FW: Re: Yet another DBT post
Bob Marcus
>mku...@aol.com (Mkuller) wrote:>
> >Now contrast this with an open-ended
> >>audio component comparison. Two amplifiers are compared playing
>Vivaldi's
> >>"Spring" as a source. In sighted listening differences are
> >>identified.
> >> In blind listening, the panelists were not reliably able to identify
>the
> >>correct amplifier as 'X'. One of the six panelist had a higher score
>within
> >>the confidence limits, but when averaged, the results fell below
>confidence.
> >
>
> >nous...@aol.com (Nousaine) wrote:>
> >This condition has never occured in any set of listener scores that I've
> >analyzed.
> >
>
>Selective memory? What about the most famous and one of the only published
>audio comparison blind tests - Greenhill, 1991, with speaker cables?
It proves Tom's point. In every test in which at least one subject got a
statistically significant score, the entire panel also got a statistically
significant score. And in the one test in which no single subject got a
statistically significant score, the panel scored just under 50%.
>
<snip>
> >>>>Because blind listening requires a
>'remember/compare/match/decide/chose'
> >>>>process where sighted listening only requires 'remember/compare', I
> >suspect
> >>>>blind listening is less sensitive.
> >>>
> >>>nousaine>
> >>>I don't see any difference between the decision-making process other
>than
> >>>un-bias controlled "listening" may often (perhaps usually) may not
>even
> >>>require the listening part.
> >>>
> >>
> >>In sighted listening, there is no 'X' to identify, the only functions
>are
> >>'remember/compare'.
> >>In a blind test you are required to 'match/chose/decide', in addition.
>That
> >>is
> >>THE difference.
> >
> >You still have to choose or grade. There's no extra work.
> >
>If you are saying there is no difference between the two types of tests in
>the
>types of brain functions involved, you are either in denial or missing an
>important part here.
Nope. Even in a blind ABX test, all you are ultimately doing is listening to
two sounds (A and X or B and X) and deciding whether they are the same or
different. That's exactly what you'd be doing if you were comparing A and B
sighted, and the same part of the brain would be used. You aren't really
"identifying" anything in an ABX test at the time you make that judgment.
You make the identification as a result of that judgment, but by then the
judgment has been made.
Furthermore, a sighted test always involves several different parts of the
brain, because you're using your eyes, as well as your memory of everything
you have ever heard, read, or thought about the products you are comparing.
To claim that sighted listening is more sensitive because it involves fewer
parts of the brain or less mental processing simply runs counter to the
facts. It is LESS sensitive precisely because it involves MORE processing in
MORE parts of the brain.
bob
_________________________________________________________________
Get a FREE online virus check for your PC here, from McAfee.
http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963
Harry Lavo
news:dFXRb.177445$na.287359@attbi_s04...
Bob-
Would you care to restate that as an opinion or an hypothesis? Else, can
you please cite some scientific "evidence" supporting this as fact (the fact
that the visual part of the brain "interferes" with the "musical analysis"
parts of the brain), as is often asked of the rest of us? And please don't
revert back to "blind vs sighted" listening. That is not what you asserted
here, e.g. "It is LESS sensitive precisely because it involves MORE
processing in MORE parts of the brain".
Harry
Bob Marcus
>"Bob Marcus" <nab...@hotmail.com> wrote in message
>news:dFXRb.177445$na.287359@attbi_s04...
> >
> > Furthermore, a sighted test always involves several different parts of
>the
> > brain, because you're using your eyes, as well as your memory of
> > everything
> > you have ever heard, read, or thought about the products you are
> > comparing.
> > To claim that sighted listening is more sensitive because it involves
> > fewer
> > parts of the brain or less mental processing simply runs counter to the
> > facts. It is LESS sensitive precisely because it involves MORE
>processing
> > in
> > MORE parts of the brain.
> >
> >
>Bob-
>
>Would you care to restate that as an opinion or an hypothesis?
No, I would not. Expectation bias is an established fact, Harry. And it
occurs precisely because the brain is simultaneously processing loads of
non-sonic information at the same time that it is trying to come to a
conclusion about the sonic information. Eliminate the sources of non-sonic
information, and you create a far more accurate test.
bob
_________________________________________________________________
High-speed users—be more efficient online with the new MSN Premium Internet
Software. http://join.msn.com/?pgmarket=en-us&page=byoa/prem&ST=1
Mkuller
>> > Furthermore, a sighted test always involves several different parts of
>>the
>> > brain, because you're using your eyes, as well as your memory of
>> > everything
>> > you have ever heard, read, or thought about the products you are
>> > comparing.
>> > To claim that sighted listening is more sensitive because it involves
>> > fewer
>> > parts of the brain or less mental processing simply runs counter to the
>> > facts. It is LESS sensitive precisely because it involves MORE
>>processing
>> > in
>> > MORE parts of the brain.
>> >
This is just plain wrong - how did you arrive at this conclusion?
Harry Lavo wrote:
>>Would you care to restate that as an opinion or an hypothesis?
>
>No, I would not. Expectation bias is an established fact, Harry.
Yes, it is. OK so far.
> And it
>occurs precisely because the brain is simultaneously processing loads of
>non-sonic information at the same time that it is trying to come to a
>conclusion about the sonic information.
That's an interesting conclusion - I would have thought it was due to listener
*expectations* of two different audible stimuli being different. In fact, in a
blind test where nothing is changed, aren't differences usually identified?
Any evidence for your statement?
Regards,
Mike
Harry Lavo
> Harry Lavo wrote:
>
> >"Bob Marcus" <nab...@hotmail.com> wrote in message
> >news:dFXRb.177445$na.287359@attbi_s04...
> > >
> > > Furthermore, a sighted test always involves several different parts of
> >the
> > > brain, because you're using your eyes, as well as your memory of
> > > everything
> > > you have ever heard, read, or thought about the products you are
> > > comparing.
> > > To claim that sighted listening is more sensitive because it involves
> > > fewer
> > > parts of the brain or less mental processing simply runs counter to
the
> > > facts. It is LESS sensitive precisely because it involves MORE
> >processing
> > > in
> > > MORE parts of the brain.
> > >
> > >
> >Bob-
> >
> >Would you care to restate that as an opinion or an hypothesis?
>
> No, I would not. Expectation bias is an established fact, Harry. And it
> occurs precisely because the brain is simultaneously processing loads of
> non-sonic information at the same time that it is trying to come to a
> conclusion about the sonic information. Eliminate the sources of non-sonic
> information, and you create a far more accurate test.
>
I take it then that there is no neurological support for your claim?
Bob Marcus
>
> >>"Bob Marcus" <nab...@hotmail.com> wrote :>
> >> > Furthermore, a sighted test always involves several different parts
>of
> >>the
> >> > brain, because you're using your eyes, as well as your memory of
> >> > everything
> >> > you have ever heard, read, or thought about the products you are
> >> > comparing.
> >> > To claim that sighted listening is more sensitive because it involves
> >> > fewer
> >> > parts of the brain or less mental processing simply runs counter to
>the
> >> > facts. It is LESS sensitive precisely because it involves MORE
> >>processing
> >> > in
> >> > MORE parts of the brain.
> >> >
> >
> >
>Harry Lavo wrote:
> >
>
> >No, I would not. Expectation bias is an established fact, Harry.
> >
> >
> >occurs precisely because the brain is simultaneously processing loads of
> >non-sonic information at the same time that it is trying to come to a
> >conclusion about the sonic information.
> >
>listener
>*expectations* of two different audible stimuli being different.
And those expectations result from the non-sonic information--seeing the
cables, or having formed a prior impression of them. Don't take the word
"expectation" literally, here. It doesn't require a conscious pre-judgment.
Indeed, many people who have consciously "expected" two things to sound the
same have perceived them differently in a listening test (and reported so
here). That doesn't mean they weren't affected by expectation bias. This
bias rears its ugly head subconsicously *during* the listening test.
I think the best way to understand this is to think of the brain as
synthesizing all of the information it has available to it--what you see,
what you hear, what you've read or heard about the product in the past--at
the time you are conducting the comparison. Most of the time in life,
synthesizing available information is exactly what you want your brain to
do. Listening comparisons may be one of the rare cases where you don't want
that synthesis--you want your brain to respond based solely on what you hear
at that moment. Alas, the survival of our primate ancestors did not depend
on the ability to isolate information from a single sensory organ, so it
wasn't a skill we developed.
>In fact, in a
>blind test where nothing is changed, aren't differences usually identified?
I'd say "often," not "usually," since I can't say it happens more or less
than 50% of the time. (Tom N. may have better data on this.) Again, this is
expectation bias, based on the non-sonic "knowledge" that something *has*
changed. Your brain has two conflicting pieces of information: the sound,
which we agree is identical from A to B; and the belief that a different
mechanism is making the sound. Not surprisingly, your brain produces
conflicting results in that case. Note that this phenomenon occurs even in
sighted listening, when someone fails to flip a switch. Several people here
have testified to that experience.
bob
_________________________________________________________________
Learn how to choose, serve, and enjoy wine at Wine @ MSN.
http://wine.msn.com/
Bob Marcus
information are processed by different parts of the brain? Do you deny that
long-term memory (of, say, equipment reviews) resides in yet another part of
the brain? And while we're at it, do you deny that we learned anything about
how the mind works before the invention of the functional MRI?
Please, Harry, this is too basic to even talk about.
bob
_________________________________________________________________
High-speed userså‚e more efficient online with the new MSN Premium Internet
Software. http://join.msn.com/?pgmarket=en-us&page=byoa/prem&ST=1
Nousaine
>Mkuller wrote:
...snip to content....
>Harry Lavo wrote:
...snip to content....
>
>>In fact, in a
>>blind test where nothing is changed, aren't differences usually identified?
>
>I'd say "often," not "usually," since I can't say it happens more or less
>than 50% of the time. (Tom N. may have better data on this.)
I conducted in the early 90s an experiment that showed that adult humans will
regularly, just over 3/4 of the time, report 'differences' when given 2
identical sound presentations. And when given 2 presentations where one of
those was 1-dB louder than the other would "prefer" the louder alternative and
describe preference choices as based on sound quality reasons (never on
loudness.)
Again, this is
>expectation bias, based on the non-sonic "knowledge" that something *has*
>changed.
I'm not certain I would call it "expectation" bias as such but it is apparently
a built-in human bias mechanism (audio professionals and enthusiasts showed no
significant difference from men-in-the-street.
Your brain has two conflicting pieces of information: the sound,
>which we agree is identical from A to B; and the belief that a different
>mechanism is making the sound. Not surprisingly, your brain produces
>conflicting results in that case. Note that this phenomenon occurs even in
>sighted listening, when someone fails to flip a switch. Several people here
>have testified to that experience.
Indeed. I began using controlled listening tests precisely because once when I
was demonstrating a "tweek" effect I obtained the expected result and only
afterward found that tweek was actually "out-of-circuit" during the
presentation.
Harry Lavo
Please do not go condescending on me. And please don't start trying to put
the onus on me. You made a statement that implied that one part of the
brain being active diminished the sensitivity of another part. I questioned
whether this was fact. You gave me your rationale instead. I asked if
their was neurological evidence of this. You haven't answered.
Yes, musical processing and emotional processing and visual processing do
take place in different parts of the brain as far as I know. Now I asked
you a question in response to your assertion that "It is LESS sensitive
precisely because it involves MORE processing in MORE parts of the brain".
That is a statement presented as fact which should be supportable by
evidence. I have asked you for it, since I believe it is your
opinion...considered...but still just your opinion. A bit like a
subjectivist "guessing" at a technical mechanism to explain their
observation. I am still waiting and you can still back off presenting it
as "fact".
normanstrong
news:8AeSb.139348$sv6.754480@attbi_s52...
I've asked this question before, but nobody has seen fit to answer it:
What is it that makes a sighted test sighted? Those of you who feel
that blind testing "obscures" audible differences that are easily
heard sighted--what do you mean by "sighted?"
Does the component in question have to actually be visible for the
advantages of sighted testing to be realized--or is it sufficient
simply to know the identify of the component? Suppose everything is
known; what it looks like, who made it and the model number. Is that
sufficient? Maybe one has to know all about the company, its
reputation and the price of the component?
Just what is it that makes a test "sighted?"
Norm Strong
Mkuller
>> >> > Furthermore, a sighted test always involves several different parts
>>of
>> >>the
>> >> > brain, because you're using your eyes, as well as your memory of
>> >> > everything
>> >> > you have ever heard, read, or thought about the products you are
>> >> > comparing.
>> >> > To claim that sighted listening is more sensitive because it involves
>> >> > fewer
>> >> > parts of the brain or less mental processing simply runs counter to
>>the
>> >> > facts. It is LESS sensitive precisely because it involves MORE
>> >>processing
>> >> > in
>> >> > MORE parts of the brain.
>> >> >
>Mkuller wrote:
>>This is just plain wrong - how did you arrive at this conclusion?
>> >
>>Harry Lavo wrote:
>> >>Would you care to restate that as an opinion or an hypothesis?
>> >
>>Marcus wrote:>
>> >No, I would not. Expectation bias is an established fact, Harry.
>> >
mkuller
>>Yes, it is. OK so far.
>> >
Marcus
>> > And it
>> >occurs precisely because the brain is simultaneously processing loads of
>> >non-sonic information at the same time that it is trying to come to a
>> >conclusion about the sonic information.
>> >
mkuller
>>That's an interesting conclusion - I would have thought it was due to
>>listener
>>*expectations* of two different audible stimuli being different.
>
Marcus
>And those expectations result from the non-sonic information--seeing the
>cables, or having formed a prior impression of them. Don't take the word
>"expectation" literally, here. It doesn't require a conscious pre-judgment.
>Indeed, many people who have consciously "expected" two things to sound the
>same have perceived them differently in a listening test (and reported so
>here). That doesn't mean they weren't affected by expectation bias. This
>bias rears its ugly head subconsicously *during* the listening test.
>
We're talking about two different audio component comparison tests - one
*sighted* and one *blind*. Unless the subjects are the Who's Tommy (deaf and
blind), they will be processing information from their five senses equally in
BOTH tests - regardless of what anyone has seen or read in the past.
Marcus
>I think the best way to understand this is to think of the brain as
>synthesizing all of the information it has available to it--what you see,
>what you hear, what you've read or heard about the product in the past--at
>the time you are conducting the comparison. Most of the time in life,
>synthesizing available information is exactly what you want your brain to
>do. Listening comparisons may be one of the rare cases where you don't want
>that synthesis--you want your brain to respond based solely on what you hear
>at that moment.
The only way to do *that* would be to perform the listening test in a sensory
isolation tank.
>Alas, the survival of our primate ancestors did not depend
>on the ability to isolate information from a single sensory organ, so it
>wasn't a skill we developed.
>
Or one that is relavant to *blind* versus *sighted* listening because the
subject is not really *blind*. Alas?
mkuller
>>In fact, in a
>>blind test where nothing is changed, aren't differences usually identified?
>
Marcus
>I'd say "often," not "usually," since I can't say it happens more or less
>than 50% of the time. (Tom N. may have better data on this.) Again, this is
>expectation bias, based on the non-sonic "knowledge" that something *has*
>changed. Your brain has two conflicting pieces of information: the sound,
>which we agree is identical from A to B; and the belief that a different
>mechanism is making the sound. Not surprisingly, your brain produces
>conflicting results in that case. Note that this phenomenon occurs even in
>sighted listening, when someone fails to flip a switch. Several people here
>have testified to that experience.
>
My point is - if it occurs at all in *blind* listening, it cannot be because of
sighted cues interferring with the brain's processing - which is what you are
claiming. Geesh.
Regards,
Mike
Bob Marcus
>"Bob Marcus" <nab...@hotmail.com> wrote in message
>news:bvca0...@enews2.newsguy.com...
> > Harry Lavo wrote:
> > >
> > > > "Bob Marcus" <nab...@hotmail.com> wrote in message
> > > > news:t9dSb.180943$xy6.868893@attbi_s02...
> > > > > Harry Lavo wrote:
> > > > >
> > > > > >"Bob Marcus" <nab...@hotmail.com> wrote in message
> > > > > >news:dFXRb.177445$na.287359@attbi_s04...
