Computational Neuroscience of Cognition

Catalog description

This course is designed to introduce a modern perspective in Neuroscience, based on the observation that neural population activity is best described using the theory of nonlinear dynamical systems. The emphasis is fourfold. First, we will consider neural networks as dynamical systems, and introduce concepts such as the state space of neural population dynamics and bifurcation as a mathematical engine for the emergence of novel functional abilities. Second, we will discuss how to strive for cross-level understanding from receptors and cell types to collective neural circuit dynamics to behavior, as well as clinical applications in Computational Psychiatry. Third, we will examine core building blocks of cognition, such as working memory, decision making, behavioral flexibility in multitasking. This is an area of research bridging with Artificial Intelligence. Fourth, we will cover connectome-based modeling of a multiregional brain, a frontier topic in close interplay with recent experimental advances.