NeuroOS

NeuroOS: Bringing Brain–Computer Interaction into Everyday Computing

NeuroOS is an interactive platform that aims to make brain–computer interfaces accessible to people who do not normally work with EEG—including researchers from other fields and the general public. Rather than presenting EEG as something that can only be inspected in a laboratory, NeuroOS lets users experience what brain signals can currently do, what they cannot yet do, and how they can become meaningful actions in the outside world.

The central experience of the demo is simple: use brain and body signals to operate a computer. A webcam-based gaze interface first allows the user to move a cursor naturally with their eyes. A blink can be used to lock the cursor onto a target, after which EEG-based steady-state visual evoked potential (SSVEP) decoding enables the user to issue a selection or click command. In this way, an intention represented in neural activity becomes an action on the computer without requiring a conventional mouse or keyboard.

The individual components of this interaction—gaze tracking, blink detection, and SSVEP decoding—are established techniques. The goal of NeuroOS is therefore not to claim a new BCI decoding algorithm. Instead, its novelty lies in turning these techniques into a casual, approachable application experience: a system that people can put on, explore, and use as naturally as they would try a new computer interface. Capabilities that are often demonstrated in specialized BCI or assistive-technology settings are reframed here as everyday interactions with a personal computer.

NeuroOS currently uses the BrainBit wearable EEG headset as its primary neural interface, while being designed to accommodate other EEG devices in the future. It is also inherently multimodal: EEG can be combined with signals such as gaze, blinks, video, and other physiological measurements. The broader goal is to provide a common interaction layer in which different brain and body signals can be connected to real-world actions.

For attendees, the demo is intended to make the current state of non-invasive BCI immediately understandable. Participants can observe their EEG, experience variability and artifacts firsthand, and then see how the same signals can close the loop from brain, to decoding, to action in the external world.

This directly connects to the workshop's emphasis on embodied intelligence and closed-loop interaction between the brain, body, and machine. NeuroOS explores what that loop could look like outside the laboratory: not only as a research instrument, but as an accessible interface through which anyone can experiment with the emerging relationship between neural signals and everyday computing.

The demo requires a Windows laptop, BrainBit headset, webcam, display, table, chairs, and power. All processing runs locally. Flickering visual stimuli used for SSVEP will be optional and accompanied by a photosensitivity notice.