Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: ABX testing vs. perceptual states (Read 23304 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

ABX testing vs. perceptual states

I'm making this a separate post although it's actually an extension of the previous one because of the possibility that it might get cut for TOS 8, however I think that what I have to say here is worthy of consideration.

One of the things I find frustrating about this site is that I can not freely discuss my possible objections to ABX testing, which revolve around my belief that the very situation of ABX forces a particular mode of mental functioning that precludes certain forms of perception. Please allow me to explain.

In the recent issues of Scientific American there have been reports on the use of MRI imaging to analyze brain function during perceptual processing. The specific processing involved in these particular tests happens to be visual, but baced on the findings I believe that we will likely find analogous phenomena in audio processing as well. In a nutshell, what has happened is that MRI has revealed that perceptual processing is much more involved than had previously been believed - instead of certain specific areas of the brain being mapped directly to specific senses processing is distributed across the brain quite widely and occurs in ways that had not previously been suspected. I'm not going to go into detail here - there's no room and it wouldn't be appropriate - go read it for yourselves. But the gist is that perceptual processing has a lot more going on than previously believed and the implications are pretty staggering. This is one part.

A second part is something that I have suspected for awhile due to my experience with perceptual states during mixing (yes, I know this sounds all new-agey hippie crapola-ish, but bear with me); recently Bob Ohlsson, a very widely respected and well known mixing and mastering engineer related that he goes into an altered state of consciousness when he is deep into working on a project. Now what is interesting about Bob's statement is that he had a girlfriend who was involved in brainwave research and just out of curiousity she hooked him up to her EEG machine when he was working - and found that he was radically altering his brain wave activity when he would enter this trancelike mixing state. This is actual scientific measurement, not audiophoole new age nonsense. Go ask Bob about it - his original comments on the subject are at The Womb.

I believe that this is an area of perceptual research that needs proper rigorous scientific investigation, and could be one reason that trained and experienced sound engineers have perceptual experiences while working on musical projects that can not be corroborated via ABX, because the circumstances of the ABX test impede the ability to "switch mental gears" into the other perceptual state. In other words, it requires a certain relaxation of the mind that does not occur when one is deliberately listening in a way that one does when taking a test. (Maybe somebody else could put it more clearly, I know what I mean but I'm not entirely sure how to express it to others.....)

Thanks for bearing with me. I really hope this doesn't get chopped without consideration, as X-Filesish as it might seem at first glance.

ABX testing vs. perceptual states

Reply #1
I'm making this a separate post although it's actually an extension of the previous one because of the possibility that it might get cut for TOS 8, however I think that what I have to say here is worthy of consideration.

One of the things I find frustrating about this site is that I can not freely discuss my possible objections to ABX testing, which revolve around my belief that the very situation of ABX forces a particular mode of mental functioning that precludes certain forms of perception. Please allow me to explain.


<snip a paraphrase of the same tired old story that golden-eared audiophiles have been telling us for decades >

The common thinking among golden-eared audiophiles and golden-eared production people is that they only want to be proven right, not serioiusly questioned. Anything that goes against their religiouis beliefs  must in their view  have some fatal flaw, whether real or imagined.

It doesn't matter whether the totem is a $150,000 analog tape machine or a $150,000 SET power amp - totems must be defended at any cost in reason or reliable evidence.


ABX testing vs. perceptual states

Reply #2
I believe that this is an area of perceptual research that needs proper rigorous scientific investigation, and could be one reason that trained and experienced sound engineers have perceptual experiences while working on musical projects that can not be corroborated via ABX, because the circumstances of the ABX test impede the ability to "switch mental gears" into the other perceptual state. In other words, it requires a certain relaxation of the mind that does not occur when one is deliberately listening in a way that one does when taking a test. (Maybe somebody else could put it more clearly, I know what I mean but I'm not entirely sure how to express it to others.....)

Thanks for bearing with me. I really hope this doesn't get chopped without consideration, as X-Filesish as it might seem at first glance.

I really don't find it odd that someone would change states of brain activity during an activity that involves creation.
What I'm not willing to assume, however, is that a 'different state,' whether it be peyote induced or not, actually (or potentially) improves someone's hearing.

Furthermore, I wouldn't at all be surprised if certain people are simply bad testers.
I mean, when absorbing coursework and research material for my studies, I am in a completely different mindset while reading and note taking than I am when recalling the studied material during an exam.