> > > > > > >
> > >parts of
> > > > > >the
> > > > > > > brain, because you're using your eyes, as well as your memory
>of
> > > > > > > everything
> > > > > > > you have ever heard, read, or thought about the products you
>are
> > > > > > > comparing.
> > > > > > > To claim that sighted listening is more sensitive because it
> > >involves
> > > > > > > fewer
> > > > > > > parts of the brain or less mental processing simply runs
>counter
> > >to
> > > > the
> > > > > > > facts. It is LESS sensitive precisely because it involves MORE
> > > > > >processing
> > > > > > > in
> > > > > > > MORE parts of the brain.
> > > > > > >
> > > > > > >
> > > > > >
> > > > > >Would you care to restate that as an opinion or an hypothesis?
> > > > >
> > >it
> > > > > occurs precisely because the brain is simultaneously processing
>loads
> > >of
> > > > > non-sonic information at the same time that it is trying to come
>to
>a
> > >non-sonic
> > > > > information, and you create a far more accurate test.
> > > > >
> > > >
> > > > I take it then that there is no neurological support for your claim?
> > >
> > What, do you deny that there is neurological evidence that visual and
>sonic
> > information are processed by different parts of the brain? Do you deny
>that
> > long-term memory (of, say, equipment reviews) resides in yet another
>part
>of
> > the brain? And while we're at it, do you deny that we learned anything
>about
> > how the mind works before the invention of the functional MRI?
>
>Please do not go condescending on me. And please don't start trying to put
>the onus on me. You made a statement that implied that one part of the
>brain being active diminished the sensitivity of another part.
Please don't go misinterpreting me. I said no such thing. Go back and read
the sentence, Harry. I did not say that "one part of the brain" was less
sensitive. I said that sighted listening was less sensitive.
L Mirabel
.
"Harry Lavo" <harry...@rcn.com> wrote in message
news:jHlSb.139060$Rc4.1115022@attbi_s54...
Having had to keep up to date with neurology in my
professional life as a consultant in internal medicine/cardiology I can
assure you that as yet we know zero about HOW the brain processes and
stores information, whether intellectual or complex sensory (eg. music)- MRI
or no MRI.
Any assured statements about how the brain does
it belong in journalism and RAHE debate.
Ludovic Mirabel
Mkuller
>I've asked this question before, but nobody has seen fit to answer it:
>What is it that makes a sighted test sighted?
>Those of you who feel
>that blind testing "obscures" audible differences that are easily
>heard sighted--what do you mean by "sighted?"
>
*Sighted* has nothing to do with the sense of sight. It merely means the
subject *knows* which component is playing. In a *blind* ABX type test, the
subject does not know which of two components is playing, he must match it to
his memory of the sound of 'A' or 'B'.
Regards,
Mike
Steven Sullivan
'Sighted' in the context of a blind protocol means:
knowing the identity of the item currently being tested/evaluated/compared, at
the moment the test datum is collected.
--
-S.
"They've got God on their side. All we've got is science and reason."
-- Dawn Hulsey, Talent Director
Steven Sullivan
You're getting wrapped up in semantic knots re: 'blinded', 'sighted'
Prior knowledge -- whether accurate or not, and whether derived
from sense impression or from verbal or written communication -- can
bias cognitive evaluation of *current* sensory input.
A simple example is when listeners are *told* that a condition has
been changed, but it actually hasn't been.
There will be a strong bias to 'hear' a difference anyway. This is
true even if the visual input hasn't changed either.
> Marcus
> >I think the best way to understand this is to think of the brain as
> >synthesizing all of the information it has available to it--what you see,
> >what you hear, what you've read or heard about the product in the past--at
> >the time you are conducting the comparison. Most of the time in life,
> >synthesizing available information is exactly what you want your brain to
> >do. Listening comparisons may be one of the rare cases where you don't want
> >that synthesis--you want your brain to respond based solely on what you hear
> >at that moment.
> The only way to do *that* would be to perform the listening test in a sensory
> isolation tank.
No, you do it by simply making one piece of information
*unavailable* to the listener at the time of audition -- namely, the
identity of the device or treatment currently playing. The only
cues to that, should be what is *heard*. Otherwise there's
a signficant chance the identification was made based on factors *other*
than the audible.
> >Alas, the survival of our primate ancestors did not depend
> >on the ability to isolate information from a single sensory organ, so it
> >wasn't a skill we developed.
> >
> Or one that is relavant to *blind* versus *sighted* listening because the
> subject is not really *blind*. Alas?
You are using the word *blind* incorrectly in this context.
> My point is - if it occurs at all in *blind* listening, it cannot be because of
> sighted cues interferring with the brain's processing - which is what you are
> claiming. Geesh.
He is claimign that non-audible information -- including, but NOT limited to,
current visual input -- can bias judgement of sense impressions. 'Sighted' merely
means knowing the *identity* of the device or sound source currently under
test. You can actually be looking at both cables during a cable comparison, but
so long as you do not know which one is currently in the circuit, it's a 'blind'
test.
Harry Lavo
quote:
"Furthermore, a sighted test always involves several different parts of the
brain...<>.....it is LESS sensitive precisely because it involves MORE
processing in MORE parts of the brain."
unquote
So following this logic, an audio only test uses less of the brain. ergo,
using more processing in more parts of the brain must interfere somehow with
this audio processing....otherwise the audio processing would be just as
valid as ever. I've asked for the scientifc studies showing how this
happens, neurologically, which it seems to me you must have in order to
assert it as fact, rather than supposition that this is the cause of
"sighted bias".
Mkuller
>> >> >> > Furthermore, a sighted test always involves several different parts
>
>> >>of
>> >> >>the
>> >> >> > brain, because you're using your eyes, as well as your memory of
>> >> >> > everything
>> >> >> > you have ever heard, read, or thought about the products you are
>> >> >> > comparing.
>> >> >> > To claim that sighted listening is more sensitive because it
>involves
>> >> >> > fewer
>> >> >> > parts of the brain or less mental processing simply runs counter to
>
>> >>the
>> >> >> > facts. It is LESS sensitive precisely because it involves MORE
>> >> >>processing
>> >> >> > in
>> >> >> > MORE parts of the brain.
>> >> >> >
>Steven Sullivan ssu...@panix.com wrote:>
>Prior knowledge -- whether accurate or not, and whether derived
>from sense impression or from verbal or written communication -- can
>bias cognitive evaluation of *current* sensory input.
>A simple example is when listeners are *told* that a condition has
>been changed, but it actually hasn't been. snipperroo
>He is claimign that non-audible information -- including, but NOT limited to,
>current visual input -- can bias judgement of sense impressions. 'Sighted'
>merely
>means knowing the *identity* of the device or sound source currently under
>test. You can actually be looking at both cables during a cable comparison,
>but
>so long as you do not know which one is currently in the circuit, it's a
>'blind'
>test.
>
While your efforts to come to Mr. Marcus' support is laudible, you are not
correct. If you carefully read exactly what Marcus says above, he is stating
that input from other senses including the eyes and the memory interferes with
brain function during *sighted* testing, but not *blind* testing. There
appears to be no basis for that claim.
Regards,
Mike
Nousaine
And in the classical ABX test the subject has visual contact with both the
components being compared.
Bob Marcus
>
>What about the "golden ear" that Greenhill mentioned?
He got a statistically significant score on 5 of the 6 tests. The entire
panel got a statistically significant score on the same 5 tests.
>What about the two
>"experienced listeners" versus the one non-experienced one in the Swedish
>audio
>group's CD player comparison?
My memory of this one is vague, but are you saying that two "experienced
listeners" heard a difference, and the one inexperienced listener didn't,
and the researchers concluded that the difference wasn't audible? That's not
my recollection at all. Maybe you need to provide a little mroe substance
than, "What about?"
bob
_________________________________________________________________
High-speed users—be more efficient online with the new MSN Premium Internet
Software. http://join.msn.com/?pgmarket=en-us&page=byoa/prem&ST=1
Bob Marcus
>So following this logic, an audio only test uses less of the brain.
Certainly in the sense that the portion(s) of the brain making the
same/different decision are not receiving certain stimuli that it/they would
be receiving in a sighted comparison.
> ergo,
>using more processing in more parts of the brain must interfere somehow
>with
>this audio processing.
Not "interfering." But in a sighted test it would be responding to different
(and at a rather practical level, more) stimuli.
>...otherwise the audio processing would be just as
>valid as ever. I've asked for the scientifc studies showing how this
>happens, neurologically,
As Dr. Mirabel has so rightly pointed out to you already, we know
approximately nothing about HOW the brain processes anything...
>which it seems to me you must have in order to
>assert it as fact, rather than supposition that this is the cause of
>"sighted bias".
...but we don't need neurological evidence to know that the brain processes
different stimuli differently, because as we change the stimuli we get
different responses. And in the case of listening tests, when we remove
certain stimuli, we get more sensitive tests.
bob
_________________________________________________________________
What are the 5 hot job markets for 2004? Click here to find out.
http://msn.careerbuilder.com/Custom/MSN/CareerAdvice/WPI_WhereWillWeFindJobsIn2004.htm?siteid=CBMSN3006&sc_extcmp=JS_wi08_dec03_hotmail1
josko
and read
> > the sentence, Harry. I did not say that "one part of the brain" was
less
> > sensitive. I said that sighted listening was less sensitive.
> >
>
> quote:
>
> "Furthermore, a sighted test always involves several different parts
of the
> brain...<>.....it is LESS sensitive precisely because it involves MORE
> processing in MORE parts of the brain."
>
> unquote
>
> So following this logic, an audio only test uses less of the brain.
ergo,
> using more processing in more parts of the brain must interfere
somehow with
> this audio processing....otherwise the audio processing would be just
as
> valid as ever. I've asked for the scientifc studies showing how this
> happens, neurologically, which it seems to me you must have in order
to
> assert it as fact, rather than supposition that this is the cause of
> "sighted bias".
>
Bob's quote is totally supported by standard information processing
theory from cognitive psychology (don't know about neurology). It's
really simple -- the more processing capacity one uses, the less
"accurate" the processing is due to some cognitive constraints (such as
limited short term memory). One consequence of such processing is that
people look for and utilize mental "short cuts" to form judgments and to
make decisions. These mental "short cuts" typically manifest themselves
in use of heuristics (simplified and therefore less precise mental rules
one executes to form a judgment or to make a decision) and in biased
processing of presented information and the information that is
available in memory (and presumably retrieved from there or not).
Obviously, some moderators (motivation, need for cognition, involvment,
need for justification, presence or absence of specific goals, .... not
to mention relative size/change/difference of incoming stimuli that are
being evaluated) do affect precision and sensitivity of any cognitive
processing. The problem is that I cannot think of any moderator that
would favor sighted evaluation of purely auditory signals over
evaluation "by ear" if the goal is to judge such signals in terms of
their auditory "content" only.
Literature? Really any basic textbook on cognitive psychology will do.
For the catalog of all kinds of biases and heuristics take a look at
Kahneman and Tversky's "Decision making under uncertainty: Heuristics
and Biases." Disclaimer: these books are not about evaluation of "cable
sound" by audiophiles; they are about fundamentals of human processing
of information. The theory discussed and proved (in probabilistic
sense, let's not forget that) in those books is directly applicable to
the issue discussed here, which is DBT. If somebody has a problem with
that, then he has a problem with at least 60 years of published research
in cognitive psychology (and behavioral decision theory) and, in the
context of "cable sound", with I-don't-know-how-many years of research
in psychoacoustics and electrical engineering.
Nousaine
Fully agreed. Another excellent book that discusses this issue is "Inevitable
Illusions: How Mistakes of Reason Rule Our Minds" by psychologist Massimo
Piattelli-Palmarini where discussion focus' on "..set of biases deeply embedded
in the human mind, distort the way we think." And IMO can profoundly distort
decision-making.
Bob Marcus
>
> > >"Bob Marcus" <nab...@hotmail.com> wrote :>
> > >> >> >> > Furthermore, a sighted test always involves several different
>parts
> > >
> > >> >>of
> > >> >> >>the
>of
> > >> >> >> > everything
> > >> >> >> > you have ever heard, read, or thought about the products you
>are
> > >> >> >> > comparing.
> > >> >> >> > To claim that sighted listening is more sensitive because it
> > >involves
> > >> >> >> > fewer
> > >> >> >> > parts of the brain or less mental processing simply runs
>counter to
> > >
> > >> >>the
> > >> >> >>processing
> > >> >> >> > in
> > >> >> >> > MORE parts of the brain.
<snip>
>If you carefully read exactly what Marcus says above, he is stating
>that input from other senses including the eyes and the memory interferes
>with
>brain function during *sighted* testing, but not *blind* testing.
And if you read REALLY carefully, you'll notice that I say no such thing.
Saying that a sighted test is less sensitive does not in any way imply any
"interference" in brain function. The brain function in deciding whether two
cables are the same or different is exactly the same. One test simply
involves fewer stimuli than the other. In particular, the lack of knowledge
about which cable is which eliminates from consideration any information
stored in long-term memory about the two cables or their
manaufacturers--price, reviews, reputations, etc.
>There
>appears to be no basis for that claim.
There is certainly no basis for the claim which you put into my mouth.
bob
_________________________________________________________________
Find high-speed ‘net deals — comparison-shop your local providers here.
https://broadband.msn.com
normanstrong
Now we're getting somewhere. By "knows" which unit is connected I
presume you don't mean the make and model number. If you'd never
heard of that company, this wouldn't constitute a sighted test, would
it? What I'm driving at is: just what is necessary before you can
claim that you've taken a valid sighted test?
Norm Strong
Nousaine
>Mkuller wrote:
>>
>>What about the "golden ear" that Greenhill mentioned?
>
>He got a statistically significant score on 5 of the 6 tests. The entire
>panel got a statistically significant score on the same 5 tests.
Which says that no positives were "hidden" by averages. IME the more likely
scenario is where a chance positive will tip the balance of a whole experiment
with a few subjects and few trials. Even so I've only found one test where that
"may" have happened. In the PSACS CD Test where a 14-bit 1st gen CD player was
compared toa Sony ES machine more than 10 years senior a single positive
subject tipped the whole experiment to positive.
I say "may" only because the presentations were only synchronized with
simultaneous finger-push on Play which is to say there were not confirmed to be
truly-synched. But even so a single subject scoring positive was enough to tip
the score.
And, don't forget if an experiment with NO single subject scoring positive can
have positive results if the average score is high enough.
>>What about the two
>>"experienced listeners" versus the one non-experienced one in the Swedish
>>audio
>>group's CD player comparison?
>
>My memory of this one is vague, but are you saying that two "experienced
>listeners" heard a difference, and the one inexperienced listener didn't,
>and the researchers concluded that the difference wasn't audible? That's not
>my recollection at all. Maybe you need to provide a little mroe substance
>than, "What about?"
>
>bob
Yes, when I said that my analysis concluded that there were no published
results where individual subjects with positive results were shown to 'hidden'
by averages the hand-waving began and the re-interpretation of what the data
actually showed was re-introduced.
Again there is no existing data that shows high-scoring subjects have been
buried in the average scores. Indeed in a high-profile, large sample size
experiment on capacitor dialectric the author concluded that music-type
analysis of individual selections 'showed' positive results (note that this was
not an analysis of individual scores) it was overlooked that that analysis
showed that there were "reverse-positive" results (scoring part X as part Y a
statistically significant number of times) as well indictating most likely
internal bias in the test; and NOT subject-hiding.
Harry Lavo
news:bveui...@enews4.newsguy.com...
Nice summary, thanks.
I am familiar with the principles you espouse above, having studied under
one of the more prominent behavioral psychologists in the Midwest while
getting my graduate degree at Northwestern.
But Bob made specific reference to "areas of the brain" which is the domain
of neuropsychological, and I called him on it, just as I am called every
time I venture a technical opinion that is judged "unsupported" by evidence.
All he had to do was say "well, I guess I was stretching it a bit." He
didn't and I guess isn't going to.
I would like to respond to your comments that you can't find any "moderator"
that would justify sighted testing over blind testing in trying to something
"by ear". If you have participated here long, you doubtless are aware of
the practical difficulties of doing a truly "tight" blind test at home. It
is primarily for this reason that most audiophiles rely on sighted listening
when making component comparisons. So given that very practical constraint,
don't your "moderators" actually come into play?. I would suggest an
audiophile truly wanting to evaluate the equipment and not being easily able
to do it blind can still have:
* motivation to want to get an accurate reading (e.g. is honestly searching
for improvement)
* cognitive need ("awareness", taking notes, analyzing the sound, pondering,
"gestalt listening"etc.)
