Contributed Talk Sessions | Poster Sessions | All Posters | Search Papers

Poster Session B: Wednesday, August 13, 1:00 – 4:00 pm, de Brug & E‑Hall

A Biologically Plausible Computational Model of Hippocampal Neurogenesis and Pattern Separation in Memory

Jiachuan Wang¹, Vachan Shetru Jagadeesh, M Ganesh Kumar², Camilo Libedinsky¹, Shih-Cheng Yen¹, Andrew Yong-Yi Tan¹, Jai S Polepalli; ¹National University of Singapore, ²Harvard University

Presenter: Jiachuan Wang

Pattern separation, essential for encoding distinct memories of overlapping contexts, relies on dentate gyrus coding shaped by entorhinal input and strong lateral inhibition. Although synaptic plasticity and adult hippocampal neurogenesis have been implicated in this process, their precise contributions remain unclear. The Cbln4-Neo1 complex, which mediates plasticity at entorhinal cortical synapses in the dentate gyrus without affecting basal signal transmission, offers a unique target for investigation. In this study, we selectively deleted Cbln4 from inputs to the mouse dentate gyrus. We found that Cbln4 is required for behavioral pattern separation and suppresses activity-dependent neurogenesis. We then developed a biologically plausible computational model incorporating an entorhinal cortex-dentate gyrus circuit in a reinforcement learning framework. Simulations suggested that either impaired synaptic plasticity or increased neurogenesis alone was sufficient to disrupt behavioral pattern separation by elevating representational similarity in the dentate gyrus. These findings highlight the role of Cbln4 in memory encoding and dissociate the contributions of synaptic plasticity and neurogenesis through computational modeling.

Topic Area: Memory, Spatial Cognition & Skill Learning

Extended Abstract: Full Text PDF