The boundaries of State-Space Granger Causality Analysis applied to BOLD simulated data: a comparative modelling and simulation approach

Abstract

AbstractBackgroundThe analysis of connectivity has become a fundamental tool in human neuroscience. Granger Causality Mapping is a data-driven method that uses Granger Causality (GC) to assess the existence and direction of influence between signals, based on temporal precedence of information. More recently, a theory of Granger causality has been developed for state-space (SS-GC) processes, but little is known about its statistical validation and application on functional magnetic resonance imaging (fMRI) data.New MethodWe implemented a new heuristic, focusing on the application of SS-GC with a distinct statistical validation technique - Time Reversed Testing - to generative synthetic models and compare it to classical multivariate computational frameworks. We also test a range of experimental parameters, including block structure, sampling frequency, noise and system mean pairwise correlation, using a statistical framework of binary classification.ResultsWe found that SS-GC with time reversed testing outperforms other frameworks. The results validate the application of SS-GC to generative models. When estimating reliable causal relations, SS-GC returns promising results, especially when considering synthetic data with an high impact of noise and sampling rate.ConclusionsSS-GC with time reversed testing offers a possible framework for future analysis of fMRI data in the context of data-driven causality analysis.HighlightsState-Space GC was combined with a statistical validation step, using a Time Reversed Testing.This novel heuristic overpowers classical GC, when applied to generative models.Correctly identified connections between variables increase with the increase of number of blocks and number of points per block.SNR and subsampling have a significant impact on the results.