On co-activation pattern analysis and non-stationarity of resting brain activity

Sex-specific resting state brain network dynamics in patients with major depressive disorder

Sex-specific neurobiological changes have been implicated in Major Depressive Disorder (MDD). Dysfunctions of the default mode network (DMN), salience network (SN) and frontoparietal network (FPN) are critical neural characteristics of MDD, however, the potential moderating role of sex on resting-state network dynamics in MDD has not been sufficiently evaluated. Thus, resting-state functional magnetic resonance imaging (fMRI) data were collected from 138 unmedicated patients with first-episode MDD (55 males) and 243 healthy controls (HCs; 106 males). Recurring functional network co-activation patterns (CAPs) were extracted, and time spent in each CAP (the total amount of volumes associated to a CAP), persistence (the average number of consecutive volumes linked to a CAP), and transitions across CAPs involving the SN, DMN and FPN were quantified. Relative to HCs, MDD patients exhibited greater persistence in a CAP involving activation of the DMN and deactivation of the FPN (DMN + FPN-). In addition, relative to the sex-matched HCs, the male MDD group spent more time in two CAPs involving the SN and DMN (i.e., DMN + SN- and DMN-SN + ) and transitioned more frequently from the DMN + FPN- CAP to the DMN + SN- CAP relative to the male HC group. Conversely, the female MDD group showed less persistence in the DMN + SN- CAP relative to the female HC group. Our findings highlight that the imbalance between SN and DMN could be a neurobiological marker supporting sex differences in MDD. Moreover, the dominance of the DMN accompanied by the deactivation of the FPN could be a sex-independent neurobiological correlate related to depression.

Thalamic contributions to psychosis susceptibility: Evidence from co-activation patterns accounting for intra-seed spatial variability (μCAPs)

The temporal variability of the thalamus in functional networks may provide valuable insights into the pathophysiology of schizophrenia. To address the complexity of the role of the thalamic nuclei in psychosis, we introduced micro-co-activation patterns (μCAPs) and employed this method on the human genetic model of schizophrenia 22q11.2 deletion syndrome (22q11.2DS). Participants underwent resting-state functional MRI and a data-driven iterative process resulting in the identification of six whole-brain μCAPs with specific activity patterns within the thalamus. Unlike conventional methods, μCAPs extract dynamic spatial patterns that reveal partially overlapping and non-mutually exclusive functional subparts. Thus, the μCAPs method detects finer foci of activity within the initial seed region, retaining valuable and clinically relevant temporal and spatial information. We found that a μCAP showing co-activation of the mediodorsal thalamus with brain-wide cortical regions was expressed significantly less frequently in patients with 22q11.2DS, and its occurrence negatively correlated with the severity of positive psychotic symptoms. Additionally, activity within the auditory-visual cortex and their respective geniculate nuclei was expressed in two different μCAPs. One of these auditory-visual μCAPs co-activated with salience areas, while the other co-activated with the default mode network (DMN). A significant shift of occurrence from the salience+visuo-auditory-thalamus to the DMN + visuo-auditory-thalamus μCAP was observed in patients with 22q11.2DS. Thus, our findings support existing research on the gatekeeping role of the thalamus for sensory information in the pathophysiology of psychosis and revisit the evidence of geniculate nuclei hyperconnectivity with the audio-visual cortex in 22q11.2DS in the context of dynamic functional connectivity, seen here as the specific hyper-occurrence of these circuits with the task-negative brain networks.

Spatiotemporal dynamics of hippocampal-cortical networks underlying the unique phenomenological properties of trauma-related intrusive memories