* involvement (it is a hobby, after all, so not life and death, but
presumably an important evaluation)
* presence of a specific goal (to fix whatever he feels could be/needs to be
improved in the system; to focus on same)
* presence of a size/change/difference sufficient to be actually audible
with careful listening.
The objectivist argument here seems to be that sighted listening is doomed
to be fraudulent, misleading, and useless. One subjectivist argument is
that sighted listening is more practical and so long as the above motivators
are present, the sighted testing has a good chance of being reasonably
objective and correct.
normanstrong
news:iszSb.145773$sv6.803215@attbi_s52...
> normanstrong <norman...@comcast.net> wrote:
>
> > Just what is it that makes a test "sighted?"
>
> 'Sighted' in the context of a blind protocol means:
> knowing the identity of the item currently being
tested/evaluated/compared, at
> the moment the test datum is collected.
If we were talking about an interconnect test, and you knew that you
were listening to the Acme 76-2 interconnect, would you call that a
valid sighted test? Since you've never heard of the Acme cable
company, and have no idea about the price or supposed quality of the
78-2 interconnect, does the test still have the benefits of sighted
testing as it's generally discussed on this newsgroup?
What I'm curious about here is what it is about sighted testing that
makes its supporters have more faith in the test results than they do
if the test is blind.
Norm Strong
Mkuller
>One test simply
>involves fewer stimuli than the other. In particular, the lack of knowledge
>about which cable is which eliminates from consideration any information
>stored in long-term memory about the two cables or their
>manaufacturers--price, reviews, reputations, etc.
>
Right. Not knowing which cable is playing may cause ALL of the past
information on cables, reviews, manufacturers, etc. stored in the memory to be
accessed looking for similarities which would further *cloud* one's perceptions
(in the words of "The Shadow") according to your "theory". Right?
Regards,
Mike
Mkuller
>...but we don't need neurological evidence to know that the brain processes
>different stimuli differently, because as we change the stimuli we get
>different responses. And in the case of listening tests, when we remove
>certain stimuli, we get more sensitive tests.
>
It would appear you are claiming *blind* listening tests are more *sensitive*
than *sighted*. If that is what you are claiming, please provide some data
which show the *sensitivity* of both tests using the same music program as a
source.
Or are merely speculating that the sensitivity is greater because that conforms
with your beliefs?
Regards,
Mike
chung
Hmmm, not sure if I have ever seen you do that.
> He
> didn't and I guess isn't going to.
He didn't need to, for good reasons.
>
> I would like to respond to your comments that you can't find any "moderator"
> that would justify sighted testing over blind testing in trying to something
> "by ear". If you have participated here long, you doubtless are aware of
> the practical difficulties of doing a truly "tight" blind test at home. It
> is primarily for this reason that most audiophiles rely on sighted listening
> when making component comparisons. So given that very practical constraint,
> don't your "moderators" actually come into play?. I would suggest an
> audiophile truly wanting to evaluate the equipment and not being easily able
> to do it blind can still have:
>
> * motivation to want to get an accurate reading (e.g. is honestly searching
> for improvement)
> * cognitive need ("awareness", taking notes, analyzing the sound, pondering,
> "gestalt listening"etc.)
> * involvement (it is a hobby, after all, so not life and death, but
> presumably an important evaluation)
> * presence of a specific goal (to fix whatever he feels could be/needs to be
> improved in the system; to focus on same)
> * presence of a size/change/difference sufficient to be actually audible
> with careful listening.
>
> The objectivist argument here seems to be that sighted listening is doomed
> to be fraudulent, misleading, and useless.
This is a huge strawman.
The objectivist argument expressed in this newsgroup has been that
sighted testing to determine subtle differences is doomed to be useless.
Whether is is fraudulent or not depends on the motives.
It is misleading if someone tries to convince others of the cable sound
and other similar effects, based on sighted testing.
And when the differences are not subtle, as in comparing speakers, no
one has said that you need DBT's to tell them apart.
> One subjectivist argument is
> that sighted listening is more practical and so long as the above motivators
> are present, the sighted testing has a good chance of being reasonably
> objective and correct.
No chance when the difference is subtle. Like comparing speaker cables
or interconnects. Or comparing two competent amps. Or comparing benefits
of the magic CD pen.
Harry Lavo
news:bvf83...@enews1.newsguy.com...
I think, Norm, he means if you know anything about the units and can
identify which is playing. In other words, you have two identical grey
boxes marked "a" and "b". One box is darker grey than the other. The
argument is that you are likely to hear differences because the boxes differ
in color. The argument would probably be further made that the "lighter"
grey box predisposes you to hear a more "yang"-like coloration to the sound
from that box than from the other one. Or if you see two real amps, you see
one is bigger, appears more expensive, is shaped more to your liking, etc.
this predisposes you to favor that amp if you know which one is playing.
Even if you know nothing more and even if they are technically identical.
So the argument is that you must not know what is playing; that way no
matter what you know about the two amps, you won't be able to tie it to the
sound of the amp that is playing.
Audio Guy
Only if one is trying to identify the cable, not if one is just
looking for differences between cables.
Audio Guy
"Harry Lavo" <harry...@rcn.com> writes:
>
> Nice summary, thanks.
>
> I am familiar with the principles you espouse above, having studied under
> one of the more prominent behavioral psychologists in the Midwest while
> getting my graduate degree at Northwestern.
snip
> I would like to respond to your comments that you can't find any "moderator"
> that would justify sighted testing over blind testing in trying to something
> "by ear". If you have participated here long, you doubtless are aware of
> the practical difficulties of doing a truly "tight" blind test at home. It
> is primarily for this reason that most audiophiles rely on sighted listening
> when making component comparisons. So given that very practical constraint,
> don't your "moderators" actually come into play?. I would suggest an
> audiophile truly wanting to evaluate the equipment and not being easily able
> to do it blind can still have:
>
> * motivation to want to get an accurate reading (e.g. is honestly searching
> for improvement)
> * cognitive need ("awareness", taking notes, analyzing the sound, pondering,
> "gestalt listening"etc.)
> * involvement (it is a hobby, after all, so not life and death, but
> presumably an important evaluation)
> * presence of a specific goal (to fix whatever he feels could be/needs to be
> improved in the system; to focus on same)
> * presence of a size/change/difference sufficient to be actually audible
> with careful listening.
>
> The objectivist argument here seems to be that sighted listening is doomed
> to be fraudulent, misleading, and useless.
Please show anyone purportedly on the objectivist side calling sighted
listening "fraudulent" unless a manufacturer/salesman is involved.
> One subjectivist argument is
> that sighted listening is more practical and so long as the above motivators
> are present, the sighted testing has a good chance of being reasonably
> objective and correct.
For one who has had significant training in psychology, please
explain how you can disparage DBTs when they are such a fundamental
part of psychology and have been since the days of "Clever Hans".
Harry Lavo
news:uYTSb.150381$sv6.837615@attbi_s52...
Perhaps fraudulent was a poor choice of words. I meant that it is
considered not to show what we think is being shown, and that any
differences we hear "have" to be due to some bias or another, not to the
"sound" per se.
> > One subjectivist argument is
> > that sighted listening is more practical and so long as the above
motivators
> > are present, the sighted testing has a good chance of being reasonably
> > objective and correct.
>
> For one who has had significant training in psychology, please
> explain how you can disparage DBTs when they are such a fundamental
> part of psychology and have been since the days of "Clever Hans".
>
I have never "disparaged" DBT's. I have said they are often impractical in
a home environment. I have also said that without a proper control test, I
am not sure that they are the best test technique for open-ended evaluation
of audio components.
I spent a good part of my life sponsoring, helped designing, and
interpreting consumer research testing, often of a very sophisticated
nature. I understand the arguments. I also understand the need for a
control test to "prove" beyond a shadow of a doubt that quick switch dbts
(especially abx)" do not incur their own confusion factors that obscure
certain aspects of open-ended evaluation. So until the proponents of such
testing prepare and present such a control test with some scientific rigor,
I will stay with my views.
Audio Guy
"Harry Lavo" <harry...@rcn.com> writes:
> "Audio Guy" <audio...@yahoo.com> wrote in message
> news:uYTSb.150381$sv6.837615@attbi_s52...
>>
>> explain how you can disparage DBTs when they are such a fundamental
>> part of psychology and have been since the days of "Clever Hans".
>>
>
> I have never "disparaged" DBT's. I have said they are often impractical in
> a home environment. I have also said that without a proper control test, I
> am not sure that they are the best test technique for open-ended evaluation
> of audio components.
And what evidence do you have that they aren't?
> I spent a good part of my life sponsoring, helped designing, and
> interpreting consumer research testing, often of a very sophisticated
> nature. I understand the arguments. I also understand the need for a
> control test to "prove" beyond a shadow of a doubt that quick switch dbts
> (especially abx)" do not incur their own confusion factors that obscure
> certain aspects of open-ended evaluation. So until the proponents of such
> testing prepare and present such a control test with some scientific rigor,
> I will stay with my views.
First of all, I don't require ABX or quick switching in order to
perform a DBT, only that it is just that double blind, and I'm quite
sure that is the opinion of the rest of the DBT proponents here.
Secondly, why would DBTs be appropriate in other areas of
investigation, but not in audio component comparisons? I believe the
shoe is on the other foot, that is, since they are appropriate for
many areas of investigation, what makes them inappropriate in this
case other than they come to different conclusions than "open-ended
sighted" evaluations? Why wouldn't Tom's "cover the terminals" blind
test during an open-ended evaluation work just as well?
chung
> "Steven Sullivan" <ssu...@panix.com> wrote in message
> news:iszSb.145773$sv6.803215@attbi_s52...
>> normanstrong <norman...@comcast.net> wrote:
>>
>> > Just what is it that makes a test "sighted?"
>>
>> 'Sighted' in the context of a blind protocol means:
>> knowing the identity of the item currently being
> tested/evaluated/compared, at
>> the moment the test datum is collected.
>
> If we were talking about an interconnect test, and you knew that you
> were listening to the Acme 76-2 interconnect, would you call that a
> valid sighted test? Since you've never heard of the Acme cable
> company, and have no idea about the price or supposed quality of the
> 78-2 interconnect, does the test still have the benefits of sighted
> testing as it's generally discussed on this newsgroup?
I believe that to an audiophile, the fact that he has not heard of the
Acme-76 is a negative connotation for that cable, especially if he is
comparing it to a well-known, well-reviewed one. In practice, it is
unusual for him to test a totally unknown cable; he most likely would
have been recommended to try it, or he knows how much it costs, where it
was made, etc.
Mkuller
>I believe that to an audiophile, the fact that he has not heard of the
>Acme-76 is a negative connotation for that cable, especially if he is
>comparing it to a well-known, well-reviewed one. In practice, it is
>unusual for him to test a totally unknown cable; he most likely would
>have been recommended to try it, or he knows how much it costs, where it
> was made, etc.
>
Now you're beginning to tread on the territory of the "mind-readers". It
doesn't matter if the listener has never heard of *either* of the two cables or
components being compared, or even if they look identical and have an"A" on one
and a "B" on the other - if he knows and can keep track of which one is
playing, then the test is *sighted*. The primary difference between this and a
blind ABX-type test is that the listener must *match* his memory of the sound
of A and B (which he knows the identify of and keep track of) to "X", an
unknown (which is either A or B).
Regards,
Mike
Steven Sullivan
> I have never "disparaged" DBT's. I have said they are often impractical in
> a home environment.
Have you ever said that sighted tests in home environment
are ALSO impractical as a means of determining audible difference?
> I have also said that without a proper control test, I
> am not sure that they are the best test technique for open-ended evaluation
> of audio components.
Have you ever expressed such doubt about sighted 'open-ended'
evaluation of audio components?
> I spent a good part of my life sponsoring, helped designing, and
> interpreting consumer research testing, often of a very sophisticated
> nature. I understand the arguments. I also understand the need for a
> control test to "prove" beyond a shadow of a doubt that quick switch dbts
> (especially abx)" do not incur their own confusion factors that obscure
> certain aspects of open-ended evaluation.
Do you understand the need for something *other than* the sighted
comparison -- the method of choice in audiophila -- when the purpose
is to identify *audible* difference? Or, to put it another way, do you
understand that sighted comparison is *inadequate* for exactly that purpose,
even though it is used for that purpose routinely in audiophilia?
S888Wheel
>from sense impression or from verbal or written communication -- can
>bias cognitive evaluation of *current* sensory input.
>A simple example is when listeners are *told* that a condition has
>been changed, but it actually hasn't been.
>There will be a strong bias to 'hear' a difference anyway. This is
>true even if the visual input hasn't changed either.
You are talking about misdirection. Yeah people are suseptable to misdirection.
One could say that sameness in components could also be imagined if the
listener believes ahead of time that no difference exists. That means every ABX
DBT that has been done with people who believe no difference exists between the
components in question is tainted with expectation bias unless a control was
used in the test for that bias such as the random insertion of a known barly
audible distortion.>
>
>No, you do it by simply making one piece of information
>*unavailable* to the listener at the time of audition -- namely, the
>identity of the device or treatment currently playing. The only
>cues to that, should be what is *heard*. Otherwise there's
>a signficant chance the identification was made based on factors *other*
>than the audible.
Do you think you know what orange juice or
cola or strawberry jello taste like? I bet you cannot successfully identify
such items by taste alone on a reliable basis. Take the testee out of the
envirement and all bets are off on sensitivity. The interaction of senses is
complex and critical in sensitivty of the senses.
chung
OK, so you are saying if we tell you which one of the two (A or B) we
are playing (as in your definition of sighted), then you can tell us
that they sound different?
Nousaine
>"Audio Guy" <audio...@yahoo.com> wrote
...snip to content ....
>>
>> For one who has had significant training in psychology, please
>> explain how you can disparage DBTs when they are such a fundamental
>> part of psychology and have been since the days of "Clever Hans".
>>
>
>I have never "disparaged" DBT's. I have said they are often impractical in
>a home environment.
Sure.
I have also said that without a proper control test, I
>am not sure that they are the best test technique for open-ended evaluation
>of audio components.
But what control test do you have for 'sighted' (re: as uncontrolled) testing
other than that we already know (such as subjects will often report difference
when given identical audio presentations) other than it is obvious they are
prone to false positives? -- It seems that your only real argument for them is
they are easier to conduct.
>I spent a good part of my life sponsoring, helped designing, and
>interpreting consumer research testing, often of a very sophisticated
>nature. I understand the arguments. I also understand the need for a
>control test to "prove" beyond a shadow of a doubt that quick switch dbts
>(especially abx)" do not incur their own confusion factors that obscure
>certain aspects of open-ended evaluation. So until the proponents of such
>testing prepare and present such a control test with some scientific rigor,
>I will stay with my views.
Fair enough but IMO you're just choosing and defnding a method you 'prefer'
rather than one that has been verified in sensitivity. Indeed it's quite easy
to show that open-listening is rife with false positives. That's OK by me; but
I'm not one to endorse a method just because it may be easier and obviously
lacking in validity for investigatting true acoustic effects.
Mkuller
>Bob's quote is totally supported by standard information processing
>theory from cognitive psychology (don't know about neurology). It's
>really simple -- the more processing capacity one uses, the less
>"accurate" the processing is due to some cognitive constraints (such as
>limited short term memory). One consequence of such processing is that
>people look for and utilize mental "short cuts" to form judgments and to
>make decisions. These mental "short cuts" typically manifest themselves
>in use of heuristics (simplified and therefore less precise mental rules
>one executes to form a judgment or to make a decision) and in biased
>processing of presented information and the information that is
>available in memory (and presumably retrieved from there or not).
>Obviously, some moderators (motivation, need for cognition, involvment,
>need for justification, presence or absence of specific goals, .... not
>to mention relative size/change/difference of incoming stimuli that are
>being evaluated) do affect precision and sensitivity of any cognitive
>processing.
>
Both Marcus and "josko" are on to something here. This is an explanation for
the lower sensitivity of *blind* audio component testing using music versus
*sighted* testing. Both involve the brain functions "remember/compare" (*right
brain*). In both tests, this function can be ralaxed and take as long as the
listener wants.
