Global and structured waves of rs-fMRI signal identified as putative propagation of spontaneous neural activity

doi:10.1016/j.neuroimage.2016.03.033

NeuroImage

Volume 133, June 2016, Pages 331-340

https://doi.org/10.1016/j.neuroimage.2016.03.033 Get rights and content

Highlights

•
Time lag structures of vascular and neural activation are examined.
•
Multiple structured propagation of the rs-fMRI signal was identified.
•
Putative neural streams showed propagation from task-positive regions to the DMN.

Abstract

Conventional resting-state fMRI (rs-fMRI) studies have focused on investigating the synchronous neural activity in functionally relevant distant regions that are termed as resting-state networks. On the other hand, less is known about the spatiotemporal dynamics of the spontaneous activity of the brain. By examining the characteristics of both rs-fMRI and vascular time lag that was measured using dynamic susceptibility contrast-enhanced perfusion weighted imaging, the present study identifies several structured propagation of the rs-fMRI signal as putative neural streams. Temporal shift of both rs-fMRI and perfusion imaging data in each voxel compared with the averaged whole-brain signal was computed using cross-correlation analysis. In contrast to the uniformity of the vascular time lag across subjects, whole-brain rs-fMRI time lag was estimated to be composed of three independent components. After regression of vascular time lag, independent component analysis was applied to rs-fMRI data. The putative neural streams showed slow propagation of the signal from task-positive regions to main nodes of the default mode network, which may represent a mode of transmission underlying the interactions among the resting-state networks.

Introduction

Resting-state functional magnetic resonance imaging (rs-fMRI) has become a leading paradigm for studying the functional organization of the brain (Raichle, 2009). The spatial patterns identified as the areas with synchronous blood oxygenation-level dependent (BOLD) fluctuations are termed as resting-state networks (RSNs) (Fox et al., 2005). These networks are closely related to anatomical connectivity among the neural subsystems that have been revealed by a wide variety of visual, sensorimotor and cognitive task paradigms (Vincent, J.L., et al., 2007, Zhang, D., et al., 2010). Furthermore, clinical studies on neurodegenerative and psychiatric diseases have revealed dysfunction of the RSNs, which emphasizes the significance of the large-scale brain networks (Fox and Greicius, 2010). While conventional analyses based on seed-based correlation or independent component analysis (ICA) are suitable for evaluating these network activities, implicit assumption in the use of the approaches is that spatial distribution of the synchronous neural activity is temporally constant. However, animal studies have revealed that spontaneous neural activity is spatiotemporally structured, and propagating waves of activity have been recorded in many species (for review see, Muller and Destexhe, 2012). Neuronal membrane potential in the cortex is known to undergo a spontaneous transition between up and down states in the absence of sensory inputs (Lampl, I., et al., 1999, Petersen, C.C., et al., 2003, Shu, Y., et al., 2003, Steriade, M., et al., 1993). Population activity of the neurons during the up state manifests as propagating waves not only within a part of the cortex (Civillico, E.F. and Contreras, D., 2012, Ferezou, I., et al., 2006, Petersen, C.C., et al., 2003, Xu, W., et al., 2007), but also throughout the entire brain (Mohajerani, M.H., et al., 2010, Stroh, A., et al., 2013). Repetitive spatiotemporal patterns of spontaneous neural activity have also been identified using rs-fMRI (Majeed, W., et al., 2009, Majeed, W., et al., 2011). More recently, a human rs-fMRI study investigated the dimensionality of the time lag structures of the spontaneous neural activity in the human brain and showed the existence of multiple spatiotemporal patterns (Mitra, A., et al., 2015, Mitra, A., et al., 2016). In these rs-fMRI studies, BOLD signal time lag was directly interpreted as time lag of neural activity. However, since BOLD is indirect measurement of neural activity through vascular response, it is necessarily influenced by vascular time lag. Indeed, another line of studies employing similar temporal shift analyses had verified that these measurements are closely related to vascular time lag in the brain; in a series of studies on low frequency oscillations of BOLD signal, Tong et al. demonstrated the propagation of physiological noise in the cerebral vasculature (Tong, Y. and Frederick, B., 2010, Tong, Y. and Frederick, B., 2012, Tong, Y., et al., 2013, Tong, Y. and Frederick, B., 2014a, Tong, Y., et al., 2014, Tong, Y. and Frederick, B., 2014b).

