoughts: An EEG Microstate Approach. Journal of Personalized Medicine , 11 (11), 1216. https://doi.org/10.3390/jpm11111216 Tomescu, M. I., Rihs, T. A., Roinishvili, M., Karahanoglu, F. I., Schneider, M., Menghetti, S., Van De Ville, D., Brand, A., Chkonia, E., Eliez, S., Herzog, M. H., Michel, C. M., & Cappe, C. (2015). Schizophrenia patients and 22q11.2 deletion syndrome adolescents at risk express the same deviant patterns of resting state EEG microstates: A candidate endophenotype of schizophrenia. Schizophrenia Research: Cognition , 2 (3), 159. https://doi.org/10.1016/J.SCOG.2015.04.005 Van De Ville, D., Britz, J., & Michel, C. M. (2010). EEG microstate sequences in healthy humans at rest reveal scale-free dynamics. Proceedings of the National Academy of Sciences of the United States of America , 107 (42), 18179–18184. https://doi.org/10.1073/PNAS.1007841107/-/DCSUPPLEMENTAL Vellante, F., Ferri, F., Baroni, G., Croce, P., Migliorati, D., Pettoruso, M., De Berardis, D., Martinotti, G., Zappasodi, F., & Giannantonio, M. Di. (2020). Euthymic bipolar disorder patients and EEG microstates: a neural signature of their abnormal self experience? Journal of Affective Disorders , 272 , 326–334. https://doi.org/10.1016/j.jad.2020.03.175 von Wegner, F., Bauer, S., Rosenow, F., Triesch, J., & Laufs, H. (2021). EEG microstate periodicity explained by rotating phase patterns of resting-state alpha oscillations. NeuroImage , 224 , 117372. https://doi.org/10.1016/J.NEUROIMAGE.2020.117372 von Wegner, F., Knaut, P., & Laufs, H. (2018). EEG microstate sequences from different clustering algorithms are information-theoretically invariant. Frontiers in Computational Neuroscience , 12 . https://doi.org/10.3389/fncom.2018.00070 von Wegner, F., & Laufs, H. (2018). Information-theoretical analysis of EEG microstate sequences in python. Frontiers in Neuroinformatics , 12 , 30. https://doi.org/10.3389/FNINF.2018.00030/BIBTEX von Wegner, F., Tagliazucchi, E., & Laufs, H. (2017). Information-theoretical analysis of resting state EEG microstate sequences - non-Markovianity, non-stationarity and periodicities. NeuroImage , 158 , 99–111. https://doi.org/10.1016/j.neuroimage.2017.06.062 von Wegner, F., Wiemers, M., Hermann, · Gesine, Tödt, I., Tagliazucchi, E., & Laufs, · Helmut. (2023). Complexity Measures for EEG Microstate Sequences: Concepts and Algorithms. Brain Topography 2023 , 1 , 1–16. https://doi.org/10.1007/S10548-023-01006-2 Wackermann, J., Lehmann, D., Michel, C. M., & Strik, W. K. (1993). Adaptive segmentation of spontaneous EEG map series into spatially defined microstates. International Journal of Psychophysiology , 14 (3), 269–283. https://doi.org/10.1016/0167-8760(93)90041-M Wang, Y., Duan, W., Dong, D., Ding, L., & Lei, X. (2022). A test-retest resting, and cognitive state EEG dataset during multiple subject-driven states. Scientific Data , 9 (1). https://doi.org/10.1038/s41597-022-01607-9 Wiemers, M. C., Laufs, H., & von Wegner, F. (2023). Frequency Analysis of EEG Microstate Sequences in Wakefulness and NREM Sleep. Brain Topography , 1 , 1–17. https://doi.org/10.1007/S10548-023-00971-Y/FIGURES/6 Zanesco, A. P. (2020). EEG Electric Field Topography is Stable During Moments of High Field Strength. Brain Topography , 33 (4), 450–460. https://doi.org/10.1007/s10548-020-00780-7 Zanesco, A. P. (2023). Normative Temporal Dynamics of Resting EEG Microstates. Brain Topography . https://doi.org/10.1007/S10548-023-01004-4 Zanesco, A. P., Denkova, E., & Jha, A. P. (2020). Self-reported mind wandering and response time variability differentiate prestimulus electroencephalogram microstate dynamics during a sustained attention task. Journal of Cognitive Neuroscience , 33 (1), 28–45. https://doi.org/10.1162/jocn_a_01636 Zanesco, A. P., Skwara, A. C., King, B. G., Powers, C., Wineberg, K., & Saron, C. D. (2021). Meditation training modulates brain electric microstates and felt states of awareness. Human Brain Mapping , hbm.25430. https://doi.org/10.1002/hbm.25430 Zhang, K., Shi, W., Wang, C., Li, Y., Liu, Z., Liu, T., Li, J., Yan, X., Wang, Q., Cao, Z., & Wang, G. (2021). Reliability of EEG microstate analysis at different electrode densities during propofol-induced transitions of brain states. NeuroImage , 231 . https://doi.org/10.1016/j.neuroimage.2021.117861 Zulliger, J., Diaz Hernandez, L., & Koenig, T. (2022). Within and Between Subject Spectral Fingerprints of EEG-Microstate Parameters. Brain Topography , 35 (3), 277–281. https://doi.org/10.1007/S10548-022-00896-Y/TABLES/1 Additional Declarations No competing interests reported. Supplementary Files Supplementary.docx Cite Share Download PDF Status: Published Journal Publication published 09 Dec, 2025 Read the published version in Cognitive Neurodynamics → Version 1 posted Editorial decision: Revision requested 10 Sep, 2025 Editor assigned by journal 10 Sep, 2025 Reviews received at journal 26 Apr, 2025 Reviewers agreed at journal 20 Apr, 2025 Reviews received at journal 10 Apr, 2025 Reviewers agreed at journal 03 Apr, 2025 Reviewers invited by journal 02 Apr, 2025 Submission checks completed at journal 22 Mar, 2025 First submitted to journal 20 Mar, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5875634","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":500114532,"identity":"d8ec5801-3432-43d9-8312-2888c82901b0","order_by":0,"name":"Povilas Tarailis","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+0lEQVRIie3Pv0vEMBTA8RcC7RLoGv+LSMGri/1XXgnU1VGw9HocOB129c9wcg4E6qK4xq0uN3eSm+Rej/uBQ66rQ75DW0o+5D2AUOifZvrDF95JSID1/rMHgvQQO6IkXDRcTV9zJECHlZkgs3h1abCqIY8/vodeZXn6aSMJ1Y2XXK/elcHOghC3qaTBileniXTaS5Qr0WBkaLASRoJXLukka/gE+a1BJGu+IZKnLQ3GmvkZoo0pHjkIWUbjLewF9EjsmV0skSfaxa2jbNzl2WmeYffmJbN4sRyGnxrituRfm/s6T1rL3FA9+Afbv+fNn9/oBScSCoVCIX9b40hO+1I+0iMAAAAASUVORK5CYII=","orcid":"","institution":"Simon Fraser University","correspondingAuthor":true,"prefix":"","firstName":"Povilas","middleName":"","lastName":"Tarailis","suffix":""},{"id":500114533,"identity":"fdd2a755-38e0-47f9-a9cd-10aea7ad22a2","order_by":1,"name":"Fiorenzo Artoni","email":"","orcid":"","institution":"University of Geneva","correspondingAuthor":false,"prefix":"","firstName":"Fiorenzo","middleName":"","lastName":"Artoni","suffix":""},{"id":500114534,"identity":"addc90a3-bcd2-4f29-9eac-745fbe5083a0","order_by":2,"name":"Thomas Koenig","email":"","orcid":"","institution":"University of Bern","correspondingAuthor":false,"prefix":"","firstName":"Thomas","middleName":"","lastName":"Koenig","suffix":""},{"id":500114535,"identity":"aa919a7f-7292-4b54-b2b7-faf5f63e4d39","order_by":3,"name":"Christoph M. Michel","email":"","orcid":"","institution":"University of Geneva","correspondingAuthor":false,"prefix":"","firstName":"Christoph","middleName":"M.","lastName":"Michel","suffix":""},{"id":500114536,"identity":"9c0a1716-c955-4288-91bf-4b0fa0c9f2bb","order_by":4,"name":"Inga Griskova-Bulanova","email":"","orcid":"","institution":"Vilnius University","correspondingAuthor":false,"prefix":"","firstName":"Inga","middleName":"","lastName":"Griskova-Bulanova","suffix":""}],"badges":[],"createdAt":"2025-01-21 19:23:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5875634/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5875634/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s11571-025-10391-2","type":"published","date":"2025-12-09T15:58:15+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":89448412,"identity":"4b9118cb-5d08-47bb-84f5-b0c7bd0028aa","added_by":"auto","created_at":"2025-08-20 05:43:26","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":44511,"visible":true,"origin":"","legend":"\u003cp\u003eA schematic representation of different scenarios of probability of sequence entropy based on probability of five microstates occurrence.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-5875634/v1/bde28af0d49d8ebee49a5996.png"},{"id":89448414,"identity":"547f6e4b-1d20-46ee-9c42-1b1268a49c0d","added_by":"auto","created_at":"2025-08-20 05:43:26","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":698104,"visible":true,"origin":"","legend":"\u003cp\u003eA) Assessment of extracted microstates with different criterion implemented in Cartool. The best solution for each criterion is colored green. B) Group level topographies for each session. C). Shared spatial variance between extracted topographies.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-5875634/v1/45b075f2d8abbe719bb82521.png"},{"id":89448413,"identity":"fbed01d7-b799-4ba4-834a-ef314ec2874a","added_by":"auto","created_at":"2025-08-20 05:43:26","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":823123,"visible":true,"origin":"","legend":"\u003cp\u003eA to E top: Gardner – Altman plots for extracted parameters where the Y axis indicates individual values for parameters between sessions (X axis, S1 and S2 on the left and S1 and S3 on the right). Bottom: Bland-Altman plots for short-term (left) and long-term (right) agreement. The dotted middle line indicates bias and two outer dotted lines indicate limits of agreement. Shaded areas indicate 95% confidence interval. F) Intraclass correlation coefficient values for extracted parameters. Error bars indicate 95 % confidence interval.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-5875634/v1/e89b62d615fa980de29a8ae8.png"},{"id":98243879,"identity":"91c9f7b8-a469-4e4e-b076-36c27416aef9","added_by":"auto","created_at":"2025-12-15 16:11:04","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2217776,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5875634/v1/4bf81953-0198-4d5e-ae09-47168e82abee.pdf"},{"id":89448424,"identity":"6953b5ee-a901-466a-8342-f7eec856b2ee","added_by":"auto","created_at":"2025-08-20 05:43:27","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":37722,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementary.docx","url":"https://assets-eu.researchsquare.com/files/rs-5875634/v1/056786242468b504fa79c216.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Short-term and long-term test-retest reliability of memory, complexity, and randomness of EEG microstates sequence","fulltext":[{"header":"Introduction","content":"\u003cp\u003eWith the increasing need and interest in potential easily accessible sensitive biomarkers of neuropsychiatric conditions, electroencephalography (EEG) has received extensive attention. One of the well-established method to analyse EEG in order to study large-scale brain cortical networks during resting state is EEG microstates (MS). The method uses multichannel approach to simultaneously quantify information from all channels; thus, the recorded oscillations are defined as the non-overlapping maps characterized by a unique spatial distribution (class) that allows to capture not only temporal but also spatial dynamics of the ongoing brain electrical activity (Khanna et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Michel \u0026amp; Koenig, \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Tarailis et al., \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). In MS analysis, the goal is to group EEG time samples into clusters using various algorithms (k-means, TAAHC, AAHC, PCA, ICA (Murray et al., \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Pascual-Marqui et al., \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e1995\u003c/span\u003e; Poulsen et al., \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; von Wegner et al., \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) so that the EEG samples that belong to the same class would have as similar topographical displays as possible. Resting state EEG MS approach gained a lot of attention in recent years as evidenced by increased number of clinical (de Bock et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Deiber et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; F\u0026eacute;rat, Arns, et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Hanoglu et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Smailovic et al., \u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), task-related (Comsa et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; D\u0026rsquo;croz-Baron et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Deolindo et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Milz et al., \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Nazare \u0026amp; Tomescu, \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Zanesco et al., \u003cspan citationid=\"CR98\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), methodological (F\u0026eacute;rat, Seeber, et al., 2022; Haydock et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Kalburgi et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Koenig et al., \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Murphy et al., \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), review and meta-analytical papers (Chivu et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Das et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Khanna et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Michel \u0026amp; Koenig, \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Rieger et al., \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Schiller et al., \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Tarailis et al., \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Zanesco, \u003cspan citationid=\"CR97\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Specifically, a handful of EEG MS studies suggested different microstate features as possible biomarkers for clinical and research settings (Chivu et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Deiber et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Rieger et al., \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Tarailis et al., \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). With this increased interest in EEG microstate analysis, the question of within-subject reliability (stability of measurements taken from the same individuals across different points in time) is crucial. If extracted parameters are not reliable, conclusions drawn from them could cause misleading interpretations and even inaccurate diagnosis.