Causal Dynamic Resonance

TL;DR

Causal Dynamic Resonance (CDR) enhances causality detection in complex nonlinear systems by adding noise to signals, improving accuracy in both simulated and real brain data, especially in epilepsy monitoring.

Contribution

This paper introduces Causal Dynamic Resonance, a novel method that improves causality detection by incorporating noise, addressing limitations of existing measures like CD-DDA.

Findings

Enhanced detection of causality in coupled systems

Reduced false positives with noise addition

Effective application to epilepsy iEEG data

Abstract

Natural dynamical systems, including the brain and climate, are highly nonlinear and complex. Determining information flow among the components that make up these dynamical systems is challenging. If the components are the result of a common process or become synchronized, causality measures typically fail, including Cross-Dynamical Delay Differential Analysis (CD-DDA). We previously introduced Dynamical Ergodicity Delay Differential Analysis (DE-DDA) to assess dynamical similarity. Here, we further increase the detection of false positives in causality by adding white noise to the signals, but not the dynamical system. Causal Dynamic Resonance (CDR) is applied to coupled R\"ossler systems where the ground truth is known and also to invasive intracranial electroencephalographic (iEEG) data from drug-resistant epilepsy patients undergoing presurgical monitoring.