A startup named Cognixion is launching a clinical trial that integrates a brain-computer interface (BCI) into Apple’s Vision Pro headset, aiming to give speech-impaired individuals a new path to communicate, tells Wired.com. Rather than implanting electrodes into the brain, as companies like Neuralink pursue, Cognixion proposes a noninvasive system made of a special headband and related software, built to pair seamlessly with the AR/VR capability of Apple’s device.
The trial plans to enroll up to 10 participants in the United States with speech disabilities stemming from paralysis, stroke, brain injury, ALS, and similar conditions. The custom headband replaces Apple’s native sensor band in the Vision Pro and includes six EEG sensors to detect signals primarily from the visual and parietal cortex. The approach hinges on reading users’ visual fixation, where they focus their gaze, to let them choose options via attention alone. Meanwhile, a neural processor at the hip handles signal decoding and interfaces with the AR system.
Under the hood, the software builds a personalized AI model per user, feeding off their speech history, writing, and style to propose phrases or parse selections generated by the BCI. For users with some eye movement or head motion ability, fallback options remain, but the system’s strength lies in enabling those with limited mobility. Cognixion previously tested a standalone headset called Axon-R and showed that some users could “speak” several hours per day through the BCI.
The company sees using Vision Pro as a shortcut to reach more users; the headset’s app ecosystem and wide capabilities could lower barriers to adoption. Apple itself recently opened the door to BCI integration through a new protocol that allows control of its devices without physical input. One big hurdle is signal quality: noninvasive BCIs operate on weaker and noisier brain signals compared with invasive implants. Experts say decoding must become faster and more reliable before such systems match implant performance. Still, if this path works, it could democratize access to assistive neurotechnology without surgery.