It was first proposed by Kenneth Intriligator and Nathan Seiberg, in their 1996 paper "Mirror symmetry in three-dimensional gauge theories",[1] as a relation between pairs of 3-dimensional gauge theories, such that the Coulomb branch of the moduli space of one is the Higgs branch of the moduli space of the other.
It was demonstrated using D-brane cartoons by Amihay Hanany and Edward Witten 4 months later,[2] where they found that it is a consequence of S-duality in type IIB string theory.
In particular, on the Higgs branch, where the squarks are massless and condense yielding nontrivial vacuum expectation values (VEVs), the vortices are massive.
On the other hand, Intriligator and Seiberg interpret the Coulomb branch of the gauge theory, where the scalar in the vector multiplet has a VEV, as being the regime where massless vortices condense.
In this theory, the instantons are 't Hooft–Polyakov magnetic monopoles whose actions are proportional to the VEV of the scalar in the vector multiplet.