Furthermore, exploiting rs-fMRI BOLD fluctuations as endogenous blood flow tracer, several independent clinical studies have consistently identified vascular time lag in patients suffering from severe hypoperfusion or ischemia (Amemiya, S., et al., 2014, Christen, T., et al., 2015, Lv, Y., et al., 2013). These studies showed that rs-fMRI signal delay compared with the global signal is highly correlated with the delay of cerebrovascular perfusion that was measured using dynamic susceptibility contrast-enhanced perfusion weighted imaging (Amemiya, S., et al., 2014, Christen, T., et al., 2015, Lv, Y., et al., 2013). Consistent with the notion that the major source of the global signal is physiological noise associated with cardiac pulsation and respiration, these studies identified signal delay in MR-defined ischemic penumbra (Amemiya, S., et al., 2014, Lv, Y., et al., 2013) as well as core areas of the infarction (Amemiya et al., 2014) where normal spontaneous neural activity is known to be absent. However, it does not necessarily preclude the contribution of neural activity in global signal. Both electroencephalographic work in humans and microelectrode recordings in anesthetized monkeys have indeed shown widespread BOLD fluctuations to be correlated with slow changes in neural activity (He, B.J., et al., 2008, Leopold, D.A., et al., 2003, Scholvinck, M.L., et al., 2010). The notion that similar temporal shift analysis can sometimes reveal time lag of neural origin (neural-weighted time lag), while at other times the lag is considered to reflect regional difference of the vascular dynamics (vascular-weighted time lag) is not contradictory but rather a consequence of the fact that BOLD reflects hemodynamic response triggered by neural activity that is combined with the noise. Consequently, it is impossible to infer neural time lag structure without knowledge of the vascular time lag. Indeed, careful examination in a study on propagating neural activity under motor task that introduced auto-regressive modeling of fMRI data showed that their model is valid under the assumption that the variability in the hemodynamic response function is less than the time scale of the information flow in the network (Garg et al., 2011). However actual effect of vascular time lag or individual difference of the vascular component remains unknown.

The present study aims to examine if there exists structured propagation of intrinsic neural activity that is reflected in the rs-fMRI data taking into account the issues. In order to examine the source of time lag, we obtained both rs-fMRI and perfusion data of the same subjects and directly compared the two measurements. In addition to the dimensionality of the rs-fMRI time lag across multiple-subject data, we also examined the dimensionality of the vascular time lag in the same group of subjects to verify if there exists wide variability in pattern of vascular time lag that could affect the neural time lag measurement of multiple-subject data. The temporal shift of each voxel was computed as relative time lag compared with the global signal because, 1) rs-fMRI signal in each voxel is highly correlated with global signal, and the temporal shift analysis that employs global signal as the reference signal can reliably detect the time lag in each voxel with high reproducibility even at the individual level (Amemiya, S., et al., 2014, Christen, T., et al., 2015, Lv, Y., et al., 2013), and because 2) if there exists neural activity propagating throughout the brain, it must be reflected in the global signal change. Thus global signal regression could eliminate the main source of information. After estimating the dimensionality of the rs-fMRI time lag, we tried separating the vascular component from the rs-fMRI time lag maps to infer putative streams of propagating neural activity.

