A Tutorial Review of Functional Connectivity Analysis Methods and Their Interpretational Pitfalls.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Systems Neuroscience

Background:

• Oscillatory neuronal activity is crucial for dynamic network coordination in the brain.
• Quantifying rhythmic neuronal interactions requires various metrics, each with limitations.

Purpose of the Study:

• To review and summarize current analysis methods for studying dynamic connections between neuronal populations.
• To provide intuition and quantitative definitions for functional connectivity metrics.
• To highlight interpretational caveats and common pitfalls in functional connectivity analysis.

Main Methods:

• Review of functional connectivity metrics: coherence, phase synchronization, phase-slope index, and Granger causality.
• Explanation of quantitative definitions and intuitive understanding of metrics.
• Illustration of common pitfalls using MATLAB scripts: common reference, signal-to-noise ratio, volume conduction, common input, and sample size bias.

Main Results:

• Detailed review of established functional connectivity metrics.
• Identification and explanation of critical challenges in analyzing neuronal data.
• Provision of practical tools (MATLAB scripts) to simulate and understand analysis problems.

Conclusions:

• Accurate analysis of neuronal network dynamics requires careful consideration of chosen metrics and potential pitfalls.
• Understanding and addressing common problems like volume conduction and common input is essential for reliable functional connectivity findings.
• This tutorial equips researchers with the knowledge and tools to perform robust electrophysiological data analysis.