Skip to content

Multivariate Polyspectral Framework Bottleneck

Home / Open Questions / Multivariate Polyspectral Framework Bottleneck

Background: The identification of genuine high-order neural interactions is confounded by linear instantaneous mixing, such as volume conduction, which masks underlying nonlinear dependencies in sensor-level data.

Question / Future Work: Extending existing antisymmetric polyspectral frameworks to multivariate and multidimensional signal scenarios, where multiple interacting source regions are defined by vectorial components, remains an open computational and theoretical challenge.

Why It Matters: Advancing from univariate-frequency-centric analysis to full multivariate/multidimensional source-level connectivity is essential for high-resolution neuroimaging and complex sensor array processing.

Evidence: Specifically, there is no established infrastructure to investigate complex nonlinear couplings among N interacting regions (multivariate) where each source is intrinsically vectorial (multidimensional).

Metadata & Links