Anecdotally, I think you're wrong; I spent three years dealing with PhD people in Economics. All of them used Microsoft Word; none used Latex. Many of them were absolutely brilliant at maths.

Many of them were doing table generation using STATA and some scripting in MS Office.

I don't think it's a stretch to believe that people could be brilliant in maths, but not so great with computers.

What generation were they? I'm thinking of recent PhDs and I know only one or two people working in technically difficult areas that don't use LaTeX. It's possible that Word's equation editor has improved a lot since I last used it, but typing up a proof is painful enough that it's going to force most people to LaTeX (or Scientific Word or Lyx, or some other frontend) pretty quickly, regardless of their computer literacy.

But I know that (for example) Hal White used a lot of Word and he's definitely brilliant at math. I appreciate your anecdote and will rethink my biases. (I do read these papers, though, and nothing's made me change my mind yet...)

They were of mixed generations; most in their 30s, and a few older than that.

If I had to make a wild guess, I'd say it had to do with the fact that most of them were from Asia. My gut feeling is that Latex has a bigger following in the "western" world than the "eastern" one. Again, wild conjecture on my part; especially since I have not yet spent any time in an Asian country myself.

This is probably true, though in China/computer science, we are seeing increased LaTeX usage. The problem is that LaTeX isn't that great of an experience for writing papers in Chinese, which doesn't have the kerning problems of western character sets anyways.