The comment about a movement pattern problem strikes my intuition as very plausible, but as usual I am compelled to seek out scientific literature to support the hypothesis.
I couldn't agree more. However, I am sorry to say, as you have no doubt worked out, the science behind rehab is not exactly mature. If physics and cosmology is at the top food chain and biology might be somewhere around the developing herbivores then rehab science is floating somewhere between the amoeba and bacteria.

When it gets complex - as it seems to be in your case, then we don't have a huge amount to fall back on. One thing about the methods I have suggested, however is they are highly unlikely to cause you harm, even if it is a bit of a gamble on your part. You could be poorer but other than that you may have nothing to loose and everything to gain.

You might like to have a look at the lit. on the Feldenkrais method:

PEDro - Search Results

If you can retrieve the systematic review you'll need to speak German or get someone to translate for you

The Alexander Method has one very good trial for chronic low back pain:

Randomised controlled trial of Alexander technique lessons, exercise, and massage (ATEAM) for chronic and recurrent back pain -- Little et al. 337: a884 -- BMJ

And here is one for Alexander method and Wikipedia reference-linkParkinson's Disease which is more my area of expertise:

PEDro - Detailed Search Results

I know I am throwing you evidence for conditions that are completely unrelated to yours. However given the paucity of research and also the unusualness of your presentation at least you may see that in a wide variety of chronic conditions that don't just get better by themselves and where it is plausable that there the nervous system's motor control has changed in response to impairments perhaps there are something in these methods.

At a more basic science level we do have more evidence for maladaptive changes motor control in a vast array musculoskeletal and neurological conditions. If you are interested I could try and hunt up a review article - but that might take a bit of time and effort.