Don't put words in my mouth: speech perception can falsely activate a brain-computer interface.
Anouck Schippers, Mariska J Vansteensel, Zac V Freudenburg and 3 others
PMID 40830488WHAT IT FOUND
A speech brain-computer interface trained to decode attempted speech falsely triggered during speech perception.
On average, 36.57% of perception epochs were misclassified as produced speech, creating a high risk of unwanted output during conversation.
Key findings
01Decoders trained on overt speech produced false positives during speech perception in all participants, with an average of 36.57% of perception epochs misclassified as produced syllables.
02Rest normalization did not resolve the issue, leaving an average of 21.14% of perception epochs falsely classified as speech.
03A separate classifier could distinguish perception from production and rest with high accuracy (91.68% for overt), suggesting a multi-stage decoding approach could mitigate false positives.
STILL TO COME
How it was doneWhat they foundWhat it means for SLPs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
The study used only five participants, all of whom could produce speech, whereas intended BCI users are often paralyzed and unable to speak. The decoders were trained on isolated syllables rather than full words or sentences, which may differ in neural complexity. The sample size for training was small, and the analysis was constrained to high-frequency band signals. The study did not test real-world conversational dynamics, only controlled laboratory tasks.
Declared interests
Funded by multiple organizations including the European Research Council, National Institute on Deafness and Other Communication Disorders, and the Dutch Brain Foundation. No specific commercial conflicts of interest were declared in the provided text.
The easy way to misread this
Do not assume that current speech BCIs are ready for daily use. The high false positive rate during speech perception means the device may generate unwanted output when the user is simply listening, which can disrupt communication and reduce user trust.