\u003c/p\u003e\u003cp\u003eIn a classical assessment pipeline, EEG microstates are evaluated by its temporal parameters \u0026ndash; duration (reflecting how long microstate class is present), occurrence (reflecting how often the microstate class is present), coverage (reflecting the proportion that microstate class takes in the recording). However, only eight studies up to date have addressed the question of test-retest reliability of EEG microstates with a focus on the temporal parameters (Antonova et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Bagdasarov et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Q. Guo et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Khanna et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Kleinert et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; J. Liu et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Popov et al., \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Zhang et al., \u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) with somewhat consistent results. Khanna et al. (\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2014\u003c/span\u003e) reported high test-retest reliability (Cronbach's α of 0.7\u0026ndash;0.9) of duration, occurrence, and coverage measures recorded approximately 48 hours apart, independent from clustering algorithm (k-means vs TAAHC) used and number of channels (30 vs 19 vs 8) included. Zhang et al. (\u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) demonstrated high reliability of temporal parameters acquired with 91, 64 and 32 channels EEG. Reliability decreased using 19 and 8 channels EEG, suggesting that it might be sensitive to the low number of channels. Liu et al., (\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) showed that EEG recordings with a duration greater than 2 minutes display high short-term (23\u0026ndash;25 hours) reliability (Intraclass correlation coefficient (ICC)\u0026thinsp;\u0026gt;\u0026thinsp;0.6) for all temporal parameters of microstates. Similar results were reported by Guo et al., (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) in healthy controls and Parkinson\u0026rsquo;s patients with high reliability of temporal parameters acquired from EEG longer than 3 minutes. Antonova et al. (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) reported mixed short-term (approximately 5 to 10 minutes apart) reliability results for temporal parameters of microstate with ICC values ranging from as low as 0.3 to as high as 0.9. Popov et al., (\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) showed mixed short-term (7\u0026ndash;9 days) reliability of duration, occurrence, coverage, global filed power (GFP) and global explained variance (GEV), outlining the decrease of reliability with subjects\u0026rsquo; age, while Bagdasarov et al (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) demonstrated high reliability in infants, interestingly showing that even microstate parameters extracted from 1 min. EEG show good to excellent reliability. Finally, Kleinert et al., (\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) reported good to excellent short-term (average interval between measures: 99 min) reliability and moderate to good long-term (average interval 63 days) reliability of temporal parameters obtained with different number of channels (64 and 30) and different clustering algorithms (k-means, AAHC). Overall, the abovementioned suggest that temporal parameters of microstates are reliable, i.e. stable across different points in time.\u003c/p\u003e\u003cp\u003eHowever, the switch between different microstates in the EEG recording reflects important information. It was suggested that microstates represent short-lasting, rapidly switching global mental states, thus were nicknamed \u0026lsquo;atom of thoughts\u0026rsquo; (Lehmann, \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e1990\u003c/span\u003e). In order to study transition patterns from one global state to the next one, in 1993, Wackermann (Wackermann et al., \u003cspan citationid=\"CR93\" class=\"CitationRef\"\u003e1993\u003c/span\u003e) introduced microstate syntax method. The method is based on the first order Markov process, where the probability of occurrence of each microstate depends only on the one preceding it, i.e. a memoryless process. While many studies reported microstate syntax sensitivity to different task-related conditions (Br\u0026eacute;chet et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Jab\u0026egrave;s et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Seitzman et al., \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) personality traits (Du et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; P. Guo et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Hu et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), levels of vigilance (Ke et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) and clinical conditions (Nishida et al., \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Vellante et al., \u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), the main limitation of this method is in its temporal scope. EEG microstate syntax considers only a single time step dynamics and assumes that transitions between microstates are stationary and the next state depends only on the one preceding it. It was suggested that quantification and evaluation of microstate sequences should extend beyond the first and even second-order Markovian transitions (Antonova et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Artoni et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) in order to capture longer-range temporal dependencies and higher-order dynamics, thereby enabling a more comprehensive characterization of brain state transitions and their functional significance. Moreover, a few studies assessing the reliability of EEG microstate parameters reported that EEG syntax displays poor to moderate within-subjects consistency (Antonova et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Kleinert et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Contradictory to EEG MS syntax assumption, a handful of studies reported that MS sequences show non-Markovian, higher order temporal structure (Artoni et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Gschwind et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; von Wegner et al., \u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), which is driven by the brain\u0026rsquo;s dominant frequency (Hermann et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; von Wegner et al., \u003cspan citationid=\"CR88\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Wiemers et al., \u003cspan citationid=\"CR95\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e(Antonova et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Bagdasarov et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Q. Guo et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Khanna et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Kleinert et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; J. Liu et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Popov et al., \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Zhang et al., \u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e2021\u003c/span\u003e)In our recent systematic review (Tarailis et al., \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) we noted that due to advances in the field of data science, microstates sequence analysis gained a lot of attention and different sequence analysis methods have been applied to study microstate sequence randomness, speed, complexity, periodicity and long-range memory (Artoni et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; von Wegner et al., \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e2018\u003c/span\u003e, \u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Several sequence analysis methods are implemented in microstate analysis toolboxes for MATLAB (Tait \u0026amp; Zhang, \u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) and python (F\u0026eacute;rat, Scheltienne, et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; von Wegner \u0026amp; Laufs, \u003cspan citationid=\"CR90\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) and are freely available to researchers.\u003c/p\u003e\u003cp\u003eThree studies already paid a great deal of attention to microstate sequences sensitivity to cluster algorithms, band pass filtering settings and electrode density. A work by (von Wegner et al., \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) addressed the question of impact of the clustering algorithm selection (AAHC vs k-means vs k-mediods vs PCA vs ICA) on outcomes of sequence analysis methods (Hurst exponent, transition matrix mixing time, entropy, entropy rate and autoinformation function) and concluded that these parameters are not affected by the clustering algorithm. Another study by (F\u0026eacute;rat, Seeber, et al., 2022) showed that although the topographies are not affected, temporal order of microstate sequences is sensitive to the band-pass filtering settings (broadband vs narrowband). Zhang et al. (\u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) reported stable entropy rate values for EEGs recorded using 91, 64, 32 and 19 channels, while this metric increased when 8 channels EEG was analysed. Yet, no study so far aimed to assess the test-retest within-subjects reliability of various sequence characteristics of EEG microstates. Thus, this work aims to evaluate how stable different outcomes reflecting short-range and long-range memory, complexity, sequence randomness and uncertainty of microstates across different points in time are. We performed EEG microstate segmentation on the publicly available EEG data recorded from 60 healthy young adults and evaluated short-term (after 90 min), and long-term (after 30 days) reliability on the outcomes. of the five microstate sequence analysis methods reflecting short-range and long-range memory, complexity, sequence randomness and uncertainty.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cp\u003eDataset\u003c/p\u003e\u003cp\u003ePreprocessed EEG files were downloaded from public repository and are available at OpenNeuro (\u0026lsquo;Dataset ds004148\u0026rsquo;; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://openneuro.org/datasets/ds004148/versions/1.0.1\u003c/span\u003e\u003cspan address=\"https://openneuro.org/datasets/ds004148/versions/1.0.1\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). EEG were obtained from 60 anonymized participants (M\u0026thinsp;=\u0026thinsp;28, F\u0026thinsp;=\u0026thinsp;32, mean age\u0026thinsp;=\u0026thinsp;20.01 years and SD\u0026thinsp;=\u0026thinsp;\u0026plusmn;\u0026thinsp;1.88, range between 18 and 28 years old). Detailed information about subjects and inclusion criteria are reported in the original work (Wang et al., \u003cspan citationid=\"CR94\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eRecording and preprocessing\u003c/p\u003e\u003cp\u003eDetailed information about experimental design, data collection and preprocessing is described in (Wang et al., \u003cspan citationid=\"CR94\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Briefly, EEG data was collected over three sessions, where session 1 (S1) and session 2 (S2) were held on the same day 90 minutes apart, and session 3 (S3) was scheduled 30 days later, matching the time of the day with the first recording (S1). Each session consisted of eyes open, eyes closed and three cognitive states, each lasting 5 minutes. In this study we used only eyes-closed EEG data.\u003c/p\u003e\u003cp\u003eEEG data was recorded using 63 or 64 channels, two of these electrodes were used to record eye movements. FCz was used as the online reference, data was sampled at 500 Hz and impedance was kept below 5kΩ. To unify the channels from the different sizes of caps during EEG recordings, EEG was reconstructed to 61 channels, filtered between 0.3 and 45 Hz using Finite Impulse Response (FIR) filter. EEG recordings were visually inspected and problematic channels were rejected. Linear interpolation was used to reconstruct missing channels. EEG signal was segmented into 4-s epochs and then all bad epochs were manually screened for removal.\u003c/p\u003e\u003cp\u003eBefore the microstate segmentation, the reference was re-projected to average, data was downsampled to 250 Hz and bandpass filtered between 1 and 30 Hz using Butterworth filter of the second order.