What happens inside a digital camera when switching from, say, size L to size M1?

Does the camera then use only a fraction of its "retina" or is there some algorithm for reducing the size to M1 (from L)?