Trauma-related intrusive memories (TR-IMs) possess unique phenomenological properties that contribute to adverse post-traumatic outcomes, positioning them as critical intervention targets. However, transdiagnostic treatments for TR-IMs are scarce, as their underlying mechanisms have been investigated separate from their unique phenomenological properties. Extant models of more general episodic memory highlight dynamic hippocampal-cortical interactions that vary along the anterior-posterior axis of the hippocampus (HPC) to support different cognitive-affective and sensory-perceptual features of memory. Extending this work into the unique properties of TR-IMs, we conducted a study of eighty-four trauma-exposed adults who completed daily ecological momentary assessments of TR-IM properties followed by resting-state functional magnetic resonance imaging (rs-fMRI). Spatiotemporal dynamics of anterior and posterior hippocampal (a/pHPC)-cortical networks were assessed using co-activation pattern analysis to investigate their associations with different properties of TR-IMs. Emotional intensity of TR-IMs was inversely associated with the frequency and persistence of an aHPC-default mode network co-activation pattern. Conversely, sensory features of TR-IMs were associated with more frequent co-activation of the HPC with sensory cortices and the ventral attention network, and the reliving of TR-IMs in the "here-and-now" was associated with more persistent co-activation of the pHPC and the visual cortex. Notably, no associations were found between HPC-cortical network dynamics and conventional symptom measures, including TR-IM frequency or retrospective recall, underscoring the utility of ecological assessments of memory properties in identifying their neural substrates. These findings provide novel insights into the neural correlates of the unique features of TR-IMs that are critical for the development of individualized, transdiagnostic treatments for this pervasive, difficult-to-treat symptom.

Uncovering a stability signature of brain dynamics associated with meditation experience using massive time-series feature extraction

Previous research has examined resting electroencephalographic (EEG) data to explore brain activity related to meditation. However, previous research has mostly examined power in different frequency bands. The practical objective of this study was to comprehensively test whether other types of time-series analysis methods are better suited to characterize brain activity related to meditation. To achieve this, we compared >7000 time-series features of the EEG signal to comprehensively characterize brain activity differences in meditators, using many measures that are novel in meditation research. Eyes-closed resting-state EEG data from 49 meditators and 46 non-meditators was decomposed into the top eight principal components (PCs). We extracted 7381 time-series features from each PC and each participant and used them to train classification algorithms to identify meditators. Highly differentiating individual features from successful classifiers were analysed in detail. Only the third PC (which had a central-parietal maximum) showed above-chance classification accuracy (67 %, pFDR = 0.007), for which 405 features significantly distinguished meditators (all pFDR < 0.05). Top-performing features indicated that meditators exhibited more consistent statistical properties across shorter subsegments of their EEG time-series (higher stationarity) and displayed an altered distributional shape of values about the mean. By contrast, classifiers trained with traditional band-power measures did not distinguish the groups (pFDR > 0.05). Our novel analysis approach suggests the key signatures of meditators' brain activity are higher temporal stability and a distribution of time-series values suggestive of longer, larger, or more frequent non-outlying voltage deviations from the mean within the third PC of their EEG data. The higher temporal stability observed in this EEG component might underpin the higher attentional stability associated with meditation. The novel time-series properties identified here have considerable potential for future exploration in meditation research and the analysis of neural dynamics more broadly.

Frequency-specific functional difference between gyri and sulci in naturalistic paradigm fMRI

Disentangling functional difference between cortical folding patterns of gyri and sulci provides novel insights into the relationship between brain structure and function. Previous studies using resting-state functional magnetic resonance imaging (rsfMRI) have revealed that sulcal signals exhibit stronger high-frequency but weaker low-frequency components compared to gyral ones, suggesting that gyri may serve as functional integration centers while sulci are segregated local processing units. In this study, we utilize naturalistic paradigm fMRI (nfMRI) to explore the functional difference between gyri and sulci as it has proven to record stronger functional integrations compared to rsfMRI. We adopt a convolutional neural network (CNN) to classify gyral and sulcal fMRI signals in the whole brain (the global model) and within functional brain networks (the local models). The frequency-specific difference between gyri and sulci is then inferred from the power spectral density (PSD) profiles of the learned filters in the CNN model. Our experimental results show that nfMRI shows higher gyral-sulcal PSD contrast effect sizes in the global model compared to rsfMRI. In the local models, the effect sizes are either increased or decreased depending on frequency bands and functional complexity of the FBNs. This study highlights the advantages of nfMRI in depicting the functional difference between gyri and sulci, and provides novel insights into unraveling the relationship between brain structure and function.