Section snippets

Participants

Fifteen subjects (eight men; age, 54.3 ± 11.4 years) who have no history of neurological or psychiatric disorder and who gave written informed consent participated in the study. All had no abnormality of the brain or the cerebrovascular system on MR imaging. All procedures were in compliance with the Declaration of Helsinki, and the institutional review board of the University of Tokyo approved the study.

Data acquisition

Ten-minute rs-fMRI and dynamic susceptibility contrast-enhanced perfusion weighted imaging—in

Stability of the rs-fMRI time lag measurement

Time lag maps computed from the data preprocessed without quadratic detrending were mostly the same with the maps obtained from the data with quadratic detrending (Fig. S1). Pearson's correlation coefficients between each pair of time lag maps of the same data with and without quadratic detrending were r = 0.99997 ± 3.57 × 10^− 5 (Fig. S1).

The split-half analysis confirmed internal consistency of the time lag measurement. Pearson's correlation coefficients between rs-fMRI time lag maps computed from

Discussion

In the present study, we confirmed the presence of multiple global and structured spatiotemporal streams of rs-fMRI signal, in contrast to the uniformity of the vascular time lag. Inferences in fMRI depend on accurate estimates of effect of hemodynamics. Given the difficulty in isolating neural from vascular time lag with rs-fMRI data alone, some investigators have tried to confirm the neural contribution by demonstrating that the manipulation of the neural state could alter the temporal shift

Conclusion

Using time shift analysis applied to rs-fMRI as well as perfusion data and by decomposing them, we have demonstrated the presence of multiple structured spatiotemporal streams of rs-fMRI signal in contrast to the uniformity of the vascular time lag across subjects. The putative neural streams showed slow propagation of the signal from task-positive regions to main nodes of the default mode network, which may represent a mode of transmission underlying the interactions among the resting-state

Acknowledgment

We would like to thank the anonymous reviewers for their helpful comments. This work was supported by Grant-in-Aid for Young Scientists (B) (25861074) and by grants from SENSHIN Medical Research Foundation to S.A.

References (53)

M.G. Bright et al.
Characterization of regional heterogeneity in cerebrovascular reactivity dynamics using novel hypocapnia task and BOLD fMRI
NeuroImage
(2009)
X.J. Chai et al.
Anticorrelations in resting state networks without global signal regression
NeuroImage
(2012)
C. Chang et al.
Effects of model-based physiological noise correction on default mode network anti-correlations and correlations
NeuroImage
(2009)
J. Cohen-Adad et al.
T2* mapping and B0 orientation-dependence at 7 T reveal cyto- and myeloarchitecture organization of the human cortex
NeuroImage
(2012)
I. Ferezou et al.
Visualizing the cortical representation of whisker touch: voltage-sensitive dye imaging in freely moving mice
Neuron
(2006)
R. Garg et al.
Full-brain auto-regressive modeling (FARM) using fMRI
NeuroImage
(2011)
I. Lampl et al.
Synchronous membrane potential fluctuations in neurons of the cat visual cortex
Neuron
(1999)
Z. Liang et al.
Anticorrelated resting-state functional connectivity in awake rat brain
NeuroImage
(2012)
W. Majeed et al.
Spatiotemporal dynamics of low frequency BOLD fluctuations in rats and humans
NeuroImage
(2011)
G. Mangeat et al.
Multivariate combination of magnetization transfer, T2* and B0 orientation to study the myelo-architecture of the in vivo human cortex
NeuroImage
(2015)

L. Muller et al.

Propagating waves in thalamus, cortex and the thalamocortical system: experiments and models

J. Physiol. Paris

(2012)

K. Murphy et al.

The impact of global signal regression on resting state correlations: are anti-correlated networks introduced?

NeuroImage

(2009)

S.J. Schreiber et al.

Dopplersonographic measurement of global cerebral circulation time using echo contrast-enhanced ultrasound in normal individuals and patients with arteriovenous malformations

Ultrasound Med. Biol.