\u003c/p\u003e\u003cp\u003eMicrostate segmentation\u003c/p\u003e\u003cp\u003eK-means clustering\u003c/p\u003e\u003cp\u003eThe microstate segmentation was performed using Cartool toolbox (Brunet et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). Cluster analysis was performed separately for each session and was performed in two stages. First, to speed up clustering and increase its reliability, individual EEGs were resampled into 28 random epochs of 7500 time frames. For each epoch, GFP was calculated as the spatial standard deviation and only topographical maps at GFP peaks were extracted and submitted to k-means clustering algorithm, ignoring their polarity. To identify the optimal number of microstate templates for each epoch, the number of clusters was set between 1 and 12 with 100 iterations for each number of clusters to maximize GEV. The spatial filter was applied to the topographical maps before cluster analysis to spatial smooth topographies and increase signal-to-noise ratio (Michel \u0026amp; Brunet, \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). If topography failed to reach minimum correlation threshold with cluster\u0026rsquo;s centroid (\u0026lt;\u0026thinsp;0.5), it remained unassigned to any cluster. The optimal number of clusters for each epoch was determined by meta-criteria implemented in Cartool (see Cartool\u0026rsquo;s Reference Guide). At the second stage, 1680 sets (28 epochs x 60 subjects) of the most dominant topographies in each session of each epoch were concatenated and submitted to the second level k-means clustering, once again ignoring polarity and applying minimal spatial correlation threshold. Concatenated data was resampled into 100 random epochs of 1000 time frames. In this stage, the number of clusters ranged between 1 and 15 and segmentation was repeated 200 times for each number of clusters. The optimal number of group level clusters was also based on meta-criteria.\u003c/p\u003e\u003cp\u003eBackfitting\u003c/p\u003e\u003cp\u003eBased on meta-criteria, the optimal number of group level topographical maps for all three sessions was 5 (Fig.\u0026nbsp;2A). These maps were backfitted to the original individual EEGs using winner-takes-all approach, where spatial correlation was calculated between the group level maps and topography at each time frame of individual EEG. Polarity was ignored, and minimal correlation threshold was applied. Temporal smoothing (window size of 20 ms, weight smoothness factor of 10) was applied to ensure that the noise during low GFP did not artificially interrupt the temporal segments of stable topography and small segments, less or equal to 5 time frames (20 ms), were rejected, split in half with the first half added to the preceding segment and the second half added to the proceeding segment. This resulted in time series of microstates, where every data frame was labelled as one of the possible microstates. Although temporal parameters and syntax are out of scope of this paper, additionally we calculated the mean GEV, duration, time coverage, occurrence of each microstate and first order Markovian syntax between each pair of microstates (Lehmann et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2005\u003c/span\u003e; Wackermann et al., \u003cspan citationid=\"CR93\" class=\"CitationRef\"\u003e1993\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eSequence analysis\u003c/p\u003e\u003cp\u003eHurst exponent\u003c/p\u003e\u003cp\u003eHurst exponent is a well-established method to assess the degree of temporal dependence. It describes the long-range dependency and provides information about the persistent or anti-persistent patterns within the time series (Gschwind et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Van De Ville et al., \u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e2010\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eHurst exponent values theoretically ranges from 0 to 1, where values H\u0026thinsp;\u0026lt;\u0026thinsp;0.5 suggest short-term correlations when recent states influence the immediate future states but without a sustained trend, values of H\u0026thinsp;=\u0026thinsp;0.5 indicates more random, less predictable sequence behavior, and values H\u0026thinsp;\u0026gt;\u0026thinsp;0.5 imply that the temporal structure of the sequence is not random and past states influence future states. Thus, a higher value of Hurst exponent indicates that sequence is more predictable, repetitive and indicates the degree of long-range temporal autocorrelations present between shorter and longer timescales (D\u0026iacute;az \u0026amp; C\u0026oacute;rdova, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eA handful of different methods exist to estimate Hurst exponent (Gschwind et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; von Wegner et al., \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e2018\u003c/span\u003e); here we used detrended fluctuation analysis (DFA).We modified publicly available code implemented in +\u0026thinsp;microstate toolbox (Tait \u0026amp; Zhang, \u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Individual symbolic microstates sequences were embedded into random walk by parting microstates into two classes and associating each class with a negative and a positive step, respectively (Van De Ville et al., \u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). 50 logarithmically spaced time scales (S) between 0.2 and 30 seconds were used. Random walk sequences were parcellated into number of epochs according to different time scales. Each epoch was detrended (F) by subtracting the local trend using least-squares fit (Jia et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Ros et al., \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). The Hurst parameter of the random walk corresponds to the slope between F and S on the log-log coordinates plane. Since from biological perspective there is no good basis how to partition microstates, we followed method explained in literature (Jia et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; von Wegner et al., \u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) for each subject, Hurst exponent was computed for every possible partition and then averaged across all partitions.\u003c/p\u003e\u003cp\u003eLempel-Ziv complexity\u003c/p\u003e\u003cp\u003eLempel-Ziv complexity (LZC) quantifies the complexity or compressibility of a sequence by identifying repeated subsequences required to explain the full sequence and does not rely on assumptions of Markovianity (Tait et al., \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). To compute LZC we used publicly available code (Tait \u0026amp; Zhang, \u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).Microstate sequence with low LZC indicates regular, repetitive, small number of transitioning patterns within the sequence, whilst high LZC suggests irregular, complex microstates sequences. Individual microstate sequences were transformed into a \u0026ldquo;Jump\u0026rdquo; sequence by setting each microstate activity to the size of 1 time frame, which allowed focusing only on the temporal sequence of the microstates, independently of their individual durations. LZC starts with scanning the sequence from left to right and finding repeating patterns in the sequence. Complexity refers to the number of unique patterns in the entire sequence. Since LZC increases with the length of the sequence, to account for the length differences between individual sequences, we normalized LZC by the length of the \u0026ldquo;Jump\u0026rdquo; sequence.\u003c/p\u003e\u003cp\u003eIn order to test optimized version, the Microsynt approach to LZC calculation, we used the LZMA2 algorithm for lossless data compression with 64MB dictionary, BT4 MatchFinder, and BCJ2 Filter (Artoni et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) The LZC for each \u0026lsquo;Jump\u0026rsquo; sequence was calculated via a sliding window approach (1000 microstates windows length, 80% window overlap) and averaged across windows. The sliding-window approach offers a solution to the issue of comparing complexity of sequences with different lengths. Rather than truncating all sequences to the length of shortest one, as in Tait et al. (\u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), sequences with different lengths can be compared without loss of data as the window size remains the same (Artoni et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). For more details about method see (Artoni et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eEntropy-related measures\u003c/p\u003e\u003cp\u003eEntropy measures uncertainty or randomness of the microstate sequence (von Wegner et al., \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). The number of times a specific microstate occurs in the sequence gives the probability distribution of microstates occurrence(von Wegner et al., \u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e2017\u003c/span\u003e, \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). The shape of this distribution can be characterized by its entropy (H):\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:H\\left(x\\right)=-\\sum\\:_{k=1}^{\\text{max}k}{p}_{k}\\bullet\\:{\\text{log}}_{2}{p}_{k}\\:\\:\\:\\:\\left(1\\right)$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003ewhere \u003cem\u003ep\u003c/em\u003e is the probability of the microstate \u003cem\u003ek\u003c/em\u003e occurrence in the sequence \u003cem\u003ex.\u003c/em\u003e The sequence\u0026rsquo;s entropy is bounded between 0 \u0026ndash; entire sequence is dominated by one single microstate, to \u003cem\u003elog\u003c/em\u003e\u003csub\u003e2\u003c/sub\u003e (\u003cem\u003emax k\u003c/em\u003e) \u0026ndash; all microstates have the same occurrence probability. This is schematically represented in (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eUnfortunately, entropy ignores the temporal structure of the microstate sequence, and a randomly shuffled sequence will result in exactly the same values. Entropy rate is an extension of Shannon\u0026rsquo;s entropy that can help to describe temporal structure of the sequence. It measures how much uncertainty or randomness per symbol in a sequence, given knowledge about the past states of the sequence (von Wegner et al., \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). It measures the entropy (H) about the next symbol in a sequence (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{x}_{t+1})\\)\u003c/span\u003e\u003c/span\u003e, given knowledge about past states \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{{x}_{t}}^{\\left(n\\right)}\\)\u003c/span\u003e\u003c/span\u003e:\u003cdiv id=\"Equb\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equb\" name=\"EquationSource\"\u003e\n$$\\:{h}_{1}=H\\left({x}_{t+1}\\right|{{x}_{t}}^{\\left(n\\right)}\\left)\\:\\:\\:\\:\\:\\right(2)$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eSince entropy rate becomes less reliable with longer memory (\u003cem\u003en\u003c/em\u003e), we limited memory to maximum 7 steps, as it is recommended in the literature (von Wegner et al., \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Wiemers et al., \u003cspan citationid=\"CR95\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Zhang et al., \u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). For both entropy and entropy rate metrics, low values indicate less complex, more stable sequences, while higher values suggest more random processes.\u003c/p\u003e\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eStatistical analysis\u003c/h2\u003e\u003cp\u003eTo test short term (90 minutes) and long term (30 days) test-retest reliability of each extracted parameter, we calculated intraclass correlation coefficient (ICC) (Shrout \u0026amp; Fleiss, \u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e1979\u003c/span\u003e) implemented in JASP statistical software (Version 0.17.3) (Love et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). ICC is a correlation coefficient that assesses the consistency between measures. ICC is similar to the Pearson\u0026rsquo;s correlation coefficient; it requires a linear relationship between the variables. The main difference is that ICC also takes into account differences in the means of the measures being considered (J. Liu et al., \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). ICC values are bounded between 0 and 1, where, following the guidelines by (Koo \u0026amp; Li, \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2016\u003c/span\u003e) based on 95% confidence interval (CI), values less than 0.5 indicate a poor, values between 0.5 and 0.74 indicate moderate, values between 0.75 and 0.89 indicate good, and values between 0.9 and 1 indicate excellent reliability. As a complementary analysis, we performed one-way ANOVA for each parameter separately (McKeown et al., \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) to test if there are any significant differences between parameters obtained from S1, S2 and S3. In case of significant ANOVA result, post-hoc paired sample t-test was performed with a Bonferroni correction to control type-I error.\u003c/p\u003e\u003cp\u003eBland-Altman (BA) was used as a simple method to visualize and evaluate a bias between the mean differences, and to estimate 95% limits of agreement (LOA) (Bland \u0026amp; Altman, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e1986\u003c/span\u003e; Giavarina, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Bias is calculated as the mean difference between two measurements (in this study \u0026ndash; sessions) and LOA are estimated as \u0026plusmn;\u0026thinsp;1.96 standard deviations of the mean difference (bias) (Bland \u0026amp; Altman, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e1986\u003c/span\u003e). An ideal agreement between two measurements is zero and the smaller the range between LOA, the better the agreement is (Gerke, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Myles \u0026amp; Cui, \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). Since evaluated parameters have different ranges, instead of default approach, the difference for each subject was expressed as percentage of the observation represented on the X axis [(Differences between sessions / mean between sessions)*100] (Giavarina, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). A custom written MATLAB code was used to graphically plot the values between two sessions.