Me thinks the problem is that we way too much engineering and far too little about the underlying biology. The reason is that we can't control the biology, so it seems better to ignore it. Bottom line: Our brain is not a computer. It has bottlenecks and serious limitations that can't be ignored.
All three of the factors I mentioned included heavy emphasis on psychoacoustics. I gave you no "engineering" answers but what causes us as humans to arrive at preferences. If by biology you ean something else, then I like to understand it. What I covered includes your ears and your brain. And countless experiences trying to correlate that with what we design and measure.
So let's consider some other studies on motor skills (and all skills, whether they be motor, language, listening, math follow the same rules) done several years ago at the Canadian Olympic training center on skill acquisition. Basically, there are two approaches to teaching a motor skill: "Top Down" and "Bottom Up" training. Top down entails teaching the entire movement and then letting the person cognitively figure out what to do; bottom up is teaching all the parts and then reassembling at the end. All the studies leading up to the seminal Canadian study had shown that bottom up led to faster skill acquisition than the top down approach. The Canadians though decided to extend the time course of the studies and lo and behold found that while bottom up resulted in faster skill aquisition, top down led to better skill retention. Same is true when doing this type of speaker testing in my mind. It's obvious that Floyd used a bottom up approach while I'd say most listeners use a top down approach. Draw you conclusion.
There is nothing obvious there or any scientific case for connecting one thing to the other. I sat through an audio test and found one speaker to sound better than another. You sit in an audio store and make the same decision. But somehow because I couldn't tell which speaker is which, I am a bottom up type of guy and you are top down???
Also we continue to just harp on the engineering and fail to pay any attention to the biological system being studied.
It is the exact opposite. It is people like Dr. Tool and Olive who start with biology: the listener and what could cause him to give us the wrong assessment. They test and tested again how humans behave in these situations and wound up isolating the right data. Others focus on this driver and that driver. This speaker technology or that speaker technology. You are a man of "but what does it sound like?" And here yo are critical of us starting from there?
Adaptation (read the early papers by Canadian researcher Hans Selye who ID'd adaptation though was wrong on several things such the non-specificity of stressors) as you mentioned is a basic principle of any biological system, whether it be a response to sound, heat, alcohol, etc. Our cells synthesize proteins in response to these stressors (stressors are after all what adaption is all about). We also fail to consider at all about how we learn and that short term memory is very unreliable and space limited. We don't even fully understand how we convert short to long term memories, not to mention those things that are unconsciously perceived.
I hear these arguments all the time. But as a person who has been in hundreds of such comparison tests with different time intervals, there is no question whatsoever what enables me to more accurately assess quality differences and as a result, what generates the least stress for me. And that is, the shortest time wins. It absolutely does. I have found differences that people could not find by shortening the comparison time to a fraction of a second. Then it was obvious what was different and everyone would then hear it.
If you have also participated in such tests with varying swap time and can demonstrate how your accuracy improves, then you have something. But just saying it doesn't get us anywhere.
Are tests perfect? No. But they are more prefect than the type of analysis you are talking about. I just went through a version of this with my team. A company had come in and left speakers for us to evaluate. My crew said they sounded really good. I listen and say they sound terrible. They don't believe it. So we bring out the Revel and put them side by side. Immediately they all agree they are a huge step down. They had all heard the Revel as that is the main product on the show floor. Yet they had forgotten its sound.
Then also there's basic issues with understanding the stress of testing and as I've talked about before, the inverted U hypothesis. In other words, there's a distinct relationship between perception and stress. So if you are a pianist, you want minimal stress and maximal perception; if you're blitzing a QB, you want to disturb that relationship by increasing the stress and reducing the QB's perceptual abilities!
When I sat through these tests, there was none of the usual stress you are talking about. Differences are big and so you are not worried about finding differences that don't exist or the other way around. You are being asked to vote. I do that all the time when I sit there evaluating speakers sighted. Here, it was much better because I could hear speakers in the same spot. Can't tell you how frustrating it is to try to simulate the same in our showroom. If this thing was cheap to build, I would build one yesterday.
But above all we ignore the basic evolution of the brain and its dual nature. We actually have two brains (often working at odds): the older brain consisting of the hind, mid and forebrain making up the limbic system and consisting basically of the hippocampus, hypothalmus and amygdala. Then there's the more recent on an evolutionary scale frontal cortex, the size of which separates us from other species. From an evolutionary standpoint, we've a lot more experience with this primitive brain and many of our automated responses come from this site. The frontal cortex, on an evolutionary scale, is far more recent and we're still trying as human to understand how to use this part of our brain. It seems to me that when we're listening, we're "talking" to these two different structures and thus the dualistic reactions we have to sound and music. I'd postulate that the "early/more automated" reactions stem from our primitive brain (and maybe things we needed for self preservation) and the more sophisticated, emotional reaction to music from our more modern brain.
This is all fine but if you can't show research on how it can be put to use and proven in audio, it means nothing. It really doesn't. In 30 years since Dr. Toole has talked about this, no one has put forward these types of research to counter it. The scientific standard whether we like it or not, is to eliminate bias and then evaluate.
I think we have to be honest here Myles. We really do. Honest answer says the findings don't match our preconceived notions. That is a fact. It is absolutely positively true. I don't think you have lived through your audio assumptions being challenged but I have been in so many that I no longer hang on to them as you do. I know as a human, my judgement of audio fidelity can be so wrong that makes you look like a fool
. I have sworn files to sound better only to do binary tests that show them to be identical. I don't go and say, "well, it was the stress of the test." It was not stress. I simply convinced myself one was better. Simple as that.
We don't have perfect tools for this business. But that is not cause for pretending science does not exist. Lots of good data does exist.
If I were you, next time you are in LA, I would ask Sean to let you take the test. Until you take it, you can't make a case for your cause. You are speaking hypothetically about something you have not experienced.
