EEG processing

The BeMoBIL pipeline allows fully automatic processing of EEG data from raw freshly loaded files to the final clean data sets with independent component analysis (ICA) information. We assume all data per participant is already merged into one .set to be available for EEGLAB.

The processing consists of two main parts with one optional previous step:

0. Removing non-experiment segments from the data. This means everything up to the start and after the end, plus all breaks should be removed. This is important as in these times participants might be showing increased movement or the EEG cap might be touched/moved, or equipment like a virtual reality head-mounted display might be added or removed. IMPORTANT: This step is impossible to automatize since it highly depends on the structure of the experiment and the available event markers. Hence, this has to be done manually, or coded in before the subsequent processing. An example for this can be found in example_bemobil_process_all_EEG_data.m.
1. Preprocessing, containing line noise removal, bad channel detection and interpolation, and average referencing.
2. AMICA processing, containing high-pass filtering, AMICA, dipole fitting, and labeling of ICs.

All processing is automatically documented and parameters are stored in the EEG.etc struct. Additionally, plots are created to allow checking the data and results. They are explained in this guide.