Cliff, you're right that blind testing would be the only unbiased way of testing that, but it can't be done in the way that you described. The most important factor in blind testing is that the listener doesn't know which alternative is being heard at the moment; obviously just having him operate a switch can't do this and isn't a meaningful comparison. Having another person making the changes manually while you left, although it would take some time, is one way of doing this.
On the basic question of bi-wiring, this has been discussed here several times. The technical analysis of bi-wiring seems to clearly indicate that it's not significantly different electrically. In my view this doesn't raise enough of a question to call for personal testing of the point.
