Spontaneous activity changes in large-scale cortical networks in older adults couple to distinct hemodynamic morphology

  1. Tatiana Sitnikova
  2. Jeremy W. Hughes
  3. Casey M. Howard
  4. Kimberly A. Stephens
  5. Mark W. Woolrich
  6. David H. Salat

Reviewed by

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

Log in to save this article

Abstract

Neurovascular coupling is a dynamic core mechanism supporting brain energy demand. Therefore, even spontaneous changes in neural activity are expected to evoke a vascular hemodynamic response (HDR). Here, we developed a novel procedure for estimating transient states in intrinsic activity of neural networks based on source-localized electroencephalogram in combination with HDR estimation based on simultaneous rapid-acquisition functional magnetic resonance imaging. We demonstrate a readily apparent spatiotemporal correspondence between electrophysiological and HDR signals, describing for the first time how features of neurovascular coupling may differ among large-scale brain networks. In the default mode network, the HDR pattern in our older adult participants was associated with a surrogate marker of cerebrovascular deterioration and predicted alterations in temporal structure of fast intrinsic electrophysiological activity linked to memory decline. These results show the potential of our technique for making inferences about neural and vascular processes in higher-level cognitive networks in healthy and at-risk populations.

Article activity feed

  1. eLife

    Reviewer #3: (Daniele Marinazzo)

    Dear authors,

    Thanks for the opportunity to read this nice paper. I appreciated the quality of the data analysis, and the quest towards associating electrophysiology and BOLD data through a data-driven transfer function, which can be interpreted as a proxy of the HRF. Also I completely agree with you that we need to move beyond a canonical response.

    There are a few issues I would like to discuss with you. I have done quite some work in this sense. On one hand this is good (and I think it's also the reason why I was invited to review this paper), on the other one there is always the risk that I have shaped my own goggles in these last years, and that I am projecting on your work some doubts and issues that I have with my own. In this case I apologize in advance, and I hope that we can have an enriching conversation.

    Please forgive me if I start by my own work; there is always the danger that reviewers try to make authors write the paper that they would write themselves, I will keep this in mind, but on the other hand I think that the best way to convey my thoughts to you is to let them flow as they come.

    So, here's our toolbox: https://www.nitrc.org/projects/rshrf. The idea behind it is that we can take peaks in the BOLD signal and take them as signatures of a pseudo neural event happening some time before at the neural level. This is in line with this work (which could also be relevant with respect to your power law figures):

    Tagliazucchi E, Balenzuela P, Fraiman D, Chialvo DR. Criticality in large-scale brain FMRI dynamics unveiled by a novel point process analysis. Front Physiol. 2012;3:15. Published 2012 Feb 8. doi:10.3389/fphys.2012.00015 and with the subsequent spatial clustering approach which has been called coactivation patterns (CAP)

    Liu X, Zhang N, Chang C, Duyn JH. Co-activation patterns in resting-state fMRI signals. Neuroimage. 2018;180(Pt B):485-494. doi:10.1016/j.neuroimage.2018.01.041 and innovation CAPs

    Karahanoğlu FI, Caballero-Gaudes C, Lazeyras F, Van de Ville D. Total activation: fMRI deconvolution through spatio-temporal regularization. Neuroimage. 2013;73:121-134. doi:10.1016/j.neuroimage.2013.01.067 Karahanoğlu FI, Van De Ville D. Transient brain activity disentangles fMRI resting-state dynamics in terms of spatially and temporally overlapping networks. Nat Commun. 2015;6:7751. Published 2015 Jul 16. doi:10.1038/ncomms8751

    Zoller DM, Bolton TAW, Karahanoglu FI, Eliez S, Schaer M, Van De Ville D. Robust Recovery of Temporal Overlap Between Network Activity Using Transient-Informed Spatio-Temporal Regression. IEEE Trans Med Imaging. 2019;38(1):291-302. doi:10.1109/TMI.2018.2863944

    We then fit these peaks with a GLM, with the time lag as a free parameter. We use several families of basis functions. In the original paper (Wu GR, Liao W, Stramaglia S, Ding JR, Chen H, Marinazzo D. A blind deconvolution approach to recover effective connectivity brain networks from resting state fMRI data. Med Image Anal. 2013;17(3):365-374. doi:10.1016/j.media.2013.01.003) we used canonical HRF and FIR (together with the rBETA, which is basically the portion of the BOLD peak exceeding a certain threshold, as in the Tagliazucchi paper above).

    We then included a mixture of gamma functions together with other families of basis functions in subsequent versions of the toolbox. Then we set up for validation of the approach with electrophysiological signatures, and that's where the doubts and pain kicked in. Some results on simultaneous EEG-fMRI, reported here (Wu G, Marinazzo D. 2015. Retrieving the Hemodynamic Response Function in resting state fMRI: methodology and applications. PeerJ PrePrints 3:e1317v1 https://doi.org/10.7287/peerj.preprints.1317v1 Wu GR, Deshpande G, Laureys S, Marinazzo D. Retrieving the Hemodynamic Response Function in resting state fMRI: Methodology and application. Conf Proc IEEE Eng Med Biol Soc. 2015;2015:6050-6053. doi:10.1109/EMBC.2015.7319771) were encouraging: for example we saw that the positive correlation between envelope of EEG and BOLD in the occipital cortex becomes more positive when we use instead the deconvolved BOLD and the EEG, while the negative correlation in the thalamus becomes more negative.

