Trouble at Rest: How Correlation Patterns and Group Differences Become Distorted After Global Signal Regression

By Ziad S. Saad, S. Gotts, Kevin Murphy, Gang Chen, Hang Joon Jo, Alex Martin, Robert W. Cox. Brain Connectivity, 2012

886 citations recorded by the source corpus.

Resting-state functional magnetic resonance imaging (RS-FMRI) holds the promise of revealing brain functional connectivity without requiring specific tasks targeting particular brain systems. RS-FMRI is being used to find differences between populations even when a specific candidate target for traditional inferences is lacking. However, the problem with RS-FMRI is a lacking definition of what constitutes noise and signal. RS-FMRI is easy to acquire but is not easy to analyze or draw inferences from. In this commentary we discuss a problem that is still treated lightly despite its significant impact on RS-FMRI inferences; global signal regression (GSReg), the practice of projecting out signal averaged over the entire brain, can change resting-state correlations in ways that dramatically alter correlation patterns and hence conclusions about brain functional connectedness. Although Murphy et al. in 2009 demonstrated that GSReg negatively biases correlations, the approach remains in wide use. We revisit this issue to argue the problem that GSReg is more than negative bias or the interpretability of negative correlations. Its usage can fundamentally alter interregional correlations within a group, or their differences between groups. We used an illustrative model to clearly convey our objections and derived equations formalizing our conclusions. We hope this creates a clear context in which counterarguments can be made. We conclude that GSReg should not be used when studying RS-FMRI because GSReg biases correlations differently in different regions depending on the underlying true interregional correlation structure. GSReg can alter local and long-range correlations, potentially spreading underlying group differences to regions that may never have had any. Conclusions also apply to substitutions of GSReg for denoising with decompositions of signals aggregated over the network's regions to the extent they cannot separate signals of interest from noise. We touch on the need for careful accounting of nuisance parameters when making group comparisons of correlation maps.

Subfields: Cognitive Neuroscience, Radiology, Nuclear Medicine and Imaging

View Paper: https://www.semanticscholar.org/paper/ca9100bd89dfce0588a094da40ce50b6bf9c1b8e

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

The Human Connectome Project’s Neuroimaging Approach (2016, methodology): “…signal will artifactually decrease correlations across the brain and, more problematically, distort the pattern of correlations for particularly large or strong networks 63,109,110 What is needed to get past the current impasse is a method that selectively separates global signal from global noise.”
Abnormal rich club organization and functional brain dynamics in schizophrenia. (2013, neutral): “9) [37-88] 57.”
Reduction of motion-related artifacts in resting state fMRI using aCompCor (2014, background): “…FC metrics compared to regression of explicit models of motioninduced signal change, concerns have been raised about the potential of GSR to remove signal of neuronal origin (Schölvinck et al., 2010) and to introduce additional confounds into rs fMRI data (Murphy et al., 2009; Saad et al., 2012).”
Toward reliable characterization of functional homogeneity in the human brain: Preprocessing, scan duration, imaging resolution and computational space (2012, background): “Global signal regression is more controversial due to potential introduction of negative correlations and potential to artifactually impact inter-individual differences in connectivity (Chai et al., 2012; Fox et al., 2009; Murphy et al., 2009; Saad et al., 2012).”
The UCLA multimodal connectivity database: a web-based platform for brain connectivity matrix sharing and analysis (2012, background): “…controls for common, non-brain sources of variation that affect the entire image (Fox et al., 2009), while opponents have shown that this step may obscure the distribution of connectivity weights and introduce artifactual anticorrelations between networks (Murphy et al., 2009; Saad et al., 2012).”
Graph analysis of the human connectome: Promise, progress, and pitfalls (2013, neutral): “Though the contributions of pre-processing techniques to the emergence of negative correlations in fMRI data remains a topic of debate (Fox et al., 2009; Murphy et al., 2009; Saad et al., 2012), it is probable that not all negative correlations are artifacts (Chang and Glover, 2009).”
EEG correlates of time-varying BOLD functional connectivity (2013, neutral): “Global signal removal was not performed (Murphy et al., 2009; Saad et al., 2012).”
Standardizing the intrinsic brain: Towards robust measurement of inter-individual variation in 1000 functional connectomes (2013, neutral): “Global signal regression (GSR), a commonly used yet controversial practice in the R-fMRI field, yields substantial increases in negative correlations (Murphy et al., 2009; Weissenbacher et al., 2009) and may distort group differences in intrinsic functional connectivity (iFC) (Saad et al., 2012).”
The Global Signal in fMRI: Nuisance or Information? (2017, neutral): “the analysis of resting-state fMRI studies where some investigators routinely remove the effects of the global signal while others argue strongly against its use (Murphy et al., 2009; Fox et al., 2009; Saad et al., 2012; Burgess et al., 2016; Murphy and Fox, 2016).”
Resting-state FMRI confounds and cleanup (2013, positive): “Indeed, by modelling group comparisons, it has been shown that GSReg can alter local and long-range correlations, potentially spreading group differences that exist only in one region to regions that show no true functional connectivity differences (Saad et al., 2012).”
A Comprehensive Assessment of Regional Variation in the Impact of Head Micromovements on Functional Connectomics (2013, neutral): “GSR, a commonly used, yet controversial practice in the R-fMRI field yields substantial increases in negative correlations (Murphy et al., 2009;Weissenbacher et al., 2009) andmaydistort groupdifferences in intrinsic functional connectivity (iFC) (Saad et al., 2012).”
Benchmarking of participant‐level confound regression strategies for the control of motion artifact in studies of functional connectivity (2017, methodology): “…2015; Satterthwaite et al., 2013; Yan et al., 2013a), other studies have emphasized the risk of removing a valuable signal (Yang et al., 2014; Hahamy et al., 2014), potentially biasing group differences (Gotts et al., 2013; Saad et al., 2012), or exacerbating distance-dependent motion artifact.”
Methods to detect, characterize, and remove motion artifact in resting state fMRI (2013, methodology): “Without GSR, correlations at all distances are positively related to mean FD and these relationships can be reduced but not eliminated by scrubbing and reprocessing.”
Hand classification of fMRI ICA noise components (2017, positive): “Unfortunately, such methods make no distinction between global noise and global signal and will thus also distort measures of functional connectivity (Saad et al., 2012).”