\u003c/p\u003e\u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003eBased on the meta-criteria, the optimal number of clusters for all 3 sessions was 5. The normalized values of each cluster evaluation parameter across all 3 sessions are presented in Fig.\u0026nbsp;2A. All five microstates\u0026rsquo; maps for each session visually corresponded very well with topographies reported in the literature (Tarailis et al., \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) and meta-microstates (Koenig et al., \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Topographical maps were labelled according to the updated labelling system (Tarailis et al., \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) where topography with right frontal to left posterior configuration is labelled as MS A, topography with left frontal to right posterior \u0026ndash; MS B, topography with frontal to occipital configuration \u0026ndash; MS C, topography with fronto-central configuration \u0026ndash; MS D and topography with posterior configuration \u0026ndash; MS E. Topographies and shared spatial variance are presented in Fig.\u0026nbsp;2B and Fig.\u0026nbsp;2C.\u003c/p\u003e\u003cp\u003eFigure\u0026nbsp;2. A) Assessment of extracted microstates with different criterion implemented in Cartool. The best solution for each criterion is colored green. B) Group level topographies for each session. C). Shared spatial variance between extracted topographies.\u003c/p\u003e\u003cp\u003eTemporal parameters of duration and occurrence fell in line with 95% prediction intervals (Zanesco, \u003cspan citationid=\"CR97\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) while mean coverages of MS C were higher for all three sessions and average time coverages for MS D were lower for all three sessions than expected. Since the stability of spatiotemporal parameters and EEG syntax are out of scope of this study, descriptive statistics and ICC values for short-term and long-term reliability of spatiotemporal parameters and EEG microstates syntax are presented in supplementary material and will not be discussed here.\u003c/p\u003e\u003cp\u003eShort-term and long-term reliability\u003c/p\u003e\u003cp\u003eDescriptive statistics of Hurst exponent, nLZC, Microsynt, entropy and entropy rate for each session are summarized in the supplementary material (S. Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eHurst exponent\u003c/p\u003e\u003cp\u003eMean values and standard deviations of the Hurst exponent for each session were as follows: 0.646\u0026thinsp;\u0026plusmn;\u0026thinsp;0.040 for S1, 0.653\u0026thinsp;\u0026plusmn;\u0026thinsp;0.039 for S2 and 0.643\u0026thinsp;\u0026plusmn;\u0026thinsp;0.043 for S3. ANOVA did not reveal any significant differences between the sessions [F(2,177)\u0026thinsp;=\u0026thinsp;0.89, p\u0026thinsp;=\u0026thinsp;0.411]. ICC value for short-term reliability of Hurst exponent showed good consistency and bias was low. For long-term reliability, ICC value was in the moderate reliability range and the bias was once again close to zero, indicating high agreement. Short-term and long-term ICC values for reliability and agreement are summarized in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Bias and LOA are visualized in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003eA. For the interpretation of BA plots refer to \u003cem\u003eStatistical analysis\u003c/em\u003e section.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eIntraclass correlation coefficient values with lower and upper 95% CI and bias with Limits of Agreement (LOA) for Hurst exponent, normalized Lempel-Ziv complexity (nLZC), Microsynt, Shannon\u0026lsquo;s entropy and entropy rate.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e\u003cp\u003eShort-term\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e\u003cp\u003eLong-term\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"6\" nameend=\"c7\" namest=\"c2\"\u003e\u003cp\u003e\u003cb\u003eReliability\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003ePoint estimate\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003eLower 95% CI\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003eUpper 95% CI\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003ePoint estimate\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003eLower 95% CI\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003eUpper 95% CI\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHurst\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.843\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.744\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.905\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.670\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.504\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.789\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003enLZC\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.831\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.733\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.896\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.651\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.447\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.776\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMicrosynt\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.896\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.714\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.952\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.793\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.676\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.871\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEntropy\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.902\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.841\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.940\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.750\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.613\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.843\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEntropy rate\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.882\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.675\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.946\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.769\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.642\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.855\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"6\" nameend=\"c7\" namest=\"c2\"\u003e\u003cp\u003e\u003cb\u003eAgreement\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003eBias\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003eLower LOA\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003eUpper LOA\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003eBias\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003eLower LOA\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003eUpper LOA\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHurst\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-0.994\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-7.273\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e5.286\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.560\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-9.254\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e10.374\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003enLZC\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.768\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-4.024\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e5.459\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1.270\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-6.251\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e8.791\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMicrosynt\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.122\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-2.191\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e4.435\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.541\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-5.329\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e6.412\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEntropy\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.724\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-5.143\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e8.591\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e2.307\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-10.343\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e14.959\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEntropy rate\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2.888\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-5.479\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e11.254\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1.629\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-13.479\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e16.737\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eNormalized Lempel-Ziv complexity\u003c/p\u003e\u003cp\u003eMean values and standard deviations of the normalized LZ complexity for each session were as follows: 0.154\u0026thinsp;\u0026plusmn;\u0026thinsp;0.006 for S2, 0.153\u0026thinsp;\u0026plusmn;\u0026thinsp;0.007 for S2 and 0.152\u0026thinsp;\u0026plusmn;\u0026thinsp;0.009 for S3. ANOVA did not show significant differences of nLZC between sessions [F(2,177)\u0026thinsp;=\u0026thinsp;1.15, p\u0026thinsp;=\u0026thinsp;0.320]. ICC value for short-term reliability of normalized LZC showed good consistency with high agreement (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). For long-term reliability, ICC value was in the moderate reliability range and the agreement was high (low bias). Short-term and long-term ICC values for reliability and agreement are summarized in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Bias and LOA for nLZC are visualized in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003eB.\u003c/p\u003e\u003cp\u003eMicrosynt\u003c/p\u003e\u003cp\u003eMean values and standard deviations of the sequence complexity as measured with Microsynt method for each session were as follows: 258\u0026thinsp;\u0026plusmn;\u0026thinsp;10.76 for S1, 255.92\u0026thinsp;\u0026plusmn;\u0026thinsp;11.34 for S2 and 257.46\u0026thinsp;\u0026plusmn;\u0026thinsp;12.33 for S3. No significant differences were obtained between sessions for Microsynt [F(2,177)\u0026thinsp;=\u0026thinsp;0.93, p\u0026thinsp;=\u0026thinsp;0.397]. ICC value for short-term reliability indicated good consistency and high agreement. For long-term reliability, ICC value was in the moderate reliability range and agreement was close to zero (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Bias and LOA for complexity evaluated with Microsynt approach are visualized in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003eC.\u003c/p\u003e\u003cp\u003eEntropy\u003c/p\u003e\u003cp\u003eMean values and standard deviations of entropy for each session were as follows: 2.097\u0026thinsp;\u0026plusmn;\u0026thinsp;0.148 for S1, 2.063\u0026thinsp;\u0026plusmn;\u0026thinsp;0.164 for S2 and 2.053\u0026thinsp;\u0026plusmn;\u0026thinsp;0.189 for S3. Again. ANOVA did not reveal any differences of the measure between the sessions [F(2,177)\u0026thinsp;=\u0026thinsp;1.13, p\u0026thinsp;=\u0026thinsp;0.325]. ICC value for short-term reliability of entropy was just above excellent consistency threshold and the bias was close to zero (high agreement). For long-term reliability, ICC value dropped to the good reliability range, while agreement stayed high (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Entropy bias and LOA with 95% CI are visualized in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003eD.\u003c/p\u003e\u003cp\u003eEntropy rate\u003c/p\u003e\u003cp\u003eMean values and standard deviations of the entropy rate for each session were as follows: 0.417\u0026thinsp;\u0026plusmn;\u0026thinsp;0.040 for S1, 0.406\u0026thinsp;\u0026plusmn;\u0026thinsp;0.042 for S2 and 0.411\u0026thinsp;\u0026plusmn;\u0026thinsp;0.046 for S3. As indicated by ANOVA, no differences in entropy rates were observed entropy rate [F(2,177)\u0026thinsp;=\u0026thinsp;1.1, p\u0026thinsp;=\u0026thinsp;0.337]. ICC value for short-term reliability of entropy rate showed good consistency with high agreement. For long-term reliability, ICC value was in the good reliability range and agreement was high (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Bias and LOA for entropy rate are visualized in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003eE.