Researchers unveil new method to identify shared brain activity patterns

Researchers have introduced a new method called MARBLE, which uses geometric deep learning to identify shared brain activity patterns among different individuals. This approach allows for better understanding of how various brains solve the same tasks. In tests with macaques and rats, MARBLE successfully decoded neural activity related to movement and navigation, showing improved accuracy over traditional machine learning methods. It maps neural data into high-dimensional spaces, enabling effective pattern recognition. MARBLE's findings could enhance brain-machine interfaces and assistive robotics by translating brain activity into recognizable patterns. The method may also benefit other scientific fields by allowing for the analysis of dynamic phenomena across multiple datasets.