Modular subgraphs in large-scale connectomes underpin spontaneous co-fluctuation events in mouse and human brains

Previous studies have adopted an edge-centric framework to study fine-scale network dynamics in human fMRI. To date, however, no studies have applied this framework to data collected from model organisms. Here, we analyze structural and functional imaging data from lightly anesthetized mice through an edge-centric lens. We find evidence of "bursty" dynamics and events - brief periods of high-amplitude network connectivity. Further, we show that on a per-frame basis events best explain static FC and can be divided into a series of hierarchically-related clusters. The co-fluctuation patterns associated with each cluster centroid link distinct anatomical areas and largely adhere to the boundaries of algorithmically detected functional brain systems. We then investigate the anatomical connectivity undergirding high-amplitude co-fluctuation patterns. We find that events induce modular bipartitions of the anatomical network of inter-areal axonal projections. Finally, we replicate these same findings in a human imaging dataset. In summary, this report recapitulates in a model organism many of the same phenomena observed in previously edge-centric analyses of human imaging data. However, unlike human subjects, the murine nervous system is amenable to invasive experimental perturbations. Thus, this study sets the stage for future investigation into the causal origins of fine-scale brain dynamics and high-amplitude co-fluctuations. Moreover, the cross-species consistency of the reported findings enhances the likelihood of future translation.

Static and dynamic fMRI-derived functional connectomes represent largely similar information

Functional connectivity (FC) of blood oxygen level-dependent (BOLD) fMRI time series can be estimated using methods that differ in sensitivity to the temporal order of time points (static vs. dynamic) and the number of regions considered in estimating a single edge (bivariate vs. multivariate). Previous research suggests that dynamic FC explains variability in FC fluctuations and behavior beyond static FC. Our aim was to systematically compare methods on both dimensions. We compared five FC methods: Pearson's/full correlation (static, bivariate), lagged correlation (dynamic, bivariate), partial correlation (static, multivariate), and multivariate AR model with and without self-connections (dynamic, multivariate). We compared these methods by (i) assessing similarities between FC matrices, (ii) by comparing node centrality measures, and (iii) by comparing the patterns of brain-behavior associations. Although FC estimates did not differ as a function of sensitivity to temporal order, we observed differences between the multivariate and bivariate FC methods. The dynamic FC estimates were highly correlated with the static FC estimates, especially when comparing group-level FC matrices. Similarly, there were high correlations between the patterns of brain-behavior associations obtained using the dynamic and static FC methods. We conclude that the dynamic FC estimates represent information largely similar to that of the static FC.

Living on the edge: network neuroscience beyond nodes

Network neuroscience has emphasized the connectional properties of neural elements - cells, populations, and regions. This has come at the expense of the anatomical and functional connections that link these elements to one another. A new perspective - namely one that emphasizes 'edges' - may prove fruitful in addressing outstanding questions in network neuroscience. We highlight one recently proposed 'edge-centric' method and review its current applications, merits, and limitations. We also seek to establish conceptual and mathematical links between this method and previously proposed approaches in the network science and neuroimaging literature. We conclude by presenting several avenues for future work to extend and refine existing edge-centric analysis.

Hierarchical organization of spontaneous co-fluctuations in densely sampled individuals using fMRI

Edge time series decompose functional connectivity into its framewise contributions. Previous studies have focused on characterizing the properties of high-amplitude frames (time points when the global co-fluctuation amplitude takes on its largest value), including their cluster structure. Less is known about middle- and low-amplitude co-fluctuations (peaks in co-fluctuation time series but of lower amplitude). Here, we directly address those questions, using data from two dense-sampling studies: the MyConnectome project and Midnight Scan Club. We develop a hierarchical clustering algorithm to group peak co-fluctuations of all magnitudes into nested and multiscale clusters based on their pairwise concordance. At a coarse scale, we find evidence of three large clusters that, collectively, engage virtually all canonical brain systems. At finer scales, however, each cluster is dissolved, giving way to increasingly refined patterns of co-fluctuations involving specific sets of brain systems. We also find an increase in global co-fluctuation magnitude with hierarchical scale. Finally, we comment on the amount of data needed to estimate co-fluctuation pattern clusters and implications for brain-behavior studies. Collectively, the findings reported here fill several gaps in current knowledge concerning the heterogeneity and richness of co-fluctuation patterns as estimated with edge time series while providing some practical guidance for future studies.

