中 / EN
2026-10-06 00:00 Papers Foundations & Methods Source language EN

Semi-Dry Smarting Mobile EEG Matches Gel Lab System on Individual Auditory ERPs

Summary An independent study found that a 24-channel wireless Smarting EEG cap with saline-sponge electrodes detected individual-level N100 and P300 auditory responses, which appear about 100 and 300 milliseconds after a sound, in 34 of 36 tests, versus 34 of 40 for a 64-channel wired gel BioSemi system in the same shielded lab. Participants heard a sound sequence containing occasional target and novel tones. The Smarting group had only 9 participants, and the researchers caution that this is not a strict hardware comparison.
Why it matters EEG research mostly reports group averages, while clinical assessment, developmental tracking and personalized interventions need a reliable response in every individual, so a gel-free mobile device performing on par is a preliminary yes for that use. The evidence is thin, though: single-electrode P300 detection found responses in only 4 of the 9 Smarting participants.

半干式移动脑电在个体水平检出听觉脑电反应,检出率与实验室凝胶系统相当
Image: MBrainTrain, CC BY 4.0

BCIwiki (bciwiki.com) — A semi-dry Smarting mobile EEG system using saline-sponge electrodes detected individual-level N100 and P300 auditory event-related potentials (ERPs) in 34 of 36 subject-level tests, compared with 34 of 40 for a wired, gel-based 64-channel BioSemi lab system, Korisky and colleagues reported in Frontiers in Human Neuroscience in 2026 (doi:10.3389/fnhum.2026.1763477). Both systems were recorded in the same shielded laboratory, with 9 and 10 participants respectively, according to MBrainTrain, which makes Smarting.

Participants performed a three-sound auditory oddball task with frequent Standard tones (70%) and rare Target tones and Novel sounds (15% each). Experiment 1 compared four EEG systems, each with different participants and partly different recording environments, and the authors caution that it is not a head-to-head hardware comparison. Smarting and BioSemi were recorded in the same room with the same stimulus delivery. Experiment 2, a test-retest reliability study with 38 participants, used BioSemi only.

Using spatial clustering across electrodes, Smarting detected responses in 94% of tests and BioSemi in 85%. The study did not test the difference statistically, and one participant shifts a percentage by 10-11 points. With single-electrode temporal clustering, Smarting detected the N100 in 78%, 100% and 89% of participants for Standard, Target and Novel sounds, versus 70% for each with BioSemi. Its P300 detection with that method was 44%, or 4 of 9 participants, the lowest of the four systems, against 8 of 9 with spatial clustering. Group ERPs from Smarting correlated with BioSemi's at r = 0.73 for Standard, r = 0.87 for Novel and r = 0.41 for Target sounds, and waveform typicality did not differ significantly (β = 0.08, p = 0.144).

In its commentary, MBrainTrain said the 9-person Smarting sample supports feasibility rather than precise performance estimates, and that cap fit strongly affects semi-dry electrodes. Smarting rejected 7.5% of trials, versus 4.7% for BioSemi and 1.8% for DSI-24, though thresholds were set per system. The company said the authors, from Bar-Ilan University and Stanford, identified Smarting as one of two research-grade mobile systems whose signal quality their results validate.

Compiled by BCIwiki from public sources

Sources · 1
mbraintrain.com 2026-10-06
Read original ↗
Suggest a correction Revisions · none 中 / EN
© 2026 BCIwiki.com Digest Topics Tips Subscribe Revisions About