After a lot of thought, I might understand the idea. Keep in mind, it is a situation where you want to expand the contrast range. You want the light part of the negative (shadows) to stay light, but you want the dark part (highlights) to get darker. Usually you will increase the development time to do this, but as Shelly says you can do it with exposure too. In a low contrast setting, when you expose for the shadows, the highlights only get enough light to reach, say for example, Zone VI and you want Zone VII. A second short exposure will not add much to the shadows, the light part of the negative, because there is not much light there. But it will add to the highlights, the dark part of the negative, where there is more light. In the negative, the darks get darker and in the print, the lights get lighter.
I know I've read about this somewhere. A quick scan of two of Picker's book didn't reveal it, but if a closer look finds it, I will post the reference.