Implanted electrodes can be used to transcribe "inner speech"

It has been long known that electronic amplification of "subvocalization" can cause words that a person says "under their breath" to be heard. The researchers in the study below have taken this further, showing that an implanted electrode grid can pick up words we think to ourselves, via traces of activation of the portion of the motor cortex responsible for the muscles which exress our speech. This takes pickup of subvocalization one step further, into the brain, if not actually into the language center itself. Note that the success rate was still a bit lower than needed for an effective communucations brain implant or those who can speak normally -- a 20% error rate clearly needs improvment. However, for those robbed of speech by bulbar weakness from conditions like ALS, this type of adaptive device can be a great boon.

The authors also explored "iiner speech" as a kind of mind reading, and found a technique that patients, who normally were training to make the machine better read words, could also learn to hide their thoughts in large part; they also implemented a code phrase "chitty chitty bang bang" to temporarily stop the device from transcribing.

==============================================================

ABSTRACT

TITLE - Inner speech in motor cortex and implications for speech neuroprostheses

AUTHORS - Kunz, Erin M.; Abramovich Krasa, Benyamin; Kamdar, Foram; Avansino, Donald T.; Hahn, Nick; Yoon, Seonghyun; Singh, Akansha; Nason-Tomaszewski, Samuel R.; Card, Nicholas S.; Jude, Justin J.; Jacques, Brandon G.; Bechefsky, Payton H.; Iacobacci, Carrina; Hochberg, Leigh R.; Rubin, Daniel B.; Williams, Ziv M.; Brandman, David M.; Stavisky, Sergey D.; AuYong, Nicholas; Pandarinath, Chethan; Druckmann, Shaul; Henderson, Jaimie M.; Willett, Francis R.

doi: 10.1016/j.cell.2025.06.015

Cell, https://doi.org/10.1016/j.cell.2025.06.015

Speech brain-computer interfaces (BCIs) show promise in restoring communication to people with paralysis but have also prompted discussions regarding their potential to decode private inner speech. Separately, inner speech may be a way to bypass the current approach of requiring speech BCI users to physically attempt speech, which is fatiguing and can slow communication. Using multi-unit recordings from four participants, we found that inner speech is robustly represented in the motor cortex and that imagined sentences can be decoded in real time. The representation of inner speech was highly correlated with attempted speech, though we also identified a neural ?motor-intent? dimension that differentiates the two. We investigated the possibility of decoding private inner speech and found that some aspects of free-form inner speech could be decoded during sequence recall and counting tasks. Finally, we demonstrate high-fidelity strategies that prevent speech BCIs from unintentionally decoding private inner speech.