However, if I don't earn a great grade, I certainly don't go and complain to the professor that I would have done much better had I been actually able to use my books and notes during the test! 
elevatorladylevitateme

ABX testing vs. perceptual states

Reply #3
I'm making this a separate post although it's actually an extension of the previous one because of the possibility that it might get cut for TOS 8, however I think that what I have to say here is worthy of consideration.

One of the things I find frustrating about this site is that I can not freely discuss my possible objections to ABX testing, which revolve around my belief that the very situation of ABX forces a particular mode of mental functioning that precludes certain forms of perception. Please allow me to explain.


<snip a paraphrase of the same tired old story that golden-eared audiophiles have been telling us for decades >

The common thinking among golden-eared audiophiles and golden-eared production people is that they only want to be proven right, not serioiusly questioned. Anything that goes against their religiouis beliefs  must in their view  have some fatal flaw, whether real or imagined.

It doesn't matter whether the totem is a $150,000 analog tape machine or a $150,000 SET power amp - totems must be defended at any cost in reason or reliable evidence.


Well, that's exactly what I expected you to say. You have absolutely ZERO interest in anything that might possibly disagree with your medieval Catholic world. And I don't mean that as an insult, it's a historical reference.

You have no interest in the fact that EKG testing showed Bob Ohlsson entering a different mental state when he mixes. Yes, this is so far uncorroborated, but don't you think that maybe somebody oughtta check it out?

You have no interest in the fact that Scientific American magazine (I assume you know who they are?) publishes information on MRI brain scans of activity related to perception that appear to open a completely new book on how the brain processes perceptual information. You reject it out of hand without even reading it.

You remind me of the old joke "I know what's true, please don't bother me with FACTS!"

So if EKG data isn't valid and MRI data isn't valid, would you mind telling me what is?

Seriously. I'd like to know.


ABX testing vs. perceptual states

Reply #4
I'm making this a separate post although it's actually an extension of the previous one because of the possibility that it might get cut for TOS 8, however I think that what I have to say here is worthy of consideration.

One of the things I find frustrating about this site is that I can not freely discuss my possible objections to ABX testing, which revolve around my belief that the very situation of ABX forces a particular mode of mental functioning that precludes certain forms of perception. Please allow me to explain.

False. You can discuss the merits of ABX. It's done many times before. The only thing you can't do is make a claim of "I hear a difference between A and B" without supporting it with an ABX test. Did you really read the TOS?

Quote
I believe that this is an area of perceptual research that needs proper rigorous scientific investigation, and could be one reason that trained and experienced sound engineers have perceptual experiences while working on musical projects that can not be corroborated via ABX, because the circumstances of the ABX test impede the ability to "switch mental gears" into the other perceptual state. In other words, it requires a certain relaxation of the mind that does not occur when one is deliberately listening in a way that one does when taking a test. (Maybe somebody else could put it more clearly, I know what I mean but I'm not entirely sure how to express it to others.....)

This seems to be akin to saying "I hear better when I pay less attention" which doesn't add up, because from experience it shows that artifacts are usually heard better when people are actually focusing and not the other way around. Quite often it is reported that artifacts found during an ABX test are not a problem with normal listening.

Sure there might be differences in MRI, but you are going way to far in to interpreting them to your liking. The reverse of you statement could be just as well be true, and that would at least be in accordance with real world experience.

The idea that an ABX test impedes on the ability to find differences is highly speculative and not supported by anything you posted so far.
"We cannot win against obsession. They care, we don't. They win."

ABX testing vs. perceptual states

Reply #5
I'm making this a separate post although it's actually an extension of the previous one because of the possibility that it might get cut for TOS 8, however I think that what I have to say here is worthy of consideration.

One of the things I find frustrating about this site is that I can not freely discuss my possible objections to ABX testing, which revolve around my belief that the very situation of ABX forces a particular mode of mental functioning that precludes certain forms of perception. Please allow me to explain.

False. You can discuss the merits of ABX. It's done many times before. The only thing you can't do is make a claim of "I hear a difference between A and B" without supporting it with an ABX test. Did you really read the TOS?

Quote
I believe that this is an area of perceptual research that needs proper rigorous scientific investigation, and could be one reason that trained and experienced sound engineers have perceptual experiences while working on musical projects that can not be corroborated via ABX, because the circumstances of the ABX test impede the ability to "switch mental gears" into the other perceptual state. In other words, it requires a certain relaxation of the mind that does not occur when one is deliberately listening in a way that one does when taking a test. (Maybe somebody else could put it more clearly, I know what I mean but I'm not entirely sure how to express it to others.....)

This seems to be akin to saying "I hear better when I pay less attention" which doesn't add up, because from experience it shows that artifacts are usually heard better when people are actually focusing and not the other way around. Quite often it is reported that artifacts found during an ABX test are not a problem with normal listening.

