I have two theories: a) it's psychosomatic, or b) it's "Stop flipping channels, kid. I said stop! Back in my day, if you didn't like what was on, you had to walk--uphill!---to the telegraph station! And send a telegram---uphill!---to the radio-broadcaster! Nowadays you kids just push those buttons---there you go again! If you don't quit it, kid, why---uh---it'll make my shrapnel act up. Yes indeed. Excruciating. Now you go play and let me watch the bowl game."
Hehe. Well here's what happened:
We're all sitting in a room talking. TV is on and I'm sitting away from my uncle with one person in between us while he's lying on the couch. He can't see either my hands or the remote while I'm playing with the remote flipping it in my hand. I put the remote on my lap upside down probably pressing a button while at it. (I was not aware of this since I'm having a conversation) Few seconds later you could see my uncle turn red, goes suddenly quiet with painful expression on his face.
"Are you using the remote control? "
Me: "uhhh... no." (the channel isn't changing obviously)
"Stop it, you're killing me! Are you sure you're not pressing it?!"
I look down on my lap and see that the button is indeed pressed.
Me: "How did you know?"
"I hurts like hell when you do it"
Me: "Want a million dollars?"
As for the double blind test I was thinking along the same lines as Orangutan:
Get two identical remotes( I already have one universal remote), test it on him, see weather it hurts. If it doesn't, get the remote that does. (His TV and my TV remotes). Put both remotes in a position where he can't see them (behind him), preferably have him blindfolded, maybe even some sound insulation, earmuffs or something just to be on the safe side. Then, ask a third party to put batteries in one of the remotes and tape them both to a board. (so that you can't tell the weight difference). After that I'd press buttons of both remote controls and write down his reactions.(30 tries or so) After the test finishes I'd untape the remotes and analise the results. A hit would be him reporting pain while the button on the working remote was pressed. Everything else would be counted as a miss.
There are a few difficulties here tho. He claims that after he feels the pain from the remote it can take a long time for the pain to subside, half an hour or more. This could make the test last forever. Also, since his foot is after all, generally messed up, it's very possible he could report pain that wasn't remote control related. Especially if the test lasts for a long time - this would be almost inevitable IMO. If this is the case, I was thinking about an alternative procedure, perhaps with less tries but in a longer time-frame for a try. Let's say I do 10 tries, 5 on each remote, each lasting let's say.. 5 seconds, and each in a 10 minute time-frame. That would give him a 1 in 120 chance of guessing the exact time of one press or 1 in 600 for five presses (not sure about this one). If he can guess all five correctly I think that would be statistically significant.