(2002)

A. Stroh et al.

Making waves: initiation and propagation of corticothalamic Ca^2 + waves in vivo

Neuron

(2013)

Y. Tong et al.

Time lag dependent multimodal processing of concurrent fMRI and near-infrared spectroscopy (NIRS) data suggests a global circulatory origin for low-frequency oscillation signals in human brain

NeuroImage

(2010)

Y. Tong et al.

Concurrent fNIRS and fMRI processing allows independent visualization of the propagation of pressure waves and bulk blood flow in the cerebral vasculature

NeuroImage

(2012)

Y. Tong et al.

Evaluating the effects of systemic low frequency oscillations measured in the periphery on the independent component analysis results of resting state networks

NeuroImage

(2013)

W. Xu et al.

Compression and reflection of visually evoked cortical waves

Neuron

(2007)

S. Amemiya et al.

Cerebral hemodynamic impairment: assessment with resting-state functional MR imaging

Radiology

(2014)

M.S. Bartlett

Face Image Analysis by Unsupervised Learning and Redundancy Reduction

(1998)

A.C. Chen et al.

Causal interactions between fronto-parietal central executive and default-mode networks in humans

Proc. Natl. Acad. Sci. U. S. A.

(2013)

T. Christen et al.

Noncontrast mapping of arterial delay and functional connectivity using resting-state functional MRI: a study in Moyamoya patients

J. Magn. Reson. Imaging

(2015)

E.F. Civillico et al.

Spatiotemporal properties of sensory responses in vivo are strongly dependent on network context

Front. Syst. Neurosci.

(2012)

D. Crandell et al.

Cerebral transit time of 99m technetium sodium pertechnetate before and after cerebral arteriography

J. Neurosurg.

(1973)

D.J. Field

What is the goal of sensory coding?

Neural Comput.

(1994)

M.D. Fox et al.

Clinical applications of resting state functional connectivity

Front. Syst. Neurosci.

(2010)

Cited by (24)

