I understand the purpose of giving pretests before a lecture, to gauge how much the students have learned. But if it's a lecture on a fairly obscure topic, it starts to seem a little more pointless...
When I was in residency, we had a lecture on SSEPs. SSEP stands for Somatosensory Evoked Potential, which is probably more than I could have told you about them prior to that lecture. Anyway, before the lecture, we had a "pretest" that asked us a bunch of multiple choice questions about SSEPs and how they're used in prognostic testing.
Me (whispering): "Have you ever heard of SSEP?"
Resident next to me: "No."
Me: "So how are we supposed to take a test about them?"
I looked at the first question and it was asking what sort of response in an SSEP is a poor prognostic sign in brain injury. I looked over at the resident next to me, who appeared to be filling in answers.
Me: "Uh... what are you basing those answers on then??"
Apparently I wasn't alone in my cluelessness, because when they graded the pretest, the residents actually scored WORSE than we would have based on chance alone.