AI Decodes Mouse Brain Signals to Predict Visual Perception

EPFL researchers have developed a machine-learning algorithm called CEBRA that can predict what mice see based on decoding their neural activity. The algorithm maps brain activity to specific frames and can predict unseen movie frames directly from brain signals alone after an initial training period. CEBRA can also be used to predict movements of the arm in primates and to reconstruct the positions of rats as they move around an arena, suggesting potential clinical applications.