Effective Connectivity-Based Neural Decoding: A Causal Interaction-Driven Approach

TL;DR

This paper introduces a geometric, model-free causality measure based on multivariate delay embedding for detecting causal interactions in time series, applied to MEG data to decode visual stimuli and analyze brain connectivity patterns.

Contribution

It presents a novel causality measure that detects linear and nonlinear interactions without prior info, and applies it to brain data to map effective connectivity and analyze topological patterns.

Findings

Effective connectivity maps differentiate visual stimuli categories.

MEG responses show more geometric patterns in structured images.

Simulation and experimental results validate the approach.

Abstract

We propose a geometric model-free causality measurebased on multivariate delay embedding that can efficiently detect linear and nonlinear causal interactions between time series with no prior information. We then exploit the proposed causal interaction measure in real MEG data analysis. The results are used to construct effective connectivity maps of brain activity to decode different categories of visual stimuli. Moreover, we discovered that the MEG-based effective connectivity maps as a response to structured images exhibit more geometric patterns, as disclosed by analyzing the evolution of toplogical structures of the underlying networks using persistent homology. Extensive simulation and experimental result have been carried out to substantiate the capabilities of the proposed approach.