Uncovering the roles of representational drift in the brain through the lens of dynamical systems and their practical implications in brain-computer interfaces

Understanding representational drift—the brain’s evolving representation of its environment—is pivotal to gaining insights into neural computation. Despite its significance, the study of representational drift has been constrained by the scarcity of suitable datasets and methods. We tackle these challenges using novel experimental techniques, including full-body motion tracking and synchronized neural recordings from freely moving rhesus macaques, coupled with statistical methods for quantifying drift and its dependency on environmental and mental factors. This research will uncover the mechanisms behind the brain’s adaptability and learning, unlocking the great potential to improve brain-computer interfaces and develop new therapeutic strategies for brain injury recovery.