Science is....not often found in these threads

[anetode]

This isn't a "science forum", it's an exile for fact-based comparisons that are not allowed elsewhere on Head-Fi.

 

[Punnisher]

Quote:

Originally Posted by Koyaan I. Sqatsi /img/forum/go_quote.gif The listener doesn't have to directly answer the question whether X=A or X=B. k

Really? Then what would a wrong answer tell the statistician? I am genuinely curious.

edit nevermind. Didn't read your post correctly.

Isn't that the point though? To directly answer that sample X is either A or B? A wrong answer would mean that in that instance, with that sample, that person could not tell the difference. If this trend continues for a significant amount of time, it can be concluded that a difference in the samples is not proven.

 

[wavoman]

Quote:

Originally Posted by SmellyGas /img/forum/go_quote.gif ... What objection would you have to a simple same/different comparison. Listen to A, listen to B. Are they the same or are they different?

No objection at all. That is exactly what I think should be done. Well almost. My protocol actually asks the following, which has been carefully designed (and I am now testing) to eliminate certain subtle kinds of response bias ... I also introduce swindles (false comparisons ... A vs A and B vs B) --

1. I prefer A to B
2. I prefer B to A
3. I hear a difference between A and B, but have no preference really
4. I do not hear any difference

This differs a bit from some other questions I have posted before -- progress.

Smelly, we have no disagreement on this point.

 

[wavoman]

Quote:

Originally Posted by Koyaan I. Sqatsi /img/forum/go_quote.gif The listener doesn't have to directly answer the question whether X=A or X=B

Some A/B/X tests let you say "I can't tell". Is that what you mean?

Others insist you pick one, A or B, after hearing X.

Can you explain a bit more? Thanks.

 

[haloxt]

Quote:

Originally Posted by Koyaan I. Sqatsi /img/forum/go_quote.gif Where can one get the details of this measurement? k

It's Friday here, so here is the BIG one | Computer Audiophile

And no I don't read JAES, but I still think there are things we could try measuring but we don't for lack of interest or financial ability.

Okay, now for some blah blah blah.

Quote:

Ok, try this one - say 50% of the world believe in some form of God and 50% believe there is no form of God. So 50% are wrong and the strength of their belief does not alter the fact one way or another.

First of all, you MUST define what are the attributes of God. When you try to do this sincerely you realize everyone has their own perception of God, perhaps a small fraction of which corresponds with whatever a monotheistic religion's holy scripture says God is. Secondly, the percentages are all wrong, for the same reason, that people have not clearly defined what is God and not God. People have vastly varying opinions on religion, but in the vast majority of cases their opinion of it is based on ignorance and arrogance, including atheists.

Quote:

Sigh, start with expectations, then add the affect of appearance, then cognitive dissonance, bung in a dash of groupthink and stir with a bit of social acceptance theory, half-bake in an oven and sprinkle with a dash of magical thinking. Voila !

The term cognitive dissonance's actual use (man, I'm really hurting watching people on this website destroy scientific terms) is for expressing how people are afraid of and avoid cognitive dissonance because they do not like contrary opinions. What cognitive dissonance (two contrasting opinions) are you trying to avoid when under the influence of auditory placebo? If you honestly believe cognitive dissonance is occurring, why do you think it is forcing believers to become pro-cablers rather than anti-cablers? So therefore I think you have applied the avoidance of cognitive dissonance to define your own definition of what the term cognitive dissonance means. And you are suffering from the fallacy of avoiding cognitive dissonance by refusing to consider the possibility of cables making a difference. Uneasy, the thought of cables making a difference makes you feel?

Quote:

Wrong way around, it is the unreliability of humans that make DBTs essential and far far better than sighted tests, I have done dozen of DBTS on myself and they work, when there is a difference of a sufficient magnitude, I have used noise levels, diff CD players and filters. Seriously try it yourself, take a piece of music and apply a few different low pass filters, apply roll-offs at 5,7,9 and 10K, I guarantee you will be able to tell the difference in a DBT, at 13K it gets harder at 15K pretty marginal and at 18K very very difficult but if you have good ears you might manage it, these are real differences and really detectable.

DBT's do not intrinsically address the fact that people will often report their senses faultily, especially when under unideal circumstances, be it stress, a time-limit, fear of failure, indigestion, etc. And DBT's also do not intrinsically address the fact that, by gum, humans are animals who are designed to process data and feed it to the conscious mind in a way that makes sense to it so it can perform its daily functions like hunting and digestion and elimination and reproduction. First there is input, then processing, and finally conscious consideration and action. The parts of the human being involved in input (senses) are always never perfect due well, some would say to God having a sense of humor, thus the brain processes this faulty data, and there is serious processing evolved from millions of years of evolution which I don't have to explain to someone like yourself, and this data is served up to the animal called man. And do you think the data of what the conscious mind of the man reports, without any consideration of the condition of the test subject or test environment, will be a irrefutable picture of what a human being is capable, or not capable, of hearing?