Modular architecture facilitates noise-driven control of synchrony in neuronal networks

High-level information processing in the mammalian cortex requires both segregated processing in specialized circuits and integration across multiple circuits. One possible way to implement these seemingly opposing demands is by flexibly switching between states with different levels of synchrony. However, the mechanisms behind the control of complex synchronization patterns in neuronal networks remain elusive. Here, we use precision neuroengineering to manipulate and stimulate networks of cortical neurons in vitro, in combination with an in silico model of spiking neurons and a mesoscopic model of stochastically coupled modules to show that (i) a modular architecture enhances the sensitivity of the network to noise delivered as external asynchronous stimulation and that (ii) the persistent depletion of synaptic resources in stimulated neurons is the underlying mechanism for this effect. Together, our results demonstrate that the inherent dynamical state in structured networks of excitable units is determined by both its modular architecture and the properties of the external inputs.

Low-frequency repetitive transcranial magnetic stimulation alters the individual functional dynamical landscape

Repetitive transcranial magnetic stimulation (rTMS) is a noninvasive approach to modulate brain activity and behavior in humans. Still, how individual resting-state brain dynamics after rTMS evolves across different functional configurations is rarely studied. Here, using resting state fMRI data from healthy subjects, we aimed to examine the effects of rTMS to individual large-scale brain dynamics. Using Topological Data Analysis based Mapper approach, we construct the precise dynamic mapping (PDM) for each participant. To reveal the relationship between PDM and canonical functional representation of the resting brain, we annotated the graph using relative activation proportion of a set of large-scale resting-state networks (RSNs) and assigned the single brain volume to corresponding RSN-dominant or a hub state (not any RSN was dominant). Our results show that (i) low-frequency rTMS could induce changed temporal evolution of brain states; (ii) rTMS didn't alter the hub-periphery configurations underlined resting-state brain dynamics; and (iii) the rTMS effects on brain dynamics differ across the left frontal and occipital lobe. In conclusion, low-frequency rTMS significantly alters the individual temporo-spatial dynamics, and our finding further suggested a potential target-dependent alteration of brain dynamics. This work provides a new perspective to comprehend the heterogeneous effect of rTMS.

Manifold learning for fMRI time-varying functional connectivity

Whole-brain functional connectivity (FC) measured with functional MRI (fMRI) evolves over time in meaningful ways at temporal scales going from years (e.g., development) to seconds [e.g., within-scan time-varying FC (tvFC)]. Yet, our ability to explore tvFC is severely constrained by its large dimensionality (several thousands). To overcome this difficulty, researchers often seek to generate low dimensional representations (e.g., 2D and 3D scatter plots) hoping those will retain important aspects of the data (e.g., relationships to behavior and disease progression). Limited prior empirical work suggests that manifold learning techniques (MLTs)-namely those seeking to infer a low dimensional non-linear surface (i.e., the manifold) where most of the data lies-are good candidates for accomplishing this task. Here we explore this possibility in detail. First, we discuss why one should expect tvFC data to lie on a low dimensional manifold. Second, we estimate what is the intrinsic dimension (ID; i.e., minimum number of latent dimensions) of tvFC data manifolds. Third, we describe the inner workings of three state-of-the-art MLTs: Laplacian Eigenmaps (LEs), T-distributed Stochastic Neighbor Embedding (T-SNE), and Uniform Manifold Approximation and Projection (UMAP). For each method, we empirically evaluate its ability to generate neuro-biologically meaningful representations of tvFC data, as well as their robustness against hyper-parameter selection. Our results show that tvFC data has an ID that ranges between 4 and 26, and that ID varies significantly between rest and task states. We also show how all three methods can effectively capture subject identity and task being performed: UMAP and T-SNE can capture these two levels of detail concurrently, but LE could only capture one at a time. We observed substantial variability in embedding quality across MLTs, and within-MLT as a function of hyper-parameter selection. To help alleviate this issue, we provide heuristics that can inform future studies. Finally, we also demonstrate the importance of feature normalization when combining data across subjects and the role that temporal autocorrelation plays in the application of MLTs to tvFC data. Overall, we conclude that while MLTs can be useful to generate summary views of labeled tvFC data, their application to unlabeled data such as resting-state remains challenging.

