Reactivation of Neural Ensembles During Memory Retrieval

Memory retrieval is thought to involve the recreation of neural activity patterns that were present during learning. This idea stems from Hebbian models of plasticity, which assume that learning-induced changes in synaptic strength increase the probability of neural reactivation during retrieval. Although neural reactivation has long served as the basis of current models of memory it has been extremely difficult to validate experimentally. While it has been possible to follow the activity of individual neurons shortly after learning it has not been possible to examine their activity days and weeks later during retrieval. We addressed this issue by using a stable form of GFP (H2B-GFP) to permanently tag neurons that are active during contextual fear conditioning. We demonstrate that neurons in the hippocampus and neocortex reactivate during the retrieval of context fear (Chapter 2). We next show that patterns of retrieval-driven reactivation shift over time in the hippocampus and amygdala, while remaining stable in the cortex (Chapter 3). Finally, we demonstrate that although the hippocampus is required for neocortical reactivation several days after learning, cortical regions can compensate when this region is inactivated following a weeks-long consolidation period (Chapter 4). Together, the results presented in this dissertation increase our understanding of the hippocampus’ role in retrieval-driven reactivation and inform current debates over the accuracy of memory storage and retrieval models.