Repository logo

Time-varying functional connectivity as Wishart processes

Accepted version

Change log



We investigated the utility of Wishart processes (WP) for estimating time-varying functional connectivity (TVFC), which captures changes in functional coupling as the correlation between brain region activity in functional magnetic resonance imaging (fMRI). The WP is a stochastic process on covariance matrices that can be used to model dynamic covariances between time series, which makes it a natural fit to this task. Recent advances in scalable approximate inference techniques and the availability of robust open-source libraries have rendered the WP practically viable for fMRI applications. We introduce a comprehensive benchmarking framework to assess WP performance compared to a selection of established TVFC estimation methods. The framework contains simulations with specified ground-truth covariance structures, a subject phenotype prediction task, a test-retest study, a brain state analysis, an external stimulus prediction task, and a novel data-driven imputation benchmark. The WP performed competitively across all the benchmarks. It outperformed a sliding window (SW) approach with adaptive cross-validated window lengths and a DCC-MGARCH baseline on the external stimulus prediction task, while being less prone to false positives in the TVFC null models.



Journal Title

Imaging Neuroscience

Conference Name

Journal ISSN


Volume Title


MIT Press

Publisher DOI

Publisher URL

Royal Society (INF\R2\202107)
Wellcome Trust (205067/Z/16/Z)
Biotechnology and Biological Sciences Research Council (BB/P021255/1)
Alan Turing Institute (Unknown)