Development of top-down cortical propagations in youth
2023, Neuron
Hierarchical processing requires activity propagating between higher- and lower-order cortical areas. However, functional neuroimaging studies have chiefly quantified fluctuations within regions over time rather than propagations occurring over space. Here, we leverage advances in neuroimaging and computer vision to track cortical activity propagations in a large sample of youth (n = 388). We delineate cortical propagations that systematically ascend and descend a cortical hierarchy in all individuals in our developmental cohort, as well as in an independent dataset of densely sampled adults. Further, we demonstrate that top-down, descending hierarchical propagations become more prevalent with greater demands for cognitive control as well as with development in youth. These findings emphasize that hierarchical processing is reflected in the directionality of propagating cortical activity and suggest top-down propagations as a potential mechanism of neurocognitive maturation in youth.
Reliability and sensitivity to alterered hemodynamics measured with resting-state fMRI metrics: Comparison with <sup>123</sup>I-IMP SPECT
2022, NeuroImage
Citation Excerpt :
Using the measurement within the cerebellum as the reference signal, we showed that while the reliability was significantly lower for fMRI than SPECT, the sensitivity to postoperative changes was higher for fMRI TDc and equivalent for fMRI Rc. Consistent with previous studies, the global mean signal of the rsfMRI showed a positive correlation with each voxel's signal throughout the cerebral gray matter in healthy subjects, with a time delay similar to that of vascular perfusion (Amemiya et al., 2016, 2019, 2020; Tong et al., 2017, 2019). Although the source of this widespread ubiquitous signal is still controversial (Amemiya et al., 2020; Drew, 2019; Liu et al., 2017; Tong et al., 2019), these findings suggest that the low-frequency fluctuations mainly result from a phenomenon triggering the vascular response throughout the brain, rather than a locally confined event or motion artifact.
Blood oxygenation level-dependent (BOLD) contrast is sensitive to local hemodynamic changes and thus is applicable to imaging perfusion or vascular reactivity. However, knowledge about its measurement characteristics compared to reference standard perfusion imaging is limited. This study longitudinally evaluated perfusion in patients with steno-occlusive disease using resting-state functional MRI (rsfMRI) acquired before and within nine days of anterior circulation revascularization in patients with large cerebral artery steno-occlusive diseases. The reliability and sensitivity to longitudinal changes of rsfMRI temporal correlation (Rc) and time delay (TDc) relative to the cerebellar signal were examined voxel-wise in comparison with single-photon emission CT (SPECT) cerebral blood flow (CBF) using the within-subject standard deviation (Sw) and intraclass correlation coefficients (ICCs). For statistical comparisons, the standard deviation (SD) of longitudinal changes within the cerebellum, the number of voxels with significant changes in the left middle cerebral artery territory ipsilateral to surgery, and their average changes relative to the cerebellar SD were evaluated. The test-retest reliability of the fMRI metrics was also similarly evaluated using the human connectome project (HCP) healthy young adult dataset. The test-retest time interval was 31 ± 18 days. Test-retest reliability was significantly higher for SPECT (cerebellar SD: −2.59 ± 0.20) than for fMRI metrics (cerebellar SD: Rc, −2.34 ± 0.24, p = 0.04; TDc, −2.19 ± 0.21, p = 0.003). Sensitivity to postoperative changes, which was evaluated as the number of voxels, was significantly higher for fMRI TDc (8.78 ± 0.72) than for Rc (7.42 ± 1.48, p = 0.03) or SPECT CBF (6.88 ± 0.67, p < 0.001). The ratio between the average Rc, TDc, and SPECT CBF changes within the left MCA target region and cerebellar SD was also significantly higher for fMRI TDc (1.21 ± 0.79) than Rc (0.48 ± 0.94, p = 0.006) or SPECT CBF (0.23 ± 0.57, p = 0.001). The measurement variability of time delay was also larger than that of temporal correlation in HCP data within the cerebellum (t = −8.7, p < 0.001) or in the whole-brain (t = −27.4, p < 0.001) gray matter. These data suggest that fMRI time delay is more sensitive to the hemodynamic changes than SPECT CBF, although the reliability is lower. The implication for fMRI connectivity studies is that temporal correlation can be significantly decreased due to altered hemodynamics, even in cases with normal CBF.
Physical exercise rectifies CUMS-induced aberrant regional homogeneity in mice accompanied by the adjustment of skeletal muscle PGC-1a/IDO1 signals and hippocampal function
2020, Behavioural Brain Research
Citation Excerpt :
