Caleb KEMERE "Computational and Experimental Approaches for Studying the Latent States of Memory"

The neural circuits that enable the storage and recall of memories of our daily experiences operate not just when we are awake and engaged with the world. Rather, important patterns of activity happen during periods when animals are in "offline" states and/or asleep. Understanding how information is represented in the brain during offline periods - when correlated sensory stimuli or motor actions are not present - requires use of computational tools. This talk will discuss unsupervised learning approaches we have adopted to understand the structure of neural activity in hippocampus during offline periods. In addition, we will touch on experimental approaches we have developed which allow us to test the causal importance of specific patterns of activity during these periods.