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe aim of EEG microstate analysis is to describe the ongoing EEG patterns by a small set of representative templates. In this case the EEG can be seen as a process with a finite set of discrete states which evolves in continuous time. With the increasing interest in microstate approach as a tool to quantify resting state EEG in both normal and pathological conditions, a number of papers addressing the methodological questions regarding the topographical stability of EEG microstates (Zanesco, \u003cspan citationid=\"CR96\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), algorithm selection (Khanna et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; von Wegner et al., \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), potential bias in group level analysis (Murphy et al., \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), the impact of band-pass filtering settings (F\u0026eacute;rat, Seeber, et al., 2022) and electrode density (Zhang et al., \u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), relationship with EEG spectral amplitude (Zulliger et al., \u003cspan citationid=\"CR101\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), non-uniform labelling system (Custo et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Tarailis et al., \u003cspan citationid=\"CR84\" class=\"CitationRef\"\u003e2021\u003c/span\u003e, \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) the impact of temporal smoothing (Hermann et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) and ICA (Artoni \u0026amp; Michel, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) were addressed. However, only eight studies (Antonova et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Bagdasarov et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Q. Guo et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Khanna et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Kleinert et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; J. Liu et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Popov et al., \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Zhang et al., \u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) specifically focused on the reliability of temporal parameters. The knowledge on the reliability of parameters evaluate is crucial for further implementation of measures as potential personalized biomarkers.\u003c/p\u003e\u003cp\u003eNotably, alongside spatiotemporal parameters, microstate sequence can be evaluated, as it is believed that the order and transition between different microstates reflect dynamics necessary for cognitive functions (Eqlimi et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Haydock et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Jia et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Lehmann et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2005\u003c/span\u003e; Wackermann et al., \u003cspan citationid=\"CR93\" class=\"CitationRef\"\u003e1993\u003c/span\u003e). It was suggested that EEG syntax is controlled by a probabilistic mechanism where transitions from one microstate to another microstate occur with unequal probabilities, and that the current microstate class and the internally or externally received information co-determine the next microstate occurrence (Lehmann et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2005\u003c/span\u003e; Wackermann et al., \u003cspan citationid=\"CR93\" class=\"CitationRef\"\u003e1993\u003c/span\u003e). A handful of studies have reported altered transition probabilities in different clinical conditions (Nishida et al., \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Vellante et al., \u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) and during different experimental settings (Br\u0026eacute;chet et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Seitzman et al., \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Unfortunately, transition probabilities have never been included in the meta-analyses. Moreover, studies assessing test-retest reliability of transitions reported mixed results on reliability estimates ranging from poor to moderate (Antonova et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Kleinert et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), which raises a question: \u0026ndash; is microstate syntax reliable? Importantly, instead of focusing on the first or even second order Markovian transition probabilities, a handful of studies applied different theoretical information parameters to study various aspects of sequence characteristics - periodical activity, complexity, speed and randomness - demonstrating that it displays non-random higher order temporal structure. However, the reliability of outcomes was never assessed.\u003c/p\u003e\u003cp\u003eIn this study, using data-driven approach to extract the optimal number of microstates, we for the first time assessed the short-term (90 minutes) and long-term (30 days) test-retest reliability of different sequence characteristics - long-range memory as evaluated with Hurst exponent, complexity as computed with two different Lempel-Ziv algorithms, and sequence randomness and uncertainty measured with entropy and entropy rate - that can be utilized to define microstate sequence dynamics and reflect physiologically-relevant aspects.\u003c/p\u003e\u003cp\u003eSequence memory and complexity\u003c/p\u003e\u003cp\u003eWe showed that sequence memory of EEG microstates as estimated using Hurst exponent displays good short-term (ICC\u0026thinsp;=\u0026thinsp;0.843) and moderate long-term (ICC\u0026thinsp;=\u0026thinsp;0.670) reliability, indicating that it could be reliably used in different experimental settings. This was further substantiated by the absence of significant differences between sessions. Hurst exponent is a well-established method to assess the self-similarity within time series. The time scale-free dynamics indicates that the system is being maximally flexible for adapting to different kinds of upcoming stimuli (Chialvo, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Ros et al., \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) due to efficient and flexible information flow between multiple sources (Avramiea et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Kello et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Shew \u0026amp; Plenz, \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). In terms of functionality, it may be related to the structural memory of the activation of neural network that constrains perception and behavior (Lewis et al., \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Linkenkaer-Hansen et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2001\u003c/span\u003e, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). Also, it may define the dynamical regime of conscious resting-state activity (Demertzi et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Tagliazucchi et al., \u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). The long-range memory of EEG microstate sequence was previously estimated with a promising results in (pre-)clinical states (Lassi et al., \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Tomescu et al., \u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e2015\u003c/span\u003e), task-related conditions (Jia et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) and different sleep stages (von Wegner et al., \u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), all demonstrating affected long-range temporal dependencies.\u003c/p\u003e\u003cp\u003eSince sequence complexity increases with the length of the sequence, some sort of normalization is required to control the difference of sequence lengths between the subjects. Here, we normalized LZC values by the length of \u0026ldquo;Jump\u0026rdquo; sequence. Similarly to Hurst exponent, it displayed good short-term (ICC\u0026thinsp;=\u0026thinsp;0.831) and moderate long-term (ICC\u0026thinsp;=\u0026thinsp;0.651) stability and no differences between sessions were observed. In the case of Microsynt method, on the contrary, no normalization for window length was necessary, and the sliding window approach implemented in Microsynt method resulted in good short-term (ICC\u0026thinsp;=\u0026thinsp;0.896) and long-term reliability (ICC\u0026thinsp;=\u0026thinsp;0.793) with no differences between sessions. Importantly, LZC appears to be sensitive to clinical conditions and states of consciousness: studies reported decreased LZC in Alzheimer\u0026rsquo;s patients (Lassi et al., \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Tait et al., \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) and decreased LZC in non-REM sleep stages (von Wegner et al., \u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) and propofol-induced loss of consciousness (Artoni et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eSequence randomness and uncertainty\u003c/p\u003e\u003cp\u003eMany different entropy measures to quantify different aspects of randomness and predictability of time series (Delgado-Bonal \u0026amp; Marshak, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Keshmiri, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) exist and can be potentially used in the context of EEG microstates sequence analysis. For example, few studies reported sample entropy changes in early course of psychosis (Murphy et al., \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), obsessive compulsive disorder (Ren et al., \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), propofol induced loss of consciousness (Z. Liu et al., \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), yet no differences between genders was observed (Niu et al., \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Recently, changes of excess entropy from wakefulness to NREM sleep stages were demonstrated (von Wegner et al., \u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Here, we focused on Shannon\u0026rsquo;s entropy and entropy rate. Entropy showed excellent short-term reliability (ICC\u0026thinsp;=\u0026thinsp;0.902), while entropy rate showed good short-term reliability (ICC\u0026thinsp;=\u0026thinsp;0.882). Both measures displayed good reliability for long-term stability (ICC\u0026thinsp;=\u0026thinsp;0.750 and 0.769, respectively; no significant differences between sessions). Although entropy did not gain a lot of attention in the field of microstates so far, however, it was reported to be sensitive to levels of sleepiness (Wiemers et al., \u003cspan citationid=\"CR95\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). On the contrary, entropy rate of microstates sequence was shown to be affected by cognitive tasks (Jia et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), levels of sleepiness (von Wegner et al., \u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Wiemers et al., \u003cspan citationid=\"CR95\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) and propofol induced loss of consciousness (Hermann et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Zhang et al., \u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eConsiderations and limitations\u003c/p\u003e\u003cp\u003eEEG data by its nature contains a lot of unwanted noise. The signal-to-noise ratio decreases even more in the spontaneous data, where noise cannot be averaged out like with event-related potentials data. Even after rejection of artefacts and artefactual ICA components there is still some noise left, which could last for a few time frames. To reduce noise, especially at low GFP moments or during polarity inversions, post-processing of temporal smoothing and/or rejection of small segments are usually applied resulting in longer, more stable, less frequently changing segments. However, it alters the microstate sequence to a certain extent. Two recent studies showed that a common practice of backfitting template maps only on GFP peaks or using only sequences where all the duplicate states are removed (\u0026lsquo;Jump\u0026rsquo; sequences), can destroy EEG microstate sequence periodicity, decrease its complexity and increase its randomness (Hermann et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; von Wegner et al., \u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Although in this study we used all time frames for backfitting (not only GFP peaks) we also applied temporal smoothing that could potentially affect certain parameters, such as entropy rate and LZC.\u003c/p\u003e\u003cp\u003eOn the other hand, unsmoothed sequence results in shorter mean duration of microstates (~\u0026thinsp;20 ms) which goes against the interpretation of EEG microstates being related with different cognitive and physiological functions and brain states (Michel \u0026amp; Koenig, \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Tarailis et al., \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). This results in the double-edged sword situation, where authors have to choose: to apply temporal smoothing to reduce the noise, obtain temporal parameters that are similar to normative (Zanesco, \u003cspan citationid=\"CR97\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), but affect some aspects of sequence characteristics, or do not apply temporal smoothing, preserve certain aspects of sequence metrics, yet also preserve unwanted noise that results in short duration and high occurrence rate. The optimal approach remains an open question (Haydock et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe temporal parameters provide certain information about specific microstate which has its own functional role and underlying sources, and changes of those parameters are more straightforward to interpret. In contrast, sequence analysis methods are not microstate specific. As a result, the functional roles of different parameters become more difficult to interpret as they reflect the overall network dynamics. Nevertheless, EEG microstate sequence analysis could be a new frontier in the field.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eWe showed that microstate sequence parameters of Hurst exponent, (Microsynt) Lempel-Ziv complexity, entropy and entropy rate display moderate and good short-term and long-term reliability with no differences between sessions, outperforming reliability estimates of transition probabilities that range from poor to moderate. Inter-class correlation values reported here strongly suggest that dynamics microstates sequence represents stable traits of neural activity and could serve as a possible neurophysiological biomarker. Thus, further application of these and other sequence analysis methods in both clinical and experimental studies is encouraged.\u0026nbsp;\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAuthor Disclosure statement\u003c/h2\u003e\n\u003cp\u003eNo competing financial interests exist.