There are various restrictions in conducting research on depression and the efficacy of treatments due to ethical and technological reasons without real-time detection in vivo. Thus, it is imperative to use noninvasive methods to improve our understanding of antidepressant effects of physical exercise.There is a rapid growth of researches that applied resting-state functional magnetic resonance imaging (rs-fMRI), a widely-recognized neuroimaging technique that noninvasively measures neural activity to study mental disorders [19,20], including depression [21], schizophrenia [22]. Amplitude of the low-frequency fluctuations (ALFF) and the regional homogeneity (ReHo) are two methods used to detect alterations in variety of mental disorders [23].
Depression is a debilitating condition with a profound influence on quality of life for millions of people globally. Physical exercise has been broadly recognized for its therapeutic effects on depression, but the mechanisms that underlie its benefits remain unknown. In the study, we investigated whether the physical exercise of could be a protection from stress-induced depression and its impact on the brain activity of Regional Homogeneity (ReHo) in mice.
Adult male C57BL/6 mice were assigned to one of the following groups: control group; exercise group, 2 h/day in a running wheel apparatus; chronic unpredictable mild stress (CUMS) group; CUMS + exercise group. rs-fMRI was applied to detect the changes of regional spontaneous activity.
Firstly, CUMS-induced depressive behavior was significantly reduced by exercise. Base on the ReHo analysis, disorders of the regional spontaneous activity in the brain of CUMS mice, primarily in the limbic system, especially in the hippocampus and PFC, motor cortex, sensory cortex, visual cortex were found. While exercise remarkably prevented the CUMS-induced chaos of brain activity in parts of the above regions, such as cortex, hippocampus and corpus callosum. These results suggested physical exercise could prevent the dysfunction of mood-regulating circuit in CUMS model. Furthermore, exercise improved skeletal muscle PGC-1a and hippocampal BDNF levels in stress mice, and reduced IDO1 in skeletal muscle.
These results suggested that exercise prevented CUMS induced depressive behaviors and brain regional spontaneous activity in mice, accompanied with the adjustment of skeletal muscle PGC-1a/IDO1 signals and hippocampal function.
Latency analysis of resting-state BOLD-fMRI reveals traveling waves in visual cortex linking task-positive and task-negative networks
2019, NeuroImage
Citation Excerpt :
Summary and main findings. Whereas most resting-state BOLD-fMRI studies currently do not exploit the latency structure within the data, the existence of latencies in resting-state BOLD-fMRI data is receiving increased attention due to faster acquisition rates and the development of new analysis methods, fueled by the discovery of propagating activity patterns and latency structure that are consistent across scans, participants, and data-sets (Majeed et al., 2009, 2011; Mitra et al., 2014, 2015a; Siegel et al., 2016; Amemiya et al., 2016; Matsui et al., 2016; Raatikainen et al., 2017; Gravel et al., 2017). Furthermore, there is both indirect as well as direct evidence that the latency structure of resting-state BOLD fluctuations can only partly be explained by spatial variability in haemodynamic responses and are of electrophysiological origin (Mitra et al., 2014; Thompson et al., 2014; Amemiya et al., 2016; Matsui et al., 2016; Vanni et al., 2017).
Due to the low temporal resolution of BOLD-fMRI, imaging studies on human brain function have almost exclusively focused on instantaneous correlations within the data. Developments in hardware and acquisition protocols, however, are offering data with higher sampling rates that allow investigating the latency structure of BOLD-fMRI data. In this study we describe a method for analyzing the latency structure within BOLD-fMRI data and apply it to resting-state data of 94 participants from the Human Connectome Project. The method shows that task-positive and task-negative networks are integrated through traveling BOLD waves within early visual cortex. The waves are initiated at the periphery of the visual field and propagate towards the fovea. This observation suggests a mechanism for the functional integration of task-positive and task-negative networks, argues for an eccentricity-based view on visual information processing, and contributes to the emerging view that resting-state BOLD-fMRI fluctuations are superpositions of inherently spatiotemporal patterns.
On time delay estimation and sampling error in resting-state fMRI
2019, NeuroImage
Citation Excerpt :
Finally, we conclude with strategies to reduce the influence of sampling error on inferences drawn from TDE. Importantly, although interregional TDE is perhaps the most straightforward approach for quantifying BOLD propagation (Fig. 1), results are pertinent to the variety of approaches that have been used to detect or exploit temporal offsets among fMRI signals (Goebel et al., 2003; Sun et al., 2005; Garg et al., 2011; Majeed et al., 2011; Friston et al., 2014; Mitra et al., 2014; Amemiya et al., 2016; Gilson et al., 2016; Raatikainen et al., 2017) (see Friston et al. (Friston et al., 2013; Friston et al., 2014) for taxonomy of these approaches). While the CCF of periodic time series is likely to feature multiple extrema, BOLD signals are aperiodic (He et al., 2010; Zarahn et al., 1997) and almost always produces a single, well-defined cross-covariance extremum for a given pair of time series, typically in the range of ±1 s.
