国家数学与交叉科学中心合肥分中心报告【Bailu Si】

报告人：Dr. Bailu Si (Department of Neurobiology, Weizmann Institute of Science)

报告题目：Dynamical models of animal navigation

报告时间：2013年7月25日上午10:00

报告地点：管理楼1611教室

报告摘要：How do animals self-localize themselves in an environment when they explore the environments with variable velocities? One mechanism is dead reckoning or path-integration. Recent experiments on rodents show that such computation may be performed by grid cells in medial entorhinal cortex.  Each grid cell fires strongly when the animal enters locations that define the vertices of a triangle grid. Some of the grid cells show grid firing patterns only when the animal runs along particular directions.

In this talk, I will present an attractor network model that accounts for the conjunctive position-by-velocity selectivity of grid cells. We propose that grid cells collectively represent arbitrary conjunctions of positions and movements of the animal. Due to asymmetric recurrent connections, the network has grid patterns as states that are able to move intrinsically with all possible directions and speeds. A velocity-tuned input will activate a subset of the population that prefers similar movements, and the pattern in the network moves with a velocity proportional to the movement of the animal in physical space, up to a fixed rotation. Thus the network 'imagines' the movement of the animal, and produces single-cell grid firing responses in space with different degree of head-direction selectivity. We propose testable predictions for new experiments to verify our model.

欢迎广大师生参加！