    Other things present in the PeerJ preprint were encouraging too (and I mention them since I think that they can be relevant to the validation of your approach): namely the retrieval of a simulated ground truth HRF within certain realistic limits of SNR and jitter, the correlation with cerebral blood flow (even though physiological regressors should always be taken into account, see: Wu GR, Marinazzo D. Sensitivity of the resting-state haemodynamic response function estimation to autonomic nervous system fluctuations. Philos Trans A Math Phys Eng Sci. 2016;374(2067):20150190. doi:10.1098/rsta.2015.0190 and this becomes even more relevant when considering aging and clinical datasets), and some similarity across resting state networks.

    So, the question is: can we really trust that peaks in M/EEG reflect the local pseudo-events that would originate the BOLD signal? Reading work by people who had thoroughly investigated this, e.g.

    Logothetis NK, Pauls J, Augath M, Trinath T, Oeltermann A. Neurophysiological investigation of the basis of the fMRI signal. Nature. 2001;412(6843):150-157. doi:10.1038/35084005

    Chen X, Sobczak F, Chen Y, et al. Mapping optogenetically-driven single-vessel fMRI with concurrent neuronal calcium recordings in the rat hippocampus. Nat Commun. 2019;10(1):5239. Published 2019 Nov 20. doi:10.1038/s41467-019-12850-x

    Yu X, He Y, Wang M, et al. Sensory and optogenetically driven single-vessel fMRI. Nat Methods. 2016;13(4):337-340. doi:10.1038/nmeth.3765

    and conversing with them, I got (almost) convinced that it's unlikely that spikes in coarsely recorded or reconstructed M/EEG signal can be one to one mapped to the HRF inducing events that we use in GLM (calcium or even better glutamate signal could be a better choice).

    Now, I like the way you associated HMM states with hemodynamic ones, thus adopting a more systemic/dynamical view, and taking fractional occupancy as a trigger. Do you think that these triggers can be better markers of BOLD-inducing neural events?

    Other issues:

    • What to make of events that are very close, and that would thus violate the assumption of linearity of the GLM?

    • Apart from hemodynamic changes, can aging be associated with different electrophysiological spectral features (both periodic and aperiodic), which in turn could influence the HMM analysis?

    • Detection of brain-behavior relationships with a non-huge dataset can be misleading, see for example this recent study:

    Towards Reproducible Brain-Wide Association Studies Scott Marek, Brenden Tervo-Clemmens, Finnegan J. Calabro, David F. Montez, Benjamin P. Kay, Alexander S. Hatoum, Meghan Rose Donohue, William Foran, Ryland L. Miller, Eric Feczko, Oscar Miranda-Dominguez, Alice M. Graham, Eric A. Earl, Anders J. Perrone, Michaela Cordova, Olivia Doyle, Lucille A. Moore, Greg Conan, Johnny Uriarte, Kathy Snider, Angela Tam, Jianzhong Chen, Dillan J. Newbold, Annie Zheng, Nicole A. Seider, Andrew N. Van, Timothy O. Laumann, Wesley K. Thompson, Deanna J. Greene, Steven E. Petersen, Thomas E. Nichols, B.T. Thomas Yeo, Deanna M. Barch, Hugh Garavan, Beatriz Luna, Damien A. Fair, Nico U.F. Dosenbach bioRxiv 2020.08.21.257758; doi: 10.1101/2020.08.21.257758

    • Why the parcellation in 38 regions? How are the results consistent/robust with finer parcellations?

    • You state that the DMN "is susceptible to aging and neurodegenerative disease". That's certainly probable, the thing is that DMN is possibly sensitive to everything and specific to a very few things.

    • Instead of a point-by-point statistical test, you could use the 3dMVM algorithm in AFNI (your reference 20) to test differences in the shape as a whole.

    • You analyse data from older subjects only. How confident can you be that you are observing effects specific to aging?

    Thanks for listening to this review version of "more of a comment than a question".

  2. eLife

    Reviewer #2:

    General assessment:

    The study investigated transient coupling between EEG and fMRI during resting state in 15 elderly participants using the previously established Hidden Markov Model approach. Key findings include: 1) deviations of the hemodynamic response function (HDR) in higher-order versus sensory brain networks, 2) Power law scaling for duration and relative frequency of states, 3) associations between state duration and HDR alterations, 4) cross-sectional associations between HDR alterations, white matter signal anomalies and memory performance.

    The work is rigorously designed and very well presented. The findings are potentially of strong significance to several neuroscience communities.

    Major concerns:

    My enthusiasm was only somewhat mitigated by methodological issues related to the sample size for cross-sectional reference and missed opportunities for more specific analysis of the EEG.

