In the Neurotonics blog I wrote up a post about virtual body experiments, and while doing so had a thought on how one might develop a study about back pain using the strategy of virtual movement illusion.
In that the back, unlike other areas of the body,
a) can't be "seen" as easily in a mirror or any other way
b) does not occupy much "cortical real estate" kinesthetically
...it would definitely need kinesthetic input.
1. First the virtual body illusion would need to be firmly established. The researchers, Henrik Ehrsson in Sweden and Lenggenhager/Blanke in Switzerland have already figured out how to do that with their cameras and kinesthetic input (stroking the back with a pen, etc.). Apparently the brain finds this ridiculously easy to "believe".
2. It shouldn't require mirrors or camera projections of the "real" back, just a projected image as per Moseley with his paraplegics (see Matthias' blog post about that). Probably any "back" would do. Maybe fitted to the patient's size and shape and color of clothing. Just a film of a generic look-alike back might do, projected ahead of the patient's eyes.
3. Then the back in the movie (the "virtual" back that the patient's brain now feels it belongs to) would bend, lift its legs up alternately, rotate, extend, sidebend, whatever.. move easily.
4. The brain should be able to overcome its pain output to the physical back, because it can "see" the back it "thinks" it is embodied into (i.e., the back in the film), moving easily.
I think this might be a really really interesting study. Moseley, please consider doing something like this. Or Ehrsson. Or Lenggenhager/Blanke.