I don't eat bicycle tires. Why should I eat a little pink one?
So is it fair to say that the light and dark areas in a NMR image represent differences in the way the matter in that location responds to the "probe" magnetic field? (Exactly what differences are represented sounds like it could be very complicated.)
As far as actually constructing the image, one of your guests said that they're measuring a return voltage from the sample. It sounds like they're not using an array detector to pick up the entire image at once, but some kind of swept detector that picks up little bits of the image in rapid succession. (Maybe they're doing something Really Clever like exciting the sample with a series of Fourier basis functions and reading off the coefficients?)
Anyway, another great episode!