Neural Gain Dysregulation in Major Depression: Multidomain EEG Signal Analysis from Spectral Power to Network Topology

Neural Gain Dysregulation in Major Depression: Multidomain EEG Signal Analysis from Spectral Power to Network Topology | 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 Neural Gain Dysregulation in Major Depression: Multidomain EEG Signal Analysis from Spectral Power to Network Topology Kassra Ghassemkhani, Blake T. Dotta This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7456351/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Electroencephalography (EEG) offers a cost-effective window into the neural dynamics of Major Depressive Disorder (MDD), yet most studies focus on isolated domains of analysis. We present a multidomain computational framework spanning spectral, complexity, temporal, and network levels to test the hypothesis that MDD reflects dysregulation of neural gain. Using a publicly available resting-state EEG dataset, we quantified relative band power, nonlinear complexity (Higuchi fractal dimension, multiscale entropy), microstate dynamics, and graph-theoretic topology. MDD patients displayed significantly elevated beta power alongside higher short-scale entropy and fractal dimension values, indicating increased fast-frequency noise. Microstate analysis revealed reduced temporal stability and more frequent transitions, while graph-theoretic measures showed reduced small-worldness, particularly in the theta band, consistent with a shift toward random connectivity. Together, these results suggest that MDD is characterized by amplified fast-frequency activity with degraded signal-to-noise structure across temporal and network scales. This multidomain framework demonstrates how gain dysregulation manifests simultaneously in oscillatory, dynamical, and topological properties of EEG, offering a computational profile of depression that may support biomarker development and treatment monitoring. EEG Major Depressive Disorder neural gain spectral power fractal dimension entropy microstates graph theory small-world networks signal complexity Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 Figure 15 Figure 16 1. Introduction Major Depressive Disorder (MDD) is characterized by depressed mood and pleasure, difficulty with concentration, and rumination [ 1 – 3 ]. These symptoms are highlighted by the neurobiological features of MDD including frontal hypoactivity as assessed by positron emission tomography (PET), a neuroimaging device [ 4 ], such a reduction can impact backpropagating inhibitory projections to other cortical regions [ 5 ]. Scalp electroencephalography (EEG), while lacking spatial resolution, has a temporal resolution capable of capturing the activity at neuronal operation speeds in the range of milliseconds compared to neuroimaging devices which operate at a larger temporal scale [ 6 ]. The electric potentials measured by scalp electrodes can reflect underlying neuronal dynamics, as some properties are constant regardless of scale such as 1/f relationship between frequency and power [ 6 , 7 ]. With EEG, researchers have a variety of signal processing techniques at their disposal. Commonly used is spectral power density, wherein the squared amplitude at a given frequency is used to quantify the power or intensity, similar to how the “loudness” of a sound wave is quantified [ 8 ]. The spectral power within a range of frequencies can be summed to then produce absolute band power, which is common in EEG research. EEG frequency bands can include delta (1–4 Hz), theta (4–8 Hz), alpha (8–13 Hz), beta (13–30 Hz), and gamma (30 + Hz), although the frequency definitions of these bands may vary by researcher. To control inherent signal amplitude differences based on hair or skull size, many employ the relative power normalization, which is the ratio between absolute power of a given band to the total power of the broadband signal [ 9 ]. Typically reported in MDD is increases in beta power [ 10 – 15 ], and reductions in alpha power [ 10 , 12 ]. The ratio between two frequencies can also be used for analysis such as with the theta-to-beta ratio (TBR), which has shown to be lower than controls in attention-deficit hyperactive disorder (ADHD) [ 16 – 18 ], which relates to a reduction in theta power with increases in beta power. Not only does the TBR relate to attentional control but has also been shown to be reduced in MDD [ 19 ]. From the findings discussed above, MDD patients typically display fast frequency abnormalities regarding signal intensity in states where slow frequencies should be more dominant such as the eyes closed resting state. The concepts of information theory can be applied to EEG signals to quantify the information content, which can be done using the various complexity measures at a researcher’s disposal. First, fractal dimension analysis is useful for quantifying the space-filling properties of EEG signals, typically assessed using Higuchi’s algorithm for fractal dimension (HFD) wherein curve lengths are assessed at different temporal scales (k) [ 20 ]. The irregularity of a time series can also be quantified using sample entropy [ 21 ], wherein the data is embedded (m) with a user defined threshold for matching (r). Sample entropy can be observed at multiple different time lags between points on a time series to produce a measure of multiscale entropy [ 22 ]. Increases in both HFD and entropy have been observed in MDD patients [ 23 – 28 ]. This may indicate that EEG activity observed in depression not only shows an increase in fast frequency intensity, but also that the information is highly unpredictable, wherein excessive noise may have been introduced. Beyond observation of information content at the level of the channels, the spatiotemporal dynamics and transmission of information can be quantified using microstate analysis and graph theory metrics, respectively. Microstates are quasi-stable topographies of the spatial distribution of scalp potentials, typically lasting approximately 60 to 120 milliseconds and commonly observed in a narrow frequency range [ 29 – 31 ]. The four canonical microstates have been shown to account for approximately 70–80% of the explained variance of resting state brain activity [ 29 ]. In MDD, microstate activations have been shown to be more rapid, such as a decrease in microstate D average duration with increases in A and B occurrence compared to control [ 32 ]. These topographies have also been source localized to network activity with A, B, C, and D corresponding to auditory, visual, salience, and frontoparietal attention networks, respectively [ 33 ]. Network topology can be assessed through graph theory metrics such as the clustering coefficient (CC) and average path length (PL). The CC represents the local interconnectivity of nodes in a network, such as the formation of triangles [ 34 ]. The PL represents the average shortest number of steps between each node pair in a graph [ 34 ]. CC highlights the local interconnectedness within a network, whereas the PL highlights the global interconnectedness [ 34 ]. The network topology of a disordered or random network would have a low CC with a low PL, whereas a latticed or orderly network would have high CC and high average PL. Biological networks contain traits from both, which is optimized for information transfer wherein there is a high CC but low PL [ 35 ], these are the properties of small worldness networks. In MDD, reduced properties consistent with small worldness have been shown in theta and alpha band connectivity networks [ 36 ]. This indicates that network topology in MDD is closer to the properties seen in random or disordered networks compared to controls. Gain control refers to amplification of a signal, typically as gain increases so does input volume and also noise. Gain modulation as it pertains to neuronal function refers to the changing of input sensitivity of a given neuron while preserving its selectivity [ 37 – 39 ]. This change in neural gain depends on the background synaptic activity in a given neuron’s local environment, which can be altered during attentional tasks or changes in cognitive load [ 40 , 41 ]. For example, it has been shown that the presence of visual stimuli can induce long-lasting changes in background activity such as with resting membrane potentials in regions relevant to the stimulus [ 41 ]. Given the symptomatology of MDD pertaining to the thought components, it is possible that even in a restful state there could be abnormalities in gain control. With the amplification of fast frequencies in MDD as observed in spectral analysis, there is potential for the introduction of excess noise or variability as quantified by HFD and multiscale entropy. Given the potential regional increases in gain, the transfer of information within and between networks also becomes disorderly as assessed by microstate analysis and graph theory. For this experiment, we utilized a publicly available dataset [ 42 ], observing changes in power and unpredictability of EEG in MDD patients to build an EEG profile consistent with MDD using the discussed metrics. 2. Methods 2.1. Dataset Description For analysis we used a publicly available dataset consisting of eyes closed resting state recordings between MDD and healthy controls (HC) compiled by Mumtaz et al. (2017) [ 42 ]. In the sample was 34 MDD patients (29 recordings after exclusion criteria), and 30 in the HC group (28 after exclusion criteria). In the MDD group there were 17 males and 17 females, with a mean age of 40.3, the HC group was age-matched (mean age of 38.3) with 21 males and 9 females [ 42 ]. MDD patients met the diagnostic criteria as per the Diagnostic and Statistical Manual-IV, and underwent a 2 week medication washout period prior to recording. Data was collected with a 256 Hz sampling rate and a bandpass filter between 0.1 Hz and 70 Hz, with a 50 Hz notch filter for the removal of power line noise. Electrode placement followed the 10–20 system, and electrodes were referenced to linked ear reference electrodes. The amplifier used for recording was the Brain Master Discovery 24e [ 42 ]. 2.2. Preprocessing All preprocessing steps were done in the EEGLAB toolbox for MATLAB, the data was not downsampled from 256 Hz. Data was filtered with a high pass of 1.5 Hz and a low pass of 30 Hz using a finite impulse response (FIR) filter. This narrow filter was applied to control for high frequency noise arising from muscle or background sources [ 43 ]. Data was then referenced to the common average reference for further noise reduction and normalization. Further artifact correction was performed using independent components analysis (ICA) for the removal of eye-related, muscle, and background noise artifacts. The multiple artifact rejection algorithm (MARA) in tandem with visual inspection was used for the rejection of components pertaining to artifacts. The corrected data was then split into 5 second non-overlapping epochs, from which a compiled 30 seconds was used for analysis of the eyes closed resting state recordings. 2.3. Relative Band Power Relative band power was calculated for each channel using the Darbeliai extension for EEGLAB. First, the spectral power density (SPD) was calculated (µV²/Hz) with a 0.1 Hz step on the broadband signal (1.5–30 Hz). The absolute power was then calculated by summing the SPD values in the defined bands: delta (1.5-4 Hz), theta (4–8 Hz), alpha (8–13 Hz), and beta (13–30 Hz). The total power was also calculated for relative power normalization wherein the absolute power of a given band is divided by the total power to produce a value between 0 and 1 highlighting how much each band contributes to the total power spectrum. The 19-channel average of relative band power was primarily used in this analysis to capture global changes in signal. 2.4. Higuchi Fractal Dimension Signal complexity was first assessed using HFD. HFD was computed using open-source MATLAB code created by Selvam (2025), which utilizes the original Higuchi algorithm [ 20 , 44 ]. HFD assesses curve lengths ranging from a step size (k) of 1 to a user-defined maximum (kmax), wherein the HFD value is the relationship between curve lengths and temporal steps typically ranging from 1 and 2. The parameter, kmax, is the most sensitive and highly variable parameter across researchers in producing HFD values as no optimal method exists to select it [ 45 ]. We used a data-driven approach to select kmax as has been done before in EEG research using HFD [ 46 , 47 ]. First, we computed the 19-channel average for HFD from all subjects across kmax values ranging from 2 to the Nyquist limit of the dataset which was 128 (Fig. 1 ). From the average HFD value from each kmax, the median HFD value was selected for this analysis which corresponded to a kmax of 62. 2.5. Multiscale Sample Entropy To validate findings from HFD, we employed a second complexity technique to quantify irregularity at multiple time scales. Multiscale entropy was computed adapting functions from the get_entropy EEGLAB plugin [ 48 ]. The embedding dimension (m) was set to 2, which reconstructs the time series into pairs of points. For sample entropy, a separate time series is constructed using m + 1, or triplets in this instance. Whether pairs or triplets match is determined by the tolerance threshold (r) which was set to 0.2, or 20% of the signal standard deviation. Sample entropy is then calculated by taking the negative logarithm of the ratio between the number of matches with the m + 1 template to the number of matches with template m. Sample entropy was then calculated at multiple temporal scale factors (𝜏) on a coarse-grained version of the signal wherein a moving average is created based on 𝜏 to shorten the signal. 𝜏 refers to the time lag between points on a time series. We used values of 𝜏 ranging from 1 to 12 to leave a minimum of 100 points at the longest 𝜏 for accurate assessment of entropy given epoch duration and sampling frequency. For multiscale entropy analysis we utilized the average from scales 1–6 which were defined as short-scale entropy and the average from scales 7–12 which was defined as long-scale entropy, this was done for each channel and the 19-channel average (Fig. 2 ). 2.6. Microstate Analysis Microstate analysis was performed using the MICROSTATELAB plugin for EEGLAB [ 49 ]. Microstate analysis was performed with a 2–20 Hz frequency range as is common in microstate analysis [ 29 , 50 ]. First, individual microstates were identified based on global field power (GFP) peaks utilizing the k-means clustering algorithm. Next, microstate maps were generated based on the individual microstate maps. The mean maps were then sorted based on the meta maps template [ 51 ]. An example of the four classes of microstates used for analysis can be seen in Fig. 3 . Each of the extracted epochs were then backfitted to the sorted grand mean microstate maps, an example of a microstate activational time series can be observed in Fig. 4 . The temporal parameters of microstate activations were then exported. Used for analysis was microstate mean duration, mean occurrence, and coverage. Mean duration refers to the average duration of microstates by class in a given epoch, and the mean duration of microstates irrespective of class for a global measure. Mean occurrence refers to the average occurrences per second of each individual class, and the mean occurrence irrespective of class to gain insight on the rate of activation. Microstate coverage refers to the percentage of time spent in one of the four microstates. 