Sure there might be differences in MRI, but you are going way to far in to interpreting them to your liking. The reverse of you statement could be just as well be true, and that would at least be in accordance with real world experience.

I'm not interpreting the MRI results any way. The particular tests in question used visual stimuli, not auditory, but appear to have ramifications over the general way the brain handles perceptual data. Essentially, what it means is that the way the brain handles perception is a whole lot more complicated than anybody had guessed.

As to your "I hear better when I pay less attention", no that isn't at all what it means. It's not a matter of "paying less attention" at all, it's a matter of "switching to a different mode of processing", kind of like a zen trance if you know anything about different forms of meditation. It's a matter of clearing the extraneous crap of of the way so you can focus your attention, for lack of a better way to put it.

When I mix I've sometimes experienced going into a trancelike state where i close my eyes and "become one with the music", my hands on the faders reacting directly to the perception of the sound - which BTW, is one of the reasons I can't mix on a typical digital board with a programmable control surface - since things aren't always in the same place I can't let go of conscious thought. Now I'm not offering that as evidence of anything in this discussion, other than that I believe I've experienced the state that Bob's friend measured him going into with the EEG machine.

ABX testing vs. perceptual states

Reply #6
CORRECTION ALERT:

I just noticed that I made a typo and wrote "EKG" when I intended to write "EEG" when discussing Bob Ohlsson's experiences. Please note this correction.

Hopefully that will make more sense.

ABX testing vs. perceptual states

Reply #7
Now I'm not offering that as evidence of anything in this discussion, other than that I believe I've experienced the state that Bob's friend measured him going into with the EEG machine.

So how are you actually going to pass this off as a critique on ABX then? You first have to demonstrate that this state is better for finding differences, and then that such a state cannot be achieved during ABX.

You did actually start this with and I quote

Quote
One of the things I find frustrating about this site is that I can not freely discuss my possible objections to ABX testing, which revolve around my belief that the very situation of ABX forces a particular mode of mental functioning that precludes certain forms of perception. Please allow me to explain.

So far we haven't gone any further than that you believe something and apparently you have spend a lot of words on things that by your own admission do not support these beliefs in any kind of way.

And you have the guts to accuse me of being a religious fanatic? Thanks but no thanks.

I forgot to post this the last time: please no more name dropping of "JJ" or whoever. If you can't defend your arguments by yourself don't participate in the discussion and if "JJ" wants to say something I'm sure he is more than qualified to speak for himself. That you seem to worship him doesn't make your (or his) statements anymore true.
"We cannot win against obsession. They care, we don't. They win."

ABX testing vs. perceptual states

Reply #8
As to your "I hear better when I pay less attention", no that isn't at all what it means. It's not a matter of "paying less attention" at all, it's a matter of "switching to a different mode of processing", kind of like a zen trance if you know anything about different forms of meditation. It's a matter of clearing the extraneous crap of of the way so you can focus your attention, for lack of a better way to put it.

Well in that case ABX testing is likely to induce such a zen like state, since you're focusing all your attention on "does A sound different from B" and there is no room for "extraneous crap".

C.
PC = TAK + LossyWAV  ::  Portable = Opus (130)

ABX testing vs. perceptual states

Reply #9
I'm not interpreting the MRI results any way. The particular tests in question used visual stimuli, not auditory, but appear to have ramifications over the general way the brain handles perceptual data. Essentially, what it means is that the way the brain handles perception is a whole lot more complicated than anybody had guessed.

While this may be true, you've yet to speculate on how said possible complications impact the validity of ABX tests. 
Please, get on with it, state your theory and open yourself for reasoned debate.  So long as you refuse to post a theory there is nothing to discuss.

Zzzzzzzz
Creature of habit.

ABX testing vs. perceptual states

Reply #10
One of the things I find frustrating about this site is that I can not freely discuss my possible objections to ABX testing, which revolve around my belief that the very situation of ABX forces a particular mode of mental functioning that precludes certain forms of perception. Please allow me to explain.


No one disagrees that concentrating on one thing means losing concentration on other things (i.e. it temporarily "precludes certain forms of perception").  You can walk and chew gum at the same time, but there's a limit, and intense concentration is still more demanding and limited.  (I'm impressed with the people who do speed chess against 10 competitors simultaneously, but it's not like they're simultaneously doing speed chess and speed music composition etc.).  Consciousness and perception are poorly understood and will never be fully understood.