Accumulating evidence indicates that resting-state functional magnetic resonance imaging (rsfMRI) signals correspond to propagating electrophysiological infra-slow activity (<0.1 Hz). Thus, pairwise correlations (zero-lag functional connectivity (FC)) and temporal delays among regional rsfMRI signals provide useful, complementary descriptions of spatiotemporal structure in infra-slow activity. However, the slow nature of fMRI signals implies that practical scan durations cannot provide sufficient independent temporal samples to stabilize either of these measures. Here, we examine factors affecting sampling variability in both time delay estimation (TDE) and FC. Although both TDE and FC accuracy are highly sensitive to data quantity, we use surrogate fMRI time series to study how the former is additionally related to the magnitude of a given pairwise correlation and, to a lesser extent, the temporal sampling rate. These contingencies are further explored in real data comprising 30-min rsfMRI scans, where sampling error (i.e., limited accuracy owing to insufficient data quantity) emerges as a significant but underappreciated challenge to FC and, even more so, to TDE. Exclusion of high-motion epochs exacerbates sampling error; thus, both sides of the bias-variance (or data quality-quantity) tradeoff associated with data exclusion should be considered when analyzing rsfMRI data. Finally, we present strategies for TDE in motion-corrupted data, for characterizing sampling error in TDE and FC, and for mitigating the influence of sampling error on lag-based analyses.
Combined spatiotemporal ICA (stICA) for continuous and dynamic lag structure analysis of MREG data
2017, NeuroImage
Citation Excerpt :
The null-distributions suggest that, when having reasonable amount of samples (nodes), the mean time lag between the DMN nodes is zero, and the probability of having either DMNpcc or DMNvmpf as a ‘preceding component’ is near equal. Moreover, the standard deviation of time lags in whole time window (Fig. 5B) is around 4 s, which suggest that all the presented first node lag results lie within the standard deviation, supporting the earlier findings that there is a consistent temporal lag structure across groups of individuals (Amemiya et al., 2016; Mitra et al., 2014). Therefore, it might be difficult to draw any deep conclusions on average lag results of whole time window since there are only 25 subjects with one lag value.
This study investigated lag structure in the resting-state fMRI by applying a novel independent component (ICA) method to magnetic resonance encephalography (MREG) data. Briefly, the spatial ICA (sICA) was used for defining the frontal and back nodes of the default mode network (DMN), and the temporal ICA (tICA), which is enabled by the high temporal resolution of MREG (TR=100ms), was used to separate both neuronal and physiological components of these two spatial map regions. Subsequently, lag structure was investigated between the frontal (DMNvmpf) and posterior (DMNpcc) DMN nodes using both conventional method with all-time points and a sliding-window approach.
A rigorous noise exclusion criterion was applied for tICs to remove physiological pulsations, motion and system artefacts. All the de-noised tICs were used to calculate the null-distributions both for expected lag variability over time and over subjects. Lag analysis was done for the three highest correlating denoised tICA pairs.
Mean time lag of 0.6 s (± 0.5 std) and mean absolute correlation of 0.69 (± 0.08) between the highest correlating tICA pairs of DMN nodes was observed throughout the whole analyzed period. In dynamic 2 min window analysis, there was large variability over subjects as ranging between 1–10 sec. Directionality varied between these highly correlating sources an average 28.8% of the possible number of direction changes.
The null models show highly consistent correlation and lag structure between DMN nodes both in continuous and dynamic analysis. The mean time lag of a null-model over time between all denoised DMN nodes was 0.0 s and, thus the probability of having either DMNpcc or DMNvmpf as a preceding component is near equal. All the lag values of highest correlating tICA pairs over subjects lie within the standard deviation range of a null-model in whole time window analysis, supporting the earlier findings that there is a consistent temporal lag structure across groups of individuals. However, in dynamic analysis, there are lag values exceeding the threshold of significance of a null-model meaning that there might be biologically meaningful variation in this measure. Taken together the variability in lag and the presence of high activity peaks during strong connectivity indicate that individual avalanches may play an important role in defining dynamic independence in resting state connectivity within networks.

View all citing articles on Scopus

View full text