Quote:

Thanks but I know how to interpret experiments, though I actually did more experiments in my non Psychology degrees, yes there are some bad tests out there, I give them less credence, Blind is necessary but not sufficient.

Blind is not necessary, but it is ideal if possible to be done properly. It is the only form of testing I believe will satisfy the greatest majority, certainly not hearsay. But anything that requires complex interpretations may mathematically prove something, and I'd accept it especially if it was really funny and creative, but then again many people won't.

Quote:

Yet they make such grandiose claims, such claims without evidence look, well almost religious. Perhaps they genuinely believe that their cables are better, perhaps some do not do such tests because they fear that punters might (A) not be able to tell cables apart or (B) might prefer the competition ? That is another set of possible interpretations.

Anything is possible, and that's the only way I like it. "If the future were known, existence would lose its charm."

 

[nick_charles]

Quote:

Originally Posted by haloxt /img/forum/go_quote.gif First of all, you MUST define what are the attributes of God. When you try to do this sincerely you realize everyone has their own perception of God, perhaps a small fraction of which corresponds with whatever a monotheistic religion's holy scripture says God is. Secondly, the percentages are all wrong, for the same reason, that people have not clearly defined what is God and not God. People have vastly varying opinions on religion, but in the vast majority of cases their opinion of it is based on ignorance and arrogance, including atheists.

In which case by your definition most people are in fact wrong, unless there is no god or gods of any form, i.e any form that would be considered above known natural laws in which case the atheists are right, in any case many are wrong and their beliefs do not make them right.


Quote:

The term cognitive dissonance's actual use (man, I'm really hurting watching people on this website destroy scientific terms) is for expressing how people are afraid of and avoid cognitive dissonance because they do not like contrary opinions. What cognitive dissonance (two contrasting opinions) are you trying to avoid when under the influence of auditory placebo?

You buy a cable and you believe it will be wonderful and it isnt, there you have dissonance so you either change your view to it isn't wonderful or you focus on a different aspect i.e it is really well made so still worth the money even if the feature you originally bought it for is no longer valid, but I cede the point about the terminology, that was careless, but I think the other parts of my recipe are valid descriptors.


Quote:

So therefore I think you have applied the avoidance of cognitive dissonance to define your own definition of what the term cognitive dissonance means. And you are suffering from the fallacy of avoiding cognitive dissonance by refusing to consider the possibility of cables making a difference. Uneasy, the thought of cables making a difference makes you feel?

Actually I spent many weeks and for me a large amount of money laboriously and rigorously testing a number of cables to see if I could find any notable diferences between them, both with listening and measurements, at that time and still I have an open mind about the possibility of cable differences , though I think it less likely now with "normal" cables. In fact there have been some positive DBTs on cables under somewhat extreme conditions.

Quote:

And do you think the data of what the conscious mind of the man reports, without any consideration of the condition of the test subject or test environment, will be a irrefutable picture of what a human being is capable, or not capable, of hearing?

Test several hundred and you can easily average out any internal variations. The internal processes only become interesting IF you first find any real evidence that anyne can detect any differences. If nobody can tell the difference it does not really matter how they fail to do so.

 

[Steve Eddy]

Quote:

Originally Posted by Punnisher /img/forum/go_quote.gif Really? Then what would a wrong answer tell the statistician? I am genuinely curious. edit nevermind. Didn't read your post correctly. Isn't that the point though? To directly answer that sample X is either A or B?

What I mean is that there has been some criticism of ABX in that listeners aren't allowed to listen and do comparisons as they would under normal conditions because they're forced to identify X. In other words, the criticism is that the listener isn't able to simply listen and see if they hear a difference or not, or if they have a preference or not because they're only asking themselves what is X.

But that's not the case. The listener may listen however they want and use whatever means they want to compare. The identity of X can then be deduced after the fact by simple logic.

Quote:

A wrong answer would mean that in that instance, with that sample, that person could not tell the difference. If this trend continues for a significant amount of time, it can be concluded that a difference in the samples is not proven.

Actually, if there was a trend of WRONG answers, that would be of just as much interest as if there was a trend of CORRECT answers.

k

 

[SmellyGas]

Quote:

Originally Posted by haloxt /img/forum/go_quote.gif It's Friday here, so here is the BIG one | Computer Audiophile

LMAO. This is a link to some kind of scatter plot - the Y-axis is completely unlabelled, and the x-axis has NO units. It is an absolute meaningless presentation of data. Obviously, the details of how the "data" was measured is not reported. If I didn't know any better, I'd think it was a joke.

