Frequently asked questions

Answers to the most common questions about the scope of the measurement, the equipment required, the data and the limits of the platform.

What is NeuroScreen?

NeuroScreen is a web platform for recording and analysing EEG on one timeline with stimulus material. It supports recording of the four channels of a Muse headband at a sampling rate of 256 Hz, event markers, session history, selected ERP analyses and multi-person sessions. It is intended for research and wellness purposes; it is not a medical device.

What do I need?

You need a compatible Muse EEG headband and a Chrome or Edge browser supporting Web Bluetooth; the page must use an HTTPS connection. No additional software or drivers are required — the headband is connected directly in the browser.

Does NeuroScreen make a diagnosis?

No. NeuroScreen is not a medical device. It reports measurement results and their changes, but it does not diagnose and it does not provide clinical interpretations or therapeutic recommendations.

What exactly is recorded?

The EEG signal is recorded from four channels corresponding to the TP9, AF7, AF8 and TP10 positions, at a sampling rate of 256 Hz. The signal, the presentation of the material and the event markers are placed on one shared timeline. Depending on the protocol the data can serve, among other things, for ERP analyses and comparisons between participants.

How long does a session take?

The duration depends on the protocol and the length of the material; it is usually a few minutes. It covers the electrode contact check, an optional baseline recording and the presentation of the material with simultaneous EEG recording.

Where is my data stored?

Data is stored on servers in the European Union. The detailed rules for processing EEG recordings, including the legal bases, the scope of consent and how to download and delete data, are described in the privacy policy.

What does it cost?

At the moment, nothing. Creating an account and using the platform for your own, non-commercial purposes is free of charge and requires no card details. The platform's resources — storage, computing time, headbands for pilots — are limited, however, so for institutional or commercial use, or use significantly beyond a typical load, we get in touch and arrange the terms individually. This also applies to the NeuroScreen Professional pilot, which may require hardware, additional storage or dedicated features. Any technical limits will be described in the terms of use before they take effect.

Can a session be run with several people?

Yes. The session starts in a shared lobby with a countdown, after which participants in different locations start the same material. Afterwards the platform reports similarity measures for the time course of the responses — for the whole group and for individual pairs. Guests join through an invitation link, without creating an account.

How are similarity measures between participants tested?

The significance of a similarity measure is determined by a permutation test: one person's recording is repeatedly shifted cyclically against the other, and the p-value says how often a chance alignment produces an equally high similarity. The method is additionally checked on control data in which no shared time course is to be expected.

What must an own protocol satisfy to get an ERP analysis?

In the marker editor you mark the material as an oddball protocol and place markers of two types: standard (the frequent stimulus) and deviant (the rare stimulus). Before publication the platform checks requirements no weaker than those of its own test: at least 48 standards and 12 deviants, a deviant share of 15–25 %, at least 1.2 s between stimuli, never two deviants in a row, at least 3 standards before the first deviant, the first stimulus no earlier than 2 s after the start and the last at least 1 s before the end. The editor shows what is missing as you work; the test is created only once the list is empty.

Why does a result on an own protocol have no percentiles?

A percentile requires a reference group measured with the same stimulus. An own protocol is unique by definition, so a comparison with a population does not exist and the platform does not pretend otherwise. The result is set only against your earlier sessions on the same version of the markers; changing the markers opens a new series of comparisons from scratch.

What files can I upload as my own material?

Video in MP4, WebM, MOV and M4V formats and audio in MP3, M4A, OGG, WAV and FLAC formats, each file up to 200 MB. An image sequence is built from up to 200 photos. One account can hold up to 20 films and 20 slideshows and 2 GB of material in total. Materials are private — only invited participants see them.

What if my browser does not support Web Bluetooth?

The platform generates a one-time link that moves the session to a compatible browser — Chrome or Edge — without logging in again. Preparing materials, markers and panels works in any modern browser; Web Bluetooth is needed only for the EEG recording itself.

How can I sign in?

With a one-time link sent by e-mail or with a password — your choice. You can optionally enable two-factor sign-in: a code by e-mail or an authenticator app, with backup codes in case you lose your device.

What languages does the platform work in?

Polish, German and English — from the public pages through the portal and the session screen to the result e-mails. The Polish version of the content is the source version.

Does the platform meet accessibility requirements?

The site has been self-assessed against WCAG 2.2 AA. The current scope of conformance and the known limitations are described in the accessibility statement.

Create a free account

Registration requires an e-mail address, a password and confirmation of the address. You can connect the headband later; without it you can prepare materials, event markers and participant panels, but you cannot record EEG.