Dear Colin,
I am glad to see you entered the current contest, I also have an entry in preparation which I could not get to sooner because I was attending a conference in Sweden. During that conference, I heard a talk by Gregor Weihs in which he tried to do an experiment suggested by Asher Peres in his paper "Proposed Test for Complex versus Quaternion Quantum Theory" which purported to distinguish between a complex vs. a quaternion formulation. In this experiment there is a particular value (designated by the letter F) which under the complex formulation comes very close to 1 but under the quaternion formulation is closer to 0. When he did the experiment he did in fact find value for F much smaller than 1, but then realized that there were certain experimental limitations (which Peres had hinted at in his paper) which conspired to lower the value, and during the talk he concluded that, because he did not know of any way to overcome these limitations, the Peres test was not really an experimental test after all.
I briefly spoke to him after the talk and it seems that the problem is that to overcome this difficulty he needs to find out independently what value F should have under the quaternion formulation with those limitations factored in in order to be able to distinguish between the two possibilities.
Anyway, if the quaternion formulation suggested by Peres is similar to yours, then you may wish to correspond with Weihs, as he is obviously interested in doing a test (and in fact has already attempted it) and see whether, given your detailed knowledge in this area, you might be able to find a way to overcome the problem. I hope you can.
All the best,
Armin