Quote:

The term cognitive dissonance's actual use (man, I'm really hurting watching people on this website destroy scientific terms) is for expressing how people are afraid of and avoid cognitive dissonance because they do not like contrary opinions. What cognitive dissonance (two contrasting opinions) are you trying to avoid when under the influence of auditory placebo? If you honestly believe cognitive dissonance is occurring, why do you think it is forcing believers to become pro-cablers rather than anti-cablers? So therefore I think you have applied the avoidance of cognitive dissonance to define your own definition of what the term cognitive dissonance means.

Actually, the term WAS used correctly, and it's pretty obvious where it's occuring. I'm not sure you fully understand what it means. A simple explanation can be found here: It's Friday here, so here is the BIG one | Computer Audiophile. But here's the breakdown anyway:

Belief #1: "I spend a LOT of money on cables."
Belief #2: "There are countless published reports of blind listening tests that fail to demonstrate LARGE, EASILY-AUDIBLE differences in cables."

These two simultaneous beliefs create an unpleasant "dissonance," in that it implies that the person spent a lot of money on cables that don't, in fact, make big differences in sound at all.

In response, the person could change one of his beliefs. Since it's kind of hard to lie to yourself about how much money you already spent, a common response would be:
"oh well, those reports are obviously invalid because *I* can hear big differences in cables" or similar.

Quote:

DBT's do not intrinsically address the fact that people will often report their senses faultily, especially when under unideal circumstances, be it stress, a time-limit, fear of failure, indigestion, etc. And DBT's also do not intrinsically address the fact that, by gum, humans are animals who are designed to process data and feed it to the conscious mind in a way that makes sense to it so it can perform its daily functions like hunting and digestion and elimination and reproduction.

This particular objection was already anticipated and easily addressed. In at least one paper, the listeners were all permitted to listen to cables UNBLINDED. They all reported big differences in sound. Next, the ONLY thing that was changed was that the cables were then hidden. Suddenly, they couldn't reliably differentiate. Now, if there truly were LARGE, EASILY-AUDIBLE differences, do you really think people's "fear of failure" and "indigestoin" could possibly obliterate their ability to hear the LARGE, EASILY-AUDIBLE differences that were reported so easily when they knew which cable was which??? HMMMMM...

Interestingly, another response to cognitive dissonance is generation of ad hoc hypotheses. Yours is clearly an ad hoc hypothesis. Since you claim to be knowledgable about cognitive dissonance, I won't offer further explanation.

 

[Steve Eddy]

Quote:

Originally Posted by haloxt /img/forum/go_quote.gif It's Friday here, so here is the BIG one | Computer Audiophile

Thanks.

Quote:

And no I don't read JAES, but I still think there are things we could try measuring but we don't for lack of interest or financial ability.

So, you don't really know what has been measured, but you think there are things we could try measuring but don't?

I'm sorry, but the logic seems to escape me there.

k

 

[ServinginEcuador]

Science has its limitations, and can't prove or disprove everything. Prove that love exists using science. We can see certain physiological and neurological reactions, but can't see, measure, or quantify love.

Other disciplines take over when science hits a brick wall. Philosophy is one that can often take up the baton when science runs out of steam.

 

[Uncle Erik]

haloxt, take a look at some of the blind wine comparison tests. It seems that telling someone that a wine costs $150 generates many more positive reactions than telling them that it is a $10 wine. The psychological biases and expectations have been pretty well sorted through. It is exactly the same with cables - telling someone that a cable is "better" and costly generates the same expectation as telling someone that they're drinking a $150 bottle of wine.

And if science isn't "capable" of understanding cables, why not put them under the lens of cost accounting?

For some unknowable, mysterious reason, cables are sold at enormous margins. Margins rarely found in any other business.

Have you ever encountered someone who lied to make money? Generally, do you think people lie for money?

How do you know cable manufacturers aren't lying to take your cash?

 

[haloxt]

Quote:

Originally Posted by nick_charles /img/forum/go_quote.gif In which case by your definition most people are in fact wrong, unless there is no god or gods of any form, i.e any form that would be considered above known natural laws in which case the atheists are right, in any case many are wrong and their beliefs do not make them right.

How can you be right about something when you aren't even arguing the same thing? What God means to an atheist is just as important as what God means to a theist when they argue, but they never get down to figuring out just what they disagree with, they just fight fight fight.

Quote:

You buy a cable and you believe it will be wonderful and it isnt, there you have dissonance so you either change your view to it isn't wonderful or you focus on a different aspect i.e it is really well made so still worth the money even if the feature you originally bought it for is no longer valid, but I cede the point about the terminology, that was careless, but I think the other parts of my recipe are valid descriptors.