However, what is the alternative to abx testing for what we want to measure by it: whether the individual perceives a difference in two competing samples of music from an original?  I see none.  How would you establish that this "other mode of mental functioning" has perceived a difference?  I don't say it's impossible, but you have to come up with something to distinguish it from placebo and guessing etc. 

If one thing you want to measure is a background or less attentive mental state, I suppose you could have a double blinded test where someone listens to one sample for three months casually in the background encoded one way, and three months encoded another way, and is asked to say which sounds better, a or b.  then again for the next six months.  and again until you have many samples.  The problem is that you would have a host of other variables there you're not controlling for, but if repeated enough times and the result was consistent, then it could be trustworthy or at least suggestive.  It wouldn't be measuring transparency (maybe there could be such a test but it would be more complicated), but it would be measuring some perceived difference.  (One problem here is the switch from intense short-term to diffuse long-term memory.)

Come up with a suggestion for a test (you don't need to have a worked out theory, but you do need a test!).  Otherwise you start to look like those psychics who always feel too much "negative psychic energy" to perform when an uncheatable test is conducted. 

p.s.  as far as eeg/mri testing, you'd have to show a lot more: it's not clear what that is measuring at the snapshot time it is taken, whether the results are unambiguous in their interpretation, and what the person perceives or experiences that corresponds to that eeg etc.  I mean even a blood pressure test can correspond to something in our subjective perception/experience, but it would not be a great measurement of what if anything I perceive in listening to music--it could be if I was listening to a police siren not knowing a test was underway and if it was taken at just the right time etc., but you see the thousand complications there.  And an eeg is much harder to interpret than blood pressure, maybe it starts to be like a lie detector test, I don't know.  Obviously we're very very very far from anything like "mind reading" on the basis of an mri/eeg, if that's even  possible.

ABX testing vs. perceptual states

Reply #11
I'm making this a separate post although it's actually an extension of the previous one because of the possibility that it might get cut for TOS 8, however I think that what I have to say here is worthy of consideration.

One of the things I find frustrating about this site is that I can not freely discuss my possible objections to ABX testing, which revolve around my belief that the very situation of ABX forces a particular mode of mental functioning that precludes certain forms of perception. Please allow me to explain.


<snip a paraphrase of the same tired old story that golden-eared audiophiles have been telling us for decades >

The common thinking among golden-eared audiophiles and golden-eared production people is that they only want to be proven right, not serioiusly questioned. Anything that goes against their religiouis beliefs  must in their view  have some fatal flaw, whether real or imagined.

It doesn't matter whether the totem is a $150,000 analog tape machine or a $150,000 SET power amp - totems must be defended at any cost in reason or reliable evidence.


Well, that's exactly what I expected you to say.


At least I shouldn't expect to be excoriated for being disingenuous! ;-)

Quote
You have absolutely ZERO interest in anything that might possibly disagree with your medieval Catholic world.


Sorry John, but you were the first to be targeted with that, and simply turning back at me won't work.  Besides, the shoe fit - ABX was a reaction to the religous fervor of the high priests of high end audio.


Quote
You have no interest in the fact that EKG testing showed Bob Ohlsson entering a different mental state when he mixes. Yes, this is so far uncorroborated, but don't you think that maybe somebody oughtta check it out?


< It's blustery where you are today? ;-) >

I would seriously hope that anybody and everubody change mental state from a resting state when they start to mix.

It seems to me that going from that  factoid to a believable criticism of ABX involves a Grand Canyon-sized leap. Grand Canyon or English Channel, depending on which side of the pond you are...  I need a dooted line to follow.

What you need to show is that switching from mixing to listening for the purpose of making a decision about an ABX test trial changes one's mental state. I don't know about EKG being a reliable indicator. Perhaps a NMR scan might give us some useful information.

I'm probably among the few people who posts here who routinely does both mxiing and ABX testing. From at least 10 years of experience doing both,  I'd say that as far as mental states go; it is about six of one, half dozen of the other. Critical listening is critical listening, at least for me. YMMV.

ABX testing vs. perceptual states

Reply #12
You have no interest in the fact that Scientific American magazine (I assume you know who they are?) publishes information on MRI brain scans of activity related to perception that appear to open a completely new book on how the brain processes perceptual information. You reject it out of hand without even reading it.

You remind me of the old joke "I know what's true, please don't bother me with FACTS!"

So if EKG data isn't valid and MRI data isn't valid, would you mind telling me what is?

Seriously. I'd like to know.

You are completely missing the point.

The point is that *you* have to make the connection between "there is something on a scan" and "ABX is invalid". If you are not willing to do this there is not much to talk about, now is there.

