Measuring information-transfer delays.

Area of Science:

• Neuroscience
• Complex Systems Analysis
• Information Theory

Background:

• Understanding information transfer timing is crucial in complex networks (gene, traffic, brain circuits).
• Axonal delays in the brain significantly impact timing-based information processing.
• Accurate inference of neural interaction delays is essential for interpreting directed interactions.

Purpose of the Study:

• To propose a robust, model-free estimator for neuronal interaction delays.
• To extend information-theoretic measures for accurate delay estimation in coupled systems.
• To validate the method's performance on simulated and real-world neural data.

Main Methods:

• Developed a robust estimator rooted in an information-theoretic framework.
• Extended transfer entropy to account for delayed source-target interactions.
• Retained conditioning on the embedded target state at the previous time step, adhering to Wiener's principle of causality.

Main Results:

• The extended transfer entropy is guaranteed to identify interaction delays between coupled systems.
• Demonstrated effective delay detection on finite data using stochastic, deterministic processes, and local field potential recordings.
• Showcased the ability to detect multiple delays and feedback loops within network dynamics.

Conclusions:

• The proposed information-theoretic approach provides a robust method for estimating neuronal interaction delays.
• The extended transfer entropy offers a model-free solution applicable to various network dynamics.
• The estimator is valuable for neuroscience and other fields analyzing network interactions.