But when you add "X" to the blind test, the listener is forced to go from
"remember/compare" to "match/chose/decide", a different brain function (*left
brain*) which interferes with the recall of subtle details in audible memory.
All of a sudden, the listener is under to pressure to make a decision within
micro-seconds before his audible memory fades. Memory of subtle details fade
first (timbre, dynamics, imaging, etc.) leaving only gross frequency response
and loudness differences recognizable.
Regards,
Mike
Harry Lavo
Yep, I do. But until and adequate control test is done to verify dbt abx
for this purpose, we are left with two flawed tests. And I choose the
sighted. You choose the abx.
Steven Sullivan
Sighted is *known* to be flawed and is not used in science.
ABX/DBT is *posited* to be flawed , by
you, though psychoacousticians seem OK with it. See the difference?
Nousaine
>>Prior knowledge -- whether accurate or not, and whether derived
>>from sense impression or from verbal or written communication -- can
>>bias cognitive evaluation of *current* sensory input.
>>A simple example is when listeners are *told* that a condition has
>>been changed, but it actually hasn't been.
>>There will be a strong bias to 'hear' a difference anyway. This is
>>true even if the visual input hasn't changed either.
>
>You are talking about misdirection. Yeah people are suseptable to
>misdirection.
What if the non-change was self-imposed and accidental, as in where one
discovers after the decision that the switch was in a different position than
supposed? Same results; same practical circumstance. Misdirection only works
when the stimulus is below the threshold of detection.
>One could say that sameness in components could also be imagined if the
>listener believes ahead of time that no difference exists. That means every
>ABX
>DBT that has been done with people who believe no difference exists between
>the
>components in question is tainted with expectation bias unless a control was
>used in the test for that bias such as the random insertion of a known barly
>audible distortion.>
The Candid Camera wine test shows that the nominal expectation bias of humans
is a false positive no matter what the sense. Even so, how do you insert a
"barely", or even "obvious", audible effect (such as wire sound) that has
never been shown to exist?
>>No, you do it by simply making one piece of information
>>*unavailable* to the listener at the time of audition -- namely, the
>>identity of the device or treatment currently playing. The only
>>cues to that, should be what is *heard*. Otherwise there's
>>a signficant chance the identification was made based on factors *other*
>>than the audible.
>
>Do you think you know what orange juice or
>cola or strawberry jello taste like? I bet you cannot successfully identify
>such items by taste alone on a reliable basis. Take the testee out of the
>envirement and all bets are off on sensitivity. The interaction of senses is
>complex and critical in sensitivty of the senses.
So blind people don't have a sense of taste when they aren't in their reference
kitchen?
Nousaine
>"Steven Sullivan" <ssu...@panix.com> wrote in message
...snip to content....
>>
>> Do you understand the need for something *other than* the sighted
>> comparison -- the method of choice in audiophila -- when the purpose
>> is to identify *audible* difference? Or, to put it another way, do you
>> understand that sighted comparison is *inadequate* for exactly that
>purpose,
>> even though it is used for that purpose routinely in audiophilia?
>
>Yep, I do. But until and adequate control test is done to verify dbt abx
>for this purpose, we are left with two flawed tests. And I choose the
>sighted. You choose the abx.
I'm wondering why anyone would choose a technique that cannot be tested for
sensitivity precisely because it requires isolation of stimulus to acoustical
cause (reduction to sound alone requires isolation of known bias mechanisms)
over anything else other than ease of deployment. It's sort of like saying that
estimating lap times at a track by "feel" is just as good as using a stopwatch.
To each his own.
Bob Marcus
>
>But when you add "X" to the blind test, the listener is forced to go from
>"remember/compare" to "match/chose/decide", a different brain function
>(*left
>brain*)
Gee, I wonder why Harry isn't challenging YOU for neurological evidence?
>which interferes with the recall of subtle details in audible memory.
>All of a sudden, the listener is under to pressure to make a decision
>within
>micro-seconds before his audible memory fades. Memory of subtle details
>fade
>first (timbre, dynamics, imaging, etc.) leaving only gross frequency
>response
>and loudness differences recognizable.
In an ABX test, you are never comparing more than two sounds at a time
(either A and X or B and X). That's why it is no more challenging mentally
than any A-B comparison, sighted or unsighted. The notion that deciding
whether A sounds different from B takes place in the right hemisphere in
sighted listening but in the left brain for blind listening is quite
preposterous.
For that matter, so is the assertion that memory of imaging fades faster
than other things. That's about the easiest thing to remember long-term. As
for timbre and dynamics, well, they're pretty closely related to frequency
response and loudness. I presume you have no evidence that memory of the
former "fades" at different rates.
bob
_________________________________________________________________
Check out the coupons and bargains on MSN Offers!
http://shopping.msn.com/softcontent/softcontent.aspx?scmId=1418
normanstrong
We're getting closer, I believe.
Suppose now you can see the 2 units under test; you know them well by
reputation. Let's say you're comparing an Adcom amp to a McIntosh.
You have a switch which is labelled A or B, A being the Adcom and B
the McIntosh. The only problem is that you cannot verify the actual
connections. You can look at the wires, but you don't know which
switch terminals are connected in the A position.
Now can you do a reliable sighted test?
Norm Strong
Steven Sullivan
> Mkuller wrote:
> >
> >But when you add "X" to the blind test, the listener is forced to go from
> >"remember/compare" to "match/chose/decide", a different brain function
> >(*left
> >brain*)
> Gee, I wonder why Harry isn't challenging YOU for neurological evidence?
> >which interferes with the recall of subtle details in audible memory.
> >All of a sudden, the listener is under to pressure to make a decision
> >within
> >micro-seconds before his audible memory fades. Memory of subtle details
> >fade
> >first (timbre, dynamics, imaging, etc.) leaving only gross frequency
> >response
> >and loudness differences recognizable.
> In an ABX test, you are never comparing more than two sounds at a time
> (either A and X or B and X). That's why it is no more challenging mentally
> than any A-B comparison, sighted or unsighted. The notion that deciding
> whether A sounds different from B takes place in the right hemisphere in
> sighted listening but in the left brain for blind listening is quite
> preposterous.
The whole 'right brain/left brain' dichotomy has become akin to adding
'quantum' to a phrase -- it's simplistic, often used inappropriately
and for commercial gain, and thus raises the pseudoscience flag for scientists.
See
http://www.rense.com/general2/rb.htm
for a recounting of a seemingly elegant 'left/right' experimental result for
visual processing, that turned out to be much messier than originally claimed.
"Whatever the story about lateralisation, simple dichotomies are out.
It is how the two sides of the brain complement
and combine that counts."
Nousaine
....snip to content .....
>But when you add "X" to the blind test, the listener is forced to go from
>"remember/compare" to "match/chose/decide", a different brain function (*left
>brain*) which interferes with the recall of subtle details in audible memory.
IMO the functions "choose, decide" are just one function. Or at a minimum his
open "remember/compare" has to have a decision function and should be
"remember/compare/choose." Otherwise open-sighted evaluation would be totally
useless for making decisions about deployment of audio gear.
Mkuller
>>But when you add "X" to the blind test, the listener is forced to go from
>>"remember/compare" to "match/chose/decide", a different brain function
>>(*left
>>brain*)
>
> "Bob Marcus" nab...@hotmail.com wrote:>
>Gee, I wonder why Harry isn't challenging YOU for neurological evidence?
>
He knew he could count on you to do that. When I say *right brain* brain and
*left brain*, I am speaking figuratively about their different functions. Like
"Women are from Venus and Men are from Mars". Come on, we're NOT really from
different planets...
>mkuller>
>>which interferes with the recall of subtle details in audible memory.
>>All of a sudden, the listener is under to pressure to make a decision
>>within
>>micro-seconds before his audible memory fades. Memory of subtle details
>>fade
>>first (timbre, dynamics, imaging, etc.) leaving only gross frequency
>>response
>>and loudness differences recognizable.
>
>Marcus>
>In an ABX test, you are never comparing more than two sounds at a time
>(either A and X or B and X). That's why it is no more challenging mentally
>than any A-B comparison, sighted or unsighted.
>
In a *sighted* listening test, you always know whether A or B is playing. You
NEVER have to "match/chose/decide" whether X is A or B. No *decision-making
brain function* involved to get in the way of recalling audible memory. That's
what differentiates *sighted* listening from *blind* listening. ?Comprende?
>Marcus>
>For that matter, so is the assertion that memory of imaging fades faster
>than other things. That's about the easiest thing to remember long-term.
>
Any evidence for that? Perhaps you could cite a published ABX test where
imaging was the deciding factor, rather than gross frequency response or
loudness.
> As
>for timbre and dynamics, well, they're pretty closely related to frequency
>response and loudness. I presume you have no evidence that memory of the
>former "fades" at different rates.
>
According to some objectivists, *everything* is related to frequency response.
Perhaps you could explain the relationship between dynamic range and
frequency response for me - I don't believe they are related. What about
phase, timing, transient response, decay, etc - still related to frequency
response?
I'm only claiming that memories of subtle audible differences fade more quickly
than memories of large, gross differences, whatever they may be. You have any
evidence to the contrary?
Regards,
Mike
Audio Guy
mku...@aol.com (Mkuller) writes:
>
> I'm only claiming that memories of subtle audible differences fade more quickly
> than memories of large, gross differences, whatever they may be. You have any
> evidence to the contrary?
OK, if subtle audible differences fade quickly, doesn't that validate
the findings of DBT'ers that quick switching is the best way to
determine differences?
Harry Lavo
news:Z4nTb.159122$Rc4.1251684@attbi_s54...
> "Harry Lavo" harry...@rcn.com wrote:
>
> >"Audio Guy" <audio...@yahoo.com> wrote
>
> ...snip to content ....
>
> >>
> >> For one who has had significant training in psychology, please
> >> explain how you can disparage DBTs when they are such a fundamental
> >> part of psychology and have been since the days of "Clever Hans".
> >>
> >
> >I have never "disparaged" DBT's. I have said they are often impractical
in
> >a home environment.
>
> Sure.
>
> I have also said that without a proper control test, I
> >am not sure that they are the best test technique for open-ended
evaluation
> >of audio components.
>
> But what control test do you have for 'sighted' (re: as uncontrolled)
testing
> other than that we already know (such as subjects will often report
difference
> when given identical audio presentations) other than it is obvious they
are
> prone to false positives? -- It seems that your only real argument for
them is
> they are easier to conduct.
>
Don't be clever here Tom. I've already spelled out in detail in this forum
a control test that would servce as such for both sighted and conventional
dbt testing. It iself was a dbt, but one that introduced many elements of
sighted listening, most fundamentally evaluative rather than comparative
listening. You and the other objectivists just go on ignoring it or the
need for such a test, and continue to pursue your strawment.
> >I spent a good part of my life sponsoring, helped designing, and
> >interpreting consumer research testing, often of a very sophisticated
> >nature. I understand the arguments. I also understand the need for a
> >control test to "prove" beyond a shadow of a doubt that quick switch dbts
> >(especially abx)" do not incur their own confusion factors that obscure
> >certain aspects of open-ended evaluation. So until the proponents of
such
> >testing prepare and present such a control test with some scientific
rigor,
> >I will stay with my views.
>
> Fair enough but IMO you're just choosing and defnding a method you
'prefer'
> rather than one that has been verified in sensitivity. Indeed it's quite
easy
> to show that open-listening is rife with false positives. That's OK by me;
but
> I'm not one to endorse a method just because it may be easier and
obviously
> lacking in validity for investigatting true acoustic effects.
>
Yes, but you also refuse to do a control test that might possibly show that
the dbt'ng you do might have some 'false negatives' so you too are just
making a preference. Only you hide it under the guise of "scientific
sensitivity". I think Mr. Radeckers post should cause you to give some
thought, Tom, to what I and others here have been trying to say all these
years.
Steven Sullivan
The first step of ABX is listening to A and B. If no difference is discerned between
A and B, then it's pointless to continue with the 'match/choose/decide'
And, actually, X/A or X/B is no different than 'remember/compare'.
> All of a sudden, the listener is under to pressure to make a decision within
> micro-seconds before his audible memory fades.
No, he's not. He can keep switching back and forth between A, B, and X for as
long as he likes. He *knows* X is either A or B, so there's no third possibility,
which *would* complicate matters.
> Memory of subtle details fade
> first (timbre, dynamics, imaging, etc.) leaving only gross frequency response
> and loudness differences recognizable.
More assertions without evidence.
Steven Sullivan
> >Prior knowledge -- whether accurate or not, and whether derived
> >from sense impression or from verbal or written communication -- can
> >bias cognitive evaluation of *current* sensory input.
> >A simple example is when listeners are *told* that a condition has
> >been changed, but it actually hasn't been.
> >There will be a strong bias to 'hear' a difference anyway. This is
> >true even if the visual input hasn't changed either.
> You are talking about misdirection.
Yes, as *one kind* of prior knowledge...the most extreme kind.
> Yeah people are suseptable to misdirection.
> One could say that sameness in components could also be imagined if the
> listener believes ahead of time that no difference exists.
One could imagine that, but perceptual psychology says the stronger
tendancy is to experience *difference*.
> That means every ABX
> DBT that has been done with people who believe no difference exists
> between the components in question is tainted with expectation bias unless a control was
> used in the test for that bias such as the random insertion of a known barly
> audible distortion.
Again, 'expectation bias' isn't necessarily conscious expectation...
it's not a matter of going into the test, saying to yourself, I *will*
hear a difference.
And, too, when are ABX tests performed to test a claim of 'no difference'?
All the ones I've sen reported involved people who claimed they could
hear a difference between A and B; if not, there would be no point in
continuing the test.
> >No, you do it by simply making one piece of information
> >*unavailable* to the listener at the time of audition -- namely, the
> >identity of the device or treatment currently playing. The only
> >cues to that, should be what is *heard*. Otherwise there's
> >a signficant chance the identification was made based on factors *other*
> >than the audible.
> Do you think you know what orange juice or
> cola or strawberry jello taste like? I bet you cannot successfully identify
> such items by taste alone on a reliable basis.
I'd require taste and smell, most likely. ANyone who's had a cold knows
that a stuffed nose reduces taste sensitivity. And that's been confirmed
medically, and the anatomical basis is known.
> Take the testee out of the
> envirement and all bets are off on sensitivity.
No, Scott. All bets aren't off. We know that certain
changes to the environment reduce sensitivity. Others could
be expected to have little or no effect. This stuff has
been studied.
> The interaction of senses is
> complex and critical in sensitivty of the senses.
In some cases, for some senses, yes.
But *that's beside the point anyway*. We're not talking about blocking
any of the senses. It's not analogous to stuffing someone's nose and
asking them to identify a flavor. We aren't talking about blindfolding the testee.
or dong the test in the dark. We aren't damping down the sense of sight.
We are only talking about keeping the identity unknown at the time of actual
listening. You can *SEE* both devices
under test, during the test, if you like. You just can't *know* which
one is playing. How could that possibly be construed as changing the
interaction of the senses?
If you have to *know* which one you are listening to,
in order to identify it reliably, how could you prove that thesis?
Harry Lavo
Open-ended sighted evaluation generally results in taking notes over a
period of time on both pieces of equipment, across several moods and
environmental factors. Then after an extended period of such evaluation
(extended being a function of what time frame is available...a weekend? ...
a week? ..a month? when the time "seems right" to make a decision,
evaluating those notes, reaching a tentative conclusion, perhaps going back
to one piece or another to check one factor or descrepancy vs. another, then
deciding.
Do you see a difference between the traditional dbt comparative testing and
sighted open-ended evaluation now?
Harry Lavo
Because the kind of sensitivity "control" test required to settle the issue
for open ended component evaluation has not been undertaken by you
proselytizers for your testing...and until you have proved that your tests
are valid for such evaluations by doing such a control, I will stick to my
evaluation of the consequences of the two error types - positive ("sighted,
slow, evaluative") vs. negative ("comparative quick-switch dbt"). I've
already outlined a test that would be that control in detail in this forum.
You are free to use it, no royalty fee. If it supports your testing, I will
become a convert. So if you really a) want to convert me and other like
minded audiophiles, and b) want to be absolutely sure your test is what you
think it is, then you will undertake such a test. Until then, as Stewart is
wont to say, it is just "gum flapping".
