While the significance of brain oscillations as an indication of synchronized neuronal activity has been widely acknowledged, our understanding of how coordinated neuronal firing patterns support behavior and memory processing is only beginning to emerge. In particular, oscillations in complex memory tasks are highly dynamic with frequent transitions between predominant frequency ranges. Different brain states that are associated with behavior are characterized by a wide range of oscillatory frequencies that likely reflect distinct underlying network mechanisms to support different phases of memory. Yet, memory-guided behavior requires the uninterrupted retention and updating of task-relevant information across numerous transitions between brain states. One of the remaining key outstanding questions is thus how information is not only retained but also organized to become task-relevant throughout these transitions. To study this question, we will focus on the hippocampal dentate-CA3 network where we have recently shown that the dentate gyrus is critical for the generation of prospective coding of future correct choices by CA3 cells during sharp-wave ripples (SWRs) in a dentate-dependent working memory task. SWRs are high-frequency oscillations that are accompanied by brief increases in firing rates during which behaviorally relevant events are replayed. While our previous work determined that the prospective coding occurred during SWRs at reward locations, we propose to next determine whether the dentate inputs to CA3 are also critical for prospective coding during theta oscillations along the path to future reward locations. Given that our previous and preliminary data support the possibility that the dentate is necessary for prospective coding in CA3 during SWRs and theta states, we propose to next identify how prospective coding is coordinated across the transition between these brain states in order to support the planning of future decisions and trajectories. We predict that the neuronal representations of all available choices are played out during SWRs, while a selection of the next choice occurs during the subsequent theta state. Finally, we will establish a causal relationship between predictive neuronal firing sequences generated during high-frequency oscillations and predictive sequences in theta states by disrupting CA3 SWRs and examining subsequent behavioral choices and subsequent spike sequences in theta cycles. Taken together, these studies will allow us to describe network mechanisms in the hippocampus that dynamically interleave across brain states to support hippocampus-dependent memory. Our work will potentially also have significant implications for therapeutic intervention in diseases with gender-dependent memory comorbidities, as we will investigate potential sex differences for dentate-CA3 network computations critical for memory formation. This fills a gap in our understanding as no studies on functional network differences have been reported for this circuit despite clear evidence for sex differences in dentate-CA3 anatomy, plasticity, and response to aging, stress and diseases such as depression and epilepsy.
Understanding how neural circuits create memories in service of ongoing behavior and decision making remains one of the greatest challenges in neuroscience research. The goal of this research project is to identify the network mechanisms that bridge memory processing across brain states in order to support memory guided behavior. Elucidating these mechanisms will not only provide a deeper understanding of normal brain function, but will also provide insight into disorders where decision making, learning and memory are disrupted.