2.7. Small Worldness Small worldness was computed using the functions from Fieldtrip and Brain Connectivity Toolbox for MATLAB [ 52 , 53 ]. First we computed the phase locking value (PLV), which quantifies the phase relationship between two signals over time providing a value between 0 and 1, with higher values indicating greater phase consistency between two signals. This was computed for the broadband signal (1.5–30 Hz), and individual frequency bands including delta (1.5-4 Hz), theta (4–8 Hz), alpha (8–13 Hz), and beta (13–30 Hz) band activity. From this, a resultant 19 x 19 adjacency matrix is then created for graph theory analysis. The adjacency matrix is first normalized using banalization by retaining the top 50% of connections. Following this, graph theory metrics are calculated including CC or tendency of a graph to form interconnected node clusters, and PL or the average number of steps from the shortest paths between node pairs. The simplified workflow can be observed in Fig. 5 . Based on the node degree, randomized networks using the Erdős–Rényi model were generated for the calculation of the small worldness index, for reproducibility the average CC and PL from 100 iterations were taken for reference. Small worldness index values are calculated by taking the ratio between the observed network to random network CC divided by the ratio between the observed network to random network PL. 2.8. Statistical Analysis All statistical analyses and boxplot creation were performed using the GraphPad Prism 10 software. Raincloud plots were created using RStudio (ggplot2 and ggrain) [ 54 ] and topographic EEG maps were created using a custom MATLAB script. Independent samples t tests were used for analysis between HC and MDD, unless F tests revealed significant differences wherein a non-parametric Mann-Whitney U test would be used. For multiple comparisons, we used false discovery rate (FDR) applied with a desired Q of 5% from the Benjamini & Hochberg method. Cohen’s d was used for effect size with an effect size cutoff of Cohen’d = 0.5. 3. Results 3.1. Relative Band Power Differences in Fast Frequencies Between MDD and HC When observing the 19-channel average for relative band power (Fig. 6 ), no significant differences were observed between MDD and HC for the delta band. While MDD displayed significantly lower relative theta (p = 0.04865) and alpha (p = 0.03252) power compared to HC, these comparisons did not pass the false discovery rate (FDR) correction. However, when observing relative beta power, MDD displayed significantly higher beta power compared to HC [t(55) = 5.663, p < 0.0001, q < 0.0001, Cohen’s d = 1.5069]. Figure 7 highlights a topographic map of the channel averages for the four calculated frequency bands. Figure 8 highlights statistical significance using t values comparing MDD to HC, wherein significantly lower alpha can be observed in sensors C3 and C4, with significance found in each channel for the beta band after the FDR correction. 3.2. Higher Higuchi Fractal Dimension Displayed in MDD With regards to signal complexity, MDD displayed significantly higher HFD values compared to HC [t(55) = 4.086, p = 0.0001, Cohen’s d = 1.0575] when comparing the 19-channel averages (Fig. 9 ). Topographic plots of the channel averages for each complexity metric (HFD, short-scale entropy, long-scale entropy) can be observed in Fig. 10 . MDD displaying greater HFD compared to HC was revealed in every channel except Fp1, Fp2, F7, F8, and O2 after FDR correction. Single channel t values between MDD vs HC for complexity metrics are shown in Fig. 11 . 3.3. Short-Scale Entropy Increases in MDD When observing the 19-channel average for short-scale entropy (Fig. 12 A), MDD displays significantly higher entropy compared to HC [t(55) = 4.204, p < 0.0001, Cohen’s d = 1.1155]. However when comparing long-scale entropy (Fig. 12 B), no significant difference was observed between MDD and HC. Single channel differences in short-scale entropy after FDR show MDD has significantly higher entropy compared to HC in each channel except T3. However no single channel differences were observed in long-scale entropy after FDR as can be seen in the t value topographic plots in Fig. 11 . 3.4. Reduced Temporal Stability of Microstates in MDD When observing mean microstate duration irrespective of class (Fig. 13 A), MDD displayed significantly lower durations compared to HC [U = 197, p = 0.0006, Cohen's d = 0.9219]. When observing individual microstate classes after FDR correction, MDD displayed significantly lower microstate durations in microstate class B [t(55) = 2.997, p = 0.0040, q = 0.0122, Cohen’s d = 0.7904] and class D [t(55) = 2.851, p = 0.0061, q = 0.0122, Cohen’s d = 0.7518] but not class A and class C compared to HC (Fig. 14 A). When observing mean microstate occurrence irrespective of class (Fig. 13 B), MDD displayed significantly higher occurrence compared to HC [t(55) = 3.654, p = 0.0006, Cohen's d = 0.9670]. When observing single classes significantly higher occurrences were displayed by MDD for microstate classes A [t(55) = 2.765, p = 0.0077, q = 0.0154, Cohen’s d = 0.7305], class B [t(55) = 2.335, p = 0.0232, q = 0.0309, Cohen’s d = 0.6176], and class C [t(55) = 3.014, p = 0.0038, q = 0.0154, Cohen’s d = 0.7978] after FDR correction (Fig. 14 B). When observing coverage, while MDD displayed significantly higher coverage of class A (p = 0.0431) and significantly lower coverage of class D (p = 0.0468), these comparisons did not pass the FDR correction (Fig. 14 C). 3.5. Reduced Small Worldness Properties of Channel Networks in MDD From the graph theory metric, small worldness index, when observing the broadband (1.5–30 Hz) graphs (Fig. 15 ), MDD displayed significantly lower small worldness index values compared to HC [t(55) = 2.545, p = 0.0138, Cohen's d = 0.6747]. Upon observing individual frequencies (Fig. 16 ), MDD also displayed significantly lower small worldness index values in the theta band [t(55) = 2.906, p = 0.0052, q = 0.021, Cohen’s d = 0.7616] compared to HC. No significant differences were observed in delta, alpha, or beta bands for small worldness index values after FDR correction between MDD and HC. 4. Discussion From the results, observed was the resting state signal in MDD not only possessing greater intensity at high frequencies but the information content also in excess at short-temporal scales. The excessive information is also highlighted in the increase in HFD, wherein the scaling behaviours between curve length and temporal scale were greater in MDD compared to HC. It is possible that the increased short-scale entropy, but not long-scale, highlights not only the power increased at fast frequencies, but that the fast frequency activity in itself is more unpredictable than the activity from HC. Not only does the information content yield greater unpredictability at the level of the channels, but also how information is transmitted between channels as assessed by graph theory, and the temporal dynamics of brain states assessed by microstate analysis. The increase in power at fast frequencies potentially reflects a tonic increase in gain or sensitivity. Similar to how increasing gain on an audio device can introduce noise, this increase in beta power was also accompanied by noise as quantified through complexity metrics. Theta and alpha are highly associated with modulation of faster rhythms or cyclic inhibitory activity [ 55 , 56 ], despite the comparisons not passing FDR correction, the decreases in both frequencies potentially relate to the irregularity in the fast frequency range due to a lack of modulation. With this introduction to excessive noise, the transmission of information from region to region can also be rendered difficult. Microstates have seen much usage in current EEG literature [ 50 , 57 – 59 ] and have been termed “atoms of thought”, as their temporal dynamics can reflect different forms of thought and cognitive processing [ 30 ]. MDD was characterized by increased occurrences with reduced durations irrespective of microstate class. This shows that microstate transitions are considerably more rapid, and given their association with network activity [ 33 ], could reflect altered connectivity and synchrony patterns. From analysis of network topology, we wanted to test unpredictability using the small worldness index. A reduced small worldness index indicates a deviation towards the network topologies observed in random networks such as Erdős–Rényi models, wherein the interconnectivity of nodes is low in tandem with low average steps [ 34 ]. The dynamics between scalp electrodes in MDD displayed reduced small worldness properties compared to controls when observing the broadband signal and theta for individual frequencies. Given the modulatory features observed with theta, a random connectivity pattern also potentially gives rise to the fast frequency unpredictability [ 55 , 60 ]. Reduced theta to beta phase amplitude coupling in the right hemisphere between frontal and parietal sensors has shown to be associated with increased psychological distress [ 60 ]. These shifts in the temporal and topological components of network activity highlight not only single regional unpredictability, but global unpredictability in region-to-region communication. If the features discussed are present, one can monitor EEG to track treatment outcomes such as following neurostimulation techniques or pharmacological intervention. Following treatment with electroconvulsive therapy reductions in entropy have been found in MDD patients who are considered ECT responders (50% reduction in Hamilton Rating Scale for Depression) compared to non-responders [ 25 ]. Microstate dynamics have also been shown to improve with treatment in MDD, wherein the occurrences of classes A and B were decreased and durations of class D were increased following 8 weeks of treatment with selective serotonin reuptake inhibitors (SSRI) [ 32 ]. SSRI treatment has also yielded results in graph theory metrics wherein average clustering coefficients for the alpha band have shown to increase in responders compared to non responders [ 61 ]. Graded potentials and neuromodulatory input are important in the control of neural gain [ 37 – 39 ]. Serotonergic abnormalities present in MDD [ 62 – 64 ] could be reducing neuromodulation and disrupting the excitatory inhibitory gradient. Following SSRI treatment, levels of gamma-aminobutyric acid (GABA) and glutamate have shown to be increased and decreased, respectively [ 65 – 67 ], possibly indicating greater excitatory tone present in MDD which increases neural gain. Further, the hypothalamic-pituitary-adrenal axis (HPA) has been suggested to display hyperactivity in MDD as evident by basal concentrations of hormones associated with the HPA circuit [ 68 – 70 ]. Research with goldfish as a model, has shown cortisol to produce features consistent with increased neural gain such as long-lasting increases in post-synaptic potential magnitudes, membrane input resistance, and reduced threshold current in cells dedicated to startle responses [ 71 ]. Neurostimulation techniques such as ECT also have efficacy in treatment resistant depression as they have the capability to reduce cortical excitability long-term [ 72 ]. Much current research on MDD using EEG following treatment incorporates techniques from a single domain of information from the signal. Future research can use a combination of techniques that quantify spectral features, complexity, and the transfer of information such as graph theory to track treatment related changes in MDD. This ensures that the changes in signal unpredictability following treatment are not only at the level of a single channel but also influence how regions communicate as well. Declarations Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Author Contribution KG: Conceptualization; Data curation; Formal analysis; Visualization; Writing – original draft; Writing – review & editing.BTD: Conceptualization; Supervision; Visualization; Writing – review & editing. Data Availability The EEG dataset analyzed in this study is publicly available from Mumtaz et al. (2017) in PLOS ONE (https://doi.org/10.1371/journal.pone.0171409). Analyses were conducted using open-source MATLAB toolboxes (EEGLAB, FieldTrip, Brain Connectivity Toolbox, MICROSTATELAB). Custom MATLAB scripts used for visualization are available from the corresponding author upon reasonable request. References Su, Y. A., & Si, T. (2022). Progress and challenges in research of the mechanisms of anhedonia in major depressive disorder. General psychiatry , 35 (1), e100724. https://doi.org/10.1136/gpsych-2021-100724 Perini, G., Cotta Ramusino, M., Sinforiani, E., et al. (2019). Cognitive impairment in depression: recent advances and novel treatments. Neuropsychiatric disease and treatment , 15 , 1249–1258. https://doi.org/10.2147/NDT.S199746 Alderman, B. L., Olson, R. L., Bates, M. E., et al. (2015). Rumination in major depressive disorder is associated with impaired neural activation during conflict monitoring. Frontiers in human neuroscience , 9 , 269. https://doi.org/10.3389/fnhum.2015.00269 Videbech P. (2000). PET measurements of brain glucose metabolism and blood flow in major depressive disorder: a critical review. Acta psychiatrica Scandinavica , 101 (1), 11–20. https://doi.org/10.1034/j.1600-0447.2000.101001011.x Bravo-Rivera, C., Diehl, M. M., Roman-Ortiz, C., et al. (2015). Long-range GABAergic neurons in the prefrontal cortex modulate behavior. Journal of neurophysiology , 114 (3), 1357–1359. https://doi.org/10.1152/jn.00861.2014 Buzsáki, G. (2006). Rhythms of the brain. Oxford University Press. Li, C. Y., Poo, M. M., & Dan, Y. (2009). Burst spiking of a single cortical neuron modifies global brain state. Science (New York, N.Y.) , 324 (5927), 643–646. https://doi.org/10.1126/science.1169957 Gyurkovics, M., Clements, G. M., Low, K. A. (2021). The impact of 1/f activity and baseline correction on the results and interpretation of time-frequency analyses of EEG/MEG data: A cautionary tale. NeuroImage , 237 , 118192. https://doi.org/10.1016/j.neuroimage.2021.118192 Nishiyori, R., Xiao, R., Vanderbilt, D., & Smith, B. A. (2021). Electroencephalography measures of relative power and coherence as reaching skill emerges in infants born preterm. Scientific reports , 11 (1), 3609. https://doi.org/10.1038/s41598-021-82329-7 Özçoban, M. A., & Tan, O. (2025). Electroencephalographic markers in Major Depressive Disorder: insights from absolute, relative power, and asymmetry analyses. Frontiers in psychiatry , 15 , 1480228. https://doi.org/10.3389/fpsyt.2024.1480228 Koshiyama, D., Kirihara, K., Usui, K., et al. (2020). Resting-state EEG beta band power predicts quality of life outcomes in patients with depressive disorders: A longitudinal investigation. Journal of affective disorders , 265 , 416–422. https://doi.org/10.1016/j.jad.2020.01.030 Tatti, E., Cinti, A., Serbina, A., et al. (2024). Resting-State EEG Alterations of Practice-Related Spectral Activity and Connectivity Patterns in Depression. Biomedicines , 12 (9), 2054. https://doi.org/10.3390/biomedicines12092054 Knott, V., Mahoney, C., Kennedy, S., & Evans, K. (2001). EEG power, frequency, asymmetry and coherence in male depression. Psychiatry research , 106 (2), 123–140. https://doi.org/10.1016/s0925-4927(00)00080-9 Sang, Q., Chen, C., & Shao, Z. (2025). Decoding depression from different brain regions using hybrid machine learning methods. Bioengineering, 12 (5), 449. https://doi.org/10.3390/bioengineering12050449 Kovacevic, N., Meghdadi, A., Berka, C., et al. (2025). Differences in resting state and task-based EEG measures between patients with major depressive disorder and healthy controls. Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology , 173 , 190–198. https://doi.org/10.1016/j.clinph.2025.03.022 Picken, C., Clarke, A. R., Barry, R. J., et al. (2020). The Theta/Beta Ratio as an Index of Cognitive Processing in Adults With the Combined Type of Attention Deficit Hyperactivity Disorder. Clinical EEG and neuroscience , 51 (3), 167–173. https://doi.org/10.1177/1550059419895142 Putman, P., Verkuil, B., Arias-Garcia, E., Pantazi, I., & van Schie, C. (2014). EEG theta/beta ratio as a potential biomarker for attentional control and resilience against deleterious effects of stress on attention. Cognitive, affective & behavioral neuroscience , 14 (2), 782–791. https://doi.org/10.3758/s13415-013-0238-7 Wang, T. S., Wang, S. S., Wang, C. L., & Wong, S. B. (2024). Theta/beta ratio in EEG correlated with attentional capacity assessed by Conners Continuous Performance Test in children with ADHD. Frontiers in psychiatry , 14 , 1305397. https://doi.org/10.3389/fpsyt.2023.1305397 Chang, J., & Choi, Y. (2023). Depression diagnosis based on electroencephalography power ratios. Brain and behavior , 13 (8), e3173. https://doi.org/10.1002/brb3.3173 Higuchi, T. (1988). Approach to an irregular time series on the basis of the fractal theory. Physica D: Nonlinear Phenomena, 31(2), 277–283. https://doi.org/10.1016/0167-2789(88)90081-4 Richman, J. S., & Moorman, J. R. (2000). Physiological time‑series analysis using approximate entropy and sample entropy. American Journal of Physiology – Heart and Circulatory Physiology, 278 (6), H2039–H2049. https://doi.org/10.1152/ajpheart.2000.278.6.H2039 Costa, M., Goldberger, A. L., & Peng, C.