\u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e\u003cp\u003eNo funding was used in support of this research.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003ePT: conceptualization, formal analysis, methodology software, visualization, writing original draft, reviewing and editing. FA: formal analysis, methodology, reviewing and editing. TK: reviewing and editing. CMM: supervision, reviewing and editing. IG-B: conceptualization, supervision, reviewing and editing.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eWe would like to thank the original authors who gathered and made available the de-identified data on which this manuscript is based. We also thank Frederic von Wegner for his help computing entropy rate.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eData is provided within the manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eAntonova, E., Holding, M., Suen, H. C., Sumich, A., Maex, R., \u0026amp; Nehaniv, C. (2022). EEG microstates: Functional significance and short-term test-retest reliability. \u003cem\u003eNeuroimage: Reports\u003c/em\u003e, \u003cem\u003e2\u003c/em\u003e(2), 100089. https://doi.org/10.1016/J.YNIRP.2022.100089\u003c/li\u003e\n \u003cli\u003eArtoni, F., Maillard, J., Britz, J., Brunet, D., Lysakowski, C., Tram\u0026egrave;r, M. R., \u0026amp; Michel, C. M. (2023). Microsynt: Exploring the syntax of EEG microstates. \u003cem\u003eNeuroImage\u003c/em\u003e, \u003cem\u003e277\u003c/em\u003e, 120196. https://doi.org/10.1016/J.NEUROIMAGE.2023.120196\u003c/li\u003e\n \u003cli\u003eArtoni, F., Maillard, J., Britz, J., Seeber, M., Lysakowski, C., Br\u0026eacute;chet, L., Tram\u0026egrave;r, M. R., \u0026amp; Michel, C. M. (2022). EEG microstate dynamics indicate a U-shaped path to propofol-induced loss of consciousness. \u003cem\u003eNeuroImage\u003c/em\u003e, \u003cem\u003e256\u003c/em\u003e, 119156. https://doi.org/10.1016/J.NEUROIMAGE.2022.119156\u003c/li\u003e\n \u003cli\u003eArtoni, F., \u0026amp; Michel, C. M. (2025). How does Independent Component Analysis Preprocessing Affect EEG Microstates? \u003cem\u003eBrain Topography\u003c/em\u003e, \u003cem\u003e38\u003c/em\u003e(2), 26. https://doi.org/10.1007/S10548-024-01098-4\u003c/li\u003e\n \u003cli\u003eAvramiea, A. E., Masood, A., Mansvelder, H. D., \u0026amp; Linkenkaer-Hansen, K. (2022). Long-Range Amplitude Coupling Is Optimized for Brain Networks That Function at Criticality. \u003cem\u003eJournal of Neuroscience\u003c/em\u003e, \u003cem\u003e42\u003c/em\u003e(11), 2221\u0026ndash;2233. https://doi.org/10.1523/JNEUROSCI.1095-21.2022\u003c/li\u003e\n \u003cli\u003eBagdasarov, A., Brunet, D., Michel, C. M., \u0026amp; Gaffrey, M. S. (2024). Microstate Analysis of Continuous Infant EEG: Tutorial and Reliability. \u003cem\u003eBrain Topography 2024\u003c/em\u003e, 1\u0026ndash;18. https://doi.org/10.1007/S10548-024-01043-5\u003c/li\u003e\n \u003cli\u003eBland, J. M., \u0026amp; Altman, D. G. (1986). Statistical methods for assessing agreement between two methods of clinical measurement. \u003cem\u003eLancet\u0026nbsp;\u003c/em\u003e, \u003cem\u003e1\u003c/em\u003e(8476), 307\u0026ndash;310. https://doi.org/doi.org/10.1016/S0140-6736(86)90837-8\u003c/li\u003e\n \u003cli\u003eBr\u0026eacute;chet, L., Brunet, D., Birot, G., Gruetter, R., Michel, C. M., \u0026amp; Jorge, J. (2019). Capturing the spatiotemporal dynamics of self-generated, task-initiated thoughts with EEG and fMRI. \u003cem\u003eNeuroImage\u003c/em\u003e, \u003cem\u003e194\u003c/em\u003e, 82\u0026ndash;92. https://doi.org/10.1016/j.neuroimage.2019.03.029\u003c/li\u003e\n \u003cli\u003eBrunet, D., Murray, M. M., \u0026amp; Michel, C. M. (2011). Spatiotemporal Analysis of Multichannel EEG: CARTOOL. \u003cem\u003eComputational Intelligence and Neuroscience\u003c/em\u003e, \u003cem\u003e2011\u003c/em\u003e, 15. https://doi.org/10.1155/2011/813870\u003c/li\u003e\n \u003cli\u003eChialvo, D. R. (2010). Emergent complex neural dynamics. \u003cem\u003eNature Physics\u003c/em\u003e, \u003cem\u003e6\u003c/em\u003e(10), 744\u0026ndash;750. https://doi.org/10.1038/nphys1803\u003c/li\u003e\n \u003cli\u003eChivu, A., Pascal, S. A., Damborsk\u0026aacute;, A., \u0026amp; Tomescu, M. I. (2023). EEG Microstates in Mood and Anxiety Disorders: A Meta-analysis. \u003cem\u003eBrain Topography\u003c/em\u003e. https://doi.org/10.1007/S10548-023-00999-0\u003c/li\u003e\n \u003cli\u003eComsa, I. M., Bekinschtein, T. A., \u0026amp; Srivas Chennu, \u0026middot;. (2019). Transient Topographical Dynamics of the Electroencephalogram Predict Brain Connectivity and Behavioural Responsiveness During Drowsiness. \u003cem\u003eBrain Topography\u003c/em\u003e, \u003cem\u003e32\u003c/em\u003e, 315\u0026ndash;331. https://doi.org/10.1007/s10548-018-0689-9\u003c/li\u003e\n \u003cli\u003eCusto, A., Van De Ville, D., Wells, W. M., Tomescu, M. I., Brunet, D., \u0026amp; Michel, C. M. (2017). Electroencephalographic Resting-State Networks: Source Localization of Microstates. \u003cem\u003eBrain Connectivity\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e(10), 671\u0026ndash;682. https://doi.org/10.1089/brain.2016.0476\u003c/li\u003e\n \u003cli\u003eDas, S., Zomorrodi, R., Enticott, P. G., Kirkovski, M., Blumberger, D. M., Rajji, T. K., \u0026amp; Desarkar, P. (2022). Resting state electroencephalography microstates in autism spectrum disorder: A mini-review. \u003cem\u003eFrontiers in Psychiatry\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e, 988939. https://doi.org/10.3389/FPSYT.2022.988939/BIBTEX\u003c/li\u003e\n \u003cli\u003eD\u0026rsquo;croz-Baron, D. F., Br\u0026eacute;chet, \u0026middot; Lucie, Baker, M., \u0026amp; Karp, T. (2021). Auditory and Visual Tasks Influence the Temporal Dynamics of EEG Microstates During Post-encoding Rest. \u003cem\u003eBrain Topography\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e, 19\u0026ndash;28. https://doi.org/10.1007/s10548-020-00802-4\u003c/li\u003e\n \u003cli\u003ede Bock, R., Mackintosh, A. J., Maier, F., Borgwardt, S., Riecher-R\u0026ouml;ssler, A., \u0026amp; Andreou, C. (2020). EEG microstates as biomarker for psychosis in ultra-high-risk patients. \u003cem\u003eTranslational Psychiatry\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e(1), 300. https://doi.org/10.1038/S41398-020-00963-7\u003c/li\u003e\n \u003cli\u003eDeiber, M. P., Piguet, C., Berchio, C., Michel, C. M., Perroud, N., \u0026amp; Ros, T. (2023). Resting-State EEG Microstates and Power Spectrum in Borderline Personality Disorder: A High-Density EEG Study. \u003cem\u003eBrain Topography\u003c/em\u003e. https://doi.org/10.1007/s10548-023-01005-3\u003c/li\u003e\n \u003cli\u003eDelgado-Bonal, A., \u0026amp; Marshak, A. (2019). Approximate Entropy and Sample Entropy: A Comprehensive Tutorial. \u003cem\u003eEntropy\u003c/em\u003e, \u003cem\u003e21\u003c/em\u003e(6). https://doi.org/10.3390/E21060541\u003c/li\u003e\n \u003cli\u003eDemertzi, A., Tagliazucchi, E., Dehaene, S., Deco, G., Barttfeld, P., Raimondo, F., Martial, C., Fern\u0026aacute;ndez-Espejo, D., Rohaut, B., Voss, H. U., Schiff, N. D., Owen, A. M., Laureys, S., Naccache, L., \u0026amp; Sitt, J. D. (2019). Human consciousness is supported by dynamic complex patterns of brain signal coordination. \u003cem\u003eScience Advances\u003c/em\u003e, \u003cem\u003e5\u003c/em\u003e(2). https://doi.org/10.1126/SCIADV.AAT7603/SUPPL_FILE/AAT7603_SM.PDF\u003c/li\u003e\n \u003cli\u003eDeolindo, C. S., Ribeiro, M. W., de Aratanha, M. A. A., Scarpari, J. R. S., Forster, C. H. Q., da Silva, R. G. A., Machado, B. S., Amaro Junior, E., K\u0026ouml;nig, T., \u0026amp; Kozasa, E. H. (2021). Microstates in complex and dynamical environments: Unraveling situational awareness in critical helicopter landing maneuvers. \u003cem\u003eHuman Brain Mapping\u003c/em\u003e, \u003cem\u003e42\u003c/em\u003e(10), 3168\u0026ndash;3181. https://doi.org/10.1002/hbm.25426\u003c/li\u003e\n \u003cli\u003eD\u0026iacute;az, H. A. M., \u0026amp; C\u0026oacute;rdova, F. (2021). On the meaning of Hurst entropy applied to EEG data series. \u003cem\u003eProcedia Computer Science\u003c/em\u003e, \u003cem\u003e199\u003c/em\u003e, 1385\u0026ndash;1392. https://doi.org/10.1016/j.procs.2022.01.175\u003c/li\u003e\n \u003cli\u003eDu, M., Peng, Y., Li, Y., Zhu, Y., Yang, S., Li, J., Zou, F., Wang, Y., Wu, X., Zhang, Y., \u0026amp; Zhang, M. (2022). Effect of trait anxiety on cognitive flexibility: Evidence from event-related potentials and resting-state EEG. \u003cem\u003eBiological Psychology\u003c/em\u003e, \u003cem\u003e170\u003c/em\u003e, 108319. https://doi.org/10.1016/J.BIOPSYCHO.2022.108319\u003c/li\u003e\n \u003cli\u003eEqlimi, E., Bockstael, A., Sch\u0026ouml;nwiesner, M., Talsma, D., \u0026amp; Botteldooren, D. (2023). Time course of EEG complexity reflects attentional engagement during listening to speech in noise. \u003cem\u003eEuropean Journal of Neuroscience\u003c/em\u003e. https://doi.org/10.1111/ejn.16159\u003c/li\u003e\n \u003cli\u003eF\u0026eacute;rat, V., Arns, M., Deiber, M. P., Hasler, R., Perroud, N., Michel, C. M., \u0026amp; Ros, T. (2022). Electroencephalographic Microstates as Novel Functional Biomarkers for Adult Attention-Deficit/Hyperactivity Disorder. \u003cem\u003eBiological Psychiatry: Cognitive Neuroscience and Neuroimaging\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e(8), 814\u0026ndash;823. https://doi.org/10.1016/J.BPSC.2021.11.006\u003c/li\u003e\n \u003cli\u003eF\u0026eacute;rat, V., Scheltienne, M., Brunet, D., Ros, T., \u0026amp; Michel, C. (2022). Pycrostates: a Python library to study EEG microstates. \u003cem\u003eJournal of Open Source Software\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e(78), 4564. https://doi.org/10.21105/joss.04564\u003c/li\u003e\n \u003cli\u003eF\u0026eacute;rat, V., Seeber, | Martin, Michel, C. M., \u0026amp; Ros, T. (2022). Beyond broadband: Towards a spectral decomposition of electroencephalography microstates. \u003cem\u003eHuman Brain Mapping\u003c/em\u003e. https://doi.org/10.1002/HBM.25834\u003c/li\u003e\n \u003cli\u003eGerke, O. (2020). Reporting standards for a bland-altman agreement analysis: A review of methodological reviews. In \u003cem\u003eDiagnostics\u003c/em\u003e (Vol. 10, Issue 5). MDPI. https://doi.org/10.3390/diagnostics10050334\u003c/li\u003e\n \u003cli\u003eGiavarina, D. (2015). Understanding Bland Altman analysis. \u003cem\u003eBiochemia Medica\u003c/em\u003e, \u003cem\u003e25\u003c/em\u003e(2), 141\u0026ndash;151. https://doi.org/10.11613/BM.2015.015\u003c/li\u003e\n \u003cli\u003eGschwind, M., Michel, C. M., \u0026amp; Van De Ville, D. (2015). Long-range dependencies make the difference-Comment on \u0026ldquo;A stochastic model for EEG microstate sequence analysis.\u0026rdquo; In \u003cem\u003eNeuroImage\u003c/em\u003e (Vol. 117, pp. 449\u0026ndash;455). Academic Press Inc. https://doi.org/10.1016/j.neuroimage.2015.05.062\u003c/li\u003e\n \u003cli\u003eGuo, P., Cui, J., Wang, Y., Zou, F., Wu, X., \u0026amp; Zhang, M. (2020). Spontaneous microstates related to effects of low socioeconomic status on neuroticism. \u003cem\u003eScientific Reports 2020 10:1\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e(1), 1\u0026ndash;8. https://doi.org/10.1038/s41598-020-72590-7\u003c/li\u003e\n \u003cli\u003eGuo, Q., Liu, S., Wang, L., Feng, K., \u0026amp; Yang, S. (2024). Analysis of microstate features for Parkinson\u0026rsquo;s disease based on reliability validation. \u003cem\u003eJournal of Neuroscience Methods\u003c/em\u003e, \u003cem\u003e406\u003c/em\u003e, 110115. https://doi.org/10.1016/J.JNEUMETH.2024.110115\u003c/li\u003e\n \u003cli\u003eHanoglu, L., Toplutas, E., Saricaoglu, M., Velioglu, H. A., Yildiz, S., \u0026amp; Yulug, B. (2022). Therapeutic Role of Repetitive Transcranial Magnetic Stimulation in Alzheimer\u0026rsquo;s and Parkinson\u0026rsquo;s Disease: Electroencephalography Microstate Correlates. \u003cem\u003eFrontiers in Neuroscience\u003c/em\u003e, \u003cem\u003e16\u003c/em\u003e. https://doi.org/10.3389/FNINS.2022.798558\u003c/li\u003e\n \u003cli\u003eHaydock, D., Kadir, S., Leech, R., Nehaniv, C. L., \u0026amp; Antonova, E. (2025). EEG Microstate Syntax Analysis: A Review of Methodological Challenges and Advances. \u003cem\u003eNeuroImage\u003c/em\u003e, 121090. https://doi.org/10.1016/J.NEUROIMAGE.2025.121090\u003c/li\u003e\n \u003cli\u003eHermann, G., T\u0026ouml;dt, I., Tagliazucchi, E., Todtenhaupt, I. K., Laufs, H., \u0026amp; von Wegner, F. (2024). Propofol Reversibly Attenuates Short-Range Microstate Ordering and 20 Hz Microstate Oscillations. \u003cem\u003eBrain Topography\u003c/em\u003e. https://doi.org/10.1007/s10548-023-01023-1\u003c/li\u003e\n \u003cli\u003eHu, N., Long, Q., Li, Q., Hu, X., Li, Y., Zhang, S., Chen, A., Huo, R., Liu, J., \u0026amp; Wang, X. (2021). The modulation of salience and central executive networks by acute stress in healthy males: An EEG microstates study. \u003cem\u003eInternational Journal of Psychophysiology\u003c/em\u003e, \u003cem\u003e169\u003c/em\u003e, 63\u0026ndash;70. https://doi.org/10.1016/J.IJPSYCHO.2021.09.001\u003c/li\u003e\n \u003cli\u003eJab\u0026egrave;s, A., Klencklen, G., Ruggeri, P., Michel, C. M., Banta Lavenex, P., \u0026amp; Lavenex, P. (2021). Resting‐State EEG Microstates Parallel Age‐Related Differences in Allocentric Spatial Working Memory Performance. \u003cem\u003eBrain Topography\u003c/em\u003e. https://doi.org/10.1007/s10548-021-00835-3\u003c/li\u003e\n \u003cli\u003eJia, W., von Wegner, F., Zhao, M., \u0026amp; Zeng, Y. (2021). Network oscillations imply the highest cognitive workload and lowest cognitive control during idea generation in open-ended creation tasks. \u003cem\u003eScientific Reports 2021 11:1\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(1), 