Next-generation neural interface: paralyzed patient transmits nearly 2 million words through thought alone
The world of neural interfaces has made a qualitative leap: scientists at the University of California, Davis, have presented the results of long-term home use of a speech BCI in a patient with amyotrophic lateral sclerosis (ALS). Over 19 months, Casey Harrell, using the implanted system, transmitted 183,060 sentences — a total of 1,960,163 words. The average speed was 56 words per minute, approaching the pace of natural speech.
Real Life, Not a Laboratory
The main difference of this study from previous ones is the focus on everyday use. For over 3,800 hours, the system operated at home, without the constant presence of researchers. Harrell communicated with family, friends, and colleagues, sent letters, participated in video calls, and even maintained full-time employment despite complete paralysis. According to the patient's own assessment, 92% of sentences were decoded as at least "mostly correct." In formal tests, accuracy exceeded 99% with a vocabulary of 125,000 English words, reaching a peak value of 99.2%. After assistants were allowed to independently connect the equipment, the average daily usage time increased from 3.7 to 9.5 hours — an indicator of true integration into life.
The Technology Behind the Scenes
In 2023, Harrell had four arrays of microelectrodes implanted in the left precentral gyrus — the brain area responsible for coordinating speech. The system reads signals from 256 cortical electrodes every 80 milliseconds. The algorithm converts neural activity into phoneme probabilities, and the language model selects the most likely sequence of words. The text is displayed on the screen in real time and can be voiced with a synthesized voice tuned to the patient's voice timbre before the illness. Notably, the same speech motor cortex is used for cursor control, which refutes previous hypotheses about the need for different brain areas for speech and movement.
From Experiment to Tool
Previously, the team demonstrated high accuracy in laboratory conditions: within 30 minutes, the system achieved 99.6% accuracy with a vocabulary of 50 words. The current work represents a transition from controlled trials to real, long-term use. Study co-author Sergey Stavisky noted that 3,800 hours of recorded brain activity became the largest individual dataset with resolution at the level of individual neurons. However, the researchers emphasize that this is a single clinical case. The system remains experimental, requires wired connections, and daily assistance from trained assistants. It is not yet known how applicable the results are to other patients, implantation sites, or electrode types.
My view: The breakthrough is obvious — we are seeing not just a laboratory prototype, but a working tool that fundamentally changes the quality of life. Nevertheless, the path to widespread adoption will be long: from miniaturizing equipment to wireless data transmission and adaptation to different neurological conditions. But it is precisely such cases that lay the foundation for a future where BCIs become not a miracle, but an everyday reality.