Mkuller
>> I'm only claiming that memories of subtle audible differences fade more
>quickly
>> than memories of large, gross differences, whatever they may be. You have
>any
>> evidence to the contrary?
>
>audio...@yahoo.com (Audio Guy) wrote:>
>OK, if subtle audible differences fade quickly, doesn't that validate
>the findings of DBT'ers that quick switching is the best way to
>determine differences?
>
For DBTs, you may be correct that quick switching is best. Using a music
program as a source, which almost all audiophiles do, you have a dynamic,
ever-changing program. Here's my experience - I'm listening to A, for example,
and as the program is changing, it switches to X. Now I have to compare my
memory of A with the different part of the program now that is playing with X.
I can remember how A sounded for perhaps a second, but since the program is
different, I'm not sure if there are any audible differences or not.
Quick switching? It only allows me to compare the parts of the program that
are the same - say the continuing vocal or cymbals - on the two components A
and X as the switch is made. But as the seconds tick down, the program is not
only different, but my memory has faded and I'm not sure I can remember what A
sounded like any more. Since I'm so busy trying to decide whether X is A or B,
I have difficulty remembering any overall gestalt or impressions, subtle
differences - anything but big differences that apparent at the moment of
switching - gross frequency response and loudness.
Regards,
Mike
Harry Lavo
> >Mkuller wrote:
> >>But when you add "X" to the blind test, the listener is forced to go
from
> >>"remember/compare" to "match/chose/decide", a different brain function
> >>(*left
> >>brain*)
> >
>
> > "Bob Marcus" nab...@hotmail.com wrote:>
> >Gee, I wonder why Harry isn't challenging YOU for neurological evidence?
> >
>
> He knew he could count on you to do that. When I say *right brain* brain
and
> *left brain*, I am speaking figuratively about their different functions.
Like
> "Women are from Venus and Men are from Mars". Come on, we're NOT really
from
> different planets...
>
Actually Jung's work indicates that human's have two main "functional axis",
one for "perceiving" and one for "judging". These two axis defined his
four developmental "functions" - "thinking and feeling" were polar opposites
defining the judgmental axis; "intuition and sensation" were the opposites
defining the percepual functions. Jung considered the mature and developed
adult as one who became aware of how these funtions worked within his own
makeup, learned which were strong, which were weak and needed to be
developed or "managed". And could use these functions as appropriate for
making judgements and decisions.
Later work in the field pretty strongly tied the perceptual functions to the
right brain, and the judgemental funtions to the left brain.
In evaluative listening you start with perceptual inputs, sensing and
intuitively perceiving, and then obtaining emotional or thinking judgemental
reaction to them. This work goes on and on, back and forth, until a stable,
well-evaluated set of perceptions/judgements emerges.
This "work" as Jung would call it, simply can't be done on a forced or
rushed basis without disasterous effect (in psychological terms). And as
Oohashi noted in his experiements, it takes finite time for the emotional
component to register in the brain if it is to do so. So it is not too
difficult to see why trying to evaluate components using quick switch,
short-segment comparisons (favoring the sensate over the intuitive) before
having to choose (favoring the thinking over the emotional judgement) might
skew results.
> >mkuller>
> >>which interferes with the recall of subtle details in audible memory.
> >>All of a sudden, the listener is under to pressure to make a decision
> >>within
> >>micro-seconds before his audible memory fades. Memory of subtle details
> >>fade
> >>first (timbre, dynamics, imaging, etc.) leaving only gross frequency
> >>response
> >>and loudness differences recognizable.
> >
>
> >Marcus>
> >In an ABX test, you are never comparing more than two sounds at a time
> >(either A and X or B and X). That's why it is no more challenging
mentally
> >than any A-B comparison, sighted or unsighted.
> >
>
> In a *sighted* listening test, you always know whether A or B is playing.
You
> NEVER have to "match/chose/decide" whether X is A or B. No
*decision-making
> brain function* involved to get in the way of recalling audible memory.
That's
> what differentiates *sighted* listening from *blind* listening.
?Comprende?
>
Actually, Mike, I think sighted testing usually results in more perceptual
evaluation before getting to the judging part, and that it is this factor
that makes the main difference.
> >Marcus>
> >For that matter, so is the assertion that memory of imaging fades faster
> >than other things. That's about the easiest thing to remember long-term.
> >
>
> Any evidence for that? Perhaps you could cite a published ABX test where
> imaging was the deciding factor, rather than gross frequency response or
> loudness.
>
Imaging is definitely perceptual and almost solely a function of what Jung
called the intuitive function ("pattern recognition" is the essence of his
definition). And intuition when attempting to be forced simply
dissappears.
>snip, nothing useful to add to the remaining conversation<
Harry Lavo
No, because those subtle differences often take time to be recognized and
enter consciousness.
Steven Sullivan
> In a *sighted* listening test, you always know whether A or B is playing. You
> NEVER have to "match/chose/decide" whether X is A or B. No *decision-making
> brain function* involved to get in the way of recalling audible memory. That's
> what differentiates *sighted* listening from *blind* listening. ?Comprende?
But any attempt at *comparison*, whether sighted or not, will involve decision
making. And no sincere claim of difference can have arisen without a mental
attempt at comparison. Capice?
jjn...@sonic.net
> In a *sighted* listening test, you always know whether A or B is playing. You
> NEVER have to "match/chose/decide" whether X is A or B. No *decision-making
> brain function* involved to get in the way of recalling audible memory. That's
> what differentiates *sighted* listening from *blind* listening. ?Comprende?
IF no difference can be detected, what on EARTH is one making a decision about?
Comprende indeed.
Audio Guy
mku...@aol.com (Mkuller) writes:
>> mku...@aol.com (Mkuller) writes:>
>>> I'm only claiming that memories of subtle audible differences fade more
>>quickly
>>> than memories of large, gross differences, whatever they may be. You have
>>any
>>> evidence to the contrary?
>>
>
>>audio...@yahoo.com (Audio Guy) wrote:>
>>OK, if subtle audible differences fade quickly, doesn't that validate
>>the findings of DBT'ers that quick switching is the best way to
>>determine differences?
>>
>
> For DBTs, you may be correct that quick switching is best. Using a music
> program as a source, which almost all audiophiles do, you have a dynamic,
> ever-changing program. Here's my experience - I'm listening to A, for example,
> and as the program is changing, it switches to X. Now I have to compare my
> memory of A with the different part of the program now that is playing with X.
> I can remember how A sounded for perhaps a second, but since the program is
> different, I'm not sure if there are any audible differences or not.
What does "it switches to X" mean? With ABX, the control of the switch
is in the users hands, he can switch wwhenver they want to, days or
weeks later if desired. Quick switching is not a requirement of ABX no
matter how many times the anti-ABXers keep saying it and refuse to
acknowledge it.
> Quick switching? It only allows me to compare the parts of the program that
> are the same - say the continuing vocal or cymbals - on the two components A
> and X as the switch is made. But as the seconds tick down, the program is not
> only different, but my memory has faded and I'm not sure I can remember what A
> sounded like any more. Since I'm so busy trying to decide whether X is A or B,
> I have difficulty remembering any overall gestalt or impressions, subtle
> differences - anything but big differences that apparent at the moment of
> switching - gross frequency response and loudness.
So back up the recording and re-listen to that part over and over
again.
But again, you are the one who said "memories of subtle audible
differences fade more quickly" above, so how does that jibe with
open-ended long term evealuations? And how can one arrive at an
"overall gestalt" if "memories of subtle audible differences fade
more quickly"? You can't have it both ways.
Audio Guy
But that is not what Mike said, and that is my point. How can
"memories of subtle audible differences fade more quickly" as Mike
posted above, yet "take time to be recognized and enter
consciousness"? I really don't see how both can be true.
Audio Guy
"Harry Lavo" <harry...@rcn.com> writes:
>
> Because the kind of sensitivity "control" test required to settle the issue
> for open ended component evaluation has not been undertaken by you
> proselytizers for your testing...and until you have proved that your tests
> are valid for such evaluations by doing such a control, I will stick to my
> evaluation of the consequences of the two error types - positive ("sighted,
> slow, evaluative") vs. negative ("comparative quick-switch dbt").
Please, please show where a DBT requires quick switching? It only
requires a lack of knowledge of which item is playing. An ABX device
allows one to listen back and forth to A and B which are known and
only decide which is X and some later time, years later if that is
your wish.
I think it's time for this strawman to be dropped permanently form
these DBT discussions.
S888Wheel
>
>Yes, as *one kind* of prior knowledge...the most extreme kind.
>
And being the most extreme kind it is likely to wrought the most extreme kind
of results.
>
>> Yeah people are suseptable to misdirection.
>> One could say that sameness in components could also be imagined if the
>> listener believes ahead of time that no difference exists.
>
>One could imagine that, but perceptual psychology says the stronger
>tendancy is to experience *difference*.
>
I would like to see a citation of any study in perceptual psychology that
suggests this is true. I'll bet if it has been studied the studies would show
the pre-existing bias would prove far more powerful than any universal tendency
to percieve a difference. People, by and large, go through life not noticing
differences that do exist every bit as much as they percieve differences that
don't exist. But like coincidences vs. noncoincidences the failure to note
small but real differences in our sensery perception goes largely unnoticed.
>
>> That means every ABX
>> DBT that has been done with people who believe no difference exists
>> between the components in question is tainted with expectation bias unless
>a control was
>> used in the test for that bias such as the random insertion of a known
>barly
>> audible distortion.
>
>Again, 'expectation bias' isn't necessarily conscious expectation...
I don't believe that. Claims of the effects of the "subconscious" mind on the
conscious mind are highly debated. If you can cite any scientific claims that
"subconscious" biases are at work in the effects of expectation bias I'd like
to read about it. I am betting that none of the studdies on the effects of
expectation bias make any claims that the expectation bias is subconscious.
>it's not a matter of going into the test, saying to yourself, I *will*
>hear a difference.
I think it very much is. That is why I think deliberate misdirection is an
obvious flaw in testing for normal expectation bias. The deck is stacked.
>
>And, too, when are ABX tests performed to test a claim of 'no difference'?
Every time Tom does such a test.
>All the ones I've sen reported involved people who claimed they could
>hear a difference between A and B; if not, there would be no point in
>continuing the test.
You are not a reference for recording all such tests.
>
>> Do you think you know what orange juice or
>> cola or strawberry jello taste like? I bet you cannot successfully
>identify
>> such items by taste alone on a reliable basis.
>
>I'd require taste and smell, most likely. ANyone who's had a cold knows
>that a stuffed nose reduces taste sensitivity. And that's been confirmed
>medically, and the anatomical basis is known.
I figured that was understood. I was refering to literal blindfolding. Try it
with amny differnt samples and see how well you do.
>
>> Take the testee out of the
>> envirement and all bets are off on sensitivity.
>
>No, Scott. All bets aren't off. We know that certain
>changes to the environment reduce sensitivity. Others could
>be expected to have little or no effect. This stuff has
>been studied.
I am sure it has been studied. I am not so sure it has been accurately reported
on RAHE.
>
>> The interaction of senses is
>> complex and critical in sensitivty of the senses.
>
>In some cases, for some senses, yes.
>
>But *that's beside the point anyway*. We're not talking about blocking
>any of the senses. It's not analogous to stuffing someone's nose and
>asking them to identify a flavor. We aren't talking about blindfolding the
>testee.
>or dong the test in the dark. We aren't damping down the sense of sight.
>We are only talking about keeping the identity unknown at the time of actual
>
>listening. You can *SEE* both devices
>under test, during the test, if you like. You just can't *know* which
>one is playing. How could that possibly be construed as changing the
>interaction of the senses?
I was talking about deliberate misdirection.
S888Wheel
>>misdirection.
>
>What if the non-change was self-imposed and accidental, as in where one
>discovers after the decision that the switch was in a different position than
>supposed?
Whether misdirection is accidental or done on purpose it is still misdirection.
I would suspect your tests using misdirection are quite incomplete. I suspect
if you had done the test twice,once with the specific directions stating that
there may or may not be an audible difference when the switch is made you would
have eliminated the effect of deliberate misdirection. If the results of such a
control test were the same as your test which used deliberate misdirection you
could then claim that the misdirection was not a factor.Certainly if you tell
the testees that there will be a difference they are more likely to to for and
find one that doesn't exist. Like I said, I bet you could easily get the same
mistakes from people who expect no difference will be present when one actually
is present.
>
>>One could say that sameness in components could also be imagined if the
>>listener believes ahead of time that no difference exists. That means every
>>ABX
>>DBT that has been done with people who believe no difference exists between
>>the
>>components in question is tainted with expectation bias unless a control was
>>used in the test for that bias such as the random insertion of a known barly
>>audible distortion.>
>
>The Candid Camera wine test shows that the nominal expectation bias of humans
>is a false positive no matter what the sense.
I don't know anything about this test. But your test involved deliberate
misdirection. I don't think it proves much. We know that people can be fooled
by magicians.
>Even so, how do you insert a
>"barely", or even "obvious", audible effect (such as wire sound) that has
>never been shown to exist?
You don't. You insert one that has been shown to exist. Maybe you didn't
understand what I was proposing.
>
>>>No, you do it by simply making one piece of information
>>>*unavailable* to the listener at the time of audition -- namely, the
>>>identity of the device or treatment currently playing. The only
>>>cues to that, should be what is *heard*. Otherwise there's
>>>a signficant chance the identification was made based on factors *other*
>>>than the audible.
>>
>>Do you think you know what orange juice or
>>cola or strawberry jello taste like? I bet you cannot successfully identify
>>such items by taste alone on a reliable basis. Take the testee out of the
>>envirement and all bets are off on sensitivity. The interaction of senses
>is
>>complex and critical in sensitivty of the senses
>
>So blind people don't have a sense of taste when they aren't in their
>reference
>kitchen?
Of course they do if they have enough experience tasting things as a blind
person. I guess you have never tried to identify things by taste while
blindfolded. Try it with many different samples and see how well you fare.
Mkuller
>>But when you add "X" to the blind test, the listener is forced to go from
>>"remember/compare" to "match/chose/decide", a different brain function
>(*left
>>brain*) which interferes with the recall of subtle details in audible
>memory.
>
>nous...@aol.com (Nousaine) wrote:>
> IMO the functions "choose, decide" are just one function. Or at a minimum
>his
>open "remember/compare" has to have a decision function and should be
>"remember/compare/choose." Otherwise open-sighted evaluation would be totally
>useless for making decisions about deployment of audio gear.
>
No, there is no decision *required* in sighted listening. Only in blind
listening is the decision process an integral function of the test. Making
decision about deployment of audio gear are completely independedent of the two
types of listening tests. Once either type of tests are completed, then those
decisions can be made, or not.
Regards,
Mike
chung
While Mr. Kuller said that subtle differences fade more quickly. So
according to him, if you listen to A first for an extended period so
that those "things" can enter consciousness, then B for an extended
time, you are likely to forget those things you heard from A, so that it
is no longer possible to determine differences. Now, which of the two of
you is right?
Steven Sullivan
er...how on earth does thsi work, Harry? You listen to A for as long as you
like. You listen to B for a long as you like. You switch between them
as often as you like, until the subtle differences are recognized and
enter your consciousness.
'Quick switching' merely means that when you *do* want
to switch between A and B (or X and A/B), the actual switching occurs
'instantanously', rather than with a long lag of no sound. You can *insert*
such a lag into the process if you like, if that's what you think it
takes for the differences to be recognized and enter consciousness.
Mkuller seems to agree that inserting such a lag would promote the loss of
accurate memory of subtle audible difference.
Why do 'objectivists' have to keep explaining this stuff to you guys over
and over and over?
Steven Sullivan
> > mku...@aol.com (Mkuller) wrote:>
> >>But when you add "X" to the blind test, the listener is forced to go from
> >>"remember/compare" to "match/chose/decide", a different brain function
> >(*left
> >>brain*) which interferes with the recall of subtle details in audible
> >memory.
> >
> >nous...@aol.com (Nousaine) wrote:>
> > IMO the functions "choose, decide" are just one function. Or at a minimum
> >his
> >open "remember/compare" has to have a decision function and should be
> >"remember/compare/choose." Otherwise open-sighted evaluation would be totally
> >useless for making decisions about deployment of audio gear.
> >
> No, there is no decision *required* in sighted listening.
