{"slug": "lab-grown-brain-organoids-sort-braille-letters-with-83-accuracy", "title": "Lab-Grown Brain Organoids Sort Braille Letters With 83% Accuracy", "summary": "Groups of three lab-grown human brain organoids identified Braille letters with 83 percent accuracy, while a single organoid reached about 61 percent, University of Bristol researchers reported Friday in Frontiers in Artificial Intelligence. The organoids, each under a millimeter and holding about 10,000 neurons, received electrical stimulation patterns from a robot tactile sensor scanning all 26 Braille letters, and a support vector machine classifier decoded their responses; the experiment was open loop, so the team reported no evidence of learning or lasting change. \"Compared with established computing chips, organoid systems still have a long way to go in terms of reliability, reproducibility, and scalability,\" said Tianyi Liu, the doctoral candidate who led the study.", "body_md": "Groups of three lab-grown human brain organoids identified letters of the Braille alphabet correctly 83 percent of the time, researchers at the University of Bristol reported on Friday in the journal Frontiers in Artificial Intelligence.\n\nA single organoid did worse, at about 61 percent. Each one is a piece of neural tissue grown from stem cells, smaller than a millimeter and holding about 10,000 neurons. The Bristol group studies how robots sense and react to touch, and it wanted to know whether living tissue could process data coming off a robot’s tactile sensor.\n\n“If we’ve been trying to imitate biological neural networks all along, why not try using real biological neural networks directly?” Tianyi Liu, a doctoral candidate at Bristol who led the study, said in an email to [Becky Ferreira of 404 Media](https://404media.co/computers-made-of-human-brain-tissue-identify-braille-letters-in-breakthrough/).\n\nThe work falls in a young field called organoid intelligence, which tries to use organoids as biological hardware for computing. Earlier experiments had organoids recognizing spoken Japanese vowels and playing the video game Pong.\n\nFor this one, a tactile sensor scanned all 26 Braille letters, which are built from patterns of raised dots. The team turned each scan into a pattern of electrical stimulation and fed it to organoids sitting on multielectrode arrays that kept them alive and recorded their activity. A support vector machine, a basic machine-learning classifier, then tried to read off which letter each response matched, according to [a news release from the journal’s publisher](https://www.frontiersin.org/news/2026/10/16/mini-brain-organoids-reading-braille-biohybrid-computing-frontiers-artificial-intelligence).\n\n“Our original goal wasn’t to prove that organoids could read Braille like humans, or that they were already ‘intelligent’ enough to understand Braille,” Ms. Liu said. “Rather, we wanted to use this task to explore whether real biological neural networks could participate in processing information from robotic tactile sensors.”\n\nWhy three organoids beat one is an open question. “We still don’t know the exact mechanism behind this improvement,” she said. Her working guess is that each organoid answers the same stimulus in a similar but not identical way, so pooling them catches details any one of them misses. She compared it to “three people looking at the same object.”\n\nThe trio also held up better against interference. When the researchers added noise to the Braille signals, the same classifier run directly on the noisy data was fooled more easily than when it decoded the organoids’ responses, according to the release. “Combining outputs from several organoids also showed an improvement in performance,” said Dr. Benjamin Ward-Cherrier, a senior lecturer at Bristol and the study’s senior author.\n\nThe claims stop short of learning. The experiment was open loop: the organoids got the inputs but no feedback on how they did. So the results show they could help sort the letters, not that they learned Braille or changed in any lasting way. “We did not demonstrate learning or long-term changes in the organoids,” Ms. Liu said in the release.\n\nThe tissue is also hard to work with. No two organoids are alike, which muddies results, and they have to be kept alive in culture, a harder job than maintaining an electronic device.\n\n“Compared with established computing chips, organoid systems still have a long way to go in terms of reliability, reproducibility, and scalability,” Ms. Liu said.\n\n“The next step is definitely to move beyond Braille,” she said in the release. The team plans to try textures, edges and other kinds of touch, and eventually closed-loop tasks “where the organoid receives feedback from what it does,” to see whether the tissue can adapt through experience.", "url": "https://wpnews.pro/news/lab-grown-brain-organoids-sort-braille-letters-with-83-accuracy", "canonical_source": "https://rews.cc/a/lab-grown-brain-organoids-sort-braille-letters-with-83-accur-099c66", "published_at": "2026-10-16 04:01:00+00:00", "updated_at": "2026-10-16 16:49:07.219409+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "robotics", "neural-networks", "ai-research"], "entities": ["University of Bristol", "Frontiers in Artificial Intelligence", "Tianyi Liu", "Benjamin Ward-Cherrier", "404 Media", "Becky Ferreira"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/lab-grown-brain-organoids-sort-braille-letters-with-83-accuracy", "markdown": "https://wpnews.pro/news/lab-grown-brain-organoids-sort-braille-letters-with-83-accuracy.md", "text": "https://wpnews.pro/news/lab-grown-brain-organoids-sort-braille-letters-with-83-accuracy.txt", "jsonld": "https://wpnews.pro/news/lab-grown-brain-organoids-sort-braille-letters-with-83-accuracy.jsonld"}}