-K. (2002). Multiscale entropy analysis of complex physiologic time series. Physical Review Letters, 89 (6), 068102. https://doi.org/10.1103/PhysRevLett.89.068102 Akar, S. A., Kara, S., Agambayev, S., & Bilgic, V. (2015). Nonlinear analysis of EEG in major depression with fractal dimensions. Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference , 2015 , 7410–7413. https://doi.org/10.1109/EMBC.2015.7320104 Yun, S., & Jeong, B. (2021). Aberrant EEG signal variability at a specific temporal scale in major depressive disorder. Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology , 132 (8), 1866–1877. https://doi.org/10.1016/j.clinph.2021.05.011 Farzan F., Atluri, S., Mei, Y. et al. (2017). Brain temporal complexity in explaining the therapeutic and cognitive effects of seizure therapy, Brain , 140 (4), 1011–1025, https://doi.org/10.1093/brain/awx030 Bachmann, M., Lass, J., Suhhova, A., & Hinrikus, H. (2013). Spectral asymmetry and Higuchi's fractal dimension measures of depression electroencephalogram. Computational and mathematical methods in medicine , 2013 , 251638. https://doi.org/10.1155/2013/251638 Čukić, M., Stokić, M., Simić, S., & Pokrajac, D. (2020). The successful discrimination of depression from EEG could be attributed to proper feature extraction and not to a particular classification method. Cognitive neurodynamics , 14 (4), 443–455. https://doi.org/10.1007/s11571-020-09581-x de Freitas, S. B., Marques, A. A., Bevilaqua, M. C., de Carvalho, M. R., Ribeiro, P., Palmer, S., Nardi, A. E., & Dias, G. P. (2016). Electroencephalographic findings in patients with major depressive disorder during cognitive or emotional tasks: a systematic review. Revista brasileira de psiquiatria (Sao Paulo, Brazil : 1999) , 38 (4), 338–346. https://doi.org/10.1590/1516-4446-2015-1834 Koenig, T., Prichep, L., Lehmann, D., et al. (2002). Millisecond by millisecond, year by year: normative EEG microstates and developmental stages. NeuroImage , 16 (1), 41–48. https://doi.org/10.1006/nimg.2002.1070 Lehmann, D., Strik, W. K., Henggeler, B., et al. (1998). Brain electric microstates and momentary conscious mind states as building blocks of spontaneous thinking: I. Visual imagery and abstract thoughts. International journal of psychophysiology : official journal of the International Organization of Psychophysiology , 29 (1), 1–11. https://doi.org/10.1016/s0167-8760(97)00098-6 Lehmann, D., Ozaki, H., & Pal, I. (1987). EEG alpha map series: brain micro-states by space-oriented adaptive segmentation. Electroencephalography and clinical neurophysiology , 67 (3), 271–288. https://doi.org/10.1016/0013-4694(87)90025-3 Lei, L., Liu, Z., Zhang, Y., et al. (2022). EEG microstates as markers of major depressive disorder and predictors of response to SSRIs therapy. Progress in neuro-psychopharmacology & biological psychiatry , 116 , 110514. https://doi.org/10.1016/j.pnpbp.2022.110514 Custo, A., Van De Ville, D., Wells, et al. (2017). Electroencephalographic Resting-State Networks: Source Localization of Microstates. Brain connectivity , 7 (10), 671–682. https://doi.org/10.1089/brain.2016.0476 Smit, D. J., Stam, C. J., Posthuma, D., et al. (2008). Heritability of "small-world" networks in the brain: a graph theoretical analysis of resting-state EEG functional connectivity. Human brain mapping , 29 (12), 1368–1378. https://doi.org/10.1002/hbm.20468 Telesford, Q. K., Joyce, K. E., Hayasaka, S., et al. (2011). The ubiquity of small-world networks. Brain connectivity , 1 (5), 367–375. https://doi.org/10.1089/brain.2011.0038 Shim, M., Im, C. H., Kim, Y. W., & Lee, S. H. (2018). Altered cortical functional network in major depressive disorder: A resting-state electroencephalogram study. NeuroImage. Clinical , 19 , 1000–1007. https://doi.org/10.1016/j.nicl.2018.06.012 Salinas, E., & Sejnowski, T. J. (2001). Gain modulation in the central nervous system: where behavior, neurophysiology, and computation meet. The Neuroscientist : a review journal bringing neurobiology, neurology and psychiatry , 7 (5), 430–440. https://doi.org/10.1177/107385840100700512 Ferguson, K. A., & Cardin, J. A. (2020). Mechanisms underlying gain modulation in the cortex. Nature reviews. Neuroscience , 21 (2), 80–92. https://doi.org/10.1038/s41583-019-0253-y Chance, F. S., Abbott, L. F., & Reyes, A. D. (2002). Gain modulation from background synaptic input. Neuron , 35 (4), 773–782. https://doi.org/10.1016/s0896-6273(02)00820-6 Thiele, A., & Bellgrove, M. A. (2018). Neuromodulation of Attention. Neuron , 97 (4), 769–785. https://doi.org/10.1016/j.neuron.2018.01.008 Cardin, J. A., Palmer, L. A., & Contreras, D. (2008). Cellular mechanisms underlying stimulus-dependent gain modulation in primary visual cortex neurons in vivo. Neuron , 59 (1), 150–160. https://doi.org/10.1016/j.neuron.2008.05.002 Mumtaz, W., Xia, L., Mohd Yasin, M. A., et al. (2017). A wavelet-based technique to predict treatment outcome for Major Depressive Disorder. PloS one , 12 (2), e0171409. https://doi.org/10.1371/journal.pone.0171409 Nunez, M. D., Nunez, P. L., & Srinivasan, R. (2016). Electroencephalography (EEG): Neurophysics, experimental methods, and signal processing. In H. Ombao, M. Lindquist, W. Thompson, & J. Aston (Eds.), Handbook of Neuroimaging Data Analysis (pp. 175–197). Chapman & Hall/CRC. https://doi.org/10.13140/RG.2.2.12706.63687 Salai Selvam V (2025). COMPLETE HIGUCHI FRACTAL DIMENSION ALGORITHM (https://www.mathworks.com/matlabcentral/fileexchange/30119-complete-higuchi-fractal-dimension-algorithm), MATLAB Central File Exchange. Wanliss, J., Arriaza, R. H., Wanliss, G., & Gordon, S. (2021). Optimization of the Higuchi Method. International Journal of Research - Granthaalayah , 9(11), 202–213. https://doi.org/10.29121/granthaalayah.v9.i11.2021.4393 Smits, F. M., Porcaro, C., Cottone, C., et al. (2016). Electroencephalographic Fractal Dimension in Healthy Ageing and Alzheimer's Disease. PloS one , 11 (2), e0149587. https://doi.org/10.1371/journal.pone.0149587 Ghassemkhani, K., Saroka, K.S. & Dotta, B.T. (2025). Evaluating EEG complexity and spectral signatures in Alzheimer’s disease and frontotemporal dementia: evidence for rostrocaudal asymmetry. npj Aging 11, 50. https://doi.org/10.1038/s41514-025-00243-y Cannard, C., & Delorme, A. (2022). An open-source EEGLAB plugin for computing entropy-based measures on MEEG signals. https://doi.org/10.31234/osf.io/xwmyk Nagabhushan Kalburgi, S., Kleinert, T., Aryan, D., et al. (2024). MICROSTATELAB: The EEGLAB Toolbox for Resting-State Microstate Analysis. Brain topography , 37 (4), 621–645. https://doi.org/10.1007/s10548-023-01003-5 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 (Pt B), 577–593. https://doi.org/10.1016/j.neuroimage.2017.11.062 Koenig, T., Diezig, S., Kalburgi, S. N., et al. (2024). EEG-Meta-Microstates: Towards a More Objective Use of Resting-State EEG Microstate Findings Across Studies. Brain topography , 37 (2), 218–231. https://doi.org/10.1007/s10548-023-00993-6 Oostenveld, R., Fries, P., Maris, E., & Schoffelen, J. M. (2011). FieldTrip: Open source software for advanced analysis of MEG, EEG, and invasive electrophysiological data. Computational intelligence and neuroscience , 2011 , 156869. https://doi.org/10.1155/2011/156869 Rubinov, M., & Sporns, O. (2010). Complex network measures of brain connectivity: uses and interpretations. NeuroImage , 52 (3), 1059–1069. https://doi.org/10.1016/j.neuroimage.2009.10.003 Allen, M., Poggiali, D., Whitaker, K., et al. (2021). Raincloud plots: A multi-platform tool for robust data visualization [version 2; peer review: 2 approved]. Wellcome Open Research, 4 (63). https://doi.org/10.12688/wellcomeopenres.15191.2 Lisman, J. E., & Jensen, O. (2013). The θ-γ neural code. Neuron , 77 (6), 1002–1016. https://doi.org/10.1016/j.neuron.2013.03.007 Sadaghiani, S., & Kleinschmidt, A. (2016). Brain Networks and α-Oscillations: Structural and Functional Foundations of Cognitive Control. Trends in cognitive sciences , 20 (11), 805–817. https://doi.org/10.1016/j.tics.2016.09.004 Lin, Y., Wu, Z., Zhang, M.,et al. (2025). Abnormalities in large-scale brain network dynamics in late-life depression with suicidal ideation: an EEG microstate analysis. Journal of psychiatry & neuroscience : JPN , 50 (2), E92–E101. https://doi.org/10.1503/jpn.240115 da Cruz, J. R., Favrod, O., Roinishvili, M., et al. (2020). EEG microstates are a candidate endophenotype for schizophrenia. Nature communications , 11 (1), 3089. https://doi.org/10.1038/s41467-020-16914-1 Tait, L., Tamagnini, F., Stothart, G., et al. (2020). EEG microstate complexity for aiding early diagnosis of Alzheimer's disease. Scientific reports , 10 (1), 17627. https://doi.org/10.1038/s41598-020-74790-7 Sacks, D. D., Schwenn, P. E., Boyes, A., et al. (2023). Longitudinal associations between resting-state, interregional theta-beta phase-amplitude coupling, psychological distress, and wellbeing in 12-15-year-old adolescents. Cerebral cortex (New York, N.Y. : 1991) , 33 (12), 8066–8074. https://doi.org/10.1093/cercor/bhad099 Huang, S. S., Yu, Y. H., Chen, H. H., et al. (2023). Functional connectivity analysis on electroencephalography signals reveals potential biomarkers for treatment response in major depression. BMC psychiatry , 23 (1), 554. https://doi.org/10.1186/s12888-023-04958-8 Blier, P., & El Mansari, M. (2013). Serotonin and beyond: therapeutics for major depression. Philosophical transactions of the Royal Society of London. Series B, Biological sciences , 368 (1615), 20120536. https://doi.org/10.1098/rstb.2012.0536 Ogawa, S., Fujii, T., Koga, N., et al. (2014). Plasma L-tryptophan concentration in major depressive disorder: new data and meta-analysis. The Journal of clinical psychiatry , 75 (9), e906–e915. https://doi.org/10.4088/JCP.13r08908 Caspi, A., Sugden, K., Moffitt, T. E., et al. (2003). Influence of life stress on depression: moderation by a polymorphism in the 5-HTT gene. Science (New York, N.Y.) , 301 (5631), 386–389. https://doi.org/10.1126/science.1083968 Bhagwagar, Z., Wylezinska, M., Taylor, M., et al. (2004). Increased brain GABA concentrations following acute administration of a selective serotonin reuptake inhibitor. The American journal of psychiatry , 161 (2), 368–370. https://doi.org/10.1176/appi.ajp.161.2.368 Sanacora, G., Treccani, G., & Popoli, M. (2012). Towards a glutamate hypothesis of depression: an emerging frontier of neuropsychopharmacology for mood disorders. Neuropharmacology , 62 (1), 63–77. https://doi.org/10.1016/j.neuropharm.2011.07.036 Küçükibrahimoğlu, E., Saygin, M. Z., Calişkan, M., et al. (2009). The change in plasma GABA, glutamine and glutamate levels in fluoxetine- or S-citalopram-treated female patients with major depression. European journal of clinical pharmacology , 65 (6), 571–577. https://doi.org/10.1007/s00228-009-0650-7 Keller, J., Gomez, R., Williams, G., et al. (2017). HPA axis in major depression: cortisol, clinical symptomatology and genetic variation predict cognition. Molecular psychiatry , 22 (4), 527–536. https://doi.org/10.1038/mp.2016.120 Knorr, U., Vinberg, M., Kessing, L. V., & Wetterslev, J. (2010). Salivary cortisol in depressed patients versus control persons: a systematic review and meta-analysis. Psychoneuroendocrinology , 35 (9), 1275–1286. https://doi.org/10.1016/j.psyneuen.2010.04.001 Coryell, W., Young, E., & Carroll, B. (2006). Hyperactivity of the hypothalamic-pituitary-adrenal axis and mortality in major depressive disorder. Psychiatry research , 142 (1), 99–104. https://doi.org/10.1016/j.psychres.2005.08.009 Bronson, D. R., & Preuss, T. (2017). Cellular Mechanisms of Cortisol-Induced Changes in Mauthner-Cell Excitability in the Startle Circuit of Goldfish. Frontiers in neural circuits , 11 , 68. https://doi.org/10.3389/fncir.2017.00068 Voineskos, D., Levinson, A. J., Sun, Y., et al. (2016). The Relationship Between Cortical Inhibition and Electroconvulsive Therapy in the Treatment of Major Depressive Disorder. Scientific reports , 6 , 37461. https://doi.org/10.1038/srep37461 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted 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-7456351","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":521610998,"identity":"7b803d8e-4768-4586-abd1-138663f85996","order_by":0,"name":"Kassra Ghassemkhani","email":"","orcid":"","institution":"Laurentian University","correspondingAuthor":false,"prefix":"","firstName":"Kassra","middleName":"","lastName":"Ghassemkhani","suffix":""},{"id":521610999,"identity":"344f7a0c-22e6-40fe-9dd4-72914b17defa","order_by":1,"name":"Blake T. Dotta","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA50lEQVRIiWNgGAWjYBACAyCWADH42aEibMRpSTBgkGyGCEgQr8XgMFQLQYeZSyQw3vj444+88WHmZ9IVFXfq+NjPHmD4UYNbi2XPAWbLGQkGhtsOs5lJnjnzTIKNJy+BsecYHocdb2CT5kkwYNx2mMFMsrHtMNAvOQbM+AIB6AWwFvvNzezfJBv/AbXwvwFq+UfYlsQNzDxAWxqAWiSAtjC2EfJLmnHyjMM8xZYNxw5Ltkm8MTjY24dbCzjEPtjI2fa3t2+82VBzmF++P8fwwY9vuLUAo/0DptgBfBpGwSgYBaNgFBAGAEmbSIGFZmNjAAAAAElFTkSuQmCC","orcid":"","institution":"Laurentian University","correspondingAuthor":true,"prefix":"","firstName":"Blake","middleName":"T.","lastName":"Dotta","suffix":""}],"badges":[],"createdAt":"2025-08-25 18:38:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7456351/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7456351/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":92472801,"identity":"d6f05c35-754a-4aba-b3d8-ba066bad9332","added_by":"auto","created_at":"2025-09-30 06:56:44","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2226467,"visible":true,"origin":"","legend":"","description":"","filename":"Ghassemkhani2025EEGMDDPaper.docx","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/24e47392354a26503e20f19a.docx"},{"id":92472796,"identity":"1bc2dcbf-e4c2-4001-91d0-6bbab6b24121","added_by":"auto","created_at":"2025-09-30 06:56:43","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4481,"visible":true,"origin":"","legend":"","description":"","filename":"b7b301c508af46a8905300bad834e647.json","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/f64f0e98d0c7b927dde164d4.json"},{"id":92472804,"identity":"c9888ec6-57fa-49d8-9227-358cab898ff1","added_by":"auto","created_at":"2025-09-30 06:56:44","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":157928,"visible":true,"origin":"","legend":"","description":"","filename":"b7b301c508af46a8905300bad834e6471enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/3a0d55c7cb6f210f2cdd7fc8.xml"},{"id":92474561,"identity":"301d7860-b0c9-4470-8c7b-b2a5e8e4ba30","added_by":"auto","created_at":"2025-09-30 07:12:44","extension":"jpeg","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":109612,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/efcbd02bdbb53604b0a31fc5.jpeg"},{"id":92473326,"identity":"eae628a1-1f89-4f26-8da9-432e60ad6603","added_by":"auto","created_at":"2025-09-30 07:04:44","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":250001,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage10.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/8c4d63122d394197eda5f072.jpeg"},{"id":92472805,"identity":"1e13ddb2-6cf6-464e-9321-d18fd7c31f1e","added_by":"auto","created_at":"2025-09-30 06:56:44","extension":"jpeg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":138005,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage11.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/ce17f640fbec4a067187aaf4.jpeg"},{"id":92472802,"identity":"e97ce063-b8ed-4de9-adf6-1631a0301c69","added_by":"auto","created_at":"2025-09-30 06:56:44","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":74093,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage12.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/4adc55af85b5dfae186ecc67.jpeg"},{"id":92472800,"identity":"6161cf83-894f-487e-a9b1-0f57d5497925","added_by":"auto","created_at":"2025-09-30 06:56:43","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":72450,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage13.