Wow. The process of sighted listening seems to be getting more and more rarified
as you guys keep describing it. It's approaching some form of Zen meditation.
And in the process bearing less and less resemblance to anything practiced
by audiophile reviewers
So, if I have it straight, a listener best perceives difference when he
isn't really trying to? And trying to, only confuses the issue? Wouldn't
this be jsut as true , then, if the listener used teh 'decision making
process' under *sighted* conditions?
Zen Master is sitting in his easy chair, listening to a system with a certain
brand of cable in it. Months of this go by. Then he switches the cables.
While keeping his mind utterly aloof from any conscious comparison-thoughts,
he begins to perceive an audible difference. The skeptical young monk
in his retinue wonders if it si real. He asks Zen Master to listen to both
again, still under sighted conditions, but consciously comparing them this time.
Zen Master can no longer tell the differecce, and whacks the monk over
the head with his cane.
Does this sound plausible to anyone?
> Only in blind
> listening is the decision process an integral function of the test. Making
> decision about deployment of audio gear are completely independedent of the two
> types of listening tests. Once either type of tests are completed, then those
> decisions can be made, or not.
> Regards,
> Mike
--
Bob Marcus
>
>Open-ended sighted evaluation generally results in taking notes over a
>period of time on both pieces of equipment, across several moods and
>environmental factors. Then after an extended period of such evaluation
>(extended being a function of what time frame is available...a weekend? ...
>a week? ..a month? when the time "seems right" to make a decision,
>evaluating those notes, reaching a tentative conclusion, perhaps going back
>to one piece or another to check one factor or descrepancy vs. another,
>then
>deciding.
>
>Do you see a difference between the traditional dbt comparative testing and
>sighted open-ended evaluation now?
Sure: Unlike DBTs, which follow the same approach used by all researchers on
human hearing perception, sighted open-ended evaluation has no empirical
basis whatsoever.
bob
_________________________________________________________________
Scope out the new MSN Plus Internet Software — optimizes dial-up to the max!
http://join.msn.com/?pgmarket=en-us&page=byoa/plus&ST=1
Steven Sullivan
> > mku...@aol.com (Mkuller) writes:>
> >> I'm only claiming that memories of subtle audible differences fade more
> >quickly
> >> than memories of large, gross differences, whatever they may be. You have
> >any
> >> evidence to the contrary?
> >
> >audio...@yahoo.com (Audio Guy) wrote:>
> >OK, if subtle audible differences fade quickly, doesn't that validate
> >the findings of DBT'ers that quick switching is the best way to
> >determine differences?
> >
> For DBTs, you may be correct that quick switching is best. Using a music
> program as a source, which almost all audiophiles do, you have a dynamic,
> ever-changing program. Here's my experience - I'm listening to A, for example,
> and as the program is changing, it switches to X. Now I have to compare my
> memory of A with the different part of the program now that is playing with X.
> I can remember how A sounded for perhaps a second, but since the program is
> different, I'm not sure if there are any audible differences or not.
How do you manage to compare stuff *sighted*? I'm sure that every aspect of
your method can be replicated in an ABX, except for the part about knowing the
identity of the currently-playing source X.
> Quick switching? It only allows me to compare the parts of the program that
> are the same - say the continuing vocal or cymbals - on the two components A
> and X as the switch is made. But as the seconds tick down, the program is not
> only different, but my memory has faded and I'm not sure I can remember what A
> sounded like any more. Since I'm so busy trying to decide whether X is A or B,
> I have difficulty remembering any overall gestalt or impressions, subtle
> differences - anything but big differences that apparent at the moment of
> switching - gross frequency response and loudness.
Again, how do you manage to compare anything sighted, if this is such an issue?
In a sighted comparison, everything you have claimed about comparing parts and about
memory, holds exactly the same as it does for ABX.
How do you decide , sighted , that two audio presentations sound different?
Steven Sullivan
> "Nousaine" <nous...@aol.com> wrote in message
> news:A7xTb.167643$nt4.750873@attbi_s51...
> > mku...@aol.com (Mkuller) wrote:
> >
> > ....snip to content .....
> >
> > >But when you add "X" to the blind test, the listener is forced to go from
> > >"remember/compare" to "match/chose/decide", a different brain function
> (*left
> > >brain*) which interferes with the recall of subtle details in audible
> memory.
> >
> > IMO the functions "choose, decide" are just one function. Or at a minimum
> his
> > open "remember/compare" has to have a decision function and should be
> > "remember/compare/choose." Otherwise open-sighted evaluation would be
> totally
> > useless for making decisions about deployment of audio gear.
> Open-ended sighted evaluation generally results in taking notes over a
> period of time on both pieces of equipment, across several moods and
> environmental factors.
Whoa..on the one hand we have one camp of subjectivist implying tha
any change from the normal environment invalidates a controlled test...
on the other we have you saying that a proper test invovles listening across
several moods and environments...
Fortunately, I don't expect all subjectivists to agree on every point,
any more than you should expect all objectivists to be an accord about
every point.
Btw, there's no reason DBT/ABX can't be done using 'note taking over
a period of time' on both DUTs...I'm pretty sure tey have been done under jsut
such conditions.
> Then after an extended period of such evaluation
> (extended being a function of what time frame is available...a weekend? ...
> a week? ..a month? when the time "seems right" to make a decision,
> evaluating those notes, reaching a tentative conclusion, perhaps going back
> to one piece or another to check one factor or descrepancy vs. another, then
> deciding.
Wasn't one of Tom's tests conducted via mail, over a period of weeks?
> Do you see a difference between the traditional dbt comparative testing and
> sighted open-ended evaluation now?
Do you see that the *documented* invalidating factor in the latter, from a
scientific POV, is the 'sighted' part? You can toss those results out.
Mkuller
>> Both Marcus and "josko" are on to something here. This is an explanation
>for
>> the lower sensitivity of *blind* audio component testing using music versus
>> *sighted* testing. Both involve the brain functions "remember/compare"
>(*right
>> brain*). In both tests, this function can be ralaxed and take as long as
>the
>> listener wants.
>
> But when you add "X" to the blind test, the listener is forced to go from
>> "remember/compare" to "match/chose/decide", a different brain function
>(*left
>> brain*) which interferes with the recall of subtle details in audible
>memory.
>
>Steven Sullivan ssu...@panix.com wrote:>
>The first step of ABX is listening to A and B. If no difference is discerned
>between
>A and B, then it's pointless to continue with the 'match/choose/decide'
Right, so far.
>And, actually, X/A or X/B is no different than 'remember/compare'.
>
The difference is that in a DBT you are *forced* to chose A or B for X.
mkuller
>> All of a sudden, the listener is under to pressure to make a decision
>within
>> micro-seconds before his audible memory fades.
>
Sullivan
>No, he's not. He can keep switching back and forth between A, B, and X for
>as
>long as he likes. He *knows* X is either A or B, so there's no third
>possibility,
>which *would* complicate matters.
>
If he keeps switching back and forth between A and B and never tries to match X
with one of them, then it's not and ABX test is it?
mkuller
>> Memory of subtle details fade
>> first (timbre, dynamics, imaging, etc.) leaving only gross frequency
>response
>> and loudness differences recognizable.
>
Sullivan
>More assertions without evidence.
>
The evidence is that no published DBT I know of has ever identified anything
but gross frequency response and loudness differences between audio components.
But most of us know there are many more differences than just those. Or are
you one of the objectivists who says that all of the subtle audible differences
are imagined and don't really exist *because* they've never been shown in a
DBT?
Regards,
Mike
Mkuller
>>> > I'm only claiming that memories of subtle audible differences fade more
>> quickly
>>> > than memories of large, gross differences, whatever they may be. You
>> have any
>>> > evidence to the contrary?
>>>
I'm *claiming* that they must fade more quickly or they would show up a
differences in DBTs. They never do.
>audio...@yahoo.com (Audio Guy) wrote:>
>>> OK, if subtle audible differences fade quickly, doesn't that validate
>>> the findings of DBT'ers that quick switching is the best way to
>>> determine differences?
>>>
>>
>"Harry Lavo" <harry...@rcn.com> writes:>
>> No, because those subtle differences often take time to be recognized and
>> enter consciousness.
>
>audio...@yahoo.com> wrote:>
>But that is not what Mike said, and that is my point. How can
>"memories of subtle audible differences fade more quickly" as Mike
>posted above, yet "take time to be recognized and enter
>consciousness"? I really don't see how both can be true.
>
Why not? They make take longer for the brain to process and to be consciously
recognized. And they made fade more quickly in the forced decision of a DBT,
too. The two aren't mutually exclusive.
Regards,
Mike
Steven Sullivan
> >Mkuller <mku...@aol.com> wrote:>
> >> Both Marcus and "josko" are on to something here. This is an explanation
> >for
> >> the lower sensitivity of *blind* audio component testing using music versus
> >> *sighted* testing. Both involve the brain functions "remember/compare"
> >(*right
> >> brain*). In both tests, this function can be ralaxed and take as long as
> >the
> >> listener wants.
> >
> > But when you add "X" to the blind test, the listener is forced to go from
> >> "remember/compare" to "match/chose/decide", a different brain function
> >(*left
> >> brain*) which interferes with the recall of subtle details in audible
> >memory.
> >
> >Steven Sullivan ssu...@panix.com wrote:>
> >The first step of ABX is listening to A and B. If no difference is discerned
> >between
> >A and B, then it's pointless to continue with the 'match/choose/decide'
> Right, so far.
> >And, actually, X/A or X/B is no different than 'remember/compare'.
> >
> The difference is that in a DBT you are *forced* to chose A or B for X.
Only in the sense that if you want to compare components...at some point,
you actually have to do the comparison and make a call, don't you?
I suspect you are overinterpreting the word 'force' in 'forced choice'.
It does not imply a time-critical , sweat-inducing task.
> mkuller
> >> All of a sudden, the listener is under to pressure to make a decision
> >within
> >> micro-seconds before his audible memory fades.
> >
> Sullivan
> >No, he's not. He can keep switching back and forth between A, B, and X for
> >as
> >long as he likes. He *knows* X is either A or B, so there's no third
> >possibility,
> >which *would* complicate matters.
> >
> If he keeps switching back and forth between A and B and never tries to match X
> with one of them, then it's not and ABX test is it?
Indeed. It's a sighted A/B *comparison*. It still involves comparing and
contrasting two presentations separated in time. Unfortunately, it's
irremediably flawed as a method for validating audible difference. Not
because it's comparison, but because it's 'sighted'.
> mkuller
> >> Memory of subtle details fade
> >> first (timbre, dynamics, imaging, etc.) leaving only gross frequency
> >response
> >> and loudness differences recognizable.
> >
> Sullivan
> >More assertions without evidence.
> >
> The evidence is that no published DBT I know of has ever identified anything
> but gross frequency response and loudness differences between audio components.
Are they 'gross' because they were detectable under DBT? Or is there some other
criterion?
> But most of us know there are many more differences than just those.
No. You *assert* that. The problem is, you don't *know*.
chung
>> mku...@aol.com (Mkuller) writes:>
>>>> > I'm only claiming that memories of subtle audible differences fade more
>>> quickly
>>>> > than memories of large, gross differences, whatever they may be. You
>>> have any
>>>> > evidence to the contrary?
>>>>
>
> I'm *claiming* that they must fade more quickly or they would show up a
> differences in DBTs. They never do.
So your claim lies in the fact that the differencse never showed up in a
DBT. And the possibility that there may be *no* subtle differences just
totally escapes you?
In other words, you are saying that if the test does not give you the
results you want, the test is flawed. In other words, you firmly believe
that those subtle differences exist, and you're disputing the validity
of DBT or ABX based on that belief.
What's wrong with this picture?
>
>>audio...@yahoo.com (Audio Guy) wrote:>
>>>> OK, if subtle audible differences fade quickly, doesn't that validate
>>>> the findings of DBT'ers that quick switching is the best way to
>>>> determine differences?
>>>>
>>>
>
>>"Harry Lavo" <harry...@rcn.com> writes:>
>>> No, because those subtle differences often take time to be recognized and
>>> enter consciousness.
>>
>
>>audio...@yahoo.com> wrote:>
>>But that is not what Mike said, and that is my point. How can
>>"memories of subtle audible differences fade more quickly" as Mike
>>posted above, yet "take time to be recognized and enter
>>consciousness"? I really don't see how both can be true.
>>
>
> Why not? They make take longer for the brain to process and to be consciously
> recognized. And they made fade more quickly in the forced decision of a DBT,
> too. The two aren't mutually exclusive.
Don't you listen to A and then B in a sighted test? Don't those sublte
differences fade more quickly, too? Are you saying that if you are not
forced to make a decision, then those differences will remain in your
head longer?
Let's help you remove the stress of having to make a decision. You can
take as long as you like to make that decision, or if you find it
difficult, just toss a coin, and go to the next trial. Would that
relieve any anxiety that you might have?
> Regards,
> Mike
>
Steven Sullivan
> "Mkuller" <mku...@aol.com> wrote in message
> news:bvm5k...@enews1.newsguy.com...
> > >Mkuller wrote:
> > >>But when you add "X" to the blind test, the listener is forced to go
> from
> > >>"remember/compare" to "match/chose/decide", a different brain function
> > >>(*left
> > >>brain*)
> > >
> >
> > > "Bob Marcus" nab...@hotmail.com wrote:>
> > >Gee, I wonder why Harry isn't challenging YOU for neurological evidence?
> > >
> >
> > He knew he could count on you to do that. When I say *right brain* brain
> and
> > *left brain*, I am speaking figuratively about their different functions.
> Like
> > "Women are from Venus and Men are from Mars". Come on, we're NOT really
> from
> > different planets...
> >
> Actually Jung's work indicates that human's have two main "functional axis",
> one for "perceiving" and one for "judging".
Jung's work, like all psychoanalysis, is considered little more than sheer
speculative pseudoscience by modern researchers.
> Actually, Mike, I think sighted testing usually results in more perceptual
> evaluation before getting to the judging part, and that it is this factor
> that makes the main difference.
Harry, ABX involves perceiving a difference between A and B first.
That takes care of the 'perceptual evaluation' part. It's onlly after
A and B sound different to the listener, that they're asked to
identify X, which they *know* is either A or B..
> Imaging is definitely perceptual and almost solely a function of what Jung
> called the intuitive function ("pattern recognition" is the essence of his
> definition). And intuition when attempting to be forced simply
> dissappears.
Your citing of Jung as an authority on perceptual psychology and brain
function is, shall we say, unlikely to impress someone whose read any
of the scientific literature published in the past forty years or so.
Steven Sullivan
> >> You are talking about misdirection.
> >
> >Yes, as *one kind* of prior knowledge...the most extreme kind.
> >
> And being the most extreme kind it is likely to wrought the most extreme kind
> of results.
Even if less extreme kinds yield 'less extreme' results ...which is not
conceded...do you agree that prior knowledge still biases sighted
comparison to some degree?
> >> Yeah people are suseptable to misdirection.
> >> One could say that sameness in components could also be imagined if the
> >> listener believes ahead of time that no difference exists.
> >
> >One could imagine that, but perceptual psychology says the stronger
> >tendancy is to experience *difference*.
> >
> I would like to see a citation of any study in perceptual psychology that
> suggests this is true. I'll bet if it has been studied the studies would show
> the pre-existing bias would prove far more powerful than any universal tendency
> to percieve a difference. People, by and large, go through life not noticing
> differences that do exist every bit as much as they percieve differences that
> don't exist.
I wold like to see a citation of any study in perceptual psychology that
suggests this is true.
> But like coincidences vs. noncoincidences the failure to note
> small but real differences in our sensery perception goes largely unnoticed.
Not when a person is comparing two things. There, the tendancy is to
report difference...perhaps it's of a piece with the 'coincidence' thing,
where peopel tend to impart *meaning* to things merely based
on temporal proximity or some other possibly spurious 'connection'.
IIRC such errors of judgement are covered in the book
Inevitable Illusions : How Mistakes of Reason Rule Our Minds
by Massimo Piattelli-Palmarini, a professor of cognitive science at UAZ.
> >> That means every ABX
> >> DBT that has been done with people who believe no difference exists
> >> between the components in question is tainted with expectation bias unless
> >a control was
> >> used in the test for that bias such as the random insertion of a known
> >barly
> >> audible distortion.
> >
> >Again, 'expectation bias' isn't necessarily conscious expectation...
