Identifying key factors for improving ICA-based decomposition of EEG data in mobile and stationary experiments.

Marius Klug,Klaus Gramann

doi:10.1111/ejn.14992

Abstract

Recent developments in EEG hardware and analyses approaches allow for recordings in both stationary and mobile settings. Irrespective of the experimental setting, EEG recordings are contaminated with noise that has to be removed before the data can be functionally interpreted. Independent component analysis (ICA) is a commonly used tool to remove artifacts such as eye movement, muscle activity, and external noise from the data and to analyze activity on the level of EEG effective brain sources. The effectiveness of filtering the data is one key preprocessing step to improve the decomposition that has been investigated previously. However, no study thus far compared the different requirements of mobile and stationary experiments regarding the preprocessing for ICA decomposition. We thus evaluated how movement in EEG experiments, the number of channels, and the high-pass filter cutoff during preprocessing influence the ICA decomposition. We found that for commonly used settings (stationary experiment, 64 channels, 0.5Hz filter), the ICA results are acceptable. However, high-pass filters of up to 2Hz cut-off frequency should be used in mobile experiments, and more channels require a higher filter to reach an optimal decomposition. Fewer brain ICs were found in mobile experiments, but cleaning the data with ICA has been proved to be important and functional even with low-density channel setups. Based on the results, we provide guidelines for different experimental settings that improve the ICA decomposition.

Highlights

Over the last decade, the development of lightweight portable electroencephalography (EEG) amplifiers and new data-driven analyses approaches led to the investigation of the neural basis of ecologically valid cognitive processes in actively behaving human participants outside established laboratory environments
The Independent component analysis (ICA) decompositions were sensitive to the different preprocessing parameters
A larger residual variance (RV) of brain independent component (IC) could be observed in the mobile condition, this difference was not very large

Summary

Introduction

The development of lightweight portable electroencephalography (EEG) amplifiers and new data-driven analyses approaches led to the investigation of the neural basis of ecologically valid cognitive processes in actively behaving human participants outside established laboratory environments. Experiments allow active behavior of participants both in the lab (De Sanctis et al, 2014; Djebbara et al, 2019; Ehinger et al, 2014; Gehrke et al, 2018; Gramann et al, 2010; Nenna et al, 2020, this issue) and in the real world, which increases our understanding of human brain dynamics accompanying embodied cognitive processes as well as the impact of real world environments (Debener et al, 2012; Ladouce et al, 2017; Protzak & Gramann, 2018; Wascher et al, 2014; Wunderlich & Gramann, 2018) While these experimental protocols provide new insights into the neural activity subserving cognition in more realistic and natural settings, they present new challenges. The ability to interpret EEG data from both classic stationary as well as MoBI experiments depends greatly on the ability to dissociate signals of interest originating in the brain from those of other sources

Objectives

Methods

Results

Conclusion