1\u0026ndash;23. https://doi.org/10.1038/s41598-021-03577-1\u003c/li\u003e\n \u003cli\u003eKalburgi, S. N., Aryan, D., Nash, K., Schiller, B., \u0026amp; Koenig, T. (2023). \u003cem\u003eMICROSTATELAB: The EEGLAB toolbox for resting-state microstate analysis\u003c/em\u003e. https://doi.org/10.21203/rs.3.rs-3097311/v1\u003c/li\u003e\n \u003cli\u003eKe, M., Li, J., \u0026amp; Wang, L. (2021). Alteration in Resting-State EEG Microstates Following 24 Hours of Total Sleep Deprivation in Healthy Young Male Subjects. \u003cem\u003eFrontiers in Human Neuroscience\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e. https://doi.org/10.3389/FNHUM.2021.636252\u003c/li\u003e\n \u003cli\u003eKello, C. T., Brown, G. D. A., Ferrer-i-Cancho, R., Holden, J. G., Linkenkaer-Hansen, K., Rhodes, T., \u0026amp; Van Orden, G. C. (2010). Scaling laws in cognitive sciences. \u003cem\u003eTrends in Cognitive Sciences\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(5), 223\u0026ndash;232. https://doi.org/10.1016/j.tics.2010.02.005\u003c/li\u003e\n \u003cli\u003eKeshmiri, S. (2020). Entropy and the Brain: An Overview. \u003cem\u003eEntropy (Basel, Switzerland)\u003c/em\u003e, \u003cem\u003e22\u003c/em\u003e(9). https://doi.org/10.3390/E22090917\u003c/li\u003e\n \u003cli\u003eKhanna, A., Pascual-Leone, A., \u0026amp; Farzan, F. (2014). Reliability of resting-state microstate features in electroencephalography. \u003cem\u003ePLoS ONE\u003c/em\u003e, \u003cem\u003e9\u003c/em\u003e(12), e114163. https://doi.org/10.1371/journal.pone.0114163\u003c/li\u003e\n \u003cli\u003eKhanna, A., Pascual-Leone, A., Michel, C. M., \u0026amp; Farzan, F. (2015). Microstates in resting-state EEG: Current status and future directions. \u003cem\u003eNeuroscience and Biobehavioral Reviews\u003c/em\u003e, \u003cem\u003e49\u003c/em\u003e, 105\u0026ndash;113. https://doi.org/10.1016/j.neubiorev.2014.12.010\u003c/li\u003e\n \u003cli\u003eKleinert, T., Koenig, T., Nash, K., \u0026amp; Wascher, \u0026middot; Edmund. (2023). On the Reliability of the EEG Microstate Approach. \u003cem\u003eBrain Topography 2023\u003c/em\u003e, \u003cem\u003e1\u003c/em\u003e, 1\u0026ndash;16. https://doi.org/10.1007/S10548-023-00982-9\u003c/li\u003e\n \u003cli\u003eKoenig, T., Diezig, S., Kalburgi, S. N., Antonova, E., Artoni, F., Brechet, L., Britz, J., Croce, P., Custo, A., Damborsk\u0026aacute;, A., Deolindo, C., Heinrichs, M., Kleinert, T., Liang, Z., Murphy, M. M., Nash, K., Nehaniv, C., Schiller, B., Smailovic, U., \u0026hellip; Michel, C. M. (2023). EEG-Meta-Microstates: Towards a More Objective Use of Resting-State EEG Microstate Findings Across Studies. \u003cem\u003eBrain Topography 2023\u003c/em\u003e, 1\u0026ndash;14. https://doi.org/10.1007/S10548-023-00993-6\u003c/li\u003e\n \u003cli\u003eKoo, T. K., \u0026amp; Li, M. Y. (2016). A Guideline of Selecting and Reporting Intraclass Correlation Coefficients for Reliability Research. \u003cem\u003eJournal of Chiropractic Medicine\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e(2), 155\u0026ndash;163. https://doi.org/10.1016/j.jcm.2016.02.012\u003c/li\u003e\n \u003cli\u003eLassi, M., Fabbiani, C., Mazzeo, S., Burali, R., Vergani, A. A., Giacomucci, G., Moschini, V., Morinelli, C., Emiliani, F., Scarpino, M., Bagnoli, S., Ingannato, A., Nacmias, B., Padiglioni, S., Micera, S., Sorbi, S., Grippo, A., Bessi, V., \u0026amp; Mazzoni, A. (2023). Degradation of EEG microstates patterns in subjective cognitive decline and mild cognitive impairment: Early biomarkers along the Alzheimer\u0026rsquo;s Disease continuum? \u003cem\u003eNeuroImage: Clinical\u003c/em\u003e, \u003cem\u003e38\u003c/em\u003e, 103407. https://doi.org/10.1016/J.NICL.2023.103407\u003c/li\u003e\n \u003cli\u003eLehmann, D. (1990). Brain Electric Microstates and Cognition: The Atoms of Thought. In \u003cem\u003eMachinery of the Mind\u003c/em\u003e (pp. 209\u0026ndash;224). Birkh\u0026auml;user Boston. https://doi.org/10.1007/978-1-4757-1083-0_10\u003c/li\u003e\n \u003cli\u003eLehmann, D., Faber, P. L., Galderisi, S., Herrmann, W. M., Kinoshita, T., Koukkou, M., Mucci, A., Pascual-Marqui, R. D., Saito, N., Wackermann, J., Winterer, G., \u0026amp; Koenig, T. (2005). EEG microstate duration and syntax in acute, medication-na\u0026iuml;ve, first-episode schizophrenia: A multi-center study. \u003cem\u003ePsychiatry Research - Neuroimaging\u003c/em\u003e, \u003cem\u003e138\u003c/em\u003e(2), 141\u0026ndash;156. https://doi.org/10.1016/j.pscychresns.2004.05.007\u003c/li\u003e\n \u003cli\u003eLewis, C. M., Baldassarre, A., Committeri, G., Romani, G. L., \u0026amp; Corbetta, M. (2009). Learning sculpts the spontaneous activity of the resting human brain. \u003cem\u003eProceedings of the National Academy of Sciences of the United States of America\u003c/em\u003e, \u003cem\u003e106\u003c/em\u003e(41), 17558\u0026ndash;17563. https://doi.org/10.1073/PNAS.0902455106/SUPPL_FILE/0902455106SI.PDF\u003c/li\u003e\n \u003cli\u003eLinkenkaer-Hansen, K., Nikouline, V. V., Palva, J. M., \u0026amp; Ilmoniemi, R. J. (2001). Long-Range Temporal Correlations and Scaling Behavior in Human Brain Oscillations. \u003cem\u003eThe Journal of Neuroscience\u003c/em\u003e, \u003cem\u003e21\u003c/em\u003e(4), 1370. https://doi.org/10.1523/JNEUROSCI.21-04-01370.2001\u003c/li\u003e\n \u003cli\u003eLinkenkaer-Hansen, K., Nikulin, V. V., Palva, J. M., Kaila, K., \u0026amp; Ilmoniemi, R. J. (2004). Stimulus-induced change in long-range temporal correlations and scaling behaviour of sensorimotor oscillations. \u003cem\u003eEuropean Journal of Neuroscience\u003c/em\u003e, \u003cem\u003e19\u003c/em\u003e(1), 203\u0026ndash;218. https://doi.org/10.1111/J.1460-9568.2004.03116.X\u003c/li\u003e\n \u003cli\u003eLiu, J., Tang, W., Chen, G., Lu, Y., Feng, C., \u0026amp; Tu, X. M. (2016). Correlation and agreement: overview and clarification of competing concepts and measures. \u003cem\u003eShanghai Archives of Psychiatry\u003c/em\u003e, \u003cem\u003e28\u003c/em\u003e(2), 115\u0026ndash;120. https://doi.org/10.11919/j.issn.1002-0829.216045\u003c/li\u003e\n \u003cli\u003eLiu, J., Xu, J., Zou, G., He, Y., Zou, Q., \u0026amp; Gao, J. H. (2020). Reliability and Individual Specificity of EEG Microstate Characteristics. \u003cem\u003eBrain Topography\u003c/em\u003e, \u003cem\u003e33\u003c/em\u003e(4), 438\u0026ndash;449. https://doi.org/10.1007/s10548-020-00777-2\u003c/li\u003e\n \u003cli\u003eLiu, Z., Si, L., Xu, W., Zhang, K., Wang, Q., Chen, B., \u0026amp; Wang, G. (2022). Characteristics of EEG Microstate Sequences During Propofol-Induced Alterations of Brain Consciousness States. \u003cem\u003eIEEE Transactions on Neural Systems and Rehabilitation Engineering : A Publication of the IEEE Engineering in Medicine and Biology Society\u003c/em\u003e, \u003cem\u003e30\u003c/em\u003e, 1631\u0026ndash;1641. https://doi.org/10.1109/TNSRE.2022.3182705\u003c/li\u003e\n \u003cli\u003eLove, J., Selker, R., Marsman, M., Jamil, T., Dropmann, D., Verhagen, J., Ly, A., Gronau, Q. F., \u0026Scaron;m\u0026iacute;ra, M., Epskamp, S., Matzke, D., Wild, A., Knight, P., Rouder, J. N., Morey, R. D., \u0026amp; Wagenmakers, E. J. (2019). JASP: Graphical statistical software for common statistical designs. \u003cem\u003eJournal of Statistical Software\u003c/em\u003e, \u003cem\u003e88\u003c/em\u003e(1). https://doi.org/10.18637/jss.v088.i02\u003c/li\u003e\n \u003cli\u003eMcKeown, D. J., Finley, A. J., Kelley, N. J., Cavanagh, J. F., Keage, H. A. D., Baumann, O., Schinazi, V. R., Moustafa, A. A., \u0026amp; Angus, D. J. (2023). Test-retest reliability of spectral parameterization by 1/f characterization using SpecParam. \u003cem\u003eCerebral Cortex (New York, NY)\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e(1), bhad482. https://doi.org/10.1093/CERCOR/BHAD482\u003c/li\u003e\n \u003cli\u003eMichel, C. M., \u0026amp; Brunet, D. (2019). EEG source imaging: A practical review of the analysis steps. \u003cem\u003eFrontiers in Neurology\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e(APR), 325. https://doi.org/10.3389/fneur.2019.00325\u003c/li\u003e\n \u003cli\u003eMichel, C. M., \u0026amp; Koenig, T. (2018). EEG microstates as a tool for studying the temporal dynamics of whole-brain neuronal networks: A review. \u003cem\u003eNeuroImage\u003c/em\u003e, \u003cem\u003e180\u003c/em\u003e, 577\u0026ndash;593. https://doi.org/10.1016/j.neuroimage.2017.11.062\u003c/li\u003e\n \u003cli\u003eMilz, P., Faber, P. L., Lehmann, D., Koenig, T., Kochi, K., \u0026amp; Pascual-Marqui, R. D. (2016). The functional significance of EEG microstates-Associations with modalities of thinking. \u003cem\u003eNeuroImage\u003c/em\u003e, \u003cem\u003e125\u003c/em\u003e, 643\u0026ndash;656. https://doi.org/10.1016/j.neuroimage.2015.08.023\u003c/li\u003e\n \u003cli\u003eMurphy, M., Stickgold, R., \u0026amp; \u0026Ouml;ng\u0026uuml;r, D. (2020). Electroencephalogram Microstate Abnormalities in Early-Course Psychosis. \u003cem\u003eBiological Psychiatry: Cognitive Neuroscience and Neuroimaging\u003c/em\u003e, \u003cem\u003e5\u003c/em\u003e(1), 35\u0026ndash;44. https://doi.org/10.1016/J.BPSC.2019.07.006\u003c/li\u003e\n \u003cli\u003eMurphy, M., Wang, J., Jiang, C., Wang, L. A., Kozhemiako, N., Wang, Y., Wang, J., Jiang, C., Gai, G., Zou, K., Wang, Z., Yu, X., Wang, G., Tan, S., Murphy, M., Hall, M. H., Zhu, W., Zhou, Z., Shen, L., \u0026hellip; Purcell, S. M. (2023). A Potential Source of Bias in Group-Level EEG Microstate Analysis. \u003cem\u003eBrain Topography\u003c/em\u003e. https://doi.org/10.1007/s10548-023-00992-7\u003c/li\u003e\n \u003cli\u003eMurray, M. M., Brunet, D., \u0026amp; Michel, C. M. (2008). Topographic ERP analyses: A step-by-step tutorial review. In \u003cem\u003eBrain Topography\u003c/em\u003e (Vol. 20, Issue 4, pp. 249\u0026ndash;264). https://doi.org/10.1007/s10548-008-0054-5\u003c/li\u003e\n \u003cli\u003eMyles, P. S., \u0026amp; Cui, J. (2007). Using the Bland-Altman method to measure agreement with repeated measures. In \u003cem\u003eBritish Journal of Anaesthesia\u003c/em\u003e (Vol. 99, Issue 3, pp. 309\u0026ndash;311). Oxford University Press. https://doi.org/10.1093/bja/aem214\u003c/li\u003e\n \u003cli\u003eNazare, K., \u0026amp; Tomescu, M. I. (2024). Valence-specific EEG microstate modulations during self-generated affective states. \u003cem\u003eFrontiers in Psychology\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e, 1300416. https://doi.org/10.3389/FPSYG.2024.1300416\u003c/li\u003e\n \u003cli\u003eNishida, K., Morishima, Y., Yoshimura, M., Isotani, T., Irisawa, S., Jann, K., Dierks, T., Strik, W., Kinoshita, T., \u0026amp; Koenig, T. (2013). \u003cem\u003eEEG microstates associated with salience and frontoparietal networks in frontotemporal dementia, schizophrenia and Alzheimer\u0026rsquo;s disease\u003c/em\u003e. https://doi.org/10.1016/j.clinph.2013.01.005\u003c/li\u003e\n \u003cli\u003eNiu, Y., Chen, X., Chen, Y., Yao, Z., Chen, X., Liu, Z., Meng, X., Liu, Y., Zhao, Z., \u0026amp; Fan, H. (2024). A gender recognition method based on EEG microstates. \u003cem\u003eComputers in Biology and Medicine\u003c/em\u003e, \u003cem\u003e173\u003c/em\u003e. https://doi.org/10.1016/J.COMPBIOMED.2024.108366\u003c/li\u003e\n \u003cli\u003ePascual-Marqui, R. D., Michel, C. M., \u0026amp; Lehmann, D. (1995). Segmentation of Brain Electrical Activity into Microstates; Model Estimation and Validation. \u003cem\u003eIEEE Transactions on Biomedical Engineering\u003c/em\u003e, \u003cem\u003e42\u003c/em\u003e(7), 658\u0026ndash;665. https://doi.org/10.1109/10.391164\u003c/li\u003e\n \u003cli\u003ePopov, T., Tr\u0026ouml;ndle, M., Baranczuk-Turska, Z., Pfeiffer, C., Haufe, S., \u0026amp; Langer, N. (2023). Test\u0026ndash;retest reliability of resting-state EEG in young and older adults. \u003cem\u003ePsychophysiology\u003c/em\u003e, \u003cem\u003e60\u003c/em\u003e(7), e14268. https://doi.org/10.1111/PSYP.14268\u003c/li\u003e\n \u003cli\u003ePoulsen, A. T., Pedroni, A., Langer, N., \u0026amp; Hansen, L. K. (2018). Microstate EEGlab toolbox: An introductory guide. \u003cem\u003eBioRxiv\u003c/em\u003e, 1\u0026ndash;30. https://doi.org/10.1101/289850\u003c/li\u003e\n \u003cli\u003eRen, H., Ran, X., Qiu, M., Lv, S., Wang, J., Wang, C., Xu, Y., Gao, Z., Ren, W., Zhou, X., Mu, J., Yu, Y., \u0026amp; Zhao, Z. (2024). Abnormal nonlinear features of EEG microstate sequence in obsessive\u0026ndash;compulsive disorder. \u003cem\u003eBMC Psychiatry\u003c/em\u003e, \u003cem\u003e24\u003c/em\u003e(1), 881. https://doi.org/10.1186/S12888-024-06334-6\u003c/li\u003e\n \u003cli\u003eRieger, K., Hernandez, D., Baenninger, L., \u0026amp; Koenig, A. (2016). 