Auditory Memory and hifi component evaluation.

I think this is a bit of an "elephant in the room" of hifi - It seems common practice to compare components, cables etc. by playing a segment of music twice in succession with a swap in between (Linn's "tune dem" is a well known variant of this) but nobody seems to address the issue of whether it's even possible for the human ear / brain to discern such minor differences reliably after such a time interval? (has anyone ever tested a Linn dealer on whether they can REALLY tell which side of the mat or belt is which using Tune Dem?)

Imagine a large, flat red square is flashed up on a screen for 1 second, then vanishes to a blank screen..........30 seconds later, a second red square is flashed up and you are asked if it is brighter or darker than the first one was...my guess is the two shades would have to be at least 5-10% different in order to reliably and consistently identify the difference between them. Probably more...
I'd be interested to learn more about how auditory memory works but reading something like this seems to suggest that short term auditory memory is quite limited... http://en.wikipedia.org/wiki/Echoic_memory

Dip in (and then out) HERE