There is a period during the learning of TK where you can't distinguish between wind and your energy. As TK is developed, you will notice the difference.
As someone who has ALWAYS wanted TK, I am intrigued. I'd like to first make a few points about the Challenge: you have until April 1 of this year to apply; after that, the prerequisites get harder. Make sure you read the rules.
I think that what's even more important than the $1,000,000 is that if you could demonstrate TK under properly controlled conditions, you'll blow open wide a whole new field of physics. I mean, WIDE. You-probably-won't-be-able-to-keep-track-of-the-grant-money wide. You'll-go-down-in-history wide. This is HUGE. One million bucks is /nuthin'/ compared to this.
But you've got to design a good, rock-solid test. And I do mean solid, because you're going to come up against a lot of outright disbelief.
Many of the other forum members have already touched on some absolutely excellent means to control your test. For example, setting up your test apparatus and just observing it. I'd actually recommend two things: 1) leave the room entirely and videotape what's going on, so as to control for the possibility that you might be subconsciously turning on your TK or 2) find someone else, whom you trust, who knows nothing about your TK, and ask them to sit and observe the paper, recording what it does (video camera still recommended). Tell them you're doing some kind of air-current test or something. That way, someone's actually breathing and affecting the air currents in the room (though I'm not sure someone who wasn't obviously trying to huff and puff would make a difference from 10 feet away. I don't know your room, tho).
Putting your test target under a glass or lucite box is another excellent idea. You mentioned earlier that putting the target under a glass blocks the TK energy, at least for novices, so I'm assuming you're past this point, yes? If you are strong enough, you should definitely do it. Make sure you put something else in the box, like the styrofoam peanuts, or tape thin long strips of paper to hang down from the top of the box so that you can demonstrate that whatever's going on in the box, it's /only/ affecting the target. Again, run the test /without you & your TK/ first, then apply your ability. Later on, you might want to further demonstrate control by having someone roll a die -- for example, a 10-sider with the results of:
1-5 -- make no change to what you're doing
6-10 -- stop the spin and make the paper rotate in the opposite direction
with the expectation that the expected result will occur within x seconds of your receiving the new instruction (where x is a number determined by you based on your ability). Those die numbers, btw, would need to be run past someone who actually knows something about randomness -- I'm not mathematically inclined and just pulled those numbers out of my kiester.
Anyway, see where I'm going with this?
As for the negative posts on this thread, don't worry about it. It's training for what you'll encounter when you publish a paper that puts science on its ear. If you need any help or suggestions for test protocols, definitely post here and I'll do my best to help (and, as you've seen, there are others who are genuinely interested in helping you design a good test). I am /totally/ rooting for you!