A formalism for the long-distance magnetic field generated by populations of neurons

TL;DR

This paper develops a mean-field theoretical framework to understand how large neuronal populations generate magnetic fields detectable by MEG, linking microscopic ionic currents to macroscopic magnetic signals.

Contribution

It introduces a mean-field model based on current dipoles to connect neuronal activity with measurable magnetic fields at a distance.

Findings

Derived a theoretical model for magnetic field generation by neuronal populations

Evaluated contributions of synaptic and action potential currents

Applicable to long-distance magnetic field measurements like MEG

Abstract

Brain activity can be measured using magnetic fields located at some distance from the brain, a technique called magneto-encephalography (MEG). The origin of such magnetic fields are the ionic currents involved in neuronal activity. While these processes are well known at the microscopic scale, it is less clear how large populations of neurons generate magnetic fields. Here, we attempt to make such a link by deriving a mean-field theory of magnetic field generation by populations of neurons. We use the concept of current dipoles to build a theory that can be applied to populations of neurons. We also evaluate the contributions of neuronal current sources such as the synaptic currents or action potential currents. This theory should be useful to calculate the magnetic field located at long distances from neuronal populations, such as in MEG.