The Mean-Field Analysis of a Heterogeneous Network of Quadratic Integrate-and-Fire (QIF) Neurons is a landmark mathematical framework in computational neuroscience that establishes an exact mathematical bridge between microscopic single-neuron activity and macroscopic brain dynamics.
Pioneered in a seminal 2015 study by Ernest Montbrió, Diego Pazó, and Alex Roxin (MPR), this approach solved a long-standing challenge in neuroscience: deriving a closed, low-dimensional set of exact “firing rate equations” for spiking neural networks without relying on heuristic approximations or artificial assumptions. 1. The Microscopic Foundation: The QIF Neuron
The Quadratic Integrate-and-Fire (QIF) model is the canonical mathematical representation for Class-I excitable neurons. For an all-to-all coupled network of neurons, the membrane potential Vjcap V sub j of individual neuron evolves according to:
τV̇j=Vj2+ηj+Jτr(t)+I(t)tau cap V dot sub j equals cap V sub j squared plus eta sub j plus cap J tau r open paren t close paren plus cap I open paren t close paren
Leave a Reply