Stop the ad hominems, cause you only look silly and a religious fanatic yourself.
"We cannot win against obsession. They care, we don't. They win."

ABX testing vs. perceptual states

Reply #13
So, despite being a highly unusual occurrence for the average subject, scientifically conducted MRI scanning is a  noncontaminating and valid way to gather information about perception...but ABX comparison is so weird and different compared to normal audiophile listening, that it's inherently flawed.

Right.

ABX testing vs. perceptual states

Reply #14
Sorry if most, or all of the following has been pointed out ad infinitum on HA already (which I rather suspect it has).

I'm seeing a lot of these arguments put forth on HA, and elsewhere, by people that their (obviously very fragile) listening skills are somehow disrupted by the mere fact of being in a test.
That is what they are saying, isn't it? That somehow the test itself is bringing in some kind of stress which puts them "off their game"?

If so, I'm sorry, but I don't buy it. I mean, sure okay, when you know you are in a test, presumably some people do feel some kind of pressure to "perform", even if it's just from yourself. But, how much is it really? Surely there must be some test situations where the pressure is very very low. A test done among friends and/or family for example. Or one you perform on yourself. And even in a situation where said stress is actually quite considerable, does it actually affect your ability to hear differences in sound sources?

Compare this supposed stress from ABXing to the mixing situation introduced by the OP. Apparently they are in a state of absolute calm during their mixing sessions, so they can enter whatever claimed mode is most favourable to appreciating sound in it's most minute detail.
But wait a minute, aren't they mixing *for* some purpose? Isn't someone, somewhere, going to hear this mix one day? And possibly critique it? Possibly even a paying customer waiting somewhere?
Are they so adept at leaving everything behind, and just focusing on their mixing that they are unaffected by any pressure to perform in this situation?
So why is this amazing ability to focus not applicable to both situations?

Another point that occurs to me - there are many people in the world who say that pressure helps them to perform better. So, even if the "stress" hypothesis is true for some people, I'd imagine they are in a minority, and that for some people it's simply no issue, and for yet others it is actually a positive.

One other, again probably redundant, point to all the "ABX deniers" : If your hallowed piece of equipment introduces such an unimaginably small difference in sound to your system that even you, it's main cheerleader, can't pick it in a listening test, due to some hazily defined "psychological" block you imagine you are experiencing, then maybe it's such a small difference that it's genuinely not worth concerning yourself with. In other words, although the difference may *actually* exist, for practical purposes it *doesn't* exist (ie: you can't hear it).

ABX testing vs. perceptual states

Reply #15
If so, I'm sorry, but I don't buy it. I mean, sure okay, when you know you are in a test, presumably some people do feel some kind of pressure to "perform", even if it's just from yourself. But, how much is it really? Surely there must be some test situations where the pressure is very very low. A test done among friends and/or family for example. Or one you perform on yourself. And even in a situation where said stress is actually quite considerable, does it actually affect your ability to hear differences in sound sources?


On top of which, why would extra pressure to perform not increase performance in some people? Comments that people hear differences in ABX that are inaudible in normal listening seem rare...

When I was still a student, I definitely performed better in exams than out of them. I still find that I can think and react to questions much faster and better when I am giving talks to a large audience than when I am just sitting in my armchair and thinking.

So, unless some physiological basis for loss of hearing ability with stress is suggested, I don't buy this. But what do I know.


ABX testing vs. perceptual states

Reply #16
But wait a minute, aren't they mixing *for* some purpose?


Exactly

This business of audiophiles claiming that they can hear differences when they are listening for enjoyment, but can't hear differences when listening for ABX has been around for many years if not decades. While I dispute it on many logical and practical grounds, I can kind of get it because audiophiles kinda sorta have a license to be flakey. They are amateurs for heaven's sake. They are consumers. They are people whose purpose in audio is to be catered to.

Seeing the same song-and-dance being claimed by professional production people is really quite a shock for me. The very nature of being a professional is being able to perform on cue, get the job done on demand. I can just see someone like Mixerman telling a client that he can't mix today because umm, his aura is off. Sure everybody has off days. But you can't be sick too many days before your reputation is down the drain.

The essence of this "I'm great except when there's an ABX Comparator in the signal path" stuff is really IMO and IME all about performance anxiety. Performance anxiety is for amateurs. Watching self-proclaimed professionals hiding behind this sort of thing is just one more thing that makes me question these people's purported non-credentials.  That, and the repeated attempts to bully us out of our belief in the Fourier Transform.