15 Years of Microstate Research in Schizophrenia \u0026ndash; Where Are We? A Meta-Analysis. \u003cem\u003eA Meta-Analysis. Front. Psychiatry\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e, 22. https://doi.org/10.3389/fpsyt.2016.00022\u003c/li\u003e\n \u003cli\u003eRos, T., Frewen, P., Th\u0026eacute;berge, J., Michela, A., Kluetsch, R., Mueller, A., Candrian, G., Jetly, R., Vuilleumier, P., \u0026amp; Lanius, R. A. (2017). Neurofeedback Tunes Scale-Free Dynamics in Spontaneous Brain Activity. \u003cem\u003eCerebral Cortex\u003c/em\u003e, \u003cem\u003e27\u003c/em\u003e(10), 4911\u0026ndash;4922. https://doi.org/10.1093/CERCOR/BHW285\u003c/li\u003e\n \u003cli\u003eSchiller, B., Sperl, M. F. J., Kleinert, T., Nash, K., \u0026amp; Gianotti, L. R. R. (2023). EEG Microstates in Social and Affective Neuroscience. \u003cem\u003eBrain Topography 2023\u003c/em\u003e, \u003cem\u003e1\u003c/em\u003e, 1\u0026ndash;17. https://doi.org/10.1007/S10548-023-00987-4\u003c/li\u003e\n \u003cli\u003eSeitzman, B. A., Abell, M., Bartley, S. C., Erickson, M. A., Bolbecker, A. R., \u0026amp; Hetrick, W. P. (2017). Cognitive manipulation of brain electric microstates. \u003cem\u003eNeuroImage\u003c/em\u003e, \u003cem\u003e146\u003c/em\u003e, 533\u0026ndash;543. https://doi.org/10.1016/j.neuroimage.2016.10.002\u003c/li\u003e\n \u003cli\u003eShew, W. L., \u0026amp; Plenz, D. (2012). The Functional Benefits of Criticality in the Cortex. \u003cem\u003eHttp://Dx.Doi.Org/10.1177/1073858412445487\u003c/em\u003e, \u003cem\u003e19\u003c/em\u003e(1), 88\u0026ndash;100. https://doi.org/10.1177/1073858412445487\u003c/li\u003e\n \u003cli\u003eShrout, P. E., \u0026amp; Fleiss, J. L. (1979). Intraclass correlations: uses in assessing rater reliability. \u003cem\u003ePsychological Bulletin\u003c/em\u003e, \u003cem\u003e86\u003c/em\u003e(2), 420\u0026ndash;428. https://doi.org/10.1037//0033-2909.86.2.420\u003c/li\u003e\n \u003cli\u003eSmailovic, U., Koenig, T., Laukka, E. J., Kalpouzos, G., Andersson, T., Winblad, B., \u0026amp; Jelic, V. (2019). EEG time signature in Alzheimer\u0026acute;s disease: Functional brain networks falling apart. \u003cem\u003eNeuroImage: Clinical\u003c/em\u003e, \u003cem\u003e24\u003c/em\u003e. https://doi.org/10.1016/j.nicl.2019.102046\u003c/li\u003e\n \u003cli\u003eTagliazucchi, E., Von Wegner, F., Morzelewski, A., Brodbeck, V., Jahnke, K., \u0026amp; Laufs, H. (2013). Breakdown of long-range temporal dependence in default mode and attention networks during deep sleep. \u003cem\u003eProceedings of the National Academy of Sciences of the United States of America\u003c/em\u003e, \u003cem\u003e110\u003c/em\u003e(38), 15419\u0026ndash;15424. https://doi.org/10.1073/pnas.1312848110\u003c/li\u003e\n \u003cli\u003eTait, L., Tamagnini, F., Stothart, G., Barvas, E., Monaldini, C., Frusciante, R., Volpini, M., Guttmann, S., Coulthard, E., Brown, J. T., Kazanina, N., \u0026amp; Goodfellow, M. (2020). EEG microstate complexity for aiding early diagnosis of Alzheimer\u0026rsquo;s disease. \u003cem\u003eScientific Reports 2020 10:1\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e(1), 1\u0026ndash;10. https://doi.org/10.1038/s41598-020-74790-7\u003c/li\u003e\n \u003cli\u003eTait, L., \u0026amp; Zhang, J. (2022). +microstate: A MATLAB toolbox for brain microstate analysis in sensor and cortical EEG/MEG. \u003cem\u003eNeuroImage\u003c/em\u003e, \u003cem\u003e258\u003c/em\u003e, 119346. https://doi.org/10.1016/J.NEUROIMAGE.2022.119346\u003c/li\u003e\n \u003cli\u003eTakarae, Y., Zanesco, A., Keehn, B., Chukoskie, L., M\u0026uuml;ller, R. A., \u0026amp; Townsend, J. (2022). EEG microstates suggest atypical resting-state network activity in high-functioning children and adolescents with autism spectrum development. \u003cem\u003eDevelopmental Science\u003c/em\u003e. https://doi.org/10.1111/desc.13231\u003c/li\u003e\n \u003cli\u003eTarailis, P., Koenig, T., Michel, C. M., \u0026amp; Gri\u0026scaron;kova-Bulanova, I. (2023). The Functional Aspects of Resting EEG Microstates: A Systematic Review. \u003cem\u003eBrain Topography\u003c/em\u003e. https://doi.org/10.1007/S10548-023-00958-9\u003c/li\u003e\n \u003cli\u003eTarailis, P., \u0026Scaron;imkutė, D., Koenig, T., \u0026amp; Gri\u0026scaron;kova-Bulanova, I. (2021). Relationship between Spatiotemporal Dynamics of the Brain at Rest and Self-Reported Spontaneous Thoughts: An EEG Microstate Approach. \u003cem\u003eJournal of Personalized Medicine\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(11), 1216. https://doi.org/10.3390/jpm11111216\u003c/li\u003e\n \u003cli\u003eTomescu, M. I., Rihs, T. A., Roinishvili, M., Karahanoglu, F. I., Schneider, M., Menghetti, S., Van De Ville, D., Brand, A., Chkonia, E., Eliez, S., Herzog, M. H., Michel, C. M., \u0026amp; Cappe, C. (2015). Schizophrenia patients and 22q11.2 deletion syndrome adolescents at risk express the same deviant patterns of resting state EEG microstates: A candidate endophenotype of schizophrenia. \u003cem\u003eSchizophrenia Research: Cognition\u003c/em\u003e, \u003cem\u003e2\u003c/em\u003e(3), 159. https://doi.org/10.1016/J.SCOG.2015.04.005\u003c/li\u003e\n \u003cli\u003eVan De Ville, D., Britz, J., \u0026amp; Michel, C. M. (2010). EEG microstate sequences in healthy humans at rest reveal scale-free dynamics. \u003cem\u003eProceedings of the National Academy of Sciences of the United States of America\u003c/em\u003e, \u003cem\u003e107\u003c/em\u003e(42), 18179\u0026ndash;18184. https://doi.org/10.1073/PNAS.1007841107/-/DCSUPPLEMENTAL\u003c/li\u003e\n \u003cli\u003eVellante, F., Ferri, F., Baroni, G., Croce, P., Migliorati, D., Pettoruso, M., De Berardis, D., Martinotti, G., Zappasodi, F., \u0026amp; Giannantonio, M. Di. (2020). Euthymic bipolar disorder patients and EEG microstates: a neural signature of their abnormal self experience? \u003cem\u003eJournal of Affective Disorders\u003c/em\u003e, \u003cem\u003e272\u003c/em\u003e, 326\u0026ndash;334. https://doi.org/10.1016/j.jad.2020.03.175\u003c/li\u003e\n \u003cli\u003evon Wegner, F., Bauer, S., Rosenow, F., Triesch, J., \u0026amp; Laufs, H. (2021). EEG microstate periodicity explained by rotating phase patterns of resting-state alpha oscillations. \u003cem\u003eNeuroImage\u003c/em\u003e, \u003cem\u003e224\u003c/em\u003e, 117372. https://doi.org/10.1016/J.NEUROIMAGE.2020.117372\u003c/li\u003e\n \u003cli\u003evon Wegner, F., Knaut, P., \u0026amp; Laufs, H. (2018). EEG microstate sequences from different clustering algorithms are information-theoretically invariant. \u003cem\u003eFrontiers in Computational Neuroscience\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e. https://doi.org/10.3389/fncom.2018.00070\u003c/li\u003e\n \u003cli\u003evon Wegner, F., \u0026amp; Laufs, H. (2018). Information-theoretical analysis of EEG microstate sequences in python. \u003cem\u003eFrontiers in Neuroinformatics\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e, 30. https://doi.org/10.3389/FNINF.2018.00030/BIBTEX\u003c/li\u003e\n \u003cli\u003evon Wegner, F., Tagliazucchi, E., \u0026amp; Laufs, H. (2017). Information-theoretical analysis of resting state EEG microstate sequences - non-Markovianity, non-stationarity and periodicities. \u003cem\u003eNeuroImage\u003c/em\u003e, \u003cem\u003e158\u003c/em\u003e, 99\u0026ndash;111. https://doi.org/10.1016/j.neuroimage.2017.06.062\u003c/li\u003e\n \u003cli\u003evon Wegner, F., Wiemers, M., Hermann, \u0026middot; Gesine, T\u0026ouml;dt, I., Tagliazucchi, E., \u0026amp; Laufs, \u0026middot; Helmut. (2023). Complexity Measures for EEG Microstate Sequences: Concepts and Algorithms. \u003cem\u003eBrain Topography 2023\u003c/em\u003e, \u003cem\u003e1\u003c/em\u003e, 1\u0026ndash;16. https://doi.org/10.1007/S10548-023-01006-2\u003c/li\u003e\n \u003cli\u003eWackermann, J., Lehmann, D., Michel, C. M., \u0026amp; Strik, W. K. (1993). Adaptive segmentation of spontaneous EEG map series into spatially defined microstates. \u003cem\u003eInternational Journal of Psychophysiology\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(3), 269\u0026ndash;283. https://doi.org/10.1016/0167-8760(93)90041-M\u003c/li\u003e\n \u003cli\u003eWang, Y., Duan, W., Dong, D., Ding, L., \u0026amp; Lei, X. (2022). A test-retest resting, and cognitive state EEG dataset during multiple subject-driven states. \u003cem\u003eScientific Data\u003c/em\u003e, \u003cem\u003e9\u003c/em\u003e(1). https://doi.org/10.1038/s41597-022-01607-9\u003c/li\u003e\n \u003cli\u003eWiemers, M. C., Laufs, H., \u0026amp; von Wegner, F. (2023). Frequency Analysis of EEG Microstate Sequences in Wakefulness and NREM Sleep. \u003cem\u003eBrain Topography\u003c/em\u003e, \u003cem\u003e1\u003c/em\u003e, 1\u0026ndash;17. https://doi.org/10.1007/S10548-023-00971-Y/FIGURES/6\u003c/li\u003e\n \u003cli\u003eZanesco, A. P. (2020). EEG Electric Field Topography is Stable During Moments of High Field Strength. \u003cem\u003eBrain Topography\u003c/em\u003e, \u003cem\u003e33\u003c/em\u003e(4), 450\u0026ndash;460. https://doi.org/10.1007/s10548-020-00780-7\u003c/li\u003e\n \u003cli\u003eZanesco, A. P. (2023). Normative Temporal Dynamics of Resting EEG Microstates. \u003cem\u003eBrain Topography\u003c/em\u003e. https://doi.org/10.1007/S10548-023-01004-4\u003c/li\u003e\n \u003cli\u003eZanesco, A. P., Denkova, E., \u0026amp; Jha, A. P. (2020). Self-reported mind wandering and response time variability differentiate prestimulus electroencephalogram microstate dynamics during a sustained attention task. \u003cem\u003eJournal of Cognitive Neuroscience\u003c/em\u003e, \u003cem\u003e33\u003c/em\u003e(1), 28\u0026ndash;45. https://doi.org/10.1162/jocn_a_01636\u003c/li\u003e\n \u003cli\u003eZanesco, A. P., Skwara, A. C., King, B. G., Powers, C., Wineberg, K., \u0026amp; Saron, C. D. (2021). Meditation training modulates brain electric microstates and felt states of awareness. \u003cem\u003eHuman Brain Mapping\u003c/em\u003e, hbm.25430. https://doi.org/10.1002/hbm.25430\u003c/li\u003e\n \u003cli\u003eZhang, K., Shi, W., Wang, C., Li, Y., Liu, Z., Liu, T., Li, J., Yan, X., Wang, Q., Cao, Z., \u0026amp; Wang, G. (2021). Reliability of EEG microstate analysis at different electrode densities during propofol-induced transitions of brain states. \u003cem\u003eNeuroImage\u003c/em\u003e, \u003cem\u003e231\u003c/em\u003e. https://doi.org/10.1016/j.neuroimage.2021.117861\u003c/li\u003e\n \u003cli\u003eZulliger, J., Diaz Hernandez, L., \u0026amp; Koenig, T. (2022). Within and Between Subject Spectral Fingerprints of EEG-Microstate Parameters. \u003cem\u003eBrain Topography\u003c/em\u003e, \u003cem\u003e35\u003c/em\u003e(3), 277\u0026ndash;281. https://doi.org/10.1007/S10548-022-00896-Y/TABLES/1\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":true,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"cognitive-neurodynamics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"cody","sideBox":"Learn more about [Cognitive Neurodynamics](http://link.springer.com/journal/11571)","snPcode":"11571","submissionUrl":"https://submission.nature.com/new-submission/11571/3","title":"Cognitive Neurodynamics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"EEG microstates, test-retest reliability, sequence analysis, complexity, long-range memory, randomness","lastPublishedDoi":"10.21203/rs.3.rs-5875634/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5875634/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eEEG microstates sequence analysis gained a lot of attention in recent years and different sequence analysis methods have been applied to study microstates sequence randomness, complexity, speed, periodicity, and long-range memory. A few reliability studies reported somewhat consistent results of temporal parameters, yet no study so far addressed the within subject stability and reliability over time of different microstate sequence metrics. Here, we performed EEG microstate segmentation on data recorded from 60 healthy young adults and evaluated short-term (90 min), and long-term (30 days) reliability and agreement of EEG microstate sequence long-range memory as estimated with Hurst exponent, complexity as evaluated with two different Lempel-Ziv complexity algorithms, and its randomness as quantified with entropy and entropy rate. Our results showed mostly good short-term reliability across all 5 metrics (0.831\u0026thinsp;\u0026lt;\u0026thinsp;ICC\u0026thinsp;\u0026lt;\u0026thinsp;0.902), and moderate to good (0.651\u0026thinsp;\u0026lt;\u0026thinsp;ICC\u0026thinsp;\u0026lt;\u0026thinsp;0.793) long-term reliability. Reliability and agreement over time demonstrated in this work strongly suggests that microstates sequence dynamics is a stable trait of neural activity that can be utilised as a possible reliable neurophysiological biomarker.\u003c/p\u003e","manuscriptTitle":"Short-term and long-term test-retest reliability of memory, complexity, and randomness of EEG microstates sequence","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-20 05:43:20","doi":"10.21203/rs.3.rs-5875634/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-09-10T12:43:29+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-09-10T05:09:31+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-26T06:07:41+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"234342993846620258369371872369157386422","date":"2025-04-20T10:27:48+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-10T19:30:09+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"180124345087318190319150068885454470222","date":"2025-04-03T12:55:35+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-04-02T17:28:03+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-03-22T06:36:21+00:00","index":"","fulltext":""},{"type":"submitted","content":"Cognitive Neurodynamics","date":"2025-03-20T20:20:26+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"cognitive-neurodynamics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"cody","sideBox":"Learn more about [Cognitive Neurodynamics](http://link.springer.com/journal/11571)","snPcode":"11571","submissionUrl":"https://submission.nature.com/new-submission/11571/3","title":"Cognitive Neurodynamics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"d1b3d9f2-9a69-4e46-8f04-8b3dcd845a43","owner":[],"postedDate":"August 20th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-12-15T16:03:16+00:00","versionOfRecord":{"articleIdentity":"rs-5875634","link":"https://doi.org/10.1007/s11571-025-10391-2","journal":{"identity":"cognitive-neurodynamics","isVorOnly":false,"title":"Cognitive Neurodynamics"},"publishedOn":"2025-12-09 15:58:15","publishedOnDateReadable":"December 9th, 2025"},"versionCreatedAt":"2025-08-20 05:43:20","video":"","vorDoi":"10.1007/s11571-025-10391-2","vorDoiUrl":"https://doi.org/10.1007/s11571-025-10391-2","workflowStages":[]},"version":"v1","identity":"rs-5875634","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5875634","identity":"rs-5875634","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}