EEG Preprocessing Pipeline Guide

To set up an EEG preprocessing pipeline, follow structured workflow steps from data import to epoching and artifact management. This guide provides recommended parameters for filtering, re-referencing, and ICA to transform raw neural recordings into analysis-ready signals.

A 1 Hz high-pass filter is recommended for ICA to ensure stable decomposition, while 0.1 Hz is used for ERP analysis to avoid distorting slow drifts. This guide provides parameter rationale to justify these specific high-pass filtering choices.

To handle artifacts in EEG data using ICA and ASR, apply artifact correction strategies with defined thresholds to identify and remove noise. This guide details when to use ICA versus ASR, artifact removal thresholds, and when to interpolate bad channels.

Choosing between average or linked mastoids referencing for EEG data depends on your specific analysis type. This guide provides method selection rationale for re-referencing choices, detailing how each option impacts your ERP, time-frequency, and connectivity analyses.

You need to interpolate bad channels during EEG preprocessing after identifying uncorrectable artifacts that remain following ICA or ASR. This guide outlines artifact removal strategies and defines specific correction thresholds indicating when channel interpolation is necessary.

You can document reproducible EEG preprocessing steps by following this guide's structured workflow from data import to epoching. It provides recommended parameter tables and literature references to justify choices like filtering and artifact correction for your research study.