ABX testing vs. perceptual states

Reply #17
Of course when someone invokes an argument from authority, proper credentials are totally fair to bring up by others.

I also don't think Scientific American is such a bad publication, last time I read it, it was still written by the scientists themselves, and certainly has been less tabloid-like than New Scientist, for instance. These aren't journals, but they do a fairly good job of explaining the science to a lay audience.

But John, would you like to at least link the article mentioned? You are jumping to a LOT of conclusions, and not even formulating a clear, testable (and falsifiable) hypothesis. You are just attempting to "poison the well" from what I can see.

ABX testing vs. perceptual states

Reply #18
I'm not interpreting the MRI results any way. The particular tests in question used visual stimuli, not auditory, but appear to have ramifications over the general way the brain handles perceptual data. Essentially, what it means is that the way the brain handles perception is a whole lot more complicated than anybody had guessed.

While this may be true, you've yet to speculate on how said possible complications impact the validity of ABX tests. 
Please, get on with it, state your theory and open yourself for reasoned debate.  So long as you refuse to post a theory there is nothing to discuss.

Zzzzzzzz


I don't have a proper theory - it's not my job. What I'm doping is pointing out that there's something here that appears to not be adequately explained by current conventional theory that bears investigation and suggesting some avenues of research to those who might be qualified to pursue them.

If I was an "audiophoole" you can be assured that I'd have lots of cockamamie theories that I'd be absolutely certain are true without any proper research. But I'm not and I don't.

What I do have is observations - both my own and of professional listeners - people who make their living with their ears, as well as my reading and other informal studies of the various sciences that are associated with perception research. I'm pretty good at taking data from diverse fields and drawing inferences that might escape the notice of specialists in any one field, and that's what I'm doing here.

ABX testing vs. perceptual states

Reply #19
No, it is your job to state in concrete terms what the relevance of your introduced evidence is.

So what, it's already been conceded by me (and others) that it is no shock that perception is more complex than X.  I say "than X" because you're far from clear on what the baseline "current conventional theory" is which you believe these findings challenge.

Not only that - you bring this evidence up in the context of
Quote
One of the things I find frustrating about this site is that I can not freely discuss my possible objections to ABX testing, which revolve around my belief that the very situation of ABX forces a particular mode of mental functioning that precludes certain forms of perception.


Ok, this is about how the validity of ABX is somehow (undefined!) dependent on our "current conventional theory" of perception and that undefined term is somehow (undefined!) challenged by the introduced evidence.

DO you have anything?  Or is this the logical fallacy it smells of?
Creature of habit.

ABX testing vs. perceptual states

Reply #20
One of the things I find frustrating about this site is that I can not freely discuss my possible objections to ABX testing, which revolve around my belief that the very situation of ABX forces a particular mode of mental functioning that precludes certain forms of perception. Please allow me to explain.


No one disagrees that concentrating on one thing means losing concentration on other things (i.e. it temporarily "precludes certain forms of perception").  You can walk and chew gum at the same time, but there's a limit, and intense concentration is still more demanding and limited.  (I'm impressed with the people who do speed chess against 10 competitors simultaneously, but it's not like they're simultaneously doing speed chess and speed music composition etc.).  Consciousness and perception are poorly understood and will never be fully understood.

However, what is the alternative to abx testing for what we want to measure by it: whether the individual perceives a difference in two competing samples of music from an original?  I see none.  How would you establish that this "other mode of mental functioning" has perceived a difference?  I don't say it's impossible, but you have to come up with something to distinguish it from placebo and guessing etc. 

If one thing you want to measure is a background or less attentive mental state, I suppose you could have a double blinded test where someone listens to one sample for three months casually in the background encoded one way, and three months encoded another way, and is asked to say which sounds better, a or b.  then again for the next six months.  and again until you have many samples.  The problem is that you would have a host of other variables there you're not controlling for, but if repeated enough times and the result was consistent, then it could be trustworthy or at least suggestive.  It wouldn't be measuring transparency (maybe there could be such a test but it would be more complicated), but it would be measuring some perceived difference.  (One problem here is the switch from intense short-term to diffuse long-term memory.)

Come up with a suggestion for a test (you don't need to have a worked out theory, but you do need a test!).  Otherwise you start to look like those psychics who always feel too much "negative psychic energy" to perform when an uncheatable test is conducted. 

p.s.  as far as eeg/mri testing, you'd have to show a lot more: it's not clear what that is measuring at the snapshot time it is taken, whether the results are unambiguous in their interpretation, and what the person perceives or experiences that corresponds to that eeg etc.  I mean even a blood pressure test can correspond to something in our subjective perception/experience, but it would not be a great measurement of what if anything I perceive in listening to music--it could be if I was listening to a police siren not knowing a test was underway and if it was taken at just the right time etc., but you see the thousand complications there.  And an eeg is much harder to interpret than blood pressure, maybe it starts to be like a lie detector test, I don't know.  Obviously we're very very very far from anything like "mind reading" on the basis of an mri/eeg, if that's even  possible.


