High and dry? Comparing active dry EEG electrodes to active and passive wet electrodes

Dry electrodes are becoming popular for both lab-based and consumer-level electrophysiological-recording technologies because they better afford the ability to move traditional lab-based research into the real world. It is unclear, however, how dry electrodes compare in data quality to traditional electrodes. The current study compared three EEG electrode types: (a) passive-wet electrodes with no onboard amplification, (b) actively amplified, wet electrodes with moderate impedance levels, and low impedance levels, and (c) active-dry electrodes with very high impedance. Participants completed a classic P3 auditory oddball task to elicit characteristic EEG signatures and event-related potentials (ERPs). Across the three electrode types, we compared single-trial noise, average ERPs, scalp topographies, ERP noise, and ERP statistical power as a function of number of trials. We extended past work showing active electrodes' insensitivity to moderate levels of interelectrode impedance when compared to passive electrodes in the same amplifier. Importantly, the new dry electrode system could reliably measure EEG spectra and ERP components comparable to traditional electrode types. As expected, however, dry active electrodes with very high interelectrode impedance exhibited marked increases in single-trial and average noise levels, which decreased statistical power, requiring more trials to detect significant effects. This power decrease must be considered as a trade-off with the ease of application and long-term use. The current results help set constraints on experimental design with novel dry electrodes, and provide important evidence needed to measure brain activity in novel settings and situations.