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/8085889a5514b9c320149611.jpeg"},{"id":92473330,"identity":"d3f1aec7-4d1c-43c1-8a1a-2de11b586c17","added_by":"auto","created_at":"2025-09-30 07:04:44","extension":"jpeg","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":144136,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage14.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/81b39e40c7146608b2c3afd5.jpeg"},{"id":92472831,"identity":"feff8e64-0c9a-42f4-abde-cb0fe2c81ea7","added_by":"auto","created_at":"2025-09-30 06:56:44","extension":"jpeg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":182591,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage15.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/04b9fc282c8ec67663c48b14.jpeg"},{"id":92472816,"identity":"f18a2179-132c-4221-a412-a4066eeb8a50","added_by":"auto","created_at":"2025-09-30 06:56:44","extension":"jpeg","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":108542,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage16.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/70be8e73cbe86467b6ab499a.jpeg"},{"id":92473324,"identity":"cfdbbdbb-2576-4885-9c42-f3b7f9fb0421","added_by":"auto","created_at":"2025-09-30 07:04:44","extension":"jpeg","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":108048,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/763414e721793c0cd194d5ea.jpeg"},{"id":92474573,"identity":"464fa3f0-2c42-4323-8cea-15e89b24d0c9","added_by":"auto","created_at":"2025-09-30 07:12:44","extension":"jpeg","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":78105,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/432f1adf7a108ecf9ad6843b.jpeg"},{"id":92472836,"identity":"12230953-ae35-41a3-85bd-002678fee3b8","added_by":"auto","created_at":"2025-09-30 06:56:45","extension":"jpeg","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":141783,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/5b0a5b53648922469ff24720.jpeg"},{"id":92472813,"identity":"7e0bc6a6-2237-4a65-9fef-ea9ce4ac5a4d","added_by":"auto","created_at":"2025-09-30 06:56:44","extension":"jpeg","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":125248,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/2aee73206f437efcaa8198a5.jpeg"},{"id":92473335,"identity":"4b99d281-9c65-417d-9982-55a3dce3670a","added_by":"auto","created_at":"2025-09-30 07:04:45","extension":"jpeg","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":135567,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/ba4f4032d28bfe699effef0e.jpeg"},{"id":92472817,"identity":"84a38156-ad5f-4966-9ddf-c6446ac6f87e","added_by":"auto","created_at":"2025-09-30 06:56:44","extension":"jpeg","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":216637,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/3a1db283b0887a74cc581f76.jpeg"},{"id":92472828,"identity":"6eb50f58-8c07-4303-be08-a54803152353","added_by":"auto","created_at":"2025-09-30 06:56:44","extension":"jpeg","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":114240,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage8.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/67e439a3b08e28e70d6b0acd.jpeg"},{"id":92472811,"identity":"f70d3aad-e291-4972-bf83-e72108168324","added_by":"auto","created_at":"2025-09-30 06:56:44","extension":"jpeg","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":184044,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage9.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/a21329f9c0be87a140840968.jpeg"},{"id":92472806,"identity":"fa8c9830-8e97-44bc-b603-12bd69b89fca","added_by":"auto","created_at":"2025-09-30 06:56:44","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":41610,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/d4b90d12866399f5eb14309f.png"},{"id":92472844,"identity":"fa004776-d01c-44ff-bfaf-1c8cc12b68a5","added_by":"auto","created_at":"2025-09-30 06:56:45","extension":"png","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":230000,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/9d32b5b6e9bd6cd1b0aae6fd.png"},{"id":92474562,"identity":"2fdd246f-70b0-4f0a-9e16-d489440e23c5","added_by":"auto","created_at":"2025-09-30 07:12:44","extension":"png","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":112991,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage11.png","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/c8f2da754192825e5fb85731.png"},{"id":92472814,"identity":"cef8588c-e646-4bcd-9561-933685033a58","added_by":"auto","created_at":"2025-09-30 06:56:44","extension":"png","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":37063,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage12.png","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/74d776ddbb92ffbd60c7f310.png"},{"id":92472840,"identity":"3a778068-edae-46e1-8220-09cbb08cdd64","added_by":"auto","created_at":"2025-09-30 06:56:45","extension":"png","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":36603,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage13.png","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/6448ee71a7773e671ef9bcc7.png"},{"id":92472851,"identity":"ec040de3-eb85-4662-b87d-4d9788837f1e","added_by":"auto","created_at":"2025-09-30 06:56:45","extension":"png","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":73086,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage14.png","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/ef311fa8dc1c39c95a9955cb.png"},{"id":92472824,"identity":"613d0950-4b2b-414c-85fc-415256aa9ea8","added_by":"auto","created_at":"2025-09-30 06:56:44","extension":"png","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":28554,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage15.png","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/bc9eb005c3511907a91866ee.png"},{"id":92472843,"identity":"b8b0f17d-38b7-4cd8-b588-757a486867cc","added_by":"auto","created_at":"2025-09-30 06:56:45","extension":"png","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":49905,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage16.png","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/9c188404b66d0131627c24aa.png"},{"id":92472827,"identity":"a60b1c3b-5e82-40cd-9b9f-621bfe19806e","added_by":"auto","created_at":"2025-09-30 06:56:44","extension":"png","order_by":27,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":54925,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/9a6eb8dbbc80a43782683049.png"},{"id":92472819,"identity":"fa627196-0b6a-46aa-96da-c6ea90c6c062","added_by":"auto","created_at":"2025-09-30 06:56:44","extension":"png","order_by":28,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":87873,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/5c198ec13803a4f3c79cbc4d.png"},{"id":92473337,"identity":"acab5425-9b38-467b-91c9-3a1a37d7ad5c","added_by":"auto","created_at":"2025-09-30 07:04:45","extension":"png","order_by":29,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":90610,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/9c3f6e4348b7174bd769818d.png"},{"id":92472846,"identity":"0a7947a3-531a-4702-a72c-8f3dc21b7e2a","added_by":"auto","created_at":"2025-09-30 06:56:45","extension":"png","order_by":30,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":78486,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/bdb0869383583dfcc80eaa5d.png"},{"id":92474585,"identity":"a4eca886-3ce3-4b2b-9a45-3f6f8f8e3f90","added_by":"auto","created_at":"2025-09-30 07:12:45","extension":"png","order_by":31,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":69794,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/86c2caa40651208a63ef0617.png"},{"id":92472833,"identity":"4ef1d9c7-b47d-4cbb-8494-550f43fc5006","added_by":"auto","created_at":"2025-09-30 06:56:45","extension":"png","order_by":32,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":211153,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/72b3d60d5d187683f0253f31.png"},{"id":92474586,"identity":"4fb64f70-e2d0-4d1f-8bb2-f575720c9f15","added_by":"auto","created_at":"2025-09-30 07:12:45","extension":"png","order_by":33,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":95410,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/d270cd50e1ba7f488230aa99.png"},{"id":92472826,"identity":"77893ece-ca2b-43b1-b476-6c0fdec74150","added_by":"auto","created_at":"2025-09-30 06:56:44","extension":"png","order_by":34,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29201,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/97e01c6bd3e7cd59e6ce530d.png"},{"id":92472852,"identity":"7c7ef5e8-b30e-4a40-a0d8-dab5becf2af1","added_by":"auto","created_at":"2025-09-30 06:56:45","extension":"xml","order_by":35,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":156011,"visible":true,"origin":"","legend":"","description":"","filename":"b7b301c508af46a8905300bad834e6471structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/8c6091ef899ceadb199c994c.xml"},{"id":92472849,"identity":"10451b0e-2c0d-420c-a4f7-0d757792c14f","added_by":"auto","created_at":"2025-09-30 06:56:45","extension":"html","order_by":36,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":169509,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/fcc2afb8a03c91ec95b19ed1.html"},{"id":92472798,"identity":"f7a64b65-8195-4418-9025-0da8d20da497","added_by":"auto","created_at":"2025-09-30 06:56:43","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":109612,"visible":true,"origin":"","legend":"\u003cp\u003eResults from testing of kmax values ranging from 2 to 128. The y-axis represents the 19 channel average for HFD by kmax which is represented on the x-axis. Solid lines indicate the mean and shaded areas around the solid line are the standard error of the mean (SEM). The orange line represents the HC group whereas the blue line represents the MDD group. Selected for analysis was kmax = 62.\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/db8c8d3acca792c136efbc71.jpeg"},{"id":92473323,"identity":"a9f1cf38-2c72-49b7-9f85-3533b5a33989","added_by":"auto","created_at":"2025-09-30 07:04:43","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":108048,"visible":true,"origin":"","legend":"\u003cp\u003eExample of sample entropy across multiple temporal scaling factors (𝜏). On the x-axis is the 19-channel average values of sample entropy by 𝜏 which is on the y-axis. Circles represent mean whereas error bars represent SEM. The blue line and means display the 19-channel average for the MDD group whereas the orange entry is the HC group. Short-scale entropy was defined as the average between scales 1-6 as seen in the blue shaded area, long-scale entropy was defined as scales 7-12 as seen in the orange shaded area.\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/77b72d3864238e962197c07c.jpeg"},{"id":92472847,"identity":"1323787a-48ec-4259-bee3-8907370e5696","added_by":"auto","created_at":"2025-09-30 06:56:45","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":149867,"visible":true,"origin":"","legend":"\u003cp\u003eExample of the 4 microstate classes used in this analysis. From left to right they are classes A, B, C, and D.\u003c/p\u003e","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/4e41ac07b6fe4faf455976b1.jpeg"},{"id":92472838,"identity":"964d8fae-527a-42cf-a586-6f7f1e9a9dac","added_by":"auto","created_at":"2025-09-30 06:56:45","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":141783,"visible":true,"origin":"","legend":"\u003cp\u003eExample of a microstate activational time series across a 5 second span from a single participant. On the x-axis is time across the 5 second epoch and on the y-axis is global field power, the colours of GFP activation by time match the microstate topographies at the top of the plot.\u003c/p\u003e","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/7b537e5408bfc9b032a390a3.jpeg"},{"id":92473325,"identity":"d63938c8-d79c-4f04-bd27-41697591b79f","added_by":"auto","created_at":"2025-09-30 07:04:44","extension":"jpeg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":125248,"visible":true,"origin":"","legend":"\u003cp\u003eSimplified workflow for graph theory. First, the PLV is calculated between each channel to generate an adjacency matrix. Then, each adjacency matrix is normalized with a top 50% proportional threshold for banalization. The construction of a graph is done on the normalized matrix for analysis of graph theory metrics such as clustering coefficient and average path length.\u003c/p\u003e","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/644a0b4367652632f83dedf5.jpeg"},{"id":92472803,"identity":"d313ef80-d676-4e20-a368-6a8a4fe5b5f0","added_by":"auto","created_at":"2025-09-30 06:56:44","extension":"jpeg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":135567,"visible":true,"origin":"","legend":"\u003cp\u003eBoxplots comparing MDD to HC for the 19-channel averaged relative band power of delta, theta, alpha, and beta. Represented by the boxes are the interquartile range with solid horizontal lines denoting the median, “+” symbols represent the mean, the whiskers outside the boxes represent the data point range, and dots represent the raw data points. Significant differences are indicated by **** indicating a FDR q value \u0026lt; 0.0001.\u003c/p\u003e","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/abd5e2898ed8329f7436912a.jpeg"},{"id":92472830,"identity":"b98c7dc0-1209-467a-9125-1a81b2d75c69","added_by":"auto","created_at":"2025-09-30 06:56:44","extension":"jpeg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":216637,"visible":true,"origin":"","legend":"\u003cp\u003eTopographic EEG plots denoting descriptive statistics between groups (HC and MDD) for the averages per channel of relative band power. Black dots represent a sensor within the 10-20 electrode placement system. The triangular structure at the top of each plot represents the nasion.\u003c/p\u003e","description":"","filename":"floatimage7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/b5fedd36fb72269218b3b9d6.jpeg"},{"id":92473328,"identity":"9c313543-b377-4364-9beb-9eac812fc095","added_by":"auto","created_at":"2025-09-30 07:04:44","extension":"jpeg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":114240,"visible":true,"origin":"","legend":"\u003cp\u003eMap of statistical significance comparing MDD to HC for relative band power. Plotted are t values from independent samples t tests. Non-significant channels after FDR correction are noted in green. Black dots represent a sensor within the 10-20 electrode placement system. The triangular structure at the top of each plot represents the nasion.\u003c/p\u003e","description":"","filename":"floatimage8.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/111ff05e6be4030e7c0d0935.jpeg"},{"id":92473336,"identity":"c12e8ba4-5600-414e-a11e-0b9ee4271d40","added_by":"auto","created_at":"2025-09-30 07:04:45","extension":"jpeg","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":184044,"visible":true,"origin":"","legend":"\u003cp\u003eRaincloud plot denoting the difference in the 19-channel average for HFD values between MDD and HC. From left to right: The large dot and error bars represent the mean and SEM, respectively. Raw data points are represented by small dots. Boxplots denote the interquartile range and lines denote the datapoint range excluding outliers. Half-violins represent the data point density. Significant differences are indicated by *** meaning p \u0026lt; 0.001.\u003c/p\u003e","description":"","filename":"floatimage9.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/383590c3ed3410bf712351d9.jpeg"},{"id":92472809,"identity":"e67cb69f-71e7-4801-a61f-f7ef877c1cd2","added_by":"auto","created_at":"2025-09-30 06:56:44","extension":"jpeg","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":250001,"visible":true,"origin":"","legend":"\u003cp\u003eTopographic EEG plots denoting descriptive statistics between groups (HC and MDD) for the complexity metrics. Black dots represent a sensor within the 10-20 electrode placement system. The triangular structure at the top of each plot represents the nasion.\u003c/p\u003e","description":"","filename":"floatimage10.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/9c327e023f8bfa2a75f5f395.jpeg"},{"id":92473329,"identity":"597cf3f7-9eba-4c86-9fca-03d3715a64ae","added_by":"auto","created_at":"2025-09-30 07:04:44","extension":"jpeg","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":138005,"visible":true,"origin":"","legend":"\u003cp\u003eMap of statistical significance comparing MDD to HC for complexity metrics. Plotted are the t values from independent samples t tests. Non-significant channels after FDR correction are noted in green. Black dots represent a sensor within the 10-20 electrode placement system. The triangular structure at the top of each plot represents the nasion.