Brain network integration, segregation and quasi-periodic activation and deactivation during tasks and rest

Previous studies have shown that a re-organization of the brain's functional connectome expressed in terms of network integration and segregation may play a pivotal role for brain function. However, it has been proven difficult to fully capture both processes independently in a single methodological framework. In this study, by starting from pair-wise assessments of instantaneous phase synchronization and community membership, we assemble spatiotemporally flexible networks that reflect changes in integration/segregation that occur at a spectrum of spatial as well as temporal scales. This is achieved by iteratively assembling smaller networks into larger units under the constraint that the smaller units should be internally integrated, i.e. belong to the same community. The assembled subnetworks can be partly overlapping and differ in size across time. Our results show that subnetwork integration and segregation occur simultaneously in the brain. During task performance, global changes in synchronization between networks arise that are tied to the underlying temporal design of the experiment. We show that a hallmark property of the dynamics of the brain's functional connectome is a presence of quasi-periodic patterns of network activation and deactivation, which during task performance becomes intertwined with the underlying temporal structure of the experimental paradigm. Additionally, we show that the degree of network integration throughout a n-back working memory task is correlated to performance.

Robust dynamic brain coactivation states estimated in individuals

A confluence of evidence indicates that brain functional connectivity is not static but rather dynamic. Capturing transient network interactions in the individual brain requires a technology that offers sufficient within-subject reliability. Here, we introduce an individualized network-based dynamic analysis technique and demonstrate that it is reliable in detecting subject-specific brain states during both resting state and a cognitively challenging language task. We evaluate the extent to which brain states show hemispheric asymmetries and how various phenotypic factors such as handedness and gender might influence network dynamics, discovering a right-lateralized brain state that occurred more frequently in men than in women and more frequently in right-handed versus left-handed individuals. Longitudinal brain state changes were also shown in 42 patients with subcortical stroke over 6 months. Our approach could quantify subject-specific dynamic brain states and has potential for use in both basic and clinical neuroscience research.

Manifold Learning for fMRI time-varying FC

Whole-brain functional connectivity ( FC ) measured with functional MRI (fMRI) evolve over time in meaningful ways at temporal scales going from years (e.g., development) to seconds (e.g., within-scan time-varying FC ( tvFC )). Yet, our ability to explore tvFC is severely constrained by its large dimensionality (several thousands). To overcome this difficulty, researchers seek to generate low dimensional representations (e.g., 2D and 3D scatter plots) expected to retain its most informative aspects (e.g., relationships to behavior, disease progression). Limited prior empirical work suggests that manifold learning techniques ( MLTs )-namely those seeking to infer a low dimensional non-linear surface (i.e., the manifold) where most of the data lies-are good candidates for accomplishing this task. Here we explore this possibility in detail. First, we discuss why one should expect tv FC data to lie on a low dimensional manifold. Second, we estimate what is the intrinsic dimension (i.e., minimum number of latent dimensions; ID ) of tvFC data manifolds. Third, we describe the inner workings of three state-of-the-art MLTs : Laplacian Eigenmaps ( LE ), T-distributed Stochastic Neighbor Embedding ( T-SNE ), and Uniform Manifold Approximation and Projection ( UMAP ). For each method, we empirically evaluate its ability to generate neuro-biologically meaningful representations of tvFC data, as well as their robustness against hyper-parameter selection. Our results show that tvFC data has an ID that ranges between 4 and 26, and that ID varies significantly between rest and task states. We also show how all three methods can effectively capture subject identity and task being performed: UMAP and T-SNE can capture these two levels of detail concurrently, but L E could only capture one at a time. We observed substantial variability in embedding quality across MLTs , and within- MLT as a function of hyper-parameter selection. To help alleviate this issue, we provide heuristics that can inform future studies. Finally, we also demonstrate the importance of feature normalization when combining data across subjects and the role that temporal autocorrelation plays in the application of MLTs to tvFC data. Overall, we conclude that while MLTs can be useful to generate summary views of labeled tvFC data, their application to unlabeled data such as resting-state remains challenging.

