I'll take moons as an example... and I'm going to reply as a high school teacher here, and I admit my reply isn't altogether satisfactory:
If a student's sophistication and outside knowledge gets in the way of assessing fundamental knowledge, they run the risk of losing credit for answers that they know.
If I ask a student how many moons Earth has, what I expect is the reply, "One." The student who replies "one" is guaranteed credit. No question about it. They're playing it safe.
The student who replies, "One; but Cruithne and other quasi-satellites and trojan moons have Earth centered orbits and could be considered moons," would probably get full credit (though I'd have some suspicion that the answer originated with the parent and not the student).
A student who answered "Two (or more)," or presented me with an essay analysing various hypothetical moons, or the works of Jules Verne, would risk losing credit. Not because I'm being punitive, but because, as I'm grading a stack of 160-220 tests, and I'd see the "Two" (or the lack of a definitive answer) before I'd notice the basis for the response.
This is true in virtually every subject, at virtually every level. I once had a paper rejected (written for publication) when I was graduate student on the basis that they could tell that what I was saying was important, but they couldn't tell whether there was a sound basis for my claims. I simply dumbed down the level of discourse and resubmitted it and it was accepted by the journal.
Last edited by moomin; 01/08/15 12:57 PM.