Kernel-nonlinear-PDC extends Partial Directed Coherence to detecting nonlinear causal coupling.

Area of Science:

• Neuroscience
• Signal Processing
• Causality Analysis

Background:

• Linear models are insufficient for detecting complex, nonlinear causal interactions in biological and other complex systems.
• Existing methods like linear Partial Directed Coherence (PDC) are limited to linear relationships.

Purpose of the Study:

• To introduce and validate a novel method, kernel-nonlinear-Partial Directed Coherence (k-PDC), for identifying nonlinear causal links.
• To demonstrate the capability of k-PDC in detecting causal relationships missed by linear approaches.

Main Methods:

• Kernel feature space representation of data to transform nonlinearities into a linear context.
• Application of asymptotic decision criteria, adapted from linear PDC, for statistical inference.
• Validation of the k-PDC method on simulated and real-world datasets.

Main Results:

• Kernel-nonlinear-Partial Directed Coherence successfully detects nonlinear causal links.
• The proposed method achieves adequate connectivity detection using established statistical criteria.
• Demonstrated superiority over linear methods in identifying complex causal pathways.

Conclusions:

• Kernel-nonlinear-Partial Directed Coherence is a powerful tool for uncovering nonlinear causality.
• This advancement expands the scope of causal inference in complex systems.
• The method offers a robust approach for analyzing nonlinear dynamics.