Applying dimension reduction to EEG data by Principal Component Analysis reduces the quality of its subsequent Independent Component decomposition

• Published in: NeuroImage (Orlando, Fla.) Vol. 175; pp. 176 - 187
• Main Authors: Artoni, Fiorenzo, Delorme, Arnaud, Makeig, Scott
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 15.07.2018; Elsevier Limited; Elsevier
• Subjects: Adult; Brain - physiology; Brain research; Brain Waves - physiology; Cognitive science; Data Interpretation, Statistical; Data processing; Decomposition; Dipolarity; EEG; Electroencephalogram, EEG; Electroencephalography; Electroencephalography - methods; Electroencephalography - standards; Electromyography; Female; Humans; Independent component analysis, ICA; Localization; Male; Neuroscience; Noise; Pattern Recognition, Visual - physiology; Principal Component Analysis; Principal component analysis, PCA; Principal components analysis; Reliability; Signal Processing, Computer-Assisted; Source localization; Young Adult; Independent component analysis, ICA; Principal component analysis, PCA; Reliability; Source localization; Electroencephalogram, EEG; Dipolarity
• ISSN: 1053-8119; 1095-9572; 1095-9572
• DOI: 10.1016/j.neuroimage.2018.03.016

Independent Component Analysis (ICA) has proven to be an effective data driven method for analyzing EEG data, separating signals from temporally and functionally independent brain and non-brain source processes and thereby increasing their definition. Dimension reduction by Principal Component Analysis (PCA) has often been recommended before ICA decomposition of EEG data, both to minimize the amount of required data and computation time. Here we compared ICA decompositions of fourteen 72-channel single subject EEG data sets obtained (i) after applying preliminary dimension reduction by PCA, (ii) after applying no such dimension reduction, or else (iii) applying PCA only. Reducing the data rank by PCA (even to remove only 1% of data variance) adversely affected both the numbers of dipolar independent components (ICs) and their stability under repeated decomposition. For example, decomposing a principal subspace retaining 95% of original data variance reduced the mean number of recovered ‘dipolar’ ICs from 30 to 10 per data set and reduced median IC stability from 90% to 76%. PCA rank reduction also decreased the numbers of near-equivalent ICs across subjects. For instance, decomposing a principal subspace retaining 95% of data variance reduced the number of subjects represented in an IC cluster accounting for frontal midline theta activity from 11 to 5. PCA rank reduction also increased uncertainty in the equivalent dipole positions and spectra of the IC brain effective sources. These results suggest that when applying ICA decomposition to EEG data, PCA rank reduction should best be avoided. •It is currently a common practice to apply dimension reduction to EEG data using PCA before performing ICA decomposition.•We tested the quality of Independent Components (ICs) after different levels of rank reduction to a principal subspace.•PCA rank reduction adversely affected dipolarity and stability of ICs accounting for brain and known non-brain processes.•PCA rank reduction also increased inter-subject variance in IC source locations (by equivalent dipole fitting) and spectra.•For EEG data at least, PCA rank reduction should be avoided or carefully tested before applying it as a preprocessing step.