And even if the answer to "can you design an experiment which [will convincingly rule out a possibility someone has in mind]" is No, e.g. because the largest N we have available is 1 and you can't do stats on that, it may *still* be worth thinking about, because Alex may wish to get points for an honest attempt to look at the issue that may be in someone's mind face on. E.g. just saying "I think it's an allergy. I can't prove that it's not partly psychological, e.g. exacerbated by the natural anxiety I feel given that my lungs are so troublesome. Regardless, the effect is real and I'd appreciate it if you acted on the assumption that it's entirely physical" would get a much better reaction from some people than the bare assertion that it's physical and no engagement with any other possibility.