Yes, but your pretty good analysis also suggests just how difficult it is to overcome the myriad bias that stand in the way of truly answering the question of "Can humans hear cable differences?" Too much muck to wade through for me to find out.

Quote:

Actually I spent many weeks and for me a large amount of money laboriously and rigorously testing a number of cables to see if I could find any notable diferences between them, both with listening and measurements, at that time and still I have an open mind about the possibility of cable differences , though I think it less likely now with "normal" cables. In fact there have been some positive DBTs on cables under somewhat extreme conditions.

Like you, I'm undecided about cables. The fact that it is either very difficult or impossible to prove it despite how many people claim they hear a difference means that there's a specter pro-cablers must continually face. Pro-cablers can only speak with certainty about cables making a difference when someone finally demonstrates a cable difference (and I won't trust unrepeatable tests funded by cable manufacturers), or a test is made so rigorous as to prove to most pro-cablers that cables don't make a difference. If such tests show up, let's assume, and pro-cablers still refuse to believe, then your religious analogies will have a lot more similarities to this cable situation.

Quote:

Test several hundred and you can easily average out any internal variations. The internal processes only become interesting IF you first find any real evidence that anyne can detect any differences. If nobody can tell the difference it does not really matter how they fail to do so.

I don't care enough to do anything for science

no money, no incentive.

SmellyGas, if you look hard enough anything can appear true. But your rationalization, although showing how cognitive dissonance works, doesn't completely jive with what nick wrote. And your claim that since people claim to hear big differences but then can't hear big differences when blindfolded does not mean the possibility of human error has been fully addressed.

Koyaan, I don't read JAES because my grasp of audio electronics is equivalent to a 5 year old's (plus I blew up two $190 soundcards putting things in backwards). But I was referring specifically to the fact that there's no special rescue team whenever an audio claim is made to verify or disprove that claim. There's many claims which simply have not been tested, whether because they're too kooky (I'll admit that one about two bit-perfect streams producing differences is pretty out there), or because there's simply no interest or financial incentive. But even testing kooky things is good because if they're wrong you can disprove it.

Uncle Erik, all valid points. But the fact that people's judgment can suffer from expectation suggests the difficulty of removing bias from truly understanding what the human tastes. And if cable manufacturers sell with huge margins, well, that's just greed, can't really deduce the truth about cable differences from that.

 

[SmellyGas]

Quote:

Originally Posted by haloxt /img/forum/go_quote.gif SmellyGas, if you look hard enough anything can appear true. But your rationalization, although showing how cognitive dissonance works, doesn't completely jive with what nick wrote.

Since there are many beliefs occuring simultaneously, it naturally follows that any two could conflict and cause "dissonance" - which means multiple instances of cognitive dissonance could be occuring simultaneously. Nick's scenario is certainly reasonable. So is mine. Definitely check out the new thread I posted. http://www.head-fi.org/forums/f133/h...esting-447387/

Quote:

And your claim that since people claim to hear big differences but then can't hear big differences when blindfolded does not mean the possibility of human error has been fully addressed.

Human error in what way? I don't understand how a human error could cause Bob to report amazing improvements in cable A vs. B, only to have them disappear completely when he is no longer aware of which cable is which. But don't worry, the blind listening tests have been repeated elsewhere, and they all have very similar results - are you going to suggest that "human error" (which doesn't really refer to anything specific) is occuring in each published study?

 

[SmellyGas]

Quote:

Originally Posted by ServinginEcuador /img/forum/go_quote.gif Science has its limitations, and can't prove or disprove everything. Prove that love exists using science. We can see certain physiological and neurological reactions, but can't see, measure, or quantify love. Other disciplines take over when science hits a brick wall. Philosophy is one that can often take up the baton when science runs out of steam.

Very rarely does a scientific experiment "prove" or "disprove" ANYTHING. It only supports or fails to support a hypothesis or theory. To give you an example, there is NO "PROOF" that cigarette smoking causes cancer in humans. To even come close to "proving" this, you would need to conduct a randomized, placebo-controlled cigarette vs. real cigarette experiment on humans and follow up after 20+ years. Has such an experiment been done? Absolutely not and for obvious reasons. Does this mean we should believe that cigarettes don't cause cancer????? After all, there is no "proof"!?

 

[wavoman]

Quote:

Originally Posted by Koyaan I. Sqatsi /img/forum/go_quote.gif But that's not the case. The listener may listen however they want and use whatever means they want to compare. The identity of X can then be deduced after the fact by simple logic

I still don't get it. Could you please explain in more detail? In A/B/X suppose I listen all I want, and decide I like A. But then X is presented and I have to identify whether X is A or B. I don't see any way around that.

Thanks.