\u003c/p\u003e","description":"","filename":"floatimage11.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/fc1654322fb2ac750712b063.jpeg"},{"id":92473333,"identity":"6d405c99-5dab-4243-bc2c-960cfc65010b","added_by":"auto","created_at":"2025-09-30 07:04:45","extension":"jpeg","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":74093,"visible":true,"origin":"","legend":"\u003cp\u003eRaincloud plots denoting the difference in the 19-channel average for short-scale (\u003cstrong\u003eA\u003c/strong\u003e) and long-scale (\u003cstrong\u003eB\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eentropy between MDD and HC. From left to right within plots: The large dot and error bars represent the mean and SEM, respectively. Raw data points are represented by small dots. Boxplots denote the interquartile range and lines denote the datapoint range excluding outliers. Half-violins represent the data point density. Significant differences are indicated by **** meaning p \u0026lt; 0.0001, “ns” indicates that no significance was found.\u003c/p\u003e","description":"","filename":"floatimage12.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/0f2a34225ae4d51528dd2cc4.jpeg"},{"id":92472812,"identity":"683de47e-4ffb-4989-ac3c-a953bc6224c6","added_by":"auto","created_at":"2025-09-30 06:56:44","extension":"jpeg","order_by":13,"title":"Figure 13","display":"","copyAsset":false,"role":"figure","size":72450,"visible":true,"origin":"","legend":"\u003cp\u003eRaincloud plot displaying the difference in mean microstate duration (\u003cstrong\u003eA\u003c/strong\u003e) and mean occurrence (\u003cstrong\u003eB\u003c/strong\u003e) irrespective of class between MDD and HC. From left to right within plots: The large dot and error bars represent the mean and SEM, respectively. Raw data points are represented by small dots. Boxplots denote the interquartile range and lines denote the datapoint range excluding outliers. Half-violins represent the data point density. Significant differences are indicated with *** meaning p \u0026lt; 0.001.\u003c/p\u003e","description":"","filename":"floatimage13.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/8f288eb7ae494023d13c24fc.jpeg"},{"id":92473334,"identity":"4f7b82ab-24e1-4337-a50c-9cbd32e0edd0","added_by":"auto","created_at":"2025-09-30 07:04:45","extension":"jpeg","order_by":14,"title":"Figure 14","display":"","copyAsset":false,"role":"figure","size":144136,"visible":true,"origin":"","legend":"\u003cp\u003eBoxplots comparing MDD to HC mean microstate duration (\u003cstrong\u003eA\u003c/strong\u003e), mean microstate occurrence (\u003cstrong\u003eB\u003c/strong\u003e), and microstate coverage (\u003cstrong\u003eC\u003c/strong\u003e). Represented by the boxes are the interquartile range with solid horizontal lines denoting the median, “+” symbols represent the mean, the whiskers outside the boxes represent the data point range, and dots represent the raw data points. Significant differences are indicated by * indicating a FDR q value \u0026lt; 0.05.\u003c/p\u003e","description":"","filename":"floatimage14.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/fb5a6dfff46ca5aa330bcc5a.jpeg"},{"id":92473332,"identity":"edfdb150-31e6-4ecc-b493-a98499de3ef3","added_by":"auto","created_at":"2025-09-30 07:04:44","extension":"jpeg","order_by":15,"title":"Figure 15","display":"","copyAsset":false,"role":"figure","size":182591,"visible":true,"origin":"","legend":"\u003cp\u003eRaincloud plot denoting the difference in small worldness index values of the broadband graphs between MDD and HC. From left to right: The large dot and error bars represent the mean and SEM, respectively. Raw data points are represented by small dots. Boxplots denote the interquartile range and lines denote the datapoint range excluding outliers. Half-violins represent the data point density. Significant differences are indicated by * meaning p \u0026lt; 0.05.\u003c/p\u003e","description":"","filename":"floatimage15.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/5550dd3dcf63d0981860e565.jpeg"},{"id":92472821,"identity":"815bf484-68be-4597-be20-3d5baf460011","added_by":"auto","created_at":"2025-09-30 06:56:44","extension":"jpeg","order_by":16,"title":"Figure 16","display":"","copyAsset":false,"role":"figure","size":108542,"visible":true,"origin":"","legend":"\u003cp\u003eBoxplots comparing MDD to HC for small worldness index values by individual frequency bands. Represented by the boxes are the interquartile range with solid horizontal lines denoting the median, “+” symbols represent the mean, the whiskers outside the boxes represent the data point range, and dots represent the raw data points. Significant differences are indicated by * indicating a FDR q value \u0026lt; 0.05.\u003c/p\u003e","description":"","filename":"floatimage16.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/66c088ec17240258ce64a33a.jpeg"},{"id":99164862,"identity":"920aa513-0ca0-4417-a82d-b13b66b5ae7b","added_by":"auto","created_at":"2025-12-29 13:54:20","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3005398,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7456351/v1/56aa18f4-43d2-46f1-80d9-2b5ec8baba4a.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Neural Gain Dysregulation in Major Depression: Multidomain EEG Signal Analysis from Spectral Power to Network Topology","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eMajor Depressive Disorder (MDD) is characterized by depressed mood and pleasure, difficulty with concentration, and rumination [\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. These symptoms are highlighted by the neurobiological features of MDD including frontal hypoactivity as assessed by positron emission tomography (PET), a neuroimaging device [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e], such a reduction can impact backpropagating inhibitory projections to other cortical regions [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Scalp electroencephalography (EEG), while lacking spatial resolution, has a temporal resolution capable of capturing the activity at neuronal operation speeds in the range of milliseconds compared to neuroimaging devices which operate at a larger temporal scale [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. The electric potentials measured by scalp electrodes can reflect underlying neuronal dynamics, as some properties are constant regardless of scale such as 1/f relationship between frequency and power [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. With EEG, researchers have a variety of signal processing techniques at their disposal.\u003c/p\u003e\u003cp\u003eCommonly used is spectral power density, wherein the squared amplitude at a given frequency is used to quantify the power or intensity, similar to how the \u0026ldquo;loudness\u0026rdquo; of a sound wave is quantified [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. The spectral power within a range of frequencies can be summed to then produce absolute band power, which is common in EEG research. EEG frequency bands can include delta (1\u0026ndash;4 Hz), theta (4\u0026ndash;8 Hz), alpha (8\u0026ndash;13 Hz), beta (13\u0026ndash;30 Hz), and gamma (30\u0026thinsp;+\u0026thinsp;Hz), although the frequency definitions of these bands may vary by researcher. To control inherent signal amplitude differences based on hair or skull size, many employ the relative power normalization, which is the ratio between absolute power of a given band to the total power of the broadband signal [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. Typically reported in MDD is increases in beta power [\u003cspan additionalcitationids=\"CR11 CR12 CR13 CR14\" citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e], and reductions in alpha power [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. The ratio between two frequencies can also be used for analysis such as with the theta-to-beta ratio (TBR), which has shown to be lower than controls in attention-deficit hyperactive disorder (ADHD) [\u003cspan additionalcitationids=\"CR17\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e], which relates to a reduction in theta power with increases in beta power. Not only does the TBR relate to attentional control but has also been shown to be reduced in MDD [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. From the findings discussed above, MDD patients typically display fast frequency abnormalities regarding signal intensity in states where slow frequencies should be more dominant such as the eyes closed resting state.\u003c/p\u003e\u003cp\u003eThe concepts of information theory can be applied to EEG signals to quantify the information content, which can be done using the various complexity measures at a researcher\u0026rsquo;s disposal. First, fractal dimension analysis is useful for quantifying the space-filling properties of EEG signals, typically assessed using Higuchi\u0026rsquo;s algorithm for fractal dimension (HFD) wherein curve lengths are assessed at different temporal scales (k) [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. The irregularity of a time series can also be quantified using sample entropy [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e], wherein the data is embedded (m) with a user defined threshold for matching (r). Sample entropy can be observed at multiple different time lags between points on a time series to produce a measure of multiscale entropy [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. Increases in both HFD and entropy have been observed in MDD patients [\u003cspan additionalcitationids=\"CR24 CR25 CR26 CR27\" citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. This may indicate that EEG activity observed in depression not only shows an increase in fast frequency intensity, but also that the information is highly unpredictable, wherein excessive noise may have been introduced.\u003c/p\u003e\u003cp\u003eBeyond observation of information content at the level of the channels, the spatiotemporal dynamics and transmission of information can be quantified using microstate analysis and graph theory metrics, respectively. Microstates are quasi-stable topographies of the spatial distribution of scalp potentials, typically lasting approximately 60 to 120 milliseconds and commonly observed in a narrow frequency range [\u003cspan additionalcitationids=\"CR30\" citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. The four canonical microstates have been shown to account for approximately 70\u0026ndash;80% of the explained variance of resting state brain activity [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. In MDD, microstate activations have been shown to be more rapid, such as a decrease in microstate D average duration with increases in A and B occurrence compared to control [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. These topographies have also been source localized to network activity with A, B, C, and D corresponding to auditory, visual, salience, and frontoparietal attention networks, respectively [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eNetwork topology can be assessed through graph theory metrics such as the clustering coefficient (CC) and average path length (PL). The CC represents the local interconnectivity of nodes in a network, such as the formation of triangles [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. The PL represents the average shortest number of steps between each node pair in a graph [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. CC highlights the local interconnectedness within a network, whereas the PL highlights the global interconnectedness [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. The network topology of a disordered or random network would have a low CC with a low PL, whereas a latticed or orderly network would have high CC and high average PL. Biological networks contain traits from both, which is optimized for information transfer wherein there is a high CC but low PL [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e], these are the properties of small worldness networks. In MDD, reduced properties consistent with small worldness have been shown in theta and alpha band connectivity networks [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. This indicates that network topology in MDD is closer to the properties seen in random or disordered networks compared to controls.\u003c/p\u003e\u003cp\u003eGain control refers to amplification of a signal, typically as gain increases so does input volume and also noise. Gain modulation as it pertains to neuronal function refers to the changing of input sensitivity of a given neuron while preserving its selectivity [\u003cspan additionalcitationids=\"CR38\" citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. This change in neural gain depends on the background synaptic activity in a given neuron\u0026rsquo;s local environment, which can be altered during attentional tasks or changes in cognitive load [\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. For example, it has been shown that the presence of visual stimuli can induce long-lasting changes in background activity such as with resting membrane potentials in regions relevant to the stimulus [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. Given the symptomatology of MDD pertaining to the thought components, it is possible that even in a restful state there could be abnormalities in gain control. With the amplification of fast frequencies in MDD as observed in spectral analysis, there is potential for the introduction of excess noise or variability as quantified by HFD and multiscale entropy. Given the potential regional increases in gain, the transfer of information within and between networks also becomes disorderly as assessed by microstate analysis and graph theory. For this experiment, we utilized a publicly available dataset [\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e], observing changes in power and unpredictability of EEG in MDD patients to build an EEG profile consistent with MDD using the discussed metrics.\u003c/p\u003e"},{"header":"2. Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1. Dataset Description\u003c/h2\u003e\u003cp\u003eFor analysis we used a publicly available dataset consisting of eyes closed resting state recordings between MDD and healthy controls (HC) compiled by Mumtaz et al. (2017) [\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. In the sample was 34 MDD patients (29 recordings after exclusion criteria), and 30 in the HC group (28 after exclusion criteria). In the MDD group there were 17 males and 17 females, with a mean age of 40.3, the HC group was age-matched (mean age of 38.3) with 21 males and 9 females [\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. MDD patients met the diagnostic criteria as per the Diagnostic and Statistical Manual-IV, and underwent a 2 week medication washout period prior to recording. Data was collected with a 256 Hz sampling rate and a bandpass filter between 0.1 Hz and 70 Hz, with a 50 Hz notch filter for the removal of power line noise. Electrode placement followed the 10\u0026ndash;20 system, and electrodes were referenced to linked ear reference electrodes. The amplifier used for recording was the Brain Master Discovery 24e [\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e].\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2. Preprocessing\u003c/h2\u003e\u003cp\u003eAll preprocessing steps were done in the EEGLAB toolbox for MATLAB, the data was not downsampled from 256 Hz. Data was filtered with a high pass of 1.5 Hz and a low pass of 30 Hz using a finite impulse response (FIR) filter. This narrow filter was applied to control for high frequency noise arising from muscle or background sources [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]. Data was then referenced to the common average reference for further noise reduction and normalization. Further artifact correction was performed using independent components analysis (ICA) for the removal of eye-related, muscle, and background noise artifacts. The multiple artifact rejection algorithm (MARA) in tandem with visual inspection was used for the rejection of components pertaining to artifacts. The corrected data was then split into 5 second non-overlapping epochs, from which a compiled 30 seconds was used for analysis of the eyes closed resting state recordings.