Alice Xue

Remembering past experiences can guide action in the present, while future behavior is often facilitated by learning from current events. Memory encoding and retrieval computations trade off, raising critical questions about how the brain balances this tension while minimizing loss of new or old knowledge. A leading computational theory from systems neuroscience posits that a mismatch between predictions based on memories of prior experiences and current experience elicits a transition from a retrieval state to an encoding state, persisting for a few seconds or longer. A second theory proposes that rapid, sub-second switches between retrieval and encoding states underlie the ability to learn from these errors. My research brings these ideas from systems neuroscience into psychology by integrating concurrent multi-modal neuroimaging and physiological recordings in an unprecedented manner. By advancing understanding of the dynamics of memory states that unfold at different timescales, this research will provide insights into neural mechanisms that can be a target of future investigation in aging and disease, as well as potential strategies to foster learning in educational settings.