Author's personal copy Independent component analysis and clustering improve signal-to-noise ratio for statistical analysis of event-related potentials Philip M. Zeman a,b, * , Bernie C. Till a,b , Nigel J. Livingston a , James W. Tanaka c , Peter F. Driessen b a CanAssist, University of Victoria, BC, Canada b Department of Electrical and Computer Engineering, University of Victoria, BC, Canada c Department of Psychology, University of Victoria, BC, Canada Accepted 7 September 2007 Available online 29 October 2007 Abstract Objective: To evaluate the effectiveness of a new method of using Independent Component Analysis (ICA) and k-means clustering to increase the signal-to-noise ratio of Event-Related Potential (ERP) measurements while permitting standard statistical comparisons to be made despite the inter-subject variations characteristic of ICA. Methods: Per-subject ICA results were used to create a channel pool, with unequal weights, that could be applied consistently across subjects. Signals derived from this and other pooling schemes, and from unpooled electrodes, were subjected to identical statistical anal- ysis of the N170 own-face effect in a Joe/No Joe face recognition paradigm wherein participants monitored for a target face (Joe) pre- sented amongst other unfamiliar faces and their own face. Results between the Joe, unfamiliar face and own face conditions were compared using Cohen’s d statistic (square root of signal-to-noise ratio) to measure effect size. Results: When the own-face condition was compared to the Joe and unfamiliar-face conditions, the channel map method increased effect size by a factor ranging from 1.2 to 2.2. These results stand in contrast to previous findings, where conventional pooling schemes failed to reveal an N170 effect to the own-face stimulus (Tanaka JW, Curran T, Porterfield A, Collins D. The activation of pre-existing and acquired face representations: the N250 ERP as an index of face familiarity. J Cogn Neurosci 2006;18:1488–97). Consistent with con- ventional pooling schemes, the channel map approach showed no reliable differences between the Joe and Unfamiliar face conditions, yielding a decrease in effect size ranging from 0.13 to 0.75. Conclusions: By increasing the signal-to-noise ratio in the measured waveforms, the channel pool method demonstrated an enhanced sensitivity to the neurophysiological response to own-face relative to other faces. Significance: By overcoming the characteristic inter-subject variations of ICA, this work allows classic ERP analysis methods to exploit the improved signal-to-noise ratio obtainable with ICA. Ó 2007 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. Keywords: EEG; ICA; Signal-to-noise ratio; Effect size; Channel pooling; N170 1. Introduction Responses to experimental manipulations, measured by Event-Related Potentials (ERPs), are considered distinct when ERP peaks differ significantly in amplitude, latency, and/or scalp topography (Dien and Santuzzi, 2005). The effects of an experimental manipulation can be made more visible by employing any of a wide variety of signal pro- cessing techniques, such as spatial-, frequency- and/or time-domain filtering, to separate the signal of interest from background noise. A key issue in the design of such filters is consistency – the ERP data must be processed consistently across subjects and conditions to avoid 1388-2457/$32.00 Ó 2007 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.clinph.2007.09.001 * Corresponding author. Address: Department of Electrical and Com- puter Engineering, University of Victoria, BC, Canada. Tel.: +1 250 589 4234; fax: +1 250 721 6611. E-mail address: pzeman@ece.uvic.ca (P.M. Zeman). www.elsevier.com/locate/clinph Clinical Neurophysiology 118 (2007) 2591–2604