Human brain organoids classified Braille letters with up to 83% accuracy

A system translated tactile sensor signals into electrical patterns that three organoids processed together to identify letters.

Editorial illustration — not from the study.

Researchers used an event-based tactile sensor mounted on a robotic arm to scan a board containing the 26 English Braille letters. They converted the sensor’s signals into electrical stimulation delivered through eight electrodes beneath each human forebrain organoid, then classified the organoids’ spike responses with machine-learning methods.

A single organoid produced an average classification accuracy of 61%. Combining responses from three organoids raised accuracy to 83% and made the system more resistant to several types of artificially added noise.

How the organoids classified letters

The study developed an encoding method that maps tactile sensor data to electrical stimulation patterns. The researchers tested how stimulation features—including pulse number, amplitude, duration and timing—affected organoid activity, measuring spike counts and changes in the spatial center of activity.

Using recordings from the Evetac tactile sensor, the system classified 26 Braille letters. Average accuracy was 61% with one organoid and 83% with a three-organoid ensemble. The three-organoid setup also showed greater robustness when Gaussian noise, uniform noise, missing data and outliers were artificially introduced.

Why the result matters

The results indicate that human forebrain organoids can serve as biological processing elements in an open-loop system: tactile information is encoded into electrical input, and the organoids’ activity is used to produce a classification. Combining several organoids improved both accuracy and resistance to added noise in this test.

The researchers present the encoding method as groundwork for future bio-hybrid computing systems, in which living neural networks are combined with sensors and electronic hardware. The study suggests a possible role for organoids in adaptive, potentially low-power computing, but it does not establish performance in a practical Braille-reading device.

Evidence and caveats

This was a laboratory classification experiment using human forebrain organoids on low-density microelectrode arrays, an event-based tactile sensor, a robotic arm and a 26-letter artificial Braille board. The system was open-loop: it translated recorded tactile inputs into stimulation and classified the resulting activity rather than testing an organoid-driven device in a real-world reading task.

The reported noise tests used artificially introduced noise. The findings therefore show performance on this experimental benchmark, not that organoids can read Braille through touch or that the system will work outside the tested setup. The excerpts also do not report a direct energy measurement or demonstrate learning through closed-loop interaction.

A peptide reduced persistent traumatic memories in mice

A study in mice found that a peptide targeting the memory-related protein CPEB3 reduced the persistence of traumatic memories and anhedonia, a loss of pleasure. The findings are an early genetic preclinical test of a possible approach for PTSD, not evidence of a treatment in people.

6 hours ago·5 views

Biology

In fruit flies, heat and pesticide stress cut fertility across generations

Acute heat and sublethal imidacloprid exposure interacted to reduce male fruit flies’ reproductive success more than either stressor alone in some conditions. Stress experienced by fathers also reduced offspring fertility, while a more diverse gut microbiome partly lessened the effects on males.

20 hours ago·5 views

Biology

A tadpole’s anti-swallowing body comes with a swimming cost

Predators can trigger different body shapes in Rana pirica tadpoles. The shapes improve swimming efficiency, but the swollen body that helps protect tadpoles from being swallowed also creates drag and reduces escape performance compared with a deeper tail alone.

1 day ago·15 views