This makes intuitive sense; the two are processes seem mutually reinforcing:

X prefers (vaguely) point of view A over B. X goes to university Y because it's known (to X, vaguely) to favor A over B. X's preference of A over B is sharpened because of being exposed to preference of A over B in his environment (via confirmation bias and the like).

In other words, given that somebody leaned to an option in a “neutral” environment, it is intuitive that when exposed to an environment that favored that option they will lean harder toward it.

[ETA:] it the post terms, this means that if the two views (A&B above) are equally “wrong” or “right” (i.e., an approximation of the truth, but with comparable accuracy), the effect doesn't make your opinion significantly more or less suspect. However, if one is closer to the truth, it would polarize “knowledge quality”: those with good initial “guesses” will be dragged closer to the truth, the others further away.

It seems to me that the latter situation is not as bad as (I suspect) it sounds: it should make it easier for outsiders (those not leaning towards either view yet) to “spot the difference”. That's because instead of a continuum of worse to better views would congeal in an opposition of two “bad” and “good” views (in respect to how accurately they approximate truth).