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.3. Relative Band Power\u003c/h2\u003e\u003cp\u003eRelative band power was calculated for each channel using the Darbeliai extension for EEGLAB. First, the spectral power density (SPD) was calculated (\u0026micro;V\u0026sup2;/Hz) with a 0.1 Hz step on the broadband signal (1.5\u0026ndash;30 Hz). The absolute power was then calculated by summing the SPD values in the defined bands: delta (1.5-4 Hz), theta (4\u0026ndash;8 Hz), alpha (8\u0026ndash;13 Hz), and beta (13\u0026ndash;30 Hz). The total power was also calculated for relative power normalization wherein the absolute power of a given band is divided by the total power to produce a value between 0 and 1 highlighting how much each band contributes to the total power spectrum. The 19-channel average of relative band power was primarily used in this analysis to capture global changes in signal.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e2.4. Higuchi Fractal Dimension\u003c/h2\u003e\u003cp\u003eSignal complexity was first assessed using HFD. HFD was computed using open-source MATLAB code created by Selvam (2025), which utilizes the original Higuchi algorithm [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]. HFD assesses curve lengths ranging from a step size (k) of 1 to a user-defined maximum (kmax), wherein the HFD value is the relationship between curve lengths and temporal steps typically ranging from 1 and 2. The parameter, kmax, is the most sensitive and highly variable parameter across researchers in producing HFD values as no optimal method exists to select it [\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. We used a data-driven approach to select kmax as has been done before in EEG research using HFD [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e]. First, we computed the 19-channel average for HFD from all subjects across kmax values ranging from 2 to the Nyquist limit of the dataset which was 128 (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). From the average HFD value from each kmax, the median HFD value was selected for this analysis which corresponded to a kmax of 62.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003e2.5. Multiscale Sample Entropy\u003c/h2\u003e\u003cp\u003eTo validate findings from HFD, we employed a second complexity technique to quantify irregularity at multiple time scales. Multiscale entropy was computed adapting functions from the get_entropy EEGLAB plugin [\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e]. The embedding dimension (m) was set to 2, which reconstructs the time series into pairs of points. For sample entropy, a separate time series is constructed using m\u0026thinsp;+\u0026thinsp;1, or triplets in this instance. Whether pairs or triplets match is determined by the tolerance threshold (r) which was set to 0.2, or 20% of the signal standard deviation. Sample entropy is then calculated by taking the negative logarithm of the ratio between the number of matches with the m\u0026thinsp;+\u0026thinsp;1 template to the number of matches with template m. Sample entropy was then calculated at multiple temporal scale factors (\u0026#120591;) on a coarse-grained version of the signal wherein a moving average is created based on \u0026#120591; to shorten the signal. \u0026#120591; refers to the time lag between points on a time series. We used values of \u0026#120591; ranging from 1 to 12 to leave a minimum of 100 points at the longest \u0026#120591; for accurate assessment of entropy given epoch duration and sampling frequency. For multiscale entropy analysis we utilized the average from scales 1\u0026ndash;6 which were defined as short-scale entropy and the average from scales 7\u0026ndash;12 which was defined as long-scale entropy, this was done for each channel and the 19-channel average (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003e2.6. Microstate Analysis\u003c/h2\u003e\u003cp\u003eMicrostate analysis was performed using the MICROSTATELAB plugin for EEGLAB [\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e]. Microstate analysis was performed with a 2\u0026ndash;20 Hz frequency range as is common in microstate analysis [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e]. First, individual microstates were identified based on global field power (GFP) peaks utilizing the k-means clustering algorithm. Next, microstate maps were generated based on the individual microstate maps. The mean maps were then sorted based on the meta maps template [\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e]. An example of the four classes of microstates used for analysis can be seen in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. Each of the extracted epochs were then backfitted to the sorted grand mean microstate maps, an example of a microstate activational time series can be observed in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e. The temporal parameters of microstate activations were then exported. Used for analysis was microstate mean duration, mean occurrence, and coverage. Mean duration refers to the average duration of microstates by class in a given epoch, and the mean duration of microstates irrespective of class for a global measure. Mean occurrence refers to the average occurrences per second of each individual class, and the mean occurrence irrespective of class to gain insight on the rate of activation. Microstate coverage refers to the percentage of time spent in one of the four microstates.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e2.7. Small Worldness\u003c/h2\u003e\u003cp\u003eSmall worldness was computed using the functions from Fieldtrip and Brain Connectivity Toolbox for MATLAB [\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e, \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e]. First we computed the phase locking value (PLV), which quantifies the phase relationship between two signals over time providing a value between 0 and 1, with higher values indicating greater phase consistency between two signals. This was computed for the broadband signal (1.5\u0026ndash;30 Hz), and individual frequency bands including delta (1.5-4 Hz), theta (4\u0026ndash;8 Hz), alpha (8\u0026ndash;13 Hz), and beta (13\u0026ndash;30 Hz) band activity. From this, a resultant 19 x 19 adjacency matrix is then created for graph theory analysis. The adjacency matrix is first normalized using banalization by retaining the top 50% of connections. Following this, graph theory metrics are calculated including CC or tendency of a graph to form interconnected node clusters, and PL or the average number of steps from the shortest paths between node pairs. The simplified workflow can be observed in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e. Based on the node degree, randomized networks using the Erdős\u0026ndash;R\u0026eacute;nyi model were generated for the calculation of the small worldness index, for reproducibility the average CC and PL from 100 iterations were taken for reference. Small worldness index values are calculated by taking the ratio between the observed network to random network CC divided by the ratio between the observed network to random network PL.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e2.8. Statistical Analysis\u003c/h2\u003e\u003cp\u003eAll statistical analyses and boxplot creation were performed using the GraphPad Prism 10 software. Raincloud plots were created using RStudio (ggplot2 and ggrain) [\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e] and topographic EEG maps were created using a custom MATLAB script. Independent samples t tests were used for analysis between HC and MDD, unless F tests revealed significant differences wherein a non-parametric Mann-Whitney U test would be used. For multiple comparisons, we used false discovery rate (FDR) applied with a desired Q of 5% from the Benjamini \u0026amp; Hochberg method. Cohen\u0026rsquo;s d was used for effect size with an effect size cutoff of Cohen\u0026rsquo;d\u0026thinsp;=\u0026thinsp;0.5.\u003c/p\u003e\u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003e3.1. Relative Band Power Differences in Fast Frequencies Between MDD and HC\u003c/h2\u003e\u003cp\u003eWhen observing the 19-channel average for relative band power (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e), no significant differences were observed between MDD and HC for the delta band. While MDD displayed significantly lower relative theta (p\u0026thinsp;=\u0026thinsp;0.04865) and alpha (p\u0026thinsp;=\u0026thinsp;0.03252) power compared to HC, these comparisons did not pass the false discovery rate (FDR) correction. However, when observing relative beta power, MDD displayed significantly higher beta power compared to HC [t(55)\u0026thinsp;=\u0026thinsp;5.663, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001, q\u0026thinsp;\u0026lt;\u0026thinsp;0.0001, Cohen\u0026rsquo;s d\u0026thinsp;=\u0026thinsp;1.5069]. Figure\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e highlights a topographic map of the channel averages for the four calculated frequency bands. Figure\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e highlights statistical significance using t values comparing MDD to HC, wherein significantly lower alpha can be observed in sensors C3 and C4, with significance found in each channel for the beta band after the FDR correction.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003e3.2. Higher Higuchi Fractal Dimension Displayed in MDD\u003c/h2\u003e\u003cp\u003eWith regards to signal complexity, MDD displayed significantly higher HFD values compared to HC [t(55)\u0026thinsp;=\u0026thinsp;4.086, p\u0026thinsp;=\u0026thinsp;0.0001, Cohen\u0026rsquo;s d\u0026thinsp;=\u0026thinsp;1.0575] when comparing the 19-channel averages (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e). Topographic plots of the channel averages for each complexity metric (HFD, short-scale entropy, long-scale entropy) can be observed in Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003e. MDD displaying greater HFD compared to HC was revealed in every channel except Fp1, Fp2, F7, F8, and O2 after FDR correction. Single channel t values between MDD vs HC for complexity metrics are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003e3.3. Short-Scale Entropy Increases in MDD\u003c/h2\u003e\u003cp\u003eWhen observing the 19-channel average for short-scale entropy (Fig.\u0026nbsp;\u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e12\u003c/span\u003eA), MDD displays significantly higher entropy compared to HC [t(55)\u0026thinsp;=\u0026thinsp;4.204, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001, Cohen\u0026rsquo;s d\u0026thinsp;=\u0026thinsp;1.1155]. However when comparing long-scale entropy (Fig.\u0026nbsp;\u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e12\u003c/span\u003eB), no significant difference was observed between MDD and HC. Single channel differences in short-scale entropy after FDR show MDD has significantly higher entropy compared to HC in each channel except T3. However no single channel differences were observed in long-scale entropy after FDR as can be seen in the t value topographic plots in Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003e3.4. Reduced Temporal Stability of Microstates in MDD\u003c/h2\u003e\u003cp\u003eWhen observing mean microstate duration irrespective of class (Fig.\u0026nbsp;\u003cspan refid=\"Fig13\" class=\"InternalRef\"\u003e13\u003c/span\u003eA), MDD displayed significantly lower durations compared to HC [U\u0026thinsp;=\u0026thinsp;197, p\u0026thinsp;=\u0026thinsp;0.0006, Cohen's d\u0026thinsp;=\u0026thinsp;0.9219]. When observing individual microstate classes after FDR correction, MDD displayed significantly lower microstate durations in microstate class B [t(55)\u0026thinsp;=\u0026thinsp;2.997, p\u0026thinsp;=\u0026thinsp;0.0040, q\u0026thinsp;=\u0026thinsp;0.0122, Cohen\u0026rsquo;s d\u0026thinsp;=\u0026thinsp;0.7904] and class D [t(55)\u0026thinsp;=\u0026thinsp;2.851, p\u0026thinsp;=\u0026thinsp;0.0061, q\u0026thinsp;=\u0026thinsp;0.0122, Cohen\u0026rsquo;s d\u0026thinsp;=\u0026thinsp;0.7518] but not class A and class C compared to HC (Fig.\u0026nbsp;\u003cspan refid=\"Fig14\" class=\"InternalRef\"\u003e14\u003c/span\u003eA). When observing mean microstate occurrence irrespective of class (Fig.\u0026nbsp;\u003cspan refid=\"Fig13\" class=\"InternalRef\"\u003e13\u003c/span\u003eB), MDD displayed significantly higher occurrence compared to HC [t(55)\u0026thinsp;=\u0026thinsp;3.654, p\u0026thinsp;=\u0026thinsp;0.0006, Cohen's d\u0026thinsp;=\u0026thinsp;0.9670]. When observing single classes significantly higher occurrences were displayed by MDD for microstate classes A [t(55)\u0026thinsp;=\u0026thinsp;2.765, p\u0026thinsp;=\u0026thinsp;0.0077, q\u0026thinsp;=\u0026thinsp;0.0154, Cohen\u0026rsquo;s d\u0026thinsp;=\u0026thinsp;0.7305], class B [t(55)\u0026thinsp;=\u0026thinsp;2.335, p\u0026thinsp;=\u0026thinsp;0.0232, q\u0026thinsp;=\u0026thinsp;0.0309, Cohen\u0026rsquo;s d\u0026thinsp;=\u0026thinsp;0.6176], and class C [t(55)\u0026thinsp;=\u0026thinsp;3.014, p\u0026thinsp;=\u0026thinsp;0.0038, q\u0026thinsp;=\u0026thinsp;0.0154, Cohen\u0026rsquo;s d\u0026thinsp;=\u0026thinsp;0.7978] after FDR correction (Fig.\u0026nbsp;\u003cspan refid=\"Fig14\" class=\"InternalRef\"\u003e14\u003c/span\u003eB). When observing coverage, while MDD displayed significantly higher coverage of class A (p\u0026thinsp;=\u0026thinsp;0.0431) and significantly lower coverage of class D (p\u0026thinsp;=\u0026thinsp;0.0468), these comparisons did not pass the FDR correction (Fig.\u0026nbsp;\u003cspan refid=\"Fig14\" class=\"InternalRef\"\u003e14\u003c/span\u003eC).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003e3.5. Reduced Small Worldness Properties of Channel Networks in MDD\u003c/h2\u003e\u003cp\u003eFrom the graph theory metric, small worldness index, when observing the broadband (1.5\u0026ndash;30 Hz) graphs (Fig.\u0026nbsp;\u003cspan refid=\"Fig15\" class=\"InternalRef\"\u003e15\u003c/span\u003e), MDD displayed significantly lower small worldness index values compared to HC [t(55)\u0026thinsp;=\u0026thinsp;2.545, p\u0026thinsp;=\u0026thinsp;0.0138, Cohen's d\u0026thinsp;=\u0026thinsp;0.6747]. Upon observing individual frequencies (Fig.\u0026nbsp;\u003cspan refid=\"Fig16\" class=\"InternalRef\"\u003e16\u003c/span\u003e), MDD also displayed significantly lower small worldness index values in the theta band [t(55)\u0026thinsp;=\u0026thinsp;2.906, p\u0026thinsp;=\u0026thinsp;0.0052, q\u0026thinsp;=\u0026thinsp;0.021, Cohen\u0026rsquo;s d\u0026thinsp;=\u0026thinsp;0.7616] compared to HC. No significant differences were observed in delta, alpha, or beta bands for small worldness index values after FDR correction between MDD and HC.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eFrom the results, observed was the resting state signal in MDD not only possessing greater intensity at high frequencies but the information content also in excess at short-temporal scales. The excessive information is also highlighted in the increase in HFD, wherein the scaling behaviours between curve length and temporal scale were greater in MDD compared to HC. It is possible that the increased short-scale entropy, but not long-scale, highlights not only the power increased at fast frequencies, but that the fast frequency activity in itself is more unpredictable than the activity from HC. Not only does the information content yield greater unpredictability at the level of the channels, but also how information is transmitted between channels as assessed by graph theory, and the temporal dynamics of brain states assessed by microstate analysis.