Dry electrodes: losing the gel
The gelled wet electrode is the enemy of daily use: it needs skin prep, dries out in an hour or two, and no one wants to wash gel from their hair before work. A dry electrode contacts the scalp directly through pins, fingers or a conductive polymer, so a consumer headset can be self-applied in seconds. The price is a much higher and less stable electrode-skin impedance, which raises the noise floor and makes the contact exquisitely sensitive to movement.
A recording is a voltage divider: the fraction of the true scalp potential you keep depends on amplifier input impedance Z_in relative to electrode impedance Z_e. Dry electrodes have large, variable Z_e, so signal is attenuated and any change in contact directly modulates the output — the root of motion sensitivity and mains pickup.
Johnson-Nyquist thermal noise: the noise voltage grows with the square root of the real contact resistance R. A dry contact with kilo- to mega-ohm impedance therefore sits on a higher noise floor than a low-impedance gelled contact — a fundamental, not merely practical, penalty.
Ear-EEG and the discreet form factor
Adherence often hinges on how a device looks and feels. Ear-EEG places electrodes in and around the ear canal, hidden inside something shaped like an earbud, which people already wear all day. The tradeoff is coverage: electrodes clustered at the ear see mainly nearby sources with small inter-electrode distances, so it excels for signals near the temporal lobe and for long-term monitoring (sleep staging, auditory attention) but is weak for whole-scalp mapping. It is a clear case of trading signal reach for the one thing that actually drives daily use — wearability.
Optics and wearable magnetometry
Two non-electrical modalities are maturing into wearables. Wearable fNIRS shines near-infrared light through the scalp and reads a slow hemodynamic proxy of activity; it is relatively robust to electrical noise and comfortable, but it is slow (seconds), shallow (cortical surface), and confounded by scalp blood and hair. Wearable magnetometry is the newer surprise: optically-pumped magnetometers (OPM-MEG) measure the brain's magnetic field at room temperature from sensors on a helmet that moves with the head, unlike a cryogenic MEG scanner. On-scalp placement puts sensors far closer to the sources than conventional scalp-distance systems, which sharpens signals.
The minimally-invasive middle: endovascular
Between non-invasive wearables and open-skull implants sits a minimally-invasive option built to reach the real world. The endovascular Stentrode mounts electrodes on a stent that is threaded through the jugular vein and lodged in a venous sinus beside motor cortex, using a familiar neuro-interventional route and avoiding a craniotomy. It records local field potentials through the vessel wall — a weaker, more filtered signal than an intracortical array — but its bet is that a less-invasive delivery is what actually lets a signal-generating implant leave the hospital and be used at home.
No single sensor wins, so a strong real-world design is often a hybrid, multimodal wearable: pairing fast-but-noisy EEG with slow-but-robust fNIRS, or adding accelerometers and eye trackers so the same motion that corrupts the brain signal can be measured and regressed out. The stack, not any one transducer, is the real product.