Spurious correlations in simultaneous EEG-fMRI driven by in-scanner movement

Abstract

Simultaneous EEG-fMRI provides an increasingly attractive research tool to investigate cognitive processes with high temporal and spatial resolution. However, artifacts in EEG data introduced by the MR-scanner still remain a major obstacle. This study employing commonly used artifact correction steps shows that head motion, one overlooked major source of artifacts in EEG-fMRI data, can cause plausible EEG effects and EEG-BOLD correlations. Specifically, low frequency EEG (<20 Hz) is strongly correlated with in-scanner movement. Accordingly, minor head motion (<0.2 mm) induces spurious effects in a twofold manner: Small differences in task-correlated motion elicit spurious low frequency effects, and, as motion concurrently influences fMRI data, EEG-BOLD correlations closely match motion-fMRI correlations. We demonstrate these effects in a memory encoding experiment showing that obtained theta power (~3-7 Hz) effects and channel-level theta-BOLD correlations reflect motion in the scanner. These findings highlight an important caveat that needs to be addressed by future EEG-fMRI studies.

Spatial Mnemonic Encoding: Theta Power Decreases and Medial Temporal Lobe BOLD Increases Co-Occur during the Usage of the Method of Loci

“…Furthermore, the increase in theta power was found in data recorded simultaneously with the fMRI recording. EEG data, especially in the lower-frequency bands (below ∼10 Hz) is highly affected by MR-related artifacts (Debener et al, 2008; Fellner et al, 2016). Consequently, considering the present results and recent studies on spatial memory encoding (Crespo-García et al, 2016; Griffiths et al, 2016), decreases in theta power during memory formation seem to be the more stable finding, contradicting the finding of this prior study.…”

“…A limitation of the present study is that an analysis of simultaneously recorded theta oscillations during fMRI was not feasible due to strong artifacts induced by the MR scanning environment (Fellner et al, 2016). Therefore, EEG and fMRI were measured separately in independent subject samples, preventing the calculation of direct EEG–fMRI correlations.…”

“…During fMRI, simultaneous EEG was recorded. Major MR scanner-induced EEG artifacts, especially in the theta frequency range, prevented reliable analysis of the simultaneously recorded EEG (for a discussion of these artifacts, see Fellner et al, 2016). …”

“…Consequently, spatial processing during the method of loci should lead to higher involvement of MTL regions, whereas successful encoding in both encoding tasks might similarly rely on MTL activity, irrespective of the spatial nature of the processing task. Note that EEG and fMRI were measured in separate groups of participants (Mukamel et al, 2005), because movement artifacts severely limit the interpretability of theta activity in simultaneous EEG-fMRI (Fellner et al, 2016). …”

Joint Estimation of Effective Brain Wave Activation Modes Using EEG/MEG Sensor Arrays and Multimodal MRI Volumes

“…This is clearly a very important question and a number of studies have appeared recently that focus on simultaneous acquisition techniques and provide attempts to confirm (or rebuff) the existence of correlations between EEG and FMRI data at different frequency bands or under different acquisition conditions (Meyer, van Oort, & Barth, 2013; Musso et al, 2010; Fellner et al, 2016; Meyer, Janssen, et al, 2013; Chang et al, 2013; Mantini et al, 2007). While we agree that the question of FMRI-EEG concordance is important, the complexity of the subject, and proving or disproving existence of the EEG-rsFMRI concordance, is clearly beyond the scope of our paper.…”