I was going over the paramaters listed in the article and the plots presented, trying, unsuccessfully, to get an additive or subtractive variation greater than about 2db to show that there "Might" be a situation existing to explain why many of us "Hear" something for at least those first few tens of seconds.
What I heard was more of an unevenness in presentation, not a major objectionable output; a part of a db variation here and another part there. Perhaps like the entire speaker settling in: Screws, glues and all! The tweets weren't mentioned in the article, they might interract in some less than obvious way. There is "something there" that lots of people are hearing that's prompting them to go out on the line against a lot of powerfull folks and put themselves up for an embarrassing rebuttle.
I think it may be just as easy to talk yourself "out of" hearing something that may exist, when backed by the currently accepted data base, than it is to hear what "may exist" and admit it!
At our current level of data aquisition and human physiology, we may not yet be capable of presenting the entire story.
