I think my above post gives those reasons. I have a large prior against this and don't think that studies that investigate popular learning styles are showing that there are no learning styles but rather that the learning styles educators currently use are not good.
I will be completely clear here: I have read none of the studies on the subject and what you are saying is a very valid criticism. I am not making a full argument here - I have no such argument. But I'm going to defy the data for now and wait until someone explains why people who hear numbers and people who see numbers best learn to count, add, etc. the same way.
Go Bayes! So if you just make your priors big enough, you never have to change your mind.
Post by fellow LW reader Razib Khan, who many here probably know from the gnxp site or perhaps from his debate with Eliezer.