Your Daily digest for NIH Director's Blog

[Article Digests for Psychology & Social Work]

NIH Director's Blog

 

(https://directorsblog.nih.gov/2024/06/06/study-suggests-computerized-brain-implant-could-one-day-decode-internal-speech-for-those-who-can-no-longer-speak/) Study Suggests Computerized Brain Implant Could One Day Decode Internal Speech for Those Who Can No Longer Speak
Jun 6th 2024, 09:00

Credit: Tom Merton/KOTO/Adobe StockThe ability to communicate using only your thoughts might sound like the stuff of science fiction. But for people who don’t have the ability to speak or move due to injury or disease, there’s great hope that this may one day be possible using brain-computer interfaces (BCIs) that can “read” relevant brain signals and translate them into written or spoken words. A research team has made a preliminary advance in this direction by showing for the first time that a computerized brain implant can decode internal speech with minimal training.

In the new NIH-supported study, researchers implanted such a device in a brain area known to be important for representing spoken words called the supramarginal gyrus in two people with tetraplegia, a condition marked by full body paralysis from the neck down due to cervical spinal cord injury. The researchers found that the device could decode several words the participants “spoke” only in their minds. While we are far from using such a device to decode whole sentences or even phrases, and the exact mechanisms of internal speech are still under study, the findings, reported in (https://www.nature.com/articles/s41562-024-01867-y) Nature Human Behavior, are notable because it had been unclear whether the brain signals involved in thinking words could be reproducibly translated.

The findings come from a team led by (https://www.bbe.caltech.edu/people/richard-a-andersen) Richard Andersen at the California Institute of Technology, Pasadena, CA, and Sarah Wandelt, now at the Feinstein Institutes for Medical Research in Manhasset, NY, and the study was supported by the NIH (https://braininitiative.nih.gov/) Brain Research Through Advancing Innovative Neurotechnologies® (BRAIN) Initiative Research Opportunities in Humans program. Though earlier research had shown that brain implants could decode vocalized, attempted, and mimed speech, it had yet to be seen whether internal speech could be similarly decoded.

An earlier advance in decoding speech signals from the brain came in 2022, when the researchers reported they could accurately predict the words that a person with tetraplegia was thinking using a BCI. In the new study, they’ve shown that the device works in a second person with tetraplegia. The finding is an indication that the approach can work in different individuals and doesn’t depend on the brain characteristics of a particular person or the precise orientation of the implant in their brain.

In the study, the researchers trained their device to recognize brain patterns associated with certain internal “spoken” words including six actual words (battlefield, cowboy, python, spoon, swimming, and telephone) and two nonsense words (nifzig and bindip). During three sessions, the researchers flashed words on a screen and asked each participant to think about “saying” the words without speaking or moving. The BCI then used measurements of brain activity during the sessions and a computer model to predict the words being “spoken” internally.  

The researchers found that in this task the device could decode the words with an average accuracy of 79% with the first participant and 23% with the second participant. They noted that the second participant had fewer unique patterns of brain activity associated with the different words, which may explain the lower results. Nevertheless, the findings show that the brain region in question generally contains signals for internal speech, although there is likely also variation among people in how thoughts of particular words are represented in patterns of brain activity. Furthermore, the device’s ability to decode nonsense words suggests that the words are represented in this part of the brain phonetically and not necessarily based on their meanings.

While there is much more to learn about how to decode internal speech more reliably across individuals, the findings offer proof-of-concept for a high-performance internal speech BCI. The new research adds to a growing portfolio of rapidly advancing technologies supported by the BRAIN Initiative that could one day routinely restore the ability to communicate for those who can no longer speak or even move, including people with brain injuries, paralysis, or diseases such as amyotrophic lateral sclerosis (ALS).

Reference:

Wandelt SK, et al. (https://pubmed.ncbi.nlm.nih.gov/38740984/) Representation of internal speech by single neurons in human supramarginal gyrus. Nature Human Behaviour. DOI: 10.1038/s41562-024-01867-y (2024).

NIH Support: NIH BRAIN Initiative

Forwarded by:
Michael Reeder LCPC
Baltimore, MD

This information is taken from free public RSS feeds published by each organization for the purpose of public distribution. Readers are linked back to the article content on each organization's website. This email is an unaffiliated unofficial redistribution of this freely provided content from the publishers. 

 

(#) unsubscribe from this feed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.clinicians-exchange.org/pipermail/article-digests-clinicians-exchange.org/attachments/20240607/7ddf4ac3/attachment.htm>