The effect of global signal regression on DCM estimates of noise and effective connectivity from resting state fMRI

Abstract

The influence of the global BOLD signal on resting state functional connectivity in fMRI data remains a topic of debate, with little consensus. In this study, we assessed the effects of global signal regression (GSR) on effective connectivity within and between resting-state networks -as estimated with dynamic causal modelling (DCM) for resting state fMRI (rsfMRI). DCM incorporates a forward (generative) model that quantifies the contribution of different types of noise (including global measurement noise), effective connectivity, and (neuro)vascular processes to functional connectivity measurements. DCM analyses were applied to two different designs; namely, longitudinal and crosssectional designs. In the modelling of longitudinal designs, we included four extensive longitudinal resting state fMRI datasets with a total number of 20 subjects. In the analysis of cross-sectional designs, we used rsfMRI data from 361 subjects from the Human Connectome Project. We hypothesized that (1) GSR would have no discernible impact on effective connectivity estimated with DCM, and (2) GSR would be reflected in the parameters representing global measurement noise. Additionally, we performed comparative analyses of the informative value of data with and without GSR. Our results showed negligible to small effects of GSR on connectivity within small (separately estimated) RSNs. For between-network connectivity, we found two important effects: the effect of GSR on betweennetwork connectivity (averaged over all connections) was negligible to small, while the effect of GSR on individual connections was non-negligible. Contrary to our expectations, we found either no effect (in the longitudinal designs) or a non-specific (cross-sectional design) effect of GSR on parameters representing (global) measurement noise. Data without GSR were found to be more informative than data with GSR; however, in small resting state networks the precision of posterior estimates was greater using data after GSR. In conclusion, GSR is a minor concern in DCM studies; however, individual between-network connections (as opposed to average between-network connectivity) and noise parameters should be interpreted quantitatively with some caution. The Kullback-Leibler divergence of the posterior from the prior, together with the precision of posterior estimates, might offer a useful measure to assess the appropriateness of GSR, when nuancing data features in resting state fMRI.

Neural Mechanisms and Psychology of Psychedelic Ego Dissolution

“…Variation in preprocessing pipelines in particular can be a significant barrier to the accurate comparison of findings across neuroimaging studies. For comparison of motion correction pipelines, see Parkes et al (2018); for global signal regression see Almgren et al (2020) and Aquino et al (2020); and for motion correction in diffusion imaging pipelines, see Oldham et al (2020). The optimization and standardization of these pipelines-in conjunction with replication and comparison between psychedelicsmay be essential.…”

“…Hierarchical predictive coding is a mechanistic account of how the brain processes information (Friston, 2008). It is supported by empirical evidence (Mumford, 1992;Rao and Ballard, 1999) and informed by our understanding of synaptic communication (Adams et al, 2013;Friston, 2020;Hobson et al, 2021). Noting the neuromodulatory role of serotonergic neurotransmission (Picard and Friston, 2014) and the influence of psychedelics upon these receptors, it provides a unifying account of psychedelic subjective and therapeutic effects.…”

Disrupted effective connectivity of the default, salience and dorsal attention networks in major depressive disorder: a study using spectral dynamic causal modelling of resting-state fMRI

“…We did not conduct global signal regression, because a previous study found that data without global signal regression are more informative for estimation of effective connectivity. 38 We used mean frame-wise displacement -which considers measures of voxel-wise differences in motion in its deri vation 39 -as a measure of the micro-head motion of each participant. 40 Participants whose mean frame-wise displacement was greater than 3 interquartile ranges from the sample median or who had fewer than 100 "good" volumes of data (frame-wise displacement threshold ≤ 0.2 mm) were to be excluded from further analysis, but no participant was excluded based on this motion threshold.…”

Fathers’ involvement in early childcare is associated with amygdala resting-state connectivity

“…In addition, cerebrospinal fluid and white matter signals were added as regressors to the model in order to reduce the influence of artifacts caused by physiological signal sources on the results. The global signal was not added to the model, as there is no consensus on the use of this regressor ( Almgren et al ., 2020 ).…”