Generating Hypotheses of Dynamic Causal Graphs in Neuroscience: Leveraging Generative Factor Models of Observed Time Series

TL;DR

This paper introduces a new method for generating hypotheses about dynamic causal graphs in neuroscience, capable of capturing complex, time-varying relationships in brain data, and demonstrates significant improvements over existing methods.

Contribution

The paper presents a novel approach modeling dynamic graphs as a superposition of static, nonlinear graphs, enabling better detection of complex neural interactions over time.

Findings

Improved f1-scores by 22-28% over baselines

Able to uncover relationships linked to behavioral states

Demonstrated effectiveness on real brain data

Abstract

The field of hypothesis generation promises to reduce costs in neuroscience by narrowing the range of interventional studies needed to study various phenomena. Existing machine learning methods can generate scientific hypotheses from complex datasets, but many approaches assume causal relationships are static over time, limiting their applicability to systems with dynamic, state-dependent behavior, such as the brain. While some techniques attempt dynamic causal discovery through factor models, they often restrict relationships to linear patterns or impose other simplifying assumptions. We propose a novel method that models dynamic graphs as a conditionally weighted superposition of static graphs, where each static graph can capture nonlinear relationships. This approach enables the detection of complex, time-varying interactions between variables beyond linear limitations. Our method improves f1-scores of predicted dynamic causal patterns by roughly 22-28% on average over baselines in some of our experiments, with some improvements reaching well over 60%. A case study on real brain data demonstrates our method's ability to uncover relationships linked to specific behavioral states, offering valuable insights into neural dynamics.