Mapping directed influence over the brain using Granger causality and fMRI

Abstract

We propose Granger causality mapping (GCM) as an approach to explore directed influences between neuronal populations (effective connectivity) in fMRI data. The method does not rely on a priori specification of a model that contains pre-selected regions and connections between them. This distinguishes it from other fMRI effective connectivity approaches that aim at testing or contrasting specific hypotheses about neuronal interactions. Instead, GCM relies on the concept of Granger causality to define the existence and direction of influence from information in the data. Temporal precedence information is exploited to compute Granger causality maps that identify voxels that are sources or targets of directed influence for any selected region-of-interest. We investigated the method by simulations and by application to fMRI data of a complex visuomotor task. The presented exploratory approach of mapping influences between a region of interest and the rest of the brain can form a useful complement to existing models of effective connectivity. D 2004 Elsevier Inc. All rights reserved.

The MVGC multivariate Granger causality toolbox: A new approach to Granger-causal inference

“…While technological developments supporting ultra-low TRs (Feinberg and Yacoub, 2012) and de-noising (Rasmussen et al, 2012) promise to alleviate these problems, in the absence of these developments a conservative methodology is recommended. Useful strategies include (i) using as short a TR as possible by compromising on coverage; (ii) examining changes in Gcausality between experimental conditions, rather than attempting to identify "ground truth" G-causality patterns; (iii) correlating changes in G-causality magnitude with behavioural variables such as reaction times across trials (or trial blocks) (Wen et al, 2012), and (iv) computing the so-called "difference of influence" term (Roebroeck et al, 2005) which may provide some robustness to HRF variation. Alternative promising approaches include estimation of state-space models which jointly parameterize functional connectivity and hemodynamic responses (Ryali et al, 2011) or blind deconvolution of the HRF to retrieve the underlying neuronal processes (Havlicek et al, 2011;Wu et al, 2013).…”

Multivariate Granger causality analysis of fMRI data

“…Accordingly, if incorporating the past values of time series X improves the future prediction of time series Y, then X is said to have a causal influence on Y [Granger, 1969]. In the case of any two time series X and Y, the efficacy of cross-prediction could be inferred either through the residual error after prediction [Roebroeck et al, 2005] or through the magnitude of the predictor coefficients . Both approaches are equivalent and the analytical relationship between them is given by Granger [1969].…”

“…With fMRI data, recent studies have applied Granger causality analysis between a target region of interest (ROI) and all other voxels in the brain to derive Granger causality maps [Abler et al, 2006;Goebel et al, 2003;Roebroeck et al, 2005]. A major limitation of applying the target ROI based approach to neuroimaging data is that it is a bivariate method and ignores interactions between other ROIs in the underlying neuronal network leading to an oversimplification of the multivariate neuronal relationships that exist during the majority of cognitive tasks.…”

Dynamic causal modelling for fMRI: A two-state model

“…Dynamical causal modelling differs from established methods for estimating effective connectivity from neurophysiological time series, which include structural equation modelling and models based on multivariate autoregressive processes McIntosh and Gonzalez-Lima, 1994;Roebroeck et al, 2005). In these models, there is no designed perturbation and the inputs are treated as unknown and stochastic.…”