What A Consumer EEG Headset Can Measure

Headbands and earbuds that claim to track focus, calm or sleep are recording genuine signals. The gap between that recording and the claims made about it is where most consumer confusion lives.

What the electrodes actually pick up

Scalp electrodes measure voltage differences produced by the summed activity of very large populations of neurons. Anything too small, too deep or not synchronized across many cells does not reach the surface at a detectable level.

The skull and scalp spread the signal out, so each electrode sees a blurred mixture from a wide area rather than the output of a specific structure.

This is a physical limit of the measurement, not a limitation of the device's processing, and no amount of software recovers information that never arrived.

Channel count decides what can be separated

Research systems use many electrodes, which allows sources to be partially separated by comparing what each one sees. A consumer device with a handful of contacts cannot perform that separation well.

With few channels, the practical output is broad frequency content over a general region, which supports coarse distinctions such as drowsy versus alert.

Claims that require locating activity precisely, or distinguishing similar mental content, exceed what the hardware can support.

Artifacts frequently exceed the signal

Jaw clenching, blinking, eye movement, heartbeat and a loose electrode all produce voltages larger than brain rhythms. Muscle activity in particular sits in the same frequency range as signals of interest.

Algorithms remove much of this, but removal is imperfect and a device that reports a clean number is making a judgment the user cannot inspect.

A common demonstration failure is a headset that appears to detect concentration when it is actually detecting a still, relaxed jaw.

Sleep staging is the strongest consumer case

Sleep produces large, slow, distinctive patterns that survive the trip through the skull and persist for minutes at a time, which suits low-channel recording well.

Consumer devices can therefore give reasonable, if imperfect, estimates of sleep structure over a night, especially compared with movement-based wrist sensors.

Clinical assessment still uses many more measures, including breathing and muscle tone, because the questions being asked are diagnostic rather than descriptive.

How to read a product claim

The useful questions are how many electrodes there are, where they sit, what is being computed, and whether the device reports the underlying measurement or only an invented score.

A score with no defined units and no published relationship to a measured quantity cannot be evaluated or compared, and it is the most common form these products take.

Anyone considering a device for a health concern rather than curiosity should raise it with a clinician, because a consumer sensor is not built to answer diagnostic questions.