Ghost interactions in MEG/EEG source space : A note of caution on inter-areal coupling measures

Show full item record



Permalink

http://hdl.handle.net/10138/234709

Citation

Palva , J M , Wang , S H , Palva , S , Zhigalov , A , Monto , S , Brookes , M J , Schoffelen , J-M & Jerbi , K 2018 , ' Ghost interactions in MEG/EEG source space : A note of caution on inter-areal coupling measures ' , NeuroImage , vol. 173 , pp. 632-643 . https://doi.org/10.1016/j.neuroimage.2018.02.032

Title: Ghost interactions in MEG/EEG source space : A note of caution on inter-areal coupling measures
Author: Palva, J. Matias; Wang, Sheng H.; Palva, Satu; Zhigalov, Alexander; Monto, Simo; Brookes, Matthew J.; Schoffelen, Jan-Mathijs; Jerbi, Karim
Contributor organization: Neuroscience Center
Matias Palva / Principal Investigator
BioMag Laboratory
Clinicum
Helsinki Institute for Information Technology
Date: 2018-06
Language: eng
Number of pages: 12
Belongs to series: NeuroImage
ISSN: 1053-8119
DOI: https://doi.org/10.1016/j.neuroimage.2018.02.032
URI: http://hdl.handle.net/10138/234709
Abstract: When combined with source modeling, magneto- (MEG) and electroencephalography (EEG) can be used to study long-range interactions among cortical processes non-invasively. Estimation of such inter-areal connectivity is nevertheless hindered by instantaneous field spread and volume conduction, which artificially introduce linear correlations and impair source separability in cortical current estimates. To overcome the inflating effects of linear source mixing inherent to standard interaction measures, alternative phase- and amplitude-correlation based connectivity measures, such as imaginary coherence and orthogonalized amplitude correlation have been proposed. Being by definition insensitive to zero-lag correlations, these techniques have become increasingly popular in the identification of correlations that cannot be attributed to field spread or volume conduction. We show here, however, that while these measures are immune to the direct effects of linear mixing, they may still reveal large numbers of spurious false positive connections through field spread in the vicinity of true interactions. This fundamental problem affects both region-of-interest-based analyses and all-to-all connectome mappings. Most importantly, beyond defining and illustrating the problem of spurious, or "ghost" interactions, we provide a rigorous quantification of this effect through extensive simulations. Additionally, we further show that signal mixing also significantly limits the separability of neuronal phase and amplitude correlations. We conclude that spurious correlations must be carefully considered in connectivity analyses in MEG/EEG source space even when using measures that are immune to zero-lag correlations.
Subject: 3112 Neurosciences
3126 Surgery, anesthesiology, intensive care, radiology
Peer reviewed: Yes
Rights: cc_by_nc_nd
Usage restriction: openAccess
Self-archived version: acceptedVersion


Files in this item

Total number of downloads: Loading...

Files Size Format View
Ghost_interactions_in_MEG_EEG_source_space.pdf 2.506Mb PDF View/Open
1_s2.0_S1053811918301290_main.pdf 1.288Mb PDF View/Open

This item appears in the following Collection(s)

Show full item record