And experiments can be interpreted very self-centeredly. Audio people often seem to do one experiment and say, "See, I told you!" Or link to one experiment and say the same thing. A theory is rightly proven when many experiments, done by different people, confirm the results. Confirmation need not be exact, but should not be contradictory. Also biased experimenters are always suspect. I liked the experiment (referenced here several pages back) that seems to show balanced cables are detectably better than single ended cables. There seemed to be no axe to grind and the experimental subjects did not have foregone conclusions. But I'd like it better if three or four other experiments confirmed it.