> I don't believe that. Claims of the effects of the "subconscious" mind on the
> conscious mind are highly debated. If you can cite any scientific claims that
> "subconscious" biases are at work in the effects of expectation bias I'd like
> to read about it. I am betting that none of the studdies on the effects of
> expectation bias make any claims that the expectation bias is subconscious.
> >it's not a matter of going into the test, saying to yourself, I *will*
> >hear a difference.
> I think it very much is. That is why I think deliberate misdirection is an
> obvious flaw in testing for normal expectation bias. The deck is stacked.
Scott, controlled comparison for audible difference does not usually
involve *deliberate misdirection*, so please abandon this straw man
argument immediately. I merely used that example as a pretty foolproof way
to highlight the inherent tendency towards *hearing* difference when some
other sort of difference exists (either real or imagined).
> >And, too, when are ABX tests performed to test a claim of 'no difference'?
> Every time Tom does such a test.
When has Tom tested a claim of *no difference*, rather than a claim of
*difference*?
> >All the ones I've sen reported involved people who claimed they could
> >hear a difference between A and B; if not, there would be no point in
> >continuing the test.
> You are not a reference for recording all such tests.
You aren't a reference for scientifically-founded ideas. It doesn't stop
you from proclaiming your lack of belief in them as if it were some sort of
dispositive argument.
> >> Do you think you know what orange juice or
> >> cola or strawberry jello taste like? I bet you cannot successfully
> >identify
> >> such items by taste alone on a reliable basis.
> >
> >I'd require taste and smell, most likely. ANyone who's had a cold knows
> >that a stuffed nose reduces taste sensitivity. And that's been confirmed
> >medically, and the anatomical basis is known.
> I figured that was understood. I was refering to literal blindfolding. Try it
> with amny differnt samples and see how well you do.
We aren't talking about literal blindfolding of ABX /DBT testees, ever.
We aren't talking about blocking any of the senses.
> >> Take the testee out of the
> >> envirement and all bets are off on sensitivity.
> >
> >No, Scott. All bets aren't off. We know that certain
> >changes to the environment reduce sensitivity. Others could
> >be expected to have little or no effect. This stuff has
> >been studied.
> I am sure it has been studied. I am not so sure it has been accurately reported
> on RAHE.
See above re : dispositive arguments. OF cousre, you could always do what
has been suggested many times to you: take the time to go do the research
in your local university library to your own satisfaction, and report back
to us.
Piatelli-Palmarini's book, btw, is avaialble cheaply from amazon.com
> >> The interaction of senses is
> >> complex and critical in sensitivty of the senses.
> >
> >In some cases, for some senses, yes.
> >
> >But *that's beside the point anyway*. We're not talking about blocking
> >any of the senses. It's not analogous to stuffing someone's nose and
> >asking them to identify a flavor. We aren't talking about blindfolding the
> >testee.
> >or dong the test in the dark. We aren't damping down the sense of sight.
> >We are only talking about keeping the identity unknown at the time of actual
> >
> >listening. You can *SEE* both devices
> >under test, during the test, if you like. You just can't *know* which
> >one is playing. How could that possibly be construed as changing the
> >interaction of the senses?
> I was talking about deliberate misdirection.
Deliberate misdirection doesn't necessarily involve hindering any senses.
If two cables or amps or CD players -- for giggles, let's make one a
hgih-end brand and the other a mass-market brand -- are hooked up to a
switchbox, with the switch and the components *always in view*, and the
proctor says the 'up' position opf the switch is DUT A and the 'down'
position is DUT B, when in fact both positions are DUT A, it's an
excellent bet that the golden ear testee will hear big differences between
the switch positions. Where have any of the senses been thwarted in this
deliberate misdirection protocol?
josko
news:bvgp4...@enews4.newsguy.com...
[text cut]
> I would like to respond to your comments that you can't find any
"moderator"
> that would justify sighted testing over blind testing in trying to
something
> "by ear". If you have participated here long, you doubtless are aware
of
> the practical difficulties of doing a truly "tight" blind test at
home. It
> is primarily for this reason that most audiophiles rely on sighted
listening
> when making component comparisons. So given that very practical
constraint,
> don't your "moderators" actually come into play?. I would suggest an
> audiophile truly wanting to evaluate the equipment and not being
easily able
> to do it blind can still have:
>
> * motivation to want to get an accurate reading (e.g. is honestly
searching
> for improvement)
Motivation biases toward finding a difference (finding a unique sonic
signature if you wish).
> * cognitive need ("awareness", taking notes, analyzing the sound,
pondering,
> "gestalt listening"etc.)
> * involvement (it is a hobby, after all, so not life and death, but
> presumably an important evaluation)
> * presence of a specific goal (to fix whatever he feels could be/needs
to be
> improved in the system; to focus on same)
All these bias toward finding a difference.
> * presence of a size/change/difference sufficient to be actually
audible
> with careful listening.
>
Theory (and literature) on these issues exist. Shell we ignore it
completely or unimaginatively question all of it all the time?
Questioning is OK if it is logical and intellectually honest. What some
subjectivists do is that they question an entire body of scientific work
without proposing anything of value. Now, your proposal is to use
evaluative listening and monitoring own emotions to judge audio
equipment. IMO, problematic strategy. Why?
In general, you claim that you look for a "gestalt" experience when you
evaluate components (gestalt listening is what you called it). How do
you cognitively reconcile "gestalt" listening with all "piecemeal"
processing that a typical audiophile, including you, does when he
critically listens to a piece of music reproduced with a new cable or a
new amplifier in your system? Piecemeal processing would include
processing of things like inner-detail resolution, bass speed, extension
and definition, soundstage depth and width, treble sweetness, air,
imaging, "midband plasticity", timbre,..... Do you really believe that a
valid cognitive mechanism that would explain differences that
audiophiles hear between two (nominally competent) cables or amplifiers
would be something like 1) piecemeal processing of individual
properties of the reproduced sound (soundstage, imaging, et al.), 2)
critical, and therefore conscious, evaluation of these individual
properties (which is something that all serious audiophiles do according
to their testimonials), 3) conscious integration of these pieces of
information into a gestalt, and 4) conscious monitoring of emotions in
reaction to the mentally constructed gestalt? Even when we disregard a
plethora of biases that plague steps 1 and 2 (and that affect perception
reliability), a typical finding in cognitive psychology, AFAIK, is that
it is cognitively very hard to do this piecemeal processing consciously
first and then to *spontaneously* have a true gestalt perception
(spontaneity is critical here because we're after *true* emotions that
are reliable measures of one's reaction to music, i.e., to sound alone)
. I would say thay you typically perceive gestalt first spontaneously,
certainly in the case of a piece of music that moves you emotionally,
and then you have to work mentally (very hard work in case of illusions)
to perceive pieces of the gestalt. Going back and forth between this
piecemeal processing of the stimulus and gestalt and at the same time
mentally recording and monitoring *true* emotions (in reaction to
gestalt) is impossible cognitively IMO. In order to have a true
emotional reaction to gestalt, a piece of music in this case, you have
to mentally stay away from all these individual properties of the
reproduced sound and literally "lose yourself" in the gestalt (i.e.,
music). However, audiophiles don't do that. They carefully listen to
those individual properties of reproduced sound and criticaly evaluate
them. Inevitably, that evalution is also biased (human nature, not an
audiophile character flaw). Moreover, simultaneous gestalt perception
is simply not there and training or experience have nothing to do with
it.
So, even if you go for gestalt and the resulting emotions, the issue is
if these emotions are reliable and valid as a piece of evidence that
supports the claim such as, for example, "CD player A is worse than CD
player B (the reference) because it has a plasticky midbass" (or
whatever the claim is). The logical answer would be: they are neither
valid nor reliable either because the person was not really evaluating
the "midbass plasticity", because he was really listening to a piece of
music in its totality, i.e., gestalt, in order to have a *truthful*
(real) emotional reaction to that gestalt *or* he was listening in a
piecemeal fashion, but if that was the case the experienced emotions are
not representative of the listening experience because simultaneous
"true" gestalt perception was not realized.
Here is another logical problem with your position: I am actually a
great believer in gestalt perception of a piece of music and emotional
reactions to it, but I can have easily both if I hear Everly Brothers'
Wishing Won't Make it So on a transistor radio or if I hear Carter
Family on a shellac record and probably if I hear Caruso's voice
reproduced on Edison's phonograph. Because of that, gestalt and
emotions have poor validity and no reliability in critical evaluation of
sound and therefore in critical evaluation of equipment used to
reproduce that sound.
> The objectivist argument here seems to be that sighted listening is
doomed
> to be fraudulent, misleading, and useless.
Not fraudulent, but misleading, and therefore useless.
> One subjectivist argument is
> that sighted listening is more practical and so long as the above
motivators
> are present, the sighted testing has a good chance of being reasonably
> objective and correct.
Not a chance. A little intellectual honesty would go a long way here
(I'm referring here to the TASes of this world). Practicality is really
not a issue if somebody is willing to spend $X,000 for a piece of cable
with a Zobel network at its ends. Also, time (of exposure to stimulus,
switching intervals, switching speed..... you name it) is not an issue
in level matched DBTs. So, please do not use that strawman again
(referring to subjectivists in general).
Keith Hughes
<snip>
>>Steven Sullivan ssu...@panix.com wrote:>
>>The first step of ABX is listening to A and B. If no difference is discerned
>>between
>>A and B, then it's pointless to continue with the 'match/choose/decide'
>
> Right, so far.
Okay! Glad you agree with the above. IF no difference is
discerned, then no need to proceed. If a difference *IS*
discerned, then you *HAVE* made the decision which, according to
you, is the problematic part of the process (the exact same
process required either sighted or blinded).
Now, the X/A comparison is *exactly* the same process you just had
no problem in performing with A/B (or you wouldn't have proceeded
would you?). The process is absolutely identical. The X/B
comparison is merely another iteration of the identical process.
So, where's the problem?
<snip>
> The difference is that in a DBT you are *forced* to chose A or B for X.
Again, there is no decision *forced* that you don't have to
perform in whatever incarnation of sighted testing you prefer. You
at some point, *MUST* decide same/different, else you have no
opinions of any kind. Whether it's A/B, A/X, or B/X, the process
is the same.
Keith Hughes
<snip>
Nousaine
This is why I prefer segments ranging from 20 to 200 seconds and Track Repeat.
Of course A/B Repeat can be used for any Track, Segment or even a whole piece.
Harry Lavo
So far from his years of testing, Tom has cited only one test that has gone
beyond a few hours or a day at most. If you are to get fifteen "samples"
into a few hours or even a day, then you are doing a lot of switching using
a lot of relatively short segments, and flipping back and forth. This is
time-pressured, comparative testing, and it is the way in reality that the
testing happens, not some theoretical month or year.
Harry Lavo
news:bvmi1...@enews4.newsguy.com...
I've responded at length in another post.
Nousaine
>
>"Nousaine" <nous...@aol.com> wrote in message
>news:A7xTb.167643$nt4.750873@attbi_s51...
>> mku...@aol.com (Mkuller) wrote:
>>
>> ....snip to content .....
>>
>> >But when you add "X" to the blind test, the listener is forced to go from
>> >"remember/compare" to "match/chose/decide", a different brain function
>(*left
>> >brain*) which interferes with the recall of subtle details in audible
>memory.
>>
>> IMO the functions "choose, decide" are just one function. Or at a minimum
>his
>> open "remember/compare" has to have a decision function and should be
>> "remember/compare/choose." Otherwise open-sighted evaluation would be
>totally
>> useless for making decisions about deployment of audio gear.
>
>Open-ended sighted evaluation generally results in taking notes over a
>period of time on both pieces of equipment, across several moods and
>environmental factors. Then after an extended period of such evaluation
>(extended being a function of what time frame is available...a weekend? ...
>a week? ..a month? when the time "seems right" to make a decision,
>evaluating those notes, reaching a tentative conclusion, perhaps going back
>to one piece or another to check one factor or descrepancy vs. another, then
>deciding.
>
>Do you see a difference between the traditional dbt comparative testing and
>sighted open-ended evaluation now?
Oh I've known the difference all along. I use both all the time. But according
to Kuller (and you) I'd have to have my notebook with me all the time so I
wouldn't "have forgotten" what my equipment sounds like from session to
session. :-)
Nousaine
>
>"Audio Guy" <audio...@yahoo.com> wrote in message
>news:GdxTb.163713$sv6.894310@attbi_s52...
>> In article <bvm5k...@enews1.newsguy.com>,
>> mku...@aol.com (Mkuller) writes:
>> >
>> > I'm only claiming that memories of subtle audible differences fade more
>quickly
>> > than memories of large, gross differences, whatever they may be. You
>have any
>> > evidence to the contrary?
>>
>> OK, if subtle audible differences fade quickly, doesn't that validate
>> the findings of DBT'ers that quick switching is the best way to
>> determine differences?
>>
>
>No, because those subtle differences often take time to be recognized and
>enter consciousness.
Ok; but what process causes experienced audiophiles to 'forget' learned and
conscious sonic attributes when nothing but a cloth is placed over speaker/amp
terminals AND an only recently encountered device is inserted/not-inserted. Why
would the subject now be unable to tell the cry of his "baby" from that of a
relative stranger?
And IF that condition suddenly causes the subject to "forget" even familiar
sound how can we be sure that Kuller isn't right and sonic memory disappears in
"microseconds?"
That's where your argument eventually leads. I believe that audible memory is
quite robust; it becomes fleeting ONLY if the effect were imagined in the first
place. Otherwise we would have to re-learn the sound of instruments for every
concert.
Harry Lavo
news:bvn3f...@enews1.newsguy.com...
Actually, I just wrote a rather lengthy response post where I relate the
process using the Jungian functions to explain how I believe evaluative
listening works. In some ways not much different from what you describe
above, although I believe the "gestalt" is built, refined, then broken down
into components, refined further into distinguishing characteristics of the
piece of gear being evaluated and then and only then does quick-switching
become an aid and a possibility for discrimination.
Harry Lavo
You don't. It's a function of not really understanding or acknowledging on
*your* part what we are saying.
For open ended evaluation, you don't know initially what you are looking
for. It make days for things to gel that "a" sounds somewhat thisway, and
"b" sounds somewhat more thatway. From extended, evaluative listening and
non-quick switching. Then a tentative conclusion is drawn. Now you know
what you are listening "for". It may be something subtle and perceptual,
such as "imaging". Once you have it firmly grasped in mind what the
signature is of "a" and how it might vary from "b", quick switching can help
precisely because it "interupts" the perception you have grasped and altered
it slightly (or not) over the flow of music.
We are talking about open-ended component evaluation. If I simply give you
two components, say "different" or "same", or "is it a" or "is it b" and
force a choice quick switching works against you because you haven't yet
really been able to determine what it is you are listening for in audio
terms. "Same" or "different" are not audio terms. They are "sound
artifact" terms on simple one or two dimensions.
Under quick switching under these circumstances, the brain seems to "panic"
in that it can't sort audio patterns quickly and has no frame of reference;
this by itself creates anxiety, which in turn creates even more confusion
and panic. I believe this is why audiophiles cite stress and fatigue in
trying to do this kind of testing when dealing with very subtle, perceptual
factors and why the test favors a "null conclusion" unless we are dealing
with straightforward factors that the sensate function can handle without
much need for the intuitive or emotional functions (volume, frequency
response).
Do I know this for sure? No. But it is reasonable and verifiable. That is
why I proposed a control test that is double-blind, relaxed, evaluative, and
leisurely. Along with testing of the same respondents using sighted,
evaluative listening and at another time relatively short, terse,
comparative ("same","different") double- blind testing as is traditionally
recommended here.
If the control test gave results similar to traditional dbt/abx, it would
verify that that traditional dbt/abx testing was a valid "shortcut" to
evaluative testing. If the control test gave results similar to sighted
open-ended evaluative testing, then it would suggest that evaluative testing
even though sighted was a more encompassing and valid approach for component
evaluation.
So if you really want to stop the "jaw flapping" and try to resolve the
differences of the two camps, first you have to acknowledge the possibility
that we might have a point, and that it is worth trying to resolve somehow.
Just as we acknowledge that traditional dbt/abx testing works fine for
simple volume and frequency response differences, and artifact detection,
which allow simple one or two dimensional evaluations.