It's not that ABX testing, used correctly, is invalid. The key phrase is "used correctly". Some people appear to believe that ABX is some kind of panacea - that if it can't be proved by ABX it doesn't exist and that is simply not true. The fact is that ABX can't prove that anthing DOESN'T exist - it can only prove that something DOES exist. ABX can demonstrate that a difference exists between A and B, and can determine that a statistical majority of people have a preference for one or the other - or that people hear a difference but there is no statistical preference.

What it can not do is prove that there is no difference.

Some people who don't understand this attempt to use a "sample population" and draw inferences when the sample is unable to discern a difference. THIS IS NOT A VALID APPLICATION OF ABX.

The fact is that if only ONE person in the world is capable of hearing a difference between A and B, reliably and over a series of repeated tests, then the difference is demonstrated to exist. Unless your ABX sample subject group is infinitely large - an impossibility - any ABX test that does not show a difference between A and B has yielded an INCONCLUSIVE RESULT. This fact may be uncomfortable to some, who would like ABX to be the perfect testing tool, but it is a fact, none the less. So when somebody like Mixerman, or Weedy, or Slipperman, or dwoz, or Bob Ohlsson says that they can hear a difference, all you can do is test them. If the test is properly constructed - i.e. if the conditions adequately duplicate the professional listening experience, there should be a correlation. But if there isn't it's STILL NOT PROOF, because you can't prove a negative. The really big problem here is that for the conditions to be valid you'd pretty much need to have the subject actually mix a project on device A and on device B under ABX conditions, which is a pretty steep task. I don't know how you'd go about it. The urge in cases like this is to simplify the test conditions, but doing so invalidates the test.

So what we need is an improved test methodology, or to simply admit that there are some phenomena that can not be conclusively tested.

Some have proposed null testing as a means of demonstrating that two devices have identical performance. The problem again is setting up the test. In many null tests I've seen that appear to prove differences, people always seem to raise objections about some parameter introducing bias which invalidates the test.

The word "transparency" seems to get bandied about a lot here. Some people are attempting to derive a "scientific" definition of transparency.  I suggest that this is an invalid quest. Transparency is a SUBJECTIVE phenomenon and is not capable of being objectified. Furthermore, it is a "property" of systems, or subsystems composed of interacting parts which themselves may or may not be "transparent". This can be demonstrated easily if one considers the case of microphones.Microphones do not detect sound the same way that the human auditory system does. Therefore any microphone based recording CAN NOT be indistinguishable from the original performance; therefore it can not be transparent, if by "transparent" we mean exactly like the source as Ethan and others would have it. When the finished work is composed of many such individual recording, all individually processed, then mixed together and processed again, the very concept of "transparency" loses all meaning. At that point we can only discuss which equipment and methodology produces the most optimally euphonious result, which is an entirely different kettle of fish.

ABX testing vs. perceptual states

Reply #21
Quote from: John Eppstein link=msg=0 date=
I don't have a proper theory - it's not my job.

Don't need a theory--but need a test, or some persuasive way of showing that you've perceived a difference in another way. 
Note that abx testing doesn't assume much of a theory either.  (Cheating is very narrowly defined--such as doing a binary/frequency analysis of the music samples, or someone else cheating for you (failure of the double-blind) etc.)  So it could be an alarm goes off in your head when listening to one of the music samples, who knows what leads you to identify it--it doesn't have to be done through intense concentration on artifacts, if you have some other powers inside you (besides the cheating power).   

Quote from: John Eppstein link=msg=0 date=
What it can not do is prove that there is no difference.

Sure but who denies that?  It's impossible to prove the non-existence of anything, unless that thing is self-contradictory.  The question is whether you have another means of proving that there is a difference.

Quote from: John Eppstein link=msg=0 date=
The fact is that if only ONE person in the world is capable of hearing a difference between A and B, reliably and over a series of repeated tests, then the difference is demonstrated to exist.