A method for reconstruction of interpretable brain networks from transient synchronization in resting-state BOLD fluctuations

Resting-state (rs) fMRI has been widely used to examine brain-wide large-scale spatiotemporal architectures, known as resting-state networks (RSNs). Recent studies have focused on the temporally evolving characteristics of RSNs, but it is unclear what temporal characteristics are reflected in the networks. To address this issue, we devised a novel method for voxel-based visualization of spatiotemporal characteristics of rs-fMRI with a time scale of tens of seconds. We first extracted clusters of dominant activity-patterns using a region-of-interest approach and then used these temporal patterns of the clusters to obtain voxel-based activation patterns related to the clusters. We found that activation patterns related to the clusters temporally evolved with a characteristic temporal structure and showed mutual temporal alternations over minutes. The voxel-based representation allowed the decoding of activation patterns of the clusters in rs-fMRI using a meta-analysis of functional activations. The activation patterns of the clusters were correlated with behavioral measures. Taken together, our analysis highlights a novel approach to examine brain activity dynamics during rest.

Resting state brain dynamics: Associations with childhood sexual abuse and major depressive disorder

Early life stress (ELS) and major depressive disorder (MDD) share neural network abnormalities. However, it is unclear how ELS and MDD may separately and/or jointly relate to brain networks, and whether neural differences exist between depressed individuals with vs without ELS. Moreover, prior work evaluated static versus dynamic network properties, a critical gap considering brain networks show changes in coordinated activity over time. Seventy-one unmedicated females with and without childhood sexual abuse (CSA) histories and/or MDD completed a resting state scan and a stress task in which cortisol and affective ratings were collected. Recurring functional network co-activation patterns (CAPs) were examined and time in CAP (number of times each CAP is expressed) and transition frequencies (transitioning between different CAPs) were computed. The effects of MDD and CSA on CAP metrics were examined and CAP metrics were correlated with depression and stress-related variables. Results showed that MDD, but not CSA, related to CAP metrics. Specifically, individuals with MDD (N = 35) relative to HCs (N = 36), spent more time in a posterior default mode (DMN)-frontoparietal network (FPN) CAP and transitioned more frequently between posterior DMN-FPN and prototypical DMN CAPs. Across groups, more time spent in a posterior DMN-FPN CAP and greater DMN-FPN and prototypical DMN CAP transition frequencies were linked to higher rumination. Imbalances between the DMN and the FPN appear central to MDD and might contribute to MDD-related cognitive dysfunction, including rumination. Unexpectedly, CSA did not modulate such dysfunctions, a finding that needs to be replicated by future studies with larger sample sizes.

Static and Dynamic Functional Connectivity Alterations in Alzheimer's Disease and Neuropsychiatric Diseases

To date, numerous studies have documented various alterations in resting brain activity in Alzheimer's disease (AD) and other neuropsychiatric diseases. In particular, disease-related alterations of functional connectivity (FC) in the resting state networks (RSN) have been documented. Altered FC in RSN is useful not only for interpreting the phenotype of diseases but also for diagnosing the diseases. More recently, several studies proposed the dynamics of resting-brain activity as a useful marker for detecting altered RSNs related to AD and other diseases. In contrast to previous studies, which focused on FC calculated using an entire fMRI scan (static FC), these newer studies focused the on temporal dynamics of FC within the scan (dynamic FC) to provide more sensitive measures to characterize RSNs. However, despite the increasing popularity of dFC, several studies cautioned that the results obtained in commonly used analyses for dFC require careful interpretation. In this mini-review, we review recent studies exploring alterations of static and dynamic functional connectivity in AD and other neuropsychiatric diseases. We then discuss how to utilize and interpret dFC for studying resting brain activity in diseases.