BCIwiki (bciwiki.com) — According to the journal record from Applied Sciences, a study by Aalborg University authors published on October 30, 2026 tested a brain-computer interface (BCI) for two-dimensional cursor control using executed or imagined tongue movements. The proof-of-concept study enrolled 10 able-bodied, BCI-naive participants, each exposed to executed-movement and motor-imagery conditions.
To address the difficulty of decoding multiple control classes in an asynchronous BCI, the researchers combined a binary BCI switch with a state machine. The system detected sensorimotor rhythms associated with executed or imagined tongue movements. Common spatial pattern filtering and linear discriminant analysis distinguished movement-related activity from idle activity. The state machine provided four directional options—up, right, down and left—which participants selected through the BCI to move the cursor toward on-screen targets.
All 10 participants achieved successful BCI control during executed tongue movements, with a completion rate of 96.0 ± 4.9% and an average completion time of 30.8 ± 17.6 s for successful targets. In the motor-imagery condition, seven of the 10 participants achieved control. The reported completion rate for that condition was 41.0 ± 36.2%, and the average completion time for successful targets among the seven participants who reached at least one target was 66.5 ± 26.8 s.
Each participant completed ten trials per condition, with a three-minute limit per trial. Touching the target counted as success; stopping on it was not required. Executed movement always preceded imagery, and fatigue was not assessed. Potential muscle artifacts from tongue movement were not controlled for, so performance during executed movements cannot be attributed entirely to brain activity. The authors interpret the findings as supporting the feasibility of combining a tongue movement-based BCI with a state machine for two-dimensional cursor control and informing asynchronous BCI development for people with severe motor impairments. The observations were made in able-bodied participants and do not establish efficacy in people with severe motor impairments.