Comparing Simultaneous Scalp EEG Recordings from the OpenBCI Cyton and Brain Products BrainAmp

TL;DR

This study compares the OpenBCI Cyton EEG device with a high-end amplifier and finds it records similar brain signals for cognitive research.

Contribution

First multi-subject validation of OpenBCI Cyton against gold-standard amplifiers across multiple ERP components.

Findings

Cyton signals showed high similarity (r2 = 97–100%) to BrainAmp signals at ERP sites.

Mean amplitudes were similar, but standard deviations differed significantly.

Results validate Cyton for cognitive neuroscience research.

Abstract

Portable, affordable electroencephalography (EEG) amplifiers could enable neuroscience-scale data collection. The open-source OpenBCI Cyton shows promise in this regard, but remains undervalidated for cognitive neuroscience ERP experiments. We simultaneously recorded eight scalp electrodes with both Cyton and gold-standard Brain Products BrainAmp amplifiers across P3b-, ERN-, and N400-eliciting tasks. Five healthy volunteers completed visual oddball (P3b), flankers (ERN), and word association (N400) tasks. We quantified within-subject signal similarity using Pearson r2, mean absolute error (MAE), mean arctangent absolute percentage error (MAAPE), and within-component window mean and standard deviation. Cyton signals showed r2 = 97–100%, MAE ≈ 1 µV, and MAAPE ≈ 20% with BrainAmp signals at ERP sites of interest. No significant differences emerged in mean amplitudes within ERP component windows across amplifiers, though standard deviations differed significantly. These results demonstrate that the Cyton records highly similar but not identical scalp EEG as research-grade equipment. This first multi-subject, concurrent scalp EEG validation across multiple ERP components validates the Cyton for cognitive neuroscience and supports broader adoption of affordable open-source tools.

Figures