Eh?
Underexposing most certainly does not increase saturation with negs.
What you are apparently talking about is pull processing, and that is a different topic involving subject brightness range and contrast index etc.
What I am talking about is a minor tweak in EI with respect to ISO, with no change in development.
"Slightly underexposing", as you originally put it, does indeed increase saturation with pos or neg films. Overexposing does the opposite. I am not talking about any changes in development...My comments about your quote were separate from my recommendation to pull.
First of all, why would overexposure make colors more saturated? That makes no sense at all. Making the whole pic slightly darker would increase the apparent saturation, not making it slightly lighter.
Second of all, why would there be any difference at all between neg or pos film? They both end up giving a darker pic if you give less exposure and a lighter one if you give more exposure. Pos versus neg is irrelevant.
Slightly overexposing a neg DOES give you a less saturated *neg*...but a lot of good that does you, since you will be making it into a print to view it and not just showing the film on a lightbox.
If you want to get technical, we are using the wrong terminology entirely in our discussion. Saturation is not affected by either of these, since saturation is purity of color; how much or now little a color is mixed with other colors. By definition, an orange is always more saturated than an orange-red, regardless of tonality. It has to do with the color response of the film, which is set by the designers of the film. . Very few photographers actually use the term this way, however. It is the intensity or richness of the colors we are talking about. It is not only in regards to photographic exposure, and is the case with slightly "darker" or slightly "lighter" than "realistic" interpretations in all imagery.