Methods in Computational Neuroscience

MCN introduces students to the computational and mathematical techniques that are used to address how the brain solves problems at levels of neural organization ranging from single membrane channels to operations of the entire brain.

Course date:
Jul 24, 2024 - Aug 20, 2024
Application due date:
Mar 06, 2024

Directors: Stefano Fusi, Columbia University; and Roozbeh Kiani, New York University

Course Description

Animals interact with a complex world, encountering a variety of challenges: They must gather data about the environment, discover useful structures in these data, store and recall information about past events, plan and guide actions, learn the consequences of these actions, etc. These are, in part, computational problems that are solved by networks of neurons, from roughly 100 cells in a small worm to 100 billion in humans. Methods in Computational Neuroscience introduces students to the computational and mathematical techniques that are used to address how the brain solves these problems at levels of neural organization ranging from single membrane channels to operations of the entire brain.

In each of the first three weeks, the course focuses on material at increasing levels of complexity (molecular/cellular, network, cognitive/behavioral), but always with an eye on these questions: Can we derive biologically plausible mechanisms that explain how nervous systems solve specific computational problems that arise in the laboratory or natural environment? Can these problems be decomposed into manageable pieces, and can we relate such mathematical decompositions to the observable properties of individual neurons and circuits? Can we identify the molecular mechanisms that provide the building blocks for these computations, as well as understand how the building blocks are organized into cells and circuits that perform useful functions?

Core presentations in weeks one to three will be given jointly by theorists and experimentalists who have worked, often together, on the same problems. In the first week, to supplement the lectures, there will be numerous optional tutorials covering topics including dynamical systems, information theory, UNIX basics, and simulation using NEURON, MATLAB, and XPP. As each week progresses, the issues brought up in these presentations will be explored in laboratory demonstrations and exercises that invite the students to follow and generalize from the paths outlined in the lectures. Exercises involve both quantitative analysis of experimental data and exploration of models through analytic and numerical techniques. To reinforce the theme of collaboration between theory and experiment, exercises are often performed in teams that combine students with theoretical and experimental backgrounds.

The fourth week of the course is reserved for student projects. These projects provide the opportunity for students to work closely with the resident faculty, to develop ideas that grew out of the lectures and seminars, and to connect these ideas with problems from the students’ own research topics.

This course is appropriate for graduate students, postdocs and faculty in a variety of fields, from zoology, ethology, and neurobiology, to physics, engineering, and mathematics. Students are expected to have a strong background in one discipline, and to have made some effort to introduce themselves to a complementary discipline. The course is limited to 24 students, who will be chosen to balance the representation of theoretical and experimental backgrounds.

This course is partially supported by the National Institute of Mental Health, National Institute for Neurological Disorders and Stroke, Simons Collaboration on the Global Brain, the National Institute for Drug Abuse, NIH and the Organization for Computational Neurosciences.