\u003c/p\u003e\u003cp\u003eThe increase in power at fast frequencies potentially reflects a tonic increase in gain or sensitivity. Similar to how increasing gain on an audio device can introduce noise, this increase in beta power was also accompanied by noise as quantified through complexity metrics. Theta and alpha are highly associated with modulation of faster rhythms or cyclic inhibitory activity [\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e, \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e], despite the comparisons not passing FDR correction, the decreases in both frequencies potentially relate to the irregularity in the fast frequency range due to a lack of modulation.\u003c/p\u003e\u003cp\u003eWith this introduction to excessive noise, the transmission of information from region to region can also be rendered difficult. Microstates have seen much usage in current EEG literature [\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e, \u003cspan additionalcitationids=\"CR58\" citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e] and have been termed \u0026ldquo;atoms of thought\u0026rdquo;, as their temporal dynamics can reflect different forms of thought and cognitive processing [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. MDD was characterized by increased occurrences with reduced durations irrespective of microstate class. This shows that microstate transitions are considerably more rapid, and given their association with network activity [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e], could reflect altered connectivity and synchrony patterns. From analysis of network topology, we wanted to test unpredictability using the small worldness index. A reduced small worldness index indicates a deviation towards the network topologies observed in random networks such as Erdős\u0026ndash;R\u0026eacute;nyi models, wherein the interconnectivity of nodes is low in tandem with low average steps [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. The dynamics between scalp electrodes in MDD displayed reduced small worldness properties compared to controls when observing the broadband signal and theta for individual frequencies. Given the modulatory features observed with theta, a random connectivity pattern also potentially gives rise to the fast frequency unpredictability [\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e, \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e]. Reduced theta to beta phase amplitude coupling in the right hemisphere between frontal and parietal sensors has shown to be associated with increased psychological distress [\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e]. These shifts in the temporal and topological components of network activity highlight not only single regional unpredictability, but global unpredictability in region-to-region communication.\u003c/p\u003e\u003cp\u003eIf the features discussed are present, one can monitor EEG to track treatment outcomes such as following neurostimulation techniques or pharmacological intervention. Following treatment with electroconvulsive therapy reductions in entropy have been found in MDD patients who are considered ECT responders (50% reduction in Hamilton Rating Scale for Depression) compared to non-responders [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. Microstate dynamics have also been shown to improve with treatment in MDD, wherein the occurrences of classes A and B were decreased and durations of class D were increased following 8 weeks of treatment with selective serotonin reuptake inhibitors (SSRI) [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. SSRI treatment has also yielded results in graph theory metrics wherein average clustering coefficients for the alpha band have shown to increase in responders compared to non responders [\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eGraded potentials and neuromodulatory input are important in the control of neural gain [\u003cspan additionalcitationids=\"CR38\" citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. Serotonergic abnormalities present in MDD [\u003cspan additionalcitationids=\"CR63\" citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e] could be reducing neuromodulation and disrupting the excitatory inhibitory gradient. Following SSRI treatment, levels of gamma-aminobutyric acid (GABA) and glutamate have shown to be increased and decreased, respectively [\u003cspan additionalcitationids=\"CR66\" citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e], possibly indicating greater excitatory tone present in MDD which increases neural gain. Further, the hypothalamic-pituitary-adrenal axis (HPA) has been suggested to display hyperactivity in MDD as evident by basal concentrations of hormones associated with the HPA circuit [\u003cspan additionalcitationids=\"CR69\" citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e]. Research with goldfish as a model, has shown cortisol to produce features consistent with increased neural gain such as long-lasting increases in post-synaptic potential magnitudes, membrane input resistance, and reduced threshold current in cells dedicated to startle responses [\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e]. Neurostimulation techniques such as ECT also have efficacy in treatment resistant depression as they have the capability to reduce cortical excitability long-term [\u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e72\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eMuch current research on MDD using EEG following treatment incorporates techniques from a single domain of information from the signal. Future research can use a combination of techniques that quantify spectral features, complexity, and the transfer of information such as graph theory to track treatment related changes in MDD. This ensures that the changes in signal unpredictability following treatment are not only at the level of a single channel but also influence how regions communicate as well.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eFunding\u003c/h2\u003e\u003cp\u003eThis research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eKG: Conceptualization; Data curation; Formal analysis; Visualization; Writing \u0026ndash; original draft; Writing \u0026ndash; review \u0026amp; editing.BTD: Conceptualization; Supervision; Visualization; Writing \u0026ndash; review \u0026amp; editing.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe EEG dataset analyzed in this study is publicly available from Mumtaz et al. (2017) in PLOS ONE (https://doi.org/10.1371/journal.pone.0171409). Analyses were conducted using open-source MATLAB toolboxes (EEGLAB, FieldTrip, Brain Connectivity Toolbox, MICROSTATELAB). Custom MATLAB scripts used for visualization are available from the corresponding author upon reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eSu, Y. A., \u0026amp; Si, T. (2022). Progress and challenges in research of the mechanisms of anhedonia in major depressive disorder. \u003cem\u003eGeneral psychiatry\u003c/em\u003e, \u003cem\u003e35\u003c/em\u003e(1), e100724. https://doi.org/10.1136/gpsych-2021-100724\u003c/li\u003e\n\u003cli\u003ePerini, G., Cotta Ramusino, M., Sinforiani, E., et al. (2019). Cognitive impairment in depression: recent advances and novel treatments. \u003cem\u003eNeuropsychiatric disease and treatment\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e, 1249\u0026ndash;1258. https://doi.org/10.2147/NDT.S199746\u003c/li\u003e\n\u003cli\u003eAlderman, B. L., Olson, R. L., Bates, M. E., et al. (2015). Rumination in major depressive disorder is associated with impaired neural activation during conflict monitoring. \u003cem\u003eFrontiers in human neuroscience\u003c/em\u003e, \u003cem\u003e9\u003c/em\u003e, 269. https://doi.org/10.3389/fnhum.2015.00269\u003c/li\u003e\n\u003cli\u003eVidebech P. (2000). PET measurements of brain glucose metabolism and blood flow in major depressive disorder: a critical review. \u003cem\u003eActa psychiatrica Scandinavica\u003c/em\u003e, \u003cem\u003e101\u003c/em\u003e(1), 11\u0026ndash;20. https://doi.org/10.1034/j.1600-0447.2000.101001011.x\u003c/li\u003e\n\u003cli\u003eBravo-Rivera, C., Diehl, M. M., Roman-Ortiz, C., et al. (2015). Long-range GABAergic neurons in the prefrontal cortex modulate behavior. \u003cem\u003eJournal of neurophysiology\u003c/em\u003e, \u003cem\u003e114\u003c/em\u003e(3), 1357\u0026ndash;1359. https://doi.org/10.1152/jn.00861.2014\u003c/li\u003e\n\u003cli\u003eBuzs\u0026aacute;ki, G. (2006). \u003cem\u003eRhythms of the brain.\u003c/em\u003e Oxford University Press.\u003c/li\u003e\n\u003cli\u003eLi, C. Y., Poo, M. M., \u0026amp; Dan, Y. (2009). Burst spiking of a single cortical neuron modifies global brain state. \u003cem\u003eScience (New York, N.Y.)\u003c/em\u003e, \u003cem\u003e324\u003c/em\u003e(5927), 643\u0026ndash;646. https://doi.org/10.1126/science.1169957\u003c/li\u003e\n\u003cli\u003eGyurkovics, M., Clements, G. M., Low, K. A. (2021). The impact of 1/f activity and baseline correction on the results and interpretation of time-frequency analyses of EEG/MEG data: A cautionary tale. \u003cem\u003eNeuroImage\u003c/em\u003e, \u003cem\u003e237\u003c/em\u003e, 118192. https://doi.org/10.1016/j.neuroimage.2021.118192\u003c/li\u003e\n\u003cli\u003eNishiyori, R., Xiao, R., Vanderbilt, D., \u0026amp; Smith, B. A. (2021). Electroencephalography measures of relative power and coherence as reaching skill emerges in infants born preterm. \u003cem\u003eScientific reports\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(1), 3609. https://doi.org/10.1038/s41598-021-82329-7\u003c/li\u003e\n\u003cli\u003e\u0026Ouml;z\u0026ccedil;oban, M. A., \u0026amp; Tan, O. (2025). Electroencephalographic markers in Major Depressive Disorder: insights from absolute, relative power, and asymmetry analyses. \u003cem\u003eFrontiers in psychiatry\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e, 1480228. https://doi.org/10.3389/fpsyt.2024.1480228\u003c/li\u003e\n\u003cli\u003eKoshiyama, D., Kirihara, K., Usui, K., et al. (2020). Resting-state EEG beta band power predicts quality of life outcomes in patients with depressive disorders: A longitudinal investigation. \u003cem\u003eJournal of affective disorders\u003c/em\u003e, \u003cem\u003e265\u003c/em\u003e, 416\u0026ndash;422. https://doi.org/10.1016/j.jad.2020.01.030\u003c/li\u003e\n\u003cli\u003eTatti, E., Cinti, A., Serbina, A., et al. (2024). Resting-State EEG Alterations of Practice-Related Spectral Activity and Connectivity Patterns in Depression. \u003cem\u003eBiomedicines\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(9), 2054. https://doi.org/10.3390/biomedicines12092054\u003c/li\u003e\n\u003cli\u003eKnott, V., Mahoney, C., Kennedy, S., \u0026amp; Evans, K. (2001). EEG power, frequency, asymmetry and coherence in male depression. \u003cem\u003ePsychiatry research\u003c/em\u003e, \u003cem\u003e106\u003c/em\u003e(2), 123\u0026ndash;140. https://doi.org/10.1016/s0925-4927(00)00080-9\u003c/li\u003e\n\u003cli\u003eSang, Q., Chen, C., \u0026amp; Shao, Z. (2025). Decoding depression from different brain regions using hybrid machine learning methods. \u003cem\u003eBioengineering, 12\u003c/em\u003e(5), 449. https://doi.org/10.3390/bioengineering12050449\u003c/li\u003e\n\u003cli\u003eKovacevic, N., Meghdadi, A., Berka, C., et al. (2025). Differences in resting state and task-based EEG measures between patients with major depressive disorder and healthy controls. \u003cem\u003eClinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology\u003c/em\u003e, \u003cem\u003e173\u003c/em\u003e, 190\u0026ndash;198. https://doi.org/10.1016/j.clinph.2025.03.022\u003c/li\u003e\n\u003cli\u003ePicken, C., Clarke, A. R., Barry, R. J., et al. (2020). The Theta/Beta Ratio as an Index of Cognitive Processing in Adults With the Combined Type of Attention Deficit Hyperactivity Disorder. \u003cem\u003eClinical EEG and neuroscience\u003c/em\u003e, \u003cem\u003e51\u003c/em\u003e(3), 167\u0026ndash;173. https://doi.org/10.1177/1550059419895142\u003c/li\u003e\n\u003cli\u003ePutman, P., Verkuil, B., Arias-Garcia, E., Pantazi, I., \u0026amp; van Schie, C. (2014). EEG theta/beta ratio as a potential biomarker for attentional control and resilience against deleterious effects of stress on attention. \u003cem\u003eCognitive, affective \u0026amp; behavioral neuroscience\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(2), 782\u0026ndash;791. https://doi.org/10.3758/s13415-013-0238-7\u003c/li\u003e\n\u003cli\u003eWang, T. S., Wang, S. S., Wang, C. L., \u0026amp; Wong, S. B. (2024). Theta/beta ratio in EEG correlated with attentional capacity assessed by Conners Continuous Performance Test in children with ADHD. \u003cem\u003eFrontiers in psychiatry\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e, 1305397. https://doi.org/10.3389/fpsyt.2023.1305397\u003c/li\u003e\n\u003cli\u003eChang, J., \u0026amp; Choi, Y. (2023). Depression diagnosis based on electroencephalography power ratios. \u003cem\u003eBrain and behavior\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(8), e3173. https://doi.org/10.1002/brb3.3173\u003c/li\u003e\n\u003cli\u003eHiguchi, T. (1988). Approach to an irregular time series on the basis of the fractal theory. Physica D: Nonlinear Phenomena, 31(2), 277\u0026ndash;283. https://doi.org/10.1016/0167-2789(88)90081-4\u003c/li\u003e\n\u003cli\u003eRichman, J. S., \u0026amp; Moorman, J. R. (2000). Physiological time‑series analysis using approximate entropy and sample entropy. \u003cem\u003eAmerican Journal of Physiology \u0026ndash; Heart and Circulatory Physiology, 278\u003c/em\u003e(6), H2039\u0026ndash;H2049. https://doi.org/10.1152/ajpheart.2000.278.6.H2039\u003c/li\u003e\n\u003cli\u003eCosta, M., Goldberger, A. L., \u0026amp; Peng, C.-K. (2002). Multiscale entropy analysis of complex physiologic time series. \u003cem\u003ePhysical Review Letters, 89\u003c/em\u003e(6), 068102. https://doi.org/10.1103/PhysRevLett.89.068102\u003c/li\u003e\n\u003cli\u003eAkar, S. A., Kara, S., Agambayev, S., \u0026amp; Bilgic, V. (2015). Nonlinear analysis of EEG in major depression with fractal dimensions. \u003cem\u003eAnnual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference\u003c/em\u003e, \u003cem\u003e2015\u003c/em\u003e, 7410\u0026ndash;7413. https://doi.org/10.1109/EMBC.2015.7320104\u003c/li\u003e\n\u003cli\u003eYun, S., \u0026amp; Jeong, B. (2021). Aberrant EEG signal variability at a specific temporal scale in major depressive disorder. \u003cem\u003eClinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology\u003c/em\u003e, \u003cem\u003e132\u003c/em\u003e(8), 1866\u0026ndash;1877. https://doi.org/10.1016/j.clinph.2021.05.011\u003c/li\u003e\n\u003cli\u003eFarzan F., Atluri, S., Mei, Y. et al. (2017). Brain temporal complexity in explaining the therapeutic and cognitive effects of seizure therapy, \u003cem\u003eBrain\u003c/em\u003e, \u003cem\u003e140\u003c/em\u003e(4), 1011\u0026ndash;1025, https://doi.org/10.1093/brain/awx030\u003c/li\u003e\n\u003cli\u003eBachmann, M., Lass, J., Suhhova, A., \u0026amp; Hinrikus, H. (2013). Spectral asymmetry and Higuchi\u0026apos;s fractal dimension measures of depression electroencephalogram. \u003cem\u003eComputational and mathematical methods in medicine\u003c/em\u003e, \u003cem\u003e2013\u003c/em\u003e, 251638. https://doi.org/10.1155/2013/251638\u003c/li\u003e\n\u003cli\u003eČukić, M., Stokić, M., Simić, S., \u0026amp; Pokrajac, D. (2020). The successful discrimination of depression from EEG could be attributed to proper feature extraction and not to a particular classification method. \u003cem\u003eCognitive neurodynamics\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(4), 443\u0026ndash;455. https://doi.org/10.1007/s11571-020-09581-x\u003c/li\u003e\n\u003cli\u003ede Freitas, S. B., Marques, A. A., Bevilaqua, M. C., de Carvalho, M. R., Ribeiro, P., Palmer, S., Nardi, A. E., \u0026amp; Dias, G. P. (2016). Electroencephalographic findings in patients with major depressive disorder during cognitive or emotional tasks: a systematic review. \u003cem\u003eRevista brasileira de psiquiatria (Sao Paulo, Brazil : 1999)\u003c/em\u003e, \u003cem\u003e38\u003c/em\u003e(4), 338\u0026ndash;346. https://doi.org/10.1590/1516-4446-2015-1834\u003c/li\u003e\n\u003cli\u003eKoenig, T., Prichep, L., Lehmann, D., et al. (2002). Millisecond by millisecond, year by year: normative EEG microstates and developmental stages. \u003cem\u003eNeuroImage\u003c/em\u003e, \u003cem\u003e16\u003c/em\u003e(1), 