I think it would be better put to say that we integrate the inputs from all of our senses at a very, very deep, automatic level, no matter how "trained" someone claims to be.
There really is no question about the difficulties in sighted testing, it's been shown in example after example, in hearing, taste, sight (bearing in mind "sighted testing" means that you don't know the identity of the probe signal, not that you can't see) and so on that we will integrate all the information available to us.
I have to say that the fellow here who took the test has been quite the gentleman compared to most (if not all) subjectivists who have embarked on this route, and deserves praise for both being willing to try as well as being willing to admit that he was at least temporarily stymied.
There are some facts about blind testing that have to be relayed. In blind auditory testing, people with no hearing impairments being tested in quiet rooms can in fact hear very, very close to the level of atmospheric noise in the room, in a blind test. That, by itself, ought to show that blind testing is not flawed. What's more, in many blind tests where there is an audible effect, I've had subjects think they were guessing, when in fact they were not guessing, and were surprised to find consistancy in their answers.
Having said all that, training is essential. Oh, and did I mention, training is absolutely essential, and that's all there is to it? Yes, I guess I did mention that training is essential.
Having said that, I've been pointed by a very gleeful audiophile to a test that shows that there is a measurable difference between two power cords. This seemed odd, until I found out that one of the power cords was #18 AWG, and the other #12 AWG. Sure enough, use a cord that's too small (and I don't want to even speculate on the UL issues around that) and it won't work right. The small cord was "special silver-plated wire", but you know, copper is this ->||<- close to silver in conductivity, and that's not very important in 60Hz applications at this current level.
Finally, among experienced scientific listeners, at least some find that they are more accurate in blind testing, because they are aware of the sensory issues, and they tend to discount their sighted results, even when the differences are obvious.