Speaker: Emilio Kropff, Principal Investigator of the Laboratory of Brain Physiology and Algorithms, Leloir Institute IIBBA-CONICET.

Time: 15:30-17:00, Feb. 28, 2025, GMT+8

Venue: Lecture Hall (Room 101), Jinguang Life Science Building, PKU

Abstract:
The mammalian hippocampal formation is essential for memory processing and spatial navigation, yet the computational principles underlying its function remain elusive. In this talk, I will discuss three aspects of hippocampal algorithms explored through experimental and computational approaches. One such principle is pattern separation, enabling the hippocampus to encode distinct memories for similar experiences. By optogenetically stimulating adult-born neurons at different maturation stages, we have gained insights into how young, immature neurons crucially contribute to this process. Another key computation is path integration, which allows for self-motion-based navigation. While entorhinal grid cells are thought to implement this function, the underlying network design remains unclear. Using computational simulations, we show that this structure may be far simpler than previously assumed, revealing that network architecture and representational space can be completely different topological objects with distinct dimensionality. Finally, I will introduce our first studies of hippocampal function in Octodon degus, a South American rodent that spontaneously develops sporadic Alzheimer's-like disease. By integrating experimental and computational approaches, our work advances the understanding of hippocampal algorithms and their broader implications for neuroscience and neurodegenerative disease models.

Source: Peking University (PKU) - Tsinghua University(THU) Joint Center for Life Sciences