41\u0026ndash;48. https://doi.org/10.1006/nimg.2002.1070\u003c/li\u003e\n\u003cli\u003eLehmann, D., Strik, W. K., Henggeler, B., et al. (1998). Brain electric microstates and momentary conscious mind states as building blocks of spontaneous thinking: I. Visual imagery and abstract thoughts. \u003cem\u003eInternational journal of psychophysiology : official journal of the International Organization of Psychophysiology\u003c/em\u003e, \u003cem\u003e29\u003c/em\u003e(1), 1\u0026ndash;11. https://doi.org/10.1016/s0167-8760(97)00098-6\u003c/li\u003e\n\u003cli\u003eLehmann, D., Ozaki, H., \u0026amp; Pal, I. (1987). EEG alpha map series: brain micro-states by space-oriented adaptive segmentation. \u003cem\u003eElectroencephalography and clinical neurophysiology\u003c/em\u003e, \u003cem\u003e67\u003c/em\u003e(3), 271\u0026ndash;288. https://doi.org/10.1016/0013-4694(87)90025-3\u003c/li\u003e\n\u003cli\u003eLei, L., Liu, Z., Zhang, Y., et al. (2022). EEG microstates as markers of major depressive disorder and predictors of response to SSRIs therapy. \u003cem\u003eProgress in neuro-psychopharmacology \u0026amp; biological psychiatry\u003c/em\u003e, \u003cem\u003e116\u003c/em\u003e, 110514. https://doi.org/10.1016/j.pnpbp.2022.110514\u003c/li\u003e\n\u003cli\u003eCusto, A., Van De Ville, D., Wells, et al. (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\u003eSmit, D. J., Stam, C. J., Posthuma, D., et al. (2008). Heritability of \u0026quot;small-world\u0026quot; networks in the brain: a graph theoretical analysis of resting-state EEG functional connectivity. \u003cem\u003eHuman brain mapping\u003c/em\u003e, \u003cem\u003e29\u003c/em\u003e(12), 1368\u0026ndash;1378. https://doi.org/10.1002/hbm.20468\u003c/li\u003e\n\u003cli\u003eTelesford, Q. K., Joyce, K. E., Hayasaka, S., et al. (2011). The ubiquity of small-world networks. \u003cem\u003eBrain connectivity\u003c/em\u003e, \u003cem\u003e1\u003c/em\u003e(5), 367\u0026ndash;375. https://doi.org/10.1089/brain.2011.0038\u003c/li\u003e\n\u003cli\u003eShim, M., Im, C. H., Kim, Y. W., \u0026amp; Lee, S. H. (2018). Altered cortical functional network in major depressive disorder: A resting-state electroencephalogram study. \u003cem\u003eNeuroImage. Clinical\u003c/em\u003e, \u003cem\u003e19\u003c/em\u003e, 1000\u0026ndash;1007. https://doi.org/10.1016/j.nicl.2018.06.012\u003c/li\u003e\n\u003cli\u003eSalinas, E., \u0026amp; Sejnowski, T. J. (2001). Gain modulation in the central nervous system: where behavior, neurophysiology, and computation meet. \u003cem\u003eThe Neuroscientist : a review journal bringing neurobiology, neurology and psychiatry\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e(5), 430\u0026ndash;440. https://doi.org/10.1177/107385840100700512\u003c/li\u003e\n\u003cli\u003eFerguson, K. A., \u0026amp; Cardin, J. A. (2020). Mechanisms underlying gain modulation in the cortex. \u003cem\u003eNature reviews. Neuroscience\u003c/em\u003e, \u003cem\u003e21\u003c/em\u003e(2), 80\u0026ndash;92. https://doi.org/10.1038/s41583-019-0253-y\u003c/li\u003e\n\u003cli\u003eChance, F. S., Abbott, L. F., \u0026amp; Reyes, A. D. (2002). Gain modulation from background synaptic input. \u003cem\u003eNeuron\u003c/em\u003e, \u003cem\u003e35\u003c/em\u003e(4), 773\u0026ndash;782. https://doi.org/10.1016/s0896-6273(02)00820-6\u003c/li\u003e\n\u003cli\u003eThiele, A., \u0026amp; Bellgrove, M. A. (2018). Neuromodulation of Attention. \u003cem\u003eNeuron\u003c/em\u003e, \u003cem\u003e97\u003c/em\u003e(4), 769\u0026ndash;785. https://doi.org/10.1016/j.neuron.2018.01.008\u003c/li\u003e\n\u003cli\u003eCardin, J. A., Palmer, L. A., \u0026amp; Contreras, D. (2008). Cellular mechanisms underlying stimulus-dependent gain modulation in primary visual cortex neurons in vivo. \u003cem\u003eNeuron\u003c/em\u003e, \u003cem\u003e59\u003c/em\u003e(1), 150\u0026ndash;160. https://doi.org/10.1016/j.neuron.2008.05.002\u003c/li\u003e\n\u003cli\u003eMumtaz, W., Xia, L., Mohd Yasin, M. A., et al. (2017). A wavelet-based technique to predict treatment outcome for Major Depressive Disorder. \u003cem\u003ePloS one\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(2), e0171409. https://doi.org/10.1371/journal.pone.0171409\u003c/li\u003e\n\u003cli\u003eNunez, M. D., Nunez, P. L., \u0026amp; Srinivasan, R. (2016). Electroencephalography (EEG): Neurophysics, experimental methods, and signal processing. In H. Ombao, M. Lindquist, W. Thompson, \u0026amp; J. Aston (Eds.), \u003cem\u003eHandbook of Neuroimaging Data Analysis\u003c/em\u003e (pp. 175\u0026ndash;197). Chapman \u0026amp; Hall/CRC. https://doi.org/10.13140/RG.2.2.12706.63687\u003c/li\u003e\n\u003cli\u003eSalai Selvam V (2025). COMPLETE HIGUCHI FRACTAL DIMENSION ALGORITHM (https://www.mathworks.com/matlabcentral/fileexchange/30119-complete-higuchi-fractal-dimension-algorithm), MATLAB Central File Exchange.\u003c/li\u003e\n\u003cli\u003eWanliss, J., Arriaza, R. H., Wanliss, G., \u0026amp; Gordon, S. (2021). Optimization of the Higuchi Method. \u003cem\u003eInternational Journal of Research - Granthaalayah\u003c/em\u003e, 9(11), 202\u0026ndash;213. https://doi.org/10.29121/granthaalayah.v9.i11.2021.4393\u003c/li\u003e\n\u003cli\u003eSmits, F. M., Porcaro, C., Cottone, C., et al. (2016). Electroencephalographic Fractal Dimension in Healthy Ageing and Alzheimer\u0026apos;s Disease. \u003cem\u003ePloS one\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(2), e0149587. https://doi.org/10.1371/journal.pone.0149587\u003c/li\u003e\n\u003cli\u003eGhassemkhani, K., Saroka, K.S. \u0026amp; Dotta, B.T. (2025). Evaluating EEG complexity and spectral signatures in Alzheimer\u0026rsquo;s disease and frontotemporal dementia: evidence for rostrocaudal asymmetry. \u003cem\u003enpj Aging\u003c/em\u003e 11, 50. https://doi.org/10.1038/s41514-025-00243-y\u003c/li\u003e\n\u003cli\u003eCannard, C., \u0026amp; Delorme, A. (2022). An open-source EEGLAB plugin for computing entropy-based measures on MEEG signals. https://doi.org/10.31234/osf.io/xwmyk\u003c/li\u003e\n\u003cli\u003eNagabhushan Kalburgi, S., Kleinert, T., Aryan, D., et al. (2024). MICROSTATELAB: The EEGLAB Toolbox for Resting-State Microstate Analysis. \u003cem\u003eBrain topography\u003c/em\u003e, \u003cem\u003e37\u003c/em\u003e(4), 621\u0026ndash;645. https://doi.org/10.1007/s10548-023-01003-5\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(Pt B), 577\u0026ndash;593. https://doi.org/10.1016/j.neuroimage.2017.11.062\u003c/li\u003e\n\u003cli\u003eKoenig, T., Diezig, S., Kalburgi, S. N., et al. (2024). EEG-Meta-Microstates: Towards a More Objective Use of Resting-State EEG Microstate Findings Across Studies. \u003cem\u003eBrain topography\u003c/em\u003e, \u003cem\u003e37\u003c/em\u003e(2), 218\u0026ndash;231. https://doi.org/10.1007/s10548-023-00993-6\u003c/li\u003e\n\u003cli\u003eOostenveld, R., Fries, P., Maris, E., \u0026amp; Schoffelen, J. M. (2011). FieldTrip: Open source software for advanced analysis of MEG, EEG, and invasive electrophysiological data. \u003cem\u003eComputational intelligence and neuroscience\u003c/em\u003e, \u003cem\u003e2011\u003c/em\u003e, 156869. https://doi.org/10.1155/2011/156869\u003c/li\u003e\n\u003cli\u003eRubinov, M., \u0026amp; Sporns, O. (2010). Complex network measures of brain connectivity: uses and interpretations. \u003cem\u003eNeuroImage\u003c/em\u003e, \u003cem\u003e52\u003c/em\u003e(3), 1059\u0026ndash;1069. https://doi.org/10.1016/j.neuroimage.2009.10.003\u003c/li\u003e\n\u003cli\u003eAllen, M., Poggiali, D., Whitaker, K., et al. (2021). Raincloud plots: A multi-platform tool for robust data visualization [version 2; peer review: 2 approved]. \u003cem\u003eWellcome Open Research, 4\u003c/em\u003e(63). https://doi.org/10.12688/wellcomeopenres.15191.2\u003c/li\u003e\n\u003cli\u003eLisman, J. E., \u0026amp; Jensen, O. (2013). The \u0026theta;-\u0026gamma; neural code. \u003cem\u003eNeuron\u003c/em\u003e, \u003cem\u003e77\u003c/em\u003e(6), 1002\u0026ndash;1016. https://doi.org/10.1016/j.neuron.2013.03.007\u003c/li\u003e\n\u003cli\u003eSadaghiani, S., \u0026amp; Kleinschmidt, A. (2016). Brain Networks and \u0026alpha;-Oscillations: Structural and Functional Foundations of Cognitive Control. \u003cem\u003eTrends in cognitive sciences\u003c/em\u003e, \u003cem\u003e20\u003c/em\u003e(11), 805\u0026ndash;817. https://doi.org/10.1016/j.tics.2016.09.004\u003c/li\u003e\n\u003cli\u003eLin, Y., Wu, Z., Zhang, M.,et al. (2025). Abnormalities in large-scale brain network dynamics in late-life depression with suicidal ideation: an EEG microstate analysis. \u003cem\u003eJournal of psychiatry \u0026amp; neuroscience : JPN\u003c/em\u003e, \u003cem\u003e50\u003c/em\u003e(2), E92\u0026ndash;E101. https://doi.org/10.1503/jpn.240115\u003c/li\u003e\n\u003cli\u003eda Cruz, J. R., Favrod, O., Roinishvili, M., et al. (2020). EEG microstates are a candidate endophenotype for schizophrenia. \u003cem\u003eNature communications\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(1), 3089. https://doi.org/10.1038/s41467-020-16914-1\u003c/li\u003e\n\u003cli\u003eTait, L., Tamagnini, F., Stothart, G., et al. (2020). EEG microstate complexity for aiding early diagnosis of Alzheimer\u0026apos;s disease. \u003cem\u003eScientific reports\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e(1), 17627. https://doi.org/10.1038/s41598-020-74790-7\u003c/li\u003e\n\u003cli\u003eSacks, D. D., Schwenn, P. E., Boyes, A., et al. (2023). Longitudinal associations between resting-state, interregional theta-beta phase-amplitude coupling, psychological distress, and wellbeing in 12-15-year-old adolescents. \u003cem\u003eCerebral cortex (New York, N.Y. : 1991)\u003c/em\u003e, \u003cem\u003e33\u003c/em\u003e(12), 8066\u0026ndash;8074. https://doi.org/10.1093/cercor/bhad099\u003c/li\u003e\n\u003cli\u003eHuang, S. S., Yu, Y. H., Chen, H. H., et al. (2023). Functional connectivity analysis on electroencephalography signals reveals potential biomarkers for treatment response in major depression. \u003cem\u003eBMC psychiatry\u003c/em\u003e, \u003cem\u003e23\u003c/em\u003e(1), 554. https://doi.org/10.1186/s12888-023-04958-8\u003c/li\u003e\n\u003cli\u003eBlier, P., \u0026amp; El Mansari, M. (2013). Serotonin and beyond: therapeutics for major depression. \u003cem\u003ePhilosophical transactions of the Royal Society of London. Series B, Biological sciences\u003c/em\u003e, \u003cem\u003e368\u003c/em\u003e(1615), 20120536. https://doi.org/10.1098/rstb.2012.0536\u003c/li\u003e\n\u003cli\u003eOgawa, S., Fujii, T., Koga, N., et al. (2014). Plasma L-tryptophan concentration in major depressive disorder: new data and meta-analysis. \u003cem\u003eThe Journal of clinical psychiatry\u003c/em\u003e, \u003cem\u003e75\u003c/em\u003e(9), e906\u0026ndash;e915. https://doi.org/10.4088/JCP.13r08908\u003c/li\u003e\n\u003cli\u003eCaspi, A., Sugden, K., Moffitt, T. E., et al. (2003). Influence of life stress on depression: moderation by a polymorphism in the 5-HTT gene. \u003cem\u003eScience (New York, N.Y.)\u003c/em\u003e, \u003cem\u003e301\u003c/em\u003e(5631), 386\u0026ndash;389. https://doi.org/10.1126/science.1083968\u003c/li\u003e\n\u003cli\u003eBhagwagar, Z., Wylezinska, M., Taylor, M., et al. (2004). Increased brain GABA concentrations following acute administration of a selective serotonin reuptake inhibitor. \u003cem\u003eThe American journal of psychiatry\u003c/em\u003e, \u003cem\u003e161\u003c/em\u003e(2), 368\u0026ndash;370. https://doi.org/10.1176/appi.ajp.161.2.368\u003c/li\u003e\n\u003cli\u003eSanacora, G., Treccani, G., \u0026amp; Popoli, M. (2012). Towards a glutamate hypothesis of depression: an emerging frontier of neuropsychopharmacology for mood disorders. \u003cem\u003eNeuropharmacology\u003c/em\u003e, \u003cem\u003e62\u003c/em\u003e(1), 63\u0026ndash;77. https://doi.org/10.1016/j.neuropharm.2011.07.036\u003c/li\u003e\n\u003cli\u003eK\u0026uuml;\u0026ccedil;\u0026uuml;kibrahimoğlu, E., Saygin, M. Z., Calişkan, M., et al. (2009). The change in plasma GABA, glutamine and glutamate levels in fluoxetine- or S-citalopram-treated female patients with major depression. \u003cem\u003eEuropean journal of clinical pharmacology\u003c/em\u003e, \u003cem\u003e65\u003c/em\u003e(6), 571\u0026ndash;577. https://doi.org/10.1007/s00228-009-0650-7\u003c/li\u003e\n\u003cli\u003eKeller, J., Gomez, R., Williams, G., et al. (2017). HPA axis in major depression: cortisol, clinical symptomatology and genetic variation predict cognition. \u003cem\u003eMolecular psychiatry\u003c/em\u003e, \u003cem\u003e22\u003c/em\u003e(4), 527\u0026ndash;536. https://doi.org/10.1038/mp.2016.120\u003c/li\u003e\n\u003cli\u003eKnorr, U., Vinberg, M., Kessing, L. V., \u0026amp; Wetterslev, J. (2010). Salivary cortisol in depressed patients versus control persons: a systematic review and meta-analysis. \u003cem\u003ePsychoneuroendocrinology\u003c/em\u003e, \u003cem\u003e35\u003c/em\u003e(9), 1275\u0026ndash;1286. https://doi.org/10.1016/j.psyneuen.2010.04.001\u003c/li\u003e\n\u003cli\u003eCoryell, W., Young, E., \u0026amp; Carroll, B. (2006). Hyperactivity of the hypothalamic-pituitary-adrenal axis and mortality in major depressive disorder. \u003cem\u003ePsychiatry research\u003c/em\u003e, \u003cem\u003e142\u003c/em\u003e(1), 99\u0026ndash;104. https://doi.org/10.1016/j.psychres.2005.08.009\u003c/li\u003e\n\u003cli\u003eBronson, D. R., \u0026amp; Preuss, T. (2017). Cellular Mechanisms of Cortisol-Induced Changes in Mauthner-Cell Excitability in the Startle Circuit of Goldfish. \u003cem\u003eFrontiers in neural circuits\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e, 68. https://doi.org/10.3389/fncir.2017.00068\u003c/li\u003e\n\u003cli\u003eVoineskos, D., Levinson, A. J., Sun, Y., et al. (2016). The Relationship Between Cortical Inhibition and Electroconvulsive Therapy in the Treatment of Major Depressive Disorder. \u003cem\u003eScientific reports\u003c/em\u003e, \u003cem\u003e6\u003c/em\u003e, 37461. https://doi.org/10.1038/srep37461\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"EEG, Major Depressive Disorder, neural gain, spectral power, fractal dimension, entropy, microstates, graph theory, small-world networks, signal complexity","lastPublishedDoi":"10.21203/rs.3.rs-7456351/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7456351/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eElectroencephalography (EEG) offers a cost-effective window into the neural dynamics of Major Depressive Disorder (MDD), yet most studies focus on isolated domains of analysis. We present a multidomain computational framework spanning spectral, complexity, temporal, and network levels to test the hypothesis that MDD reflects dysregulation of neural gain. Using a publicly available resting-state EEG dataset, we quantified relative band power, nonlinear complexity (Higuchi fractal dimension, multiscale entropy), microstate dynamics, and graph-theoretic topology. MDD patients displayed significantly elevated beta power alongside higher short-scale entropy and fractal dimension values, indicating increased fast-frequency noise. Microstate analysis revealed reduced temporal stability and more frequent transitions, while graph-theoretic measures showed reduced small-worldness, particularly in the theta band, consistent with a shift toward random connectivity. Together, these results suggest that MDD is characterized by amplified fast-frequency activity with degraded signal-to-noise structure across temporal and network scales. This multidomain framework demonstrates how gain dysregulation manifests simultaneously in oscillatory, dynamical, and topological properties of EEG, offering a computational profile of depression that may support biomarker development and treatment monitoring.\u003c/p\u003e","manuscriptTitle":"Neural Gain Dysregulation in Major Depression: Multidomain EEG Signal Analysis from Spectral Power to Network Topology","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-30 06:56:39","doi":"10.21203/rs.3.rs-7456351/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"92aad2d7-820f-442d-ba93-ac95eb607bca","owner":[],"postedDate":"September 30th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-12-29T13:53:28+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-30 06:56:39","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7456351","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7456351","identity":"rs-7456351","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}