The large-scale activity of the brain exhibits rich and complex patterns, but the spatiotemporal dynamics of these patterns and their roles in brain functions remain unclear. In this talk, we will demonstrate that brain activity forms propagating wave patterns, such as spirals or vortices. These wave patterns and their interactions play a crucial role in coordinating large-scale activity flow between distributed functional regions. This coordination mechanism enables the flexible reconfiguration of task-driven activity flow between bottom-up and top-down directions during cognitive processing. Additionally, we will introduce a theory termed Fractional Neural Sampling (FNS) to explain how wave patterns implement sampling-based probabilistic computation. We will show that FNS provides a unified account of various brain functions, from visual perception inference to attentional sampling and interarea communication, and it has important implications for understanding the dynamics of deep learning.

报告人简介

Dr. Pulin Gong (龚璞林) is an Associate Professor at the University of Sydney, where he leads the Theoretical and Computational Neuroscience Group. He also serves as an Associate Editor for the Journal of Computational Neuroscience. Before his tenure at the University of Sydney, Dr. Gong was a staff scientist at the RIKEN Brain Science Institute in Japan. Dr. Gong's research focuses on investigating the self-organizing mechanisms of brain spatiotemporal dynamics and uncovering the underlying principles that govern how these dynamics implement neural computations.

邀请人

周海军 研究员