    1. Statistical power analysis has been conducted prior to data collection, which is very laudable. Nevertheless, n=15 is a very small sample for cross-sectional inference and commonly leads to false positives despite large observed effect sizes and small p-values (it takes easily up to 200 samples to detect true zero correlations). On the other hand, the within-subject results are far more well-posed from a statistical view, hence, more strongly supported by the data.

    Recommendations:

    • The issue should be non-defensively addressed in a well-identified section or paragraph inside the discussion. The sample size should be mentioned in the abstract too.

    • The authors could put more emphasis on the participants as replication units for observations. For the theoretical perspective, the work by Smith and Little may be of help here: https://link.springer.com/article/10.3758/s13423-018-1451-8. In terms of methods, more emphasis should be put on demonstrating representativeness for example using prevalence statistics (see e.g. Donnhäuser, Florin & Baillet https://doi.org/10.1371/journal.pcbi.1005990)

    • Supplements should display the most important findings for each subject to reveal representatives of the group averages.

    • For state duration analysis (boxplots) linear mixed effect models (varying slope models) may be an interesting option to inject additional uncertainty into the estimates and allow for partial pooling through shrinkage of subject-level effects.

    • Show more raw signals / topographies to build some trust for the input data. It could be worthwhile to show topographic displays for the main states reported in characteristic frequencies. See also next concern.

    1. The authors seem to have missed an important opportunity to pinpoint the characteristic drivers in terms of EEG frequency bands. The current analysis is based on broadband signals between 4 and 30 Hz, which seems untypical and reduces the specificity of the analysis. Analyzing the spectral drivers of the different state would not only enrich the results in terms of EEG but also provide a more nuanced interpretation. Are the VisN and DAN-states potentially related to changes in alpha power, potentially induced by spontaneous opening and closing of the eyes? What is the most characteristic spectral of the DMN state? ... etc.

    Recommendations:

    • Display the power spectrum indexed by state, ideally for each subject. This would allow inspecting modulation of the power spectra by the state and reveal the characteristic spectral signature without re-analysis.

    • Repeat essential analyses after bandpass filtering in alpha or beta range. For example, if main results look very similar after filtering 8-12 one can conclude that most observations are related to alpha band power.

    • While artifacts have been removed using ICA and the network states do not look like source-localized EOG artifacts, some of the spectral changes e.g. in DAN/VisN might be attributed to transient visual deprivation. This could be investigated by performing control analysis regressing the EOG-channels amplitudes against the HMM states. These results could also enhance the discussion regarding activation/deactivation.

  3. eLife

    Reviewer #1:

    This manuscript uses simultaneous fMRI-EEG to understand the haemodynamic correlates of electrophysiological markers of brain network dynamics. The approach is both interesting and innovative. Many different analyses are conducted, but the manuscript is in general quite hard to follow. There are grammatical errors throughout, sentences/paragraphs are long and dense, and they often use vague/imprecise language or rely on (often) undefined jargon. For example, sentences such as the following example are very difficult to decipher and are found throughout the manuscript: "if replicated, an association between high positive BOLD responsiveness and a DAN electrophysiological state, characterized by low amplitude (i.e., desynchronized) activity deviating from energetically optimal spontaneous patterns, would be consistent with prior evidence that the DMN and DAN represent alternate regimes of intrinsic brain function". As a result, the reader has to work very hard to follow what has been done and to understand the key messages of the paper.

    Much is made of a potential power-law scaling of lifetime/interval times as being indicative of critical dynamics. A power-law distribution does not guarantee criticality, and could arise through other properties. Moreover, to accurately determine whether the proposed power-law is indicative of a scale-free system, the empirical property must be assessed over several orders of magnitude. It appears that only the 25-250 ms range was considered here.

    The KS statistic is used to quantify the distance between the empirical and power-law distributions, which is then used as a marker of the degree of criticality. It is unclear that this metric is appropriate, given that transitions in and out of criticality can be highly non-linear. Moreover, the physiological significance of having some networks in a critical state while others are not is unclear. Each network is part of a broader system (i.e., the brain). How should one interpret apparent gradations of criticality in different parts of the system?

    The sample size is small. I appreciate the complexity of the experimental paradigm, but the correlations do not appear to be robust. The scatterplots mask this to some extent by overlaying results from different brain regions, but close inspection suggests that the correlations in Fig 6 are driven by 2-3 observations with negative BOLD responses, the correlations in Fig 7A-B are driven by two subjects with positive WMSA volume, and Fig 7B is driven by 3 or so subjects with negative power-law fit values (indeed, x~0 in this plot is associated with a wide range of recall scores). Some correction for multiple comparisons is also required given the number of tests performed.

    Figure 1 - panel labels would make it much easier to understand what is shown in this figure relative to the caption.

    Figure 2- the aDMN does not look like the DMN at all. It is just the frontal lobe. Similarly, the putative DAN is not the DAN, but the lateral and medial parietal cortex, and cuneus.

    P6, Line 11 - please define "simulation testing"

  4. eLife

    Summary: All reviewers appreciated the technical innovation of the work, but they also shared concerns about the robustness of some of the analyses, results, and content of the manuscript.

  5. Version published to 10.1101/2020.05.05.079749 on bioRxiv