Again a straw man--who denies this?  Even if one DOG hears the difference, the difference exists.  But the abx test is mainly to establish what YOU can perceive (rather than what you THINK you perceive but do not; due to bias, it's not always clear what one truly has perceived, e.g. the problem of false eyewitness testimony).  Secondarily also to establish what other people can truly perceive, in a way that persuades others that they have perceived a difference based on listening (rather than guessed or cheated etc.).  But there may be some genetic freak living in a remote desert who can hear doglike frequencies or has unusual discernment--no one rules that out.  I'm sure if he had an internet connection, he'd be welcome to join here and abx.  But there are certain known physical limitations of the human ear, of the way sounds behave and how they cancel or mask each other etc., that makes some things impossible given what we know.  Maybe someone will turn up who hears blue--we are not limited by any theory, just suggest a test to see if he can actually hear blue.  Otherwise, it's unknowable to us -- and maybe to him also (he might come to think he's crazy and perhaps for good reason).   

Quote from: John Eppstein link=msg=0 date=
The word "transparency" seems to get bandied about a lot here. Some people are attempting to derive a "scientific" definition of transparency. I suggest that this is an invalid quest. Transparency is a SUBJECTIVE phenomenon and is not capable of being objectified. Furthermore, it is a "property" of systems .... This can be demonstrated easily if one considers the case of microphones.Microphones do not detect sound the same way that the human auditory system does. Therefore any microphone based recording CAN NOT be indistinguishable from the original performance; therefore it can not be transparent, if by "transparent" we mean exactly like the source as Ethan and others would have it

The "source" is not the performance.  The source is the recording.  You are abx'ing against the recording, not the original performance.  Transparency in the extreme sense you seem to mean (identity of total "experience" or state) may or may not be subjective (eeg implies objective evidence), but it's not provable (barring near omniscience) when the subject himself is not conscious of what his experience is or means, at least not enough to pass a simple abx test (or any other test to distinguish from guessing and placebo/bias effects).  But transparency in the limited sense we mean in this forum is objectively provable--simply put, without cheating (very narrowly defined), have you some way of detecting/perceiving and communicating to us whether two music samples differ (or whether one sample rather than another is closer to a third sample etc.).

ABX testing vs. perceptual states

Reply #22
...when somebody like Mixerman, or Weedy, or Slipperman, or dwoz, or Bob Ohlsson says that they can hear a difference, all you can do is test them. If the test is properly constructed - i.e. if the conditions adequately duplicate the professional listening experience, there should be a correlation. But if there isn't it's STILL NOT PROOF, because you can't prove a negative...

They are the ones who have to prove that they heard a difference, instead of saying they can without even trying or, sometimes, absurdly doubting the test.

ABX testing vs. perceptual states

Reply #23
I don't have a proper theory - it's not my job.

But you're acting like you do.

Quote
What I'm doping is pointing out that there's something here that appears to not be adequately explained by current conventional theory that bears investigation

By mentioning an unquoted article, without a direct link to the source, or any details for what it's worth, and which talks about something else (another sense), and which is itself an interpretation for laypeople (albeit presumably from the scientists themselves).

Quote
and suggesting some avenues of research to those who might be qualified to pursue them.

Have you suggested anything concrete or useful at all?

Quote
If I was an "audiophoole" you can be assured that I'd have lots of cockamamie theories that I'd be absolutely certain are true without any proper research. But I'm not and I don't.
You don't have to be absolutely certain of anything to be an audiophool. Just casting vacuous unfounded doubt and "skepticism" on well proven testing methods should be about enough.

Quote
What I do have is observations - both my own and of professional listeners - people who make their living with their ears, as well as my reading and other informal studies of the various sciences that are associated with perception research. I'm pretty good at taking data from diverse fields and drawing inferences that might escape the notice of specialists in any one field, and that's what I'm doing here.

Anecdotes and argument from authority (worse, own authority), won't make your argument any more credible, or sensible.

ABX testing vs. perceptual states

Reply #24
The "source" is not the performance.  The source is the recording.

No. Not when we're evaluating recording equipment and what we're looking at is the cumulative effect that the recording chain and process has on the resulting recording. The recording is not the source, it is the product.

And here we have the essence of one of the problems with Ethan's paper and with your whole mindset - you're looking at gear used for recording the same way that you evaluate gear used for hi-fi playback, but it's not really the same thing. The tools are similar in "appearance" and perhaps even in the gross details of operation - but not in use. It's like comparing dentist's pliers used to pull teeth to fine jeweler's pliers used in the fabrication of fine filigree. Yes they both belong to the general tool category of "pliers" - but the way they are used and the result of the use is vastly different. One is constructive and intended for very detailed assembly. The other assembles nothing and, if anything, is destructive in function, albeit beneficial.