Post-Mortem Human Brain Tissue Trained to Control Robotic Piano-Playing Hand
French researchers have successfully connected post-mortem adult human brain explants (OPAB) to a robotic hand, training the hybrid system to mimic piano playing. By keeping the brain slices viable in a culture dish and linking them to electrodes, the system learned to press keys in response to auditory feedback. This research advances wetware computing and biohybrid AI, which aim to leverage the high energy efficiency and rapid learning capabilities of biological neurons. It demonstrates the potential of using mature, structured human brain tissue rather than simplified 2D cultures or immature organoids for biological computing. The system utilized post-mortem adult brain tissue placed on a microelectrode array, where three electrodes were mapped to control three robotic fingers representing the notes LA, TI, and DO. A closed-loop system was established by using a microphone and sound decoder to feed auditory signals back to the brain tissue.
## BACKGROUND
Wetware computing and organoid intelligence (OI) are emerging fields that integrate living biological materials, like neurons or brain organoids, with electronic hardware to process information. Unlike traditional silicon-based computers, biological systems operate with extremely low power consumption (around 20 watts for a human brain) and excel at learning from very few examples.