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

BackgroundThe 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. HighlightsO_LIState-Space GC was combined with a statistical validation step, using a Time Reversed Testing. C_LIO_LIThis novel heuristic overpowers classical GC, when applied to generative models. C_LIO_LICorrectly identified connections between variables increase with the increase of number of blocks and number of points per block. C_LIO_LISNR and subsampling have a significant impact on the results. C_LI