报告题目：Neural Dynamics for Perception, Cognition, and Navigation

报告人：杨冬平 研究员，（之江实验室）

报告时间：2025年11月6日，15：00

报告地点：致知楼学术沙龙室3515

报告摘要：

Brains exhibit complex and variable internal activity yet produce reliable, adaptive responses to external inputs. A central open question is how neural networks achieve stable representations under such variability. We show that neural heterogeneity plays a key role by disrupting rigid activity patterns, making neuronal dynamics more independent and flexible, with heterogeneous firing that enhances sensitivity to external signals. Under stimulation, this sensitivity yields stable, input-aligned transient spatiotemporal patterns that support robust representation. Moreover, cognition requires mapping sensory inputs onto low-dimensional manifolds within high-dimensional activity. We find that such manifold mappings enable representation transfer and cognitive generalization. Finally, in spatial navigation, we reveal that hippocampal theta oscillations and their coupling with movement balance reward prediction with threat vigilance during navigation. Our model further explains drug-induced impairments in navigation and clarifies the distinct roles of oscillatory components: theta oscillations support self-location awareness, gamma oscillations enhance predictive processing, and their coupling ensures coordinated temporal integration. Together, these findings illuminate the dynamical mechanisms by which neural systems achieve both reliable representation and adaptive vigilance, providing a unified computational framework for understanding Perception, Cognition, and goal-directed Navigation in the brain.

报告人简介：

杨冬平博士，之江实验室研究专家，专注于脑活动的生物物理与动力学机制研究，主要研究方向包括神经形态智能、睡眠调节与癫痫动力学。他运用平均场理论及非线性动力学建模方法，致力于揭示脑功能背后神经活动的时空多尺度特性。作为第一作者或通讯作者，在多个顶尖期刊发表重要成果，包括Science Advances、PNAS、Intelligent Computing、PLOS Computational Biology、Physical Review E、Chaos、IEEE Transactions on Neural Systems and Rehabilitation Engineering、Journal of Neural Engineering，并在IEEE EMBC（2023年）等权威会议上进行学术交流。目前主持国家自然科学基金面上项目及浙江省自然科学基金重点项目。