A convenient method for detecting electrolyte bridges in multichannel electroencephalogram and event-related potential recordings

Craig E. Tenke, Jürgen Kayser

Department of Biopsychology, New York State Psychiatric Institute, New York, NY 10032, USA

Received 8 September 2000; revised 28 November 2000; accepted 4 December 2000. 

Abstract

Dense electrode arrays offer numerous advantages over single channel EEG/ERP recordings, but also exaggerate the influence of common error sources arising from the preparation of scalp placements. Even with conventional low density recordings (e.g., 30-channel Electro Cap), overapplication of electrode gel may result in electrolyte leakage and create low impedance bridges, particularly at vertically-aligned sites (e.g., inferior-lateral). The ensuing electrical short produces an artificial similarity of ERPs at neighboring sites that distorts the ERP topography. This artifact is not immediately apparent in group averages, and may even go undetected after visual inspection of the individual ERP waveforms. Besides adding noise variance to the topography, this error source also has the capacity to introduce systematic, localized artifacts (e.g., add or remove evidence of lateralized activity). Electrolyte bridges causing these artifacts can be easily detected by a simple variant of the Hjorth algorithm (intrinsic Hjorth), in which spatial interelectrode distances are replaced by an electrical analog of distance (i.e., the variances of the difference waveforms for all pairwise combinations of electrodes). When a low impedance bridge exists, the Hjorth algorithm identifies all affected sites as flat lines that are readily distinguishable from Hjorth waveforms at unbridged electrodes.

Keywords: EEG, event-related potential (ERP), methodology, electrolyte bridge, dense electrode array (DEA), intrinsic Hjorth