Nousaine
>> mku...@aol.com (Mkuller) wrote:>
>>>But when you add "X" to the blind test, the listener is forced to go from
>>>"remember/compare" to "match/chose/decide", a different brain function
>>(*left
>>>brain*) which interferes with the recall of subtle details in audible
>>memory.
>>
>
>>nous...@aol.com (Nousaine) wrote:>
>> IMO the functions "choose, decide" are just one function. Or at a minimum
>>his
>>open "remember/compare" has to have a decision function and should be
>>"remember/compare/choose." Otherwise open-sighted evaluation would be
>totally
>>useless for making decisions about deployment of audio gear.
>>
>
>No, there is no decision *required* in sighted listening. Only in blind
>listening is the decision process an integral function of the test.
I'll agree with that. But if your auditory memory fades in microseconds how can
one form any kind of assessment of what he just "heard" unless he makes an
immediate comparative decision about which was better? Or might be useful?
Making
>decision about deployment of audio gear are completely independedent of the
>two
>types of listening tests.
Sure; those who use blind tests can be more comfortable that decisions are more
likely to be based on true sound quality, while those who use sighted tests
only (and ignore what's been experimentally shown about amps/wires/bits in
conntrolled tests) will be doomed to be forever insecure sonically.
Once either type of tests are completed, then
>those
>decisions can be made, or not.
>Regards,
>Mike
Of course, but that still doesn't answer the question as to how sighted (or any
other kind of ....) listening can be useful in any manner given 1) the human
proclivity for false sonic positives plus 2) being saddled with a sonic memory
microseconds in length. It seems to me that based on sound alone no musician is
capable of assessing his own or others work.
Harry Lavo
news:bvn3h...@enews1.newsguy.com...
The difference is "how" that difference is arrived at. "Choice"
(comparative) vs. "Evaluative" (organically or experientially) arrived at.
Harry Lavo
See some of my other responses. Jung's personality work is incorporated
directly into the Myers-Briggs personality test, the most examined,
validated, and useful personality test in use today and for the last
thirty-five years. His concepts and terminology underlie almost all modern
psychology...and the pharmacologizing of psychology doesn't invalidate his
personality theory which provides validated insights into decision patterns
and job placement/success rates among other things. Pharmacology doesn't
even begin to get into those areas.
>
>snip, duplicate post answered elsewhere<
Harry Lavo
The simple answer is your understanding/answer to the issue is far too
simplistic. I've explained the process and why a control test is needed in
two other posts. Once you read them, I hope you'll better undertstand where
we (I at least) am coming from and why "simply throwing a blanket over the
terminals" is not an answer.
S888Wheel
>> >
>> >Yes, as *one kind* of prior knowledge...the most extreme kind.
>> >
>
>> And being the most extreme kind it is likely to wrought the most extreme
>kind
>> of results.
>
>
>Even if less extreme kinds yield 'less extreme' results ...which is not
>conceded...do you agree that prior knowledge still biases sighted
>comparison to some degree?
>
Yes, if the knowledge has some meaning to the testee.
>
>> >> Yeah people are suseptable to misdirection.
>> >> One could say that sameness in components could also be imagined if the
>> >> listener believes ahead of time that no difference exists.
>> >
>> >One could imagine that, but perceptual psychology says the stronger
>> >tendancy is to experience *difference*.
>> >
>
>> I would like to see a citation of any study in perceptual psychology that
>> suggests this is true. I'll bet if it has been studied the studies would
>show
>> the pre-existing bias would prove far more powerful than any universal
>tendency
>> to percieve a difference. People, by and large, go through life not
>noticing
>> differences that do exist every bit as much as they percieve differences
>that
>> don't exist.
>
>
>I wold like to see a citation of any study in perceptual psychology that
>suggests this is true.
Hmm I asked for citations to support your claim and none were given. Unlike
your assertion I do not claim mine is based on any scientific study. Do you
really disagree with the assertion that people go through life not noticing
subtle differences in their day to day life and this lack of notice goes
unnoted for the same reason noncoincidences go unnoted?
>
>> But like coincidences vs. noncoincidences the failure to note
>> small but real differences in our sensery perception goes largely
>unnoticed.
>
>
>Not when a person is comparing two things. There, the tendancy is to
>report difference...perhaps it's of a piece with the 'coincidence' thing,
>where peopel tend to impart *meaning* to things merely based
>on temporal proximity or some other possibly spurious 'connection'.
>IIRC such errors of judgement are covered in the book
>Inevitable Illusions : How Mistakes of Reason Rule Our Minds
>by Massimo Piattelli-Palmarini, a professor of cognitive science at UAZ.
>
I don't think your assertion is so universal. Many people involved in QC in
many different areas of production make unbias controled evaluations and do a
very good job of it.
>
>> >> That means every ABX
>> >> DBT that has been done with people who believe no difference exists
>> >> between the components in question is tainted with expectation bias
>unless
>> >a control was
>> >> used in the test for that bias such as the random insertion of a known
>> >barly
>> >> audible distortion.
>
>> >
>> >Again, 'expectation bias' isn't necessarily conscious expectation...
>
>> I don't believe that. Claims of the effects of the "subconscious" mind on
>the
>> conscious mind are highly debated. If you can cite any scientific claims
>that
>> "subconscious" biases are at work in the effects of expectation bias I'd
>like
>> to read about it. I am betting that none of the studdies on the effects of
>> expectation bias make any claims that the expectation bias is subconscious.
>
>
>> >it's not a matter of going into the test, saying to yourself, I *will*
>> >hear a difference.
>
>
>> I think it very much is. That is why I think deliberate misdirection is an
>> obvious flaw in testing for normal expectation bias. The deck is stacked.
>
>Scott, controlled comparison for audible difference does not usually
>involve *deliberate misdirection*, so please abandon this straw man
>argument immediately.
There is no straw man argument since I was specifically refering to Tom's tests
which did involve misdirection.
> I merely used that example as a pretty foolproof way
>to highlight the inherent tendency towards *hearing* difference when some
>other sort of difference exists (either real or imagined).
If you don't want me to address a specific example then maybe it shouldn't be
used.
>
>> >And, too, when are ABX tests performed to test a claim of 'no difference'?
>
>> Every time Tom does such a test.
>
>When has Tom tested a claim of *no difference*, rather than a claim of
>*difference*?
This smells of a semantic argument. If you are testing one you are testing the
other. Tom's position of no difference is well known.
>
>> >All the ones I've sen reported involved people who claimed they could
>> >hear a difference between A and B; if not, there would be no point in
>> >continuing the test.
>
>> You are not a reference for recording all such tests.
>
>You aren't a reference for scientifically-founded ideas. It doesn't stop
>you from proclaiming your lack of belief in them as if it were some sort of
>dispositive argument.
>
Complete nonsense. I have NEVER claimed any lack of belief in scientifically
founded ideas. I think many such ideas have been grossly misrepresented on RAO
though. I have spoken out about those misrepresentations.
>
>> >> Do you think you know what orange juice or
>> >> cola or strawberry jello taste like? I bet you cannot successfully
>> >identify
>> >> such items by taste alone on a reliable basis.
>> >
>> >I'd require taste and smell, most likely. ANyone who's had a cold knows
>> >that a stuffed nose reduces taste sensitivity. And that's been confirmed
>> >medically, and the anatomical basis is known.
>
>> I figured that was understood. I was refering to literal blindfolding. Try
>it
>> with amny differnt samples and see how well you do.
>
>We aren't talking about literal blindfolding of ABX /DBT testees, ever.
>We aren't talking about blocking any of the senses.
I know. I was using that as an extreme example of how such changes can
profoundly affect sensitivity. I am not claiming that ABX DBTs do affect
sensitivity but that it may do so when listening to music. It does inherently
change the way people listen. It does not surprise me that people can hear
differences more easily with pink noise than with music in quick switching time
synced ABX DBTs. When one listens to music and uses quick switching one is
never comparing the same source. At least with pink noise the sourse doesn't
change. I think it would be interesting to do ABX DBTs with source samples of
music that are replayed in their entirety with each switch rather than a real
time switch of an on going musical source that is ever changing in sound. It
does seem to me that pink noise and a few other test signals that do not change
or can be replayed in their entirety would make for one less variable and would
make for a far more sensitive test.
>
>> >> Take the testee out of the
>> >> envirement and all bets are off on sensitivity.
>> >
>> >No, Scott. All bets aren't off. We know that certain
>> >changes to the environment reduce sensitivity. Others could
>> >be expected to have little or no effect. This stuff has
>> >been studied.
>
>> I am sure it has been studied. I am not so sure it has been accurately
>reported
>> on RAHE.
>
>See above re : dispositive arguments. OF cousre, you could always do what
>has been suggested many times to you: take the time to go do the research
>in your local university library to your own satisfaction, and report back
>to us.
>
>Piatelli-Palmarini's book, btw, is avaialble cheaply from amazon.com
>
I have made such purchases in the past only to find out that indeed the
information had been misrepresented on RAHE. I have no interest in buying books
only to find out I am right. If I am wrong it is easy enough to quote a passage
from any such book that shows it to be so.
>
>> >> The interaction of senses is
>> >> complex and critical in sensitivty of the senses.
>> >
>> >In some cases, for some senses, yes.
>> >
>> >But *that's beside the point anyway*. We're not talking about blocking
>> >any of the senses. It's not analogous to stuffing someone's nose and
>> >asking them to identify a flavor. We aren't talking about blindfolding
>the
>> >testee.
>> >or dong the test in the dark. We aren't damping down the sense of sight.
>> >We are only talking about keeping the identity unknown at the time of
>actual
>> >
>> >listening. You can *SEE* both devices
>> >under test, during the test, if you like. You just can't *know* which
>> >one is playing. How could that possibly be construed as changing the
>> >interaction of the senses?
>
>> I was talking about deliberate misdirection.
>
>Deliberate misdirection doesn't necessarily involve hindering any senses.
It sure does indirectly.
>
>If two cables or amps or CD players -- for giggles, let's make one a
>hgih-end brand and the other a mass-market brand -- are hooked up to a
>switchbox, with the switch and the components *always in view*, and the
>proctor says the 'up' position opf the switch is DUT A and the 'down'
>position is DUT B, when in fact both positions are DUT A, it's an
>excellent bet that the golden ear testee will hear big differences between
>the switch positions. Where have any of the senses been thwarted in this
>deliberate misdirection protocol?
>
This has nothing to do with the specific tests that Tom did which he claims as
proof that people will so regularly imagine differences. That is the specific
misdirection that I am saying is stacking the deck and possibly giving skewed
results. I am saying that the same misdirection in reverse would probably
wrought similar mistaken results with testees failing to detect real
differences. That's all.
Audio Guy
But I don't believe that has been your argument with DBTs. Over and
over the phrase "quick-switching" is attached to DBT by you and
others. But that is not a requirement of a audio double blind test
yet you almost always use it as a means to show what is wrong with
them. As I asked above it is time to separate the two and for you to
acknowledge that time limits are not a prerequisite to an audio DBT.
In fact only a lack of knowledge which item is in use is the only
requirement and so deflates your argument.
So if he was to perform many more of these long term DBTs that all
came up with the same "no difference" results that would satisfy you?
How about performing some yourself? That is the scientific method in
that if one disbelieves the results of an experiment then one runes
their own and presents their results. It is not up to the original
researcher to run a test proposed by someone else in order to
re-prove or disprove their results, even though Tom has done just
that many times over.
Harry Lavo
I start by not confusing "sound different" (one-dimensional, comparative)
with "audio evaluation".(multi-dimensional, evaluative). See my lengthy
post elsewhere.
Harry Lavo
>snip, to clarify discussion below<
> > The evidence is that no published DBT I know of has ever identified
anything
> > but gross frequency response and loudness differences between audio
components.
>
> Are they 'gross' because they were detectable under DBT? Or is there some
other
> criterion?
>
See my lengthy post elsewhere. Better perhaps to call them
"one-dimensional" and requiring only a relatively crude sensate response,
rather than a perceptual intuitive and emotional judgemental response.
That makes them simple and relatively crude as far as audio evaluation is
concerned.
> > But most of us know there are many more differences than just those.
>
> No. You *assert* that. The problem is, you don't *know*.
You are an audio hobbiest, are you not? Then you know as well. Com'on,
Steven!
Harry Lavo
The Myers-Briggs test, based on Jung's work, is the single most studied,
tested, validated, and applied personality test in use. It is extremely
predictive of many things, job placement/success and decision making
patterns among them. He is acknowledged as the father of modern psychology,
as you probably know. Some of his work went "out of fashion" years ago; his
personality theory has never gone out of fashion. Indeed the concepts of
thinking, feeling, intuitive, sensate, introverted, extroverted as parts of
personality laid the foundation for virtually all who have followed. It is
the Adlers and others who have fallen by the wayside; Jung's personality
theory continues to be used as a reference point every day.
Nousaine
>"Nousaine" <nous...@aol.com> wrote in message
>news:Z4nTb.159122$Rc4.1251684@attbi_s54...
>> "Harry Lavo" harry...@rcn.com wrote:
>>
>> >"Audio Guy" <audio...@yahoo.com> wrote
>>
>> ...snip to content ....
>>
>> >>
>> >> For one who has had significant training in psychology, please
>> >> explain how you can disparage DBTs when they are such a fundamental
>> >> part of psychology and have been since the days of "Clever Hans".
>> >>
>> >
>> >I have never "disparaged" DBT's. I have said they are often impractical
>in
>> >a home environment.
>>
>> Sure.
>>
>> I have also said that without a proper control test, I
>> >am not sure that they are the best test technique for open-ended
>evaluation
>> >of audio components.
>>
>> But what control test do you have for 'sighted' (re: as uncontrolled)
>testing
>> other than that we already know (such as subjects will often report
>difference
>> when given identical audio presentations) other than it is obvious they
>are
>> prone to false positives? -- It seems that your only real argument for
>them is
>> they are easier to conduct.
>>
>
> Don't be clever here Tom. I've already spelled out in detail in this forum
>a control test that would servce as such for both sighted and conventional
>dbt testing. It iself was a dbt, but one that introduced many elements of
>sighted listening, most fundamentally evaluative rather than comparative
>listening. You and the other objectivists just go on ignoring it or the
>need for such a test, and continue to pursue your strawment.
You've postulated an experiment that YOU refuse to implement. I can make up
erxperiments for you too. I've even challenged you to one where you could prove
your point. But, you won't. So why should I devote resources to an experiment
that seems un-needed and un-necessary.
However in the interest of fact-finding I'll offer you this experiment. With
50/50 cost sharing I'll proctor an experiment at your facility where you (as
subject) can show your ability to hear amps or wires (your pick) under bias
controlled conditions in your system. Protocol completely optional as long as
bias controls are followed.
>
>> >I spent a good part of my life sponsoring, helped designing, and
>> >interpreting consumer research testing, often of a very sophisticated
>> >nature. I understand the arguments. I also understand the need for a
>> >control test to "prove" beyond a shadow of a doubt that quick switch dbts
>> >(especially abx)" do not incur their own confusion factors that obscure
>> >certain aspects of open-ended evaluation. So until the proponents of
>such
>> >testing prepare and present such a control test with some scientific
>rigor,
>> >I will stay with my views.
So why not instead show how amp/wire sound exists with nominally competent
components instead? Don't like ABX? Great we'll use cable swaps? Don't like
short segments? Great; you choose the length.
Your system, your protocol (only subject to blindness and ordinary bias
control), your programs and I'll supply a comparative piece (amp and EQ if
needed.) No time limits. Shared expense.
>> Fair enough but IMO you're just choosing and defnding a method you
>'prefer'
>> rather than one that has been verified in sensitivity. Indeed it's quite
>easy
>> to show that open-listening is rife with false positives. That's OK by me;
>but
>> I'm not one to endorse a method just because it may be easier and
>obviously
>> lacking in validity for investigatting true acoustic effects.
>>
>
>Yes, but you also refuse to do a control test that might possibly show that
>the dbt'ng you do might have some 'false negatives' so you too are just
>making a preference. Only you hide it under the guise of "scientific
>sensitivity".
I've done all these tests Harry. I'm inviting you to as test-party.
I think Mr. Radeckers post should cause you to give some
>thought, Tom, to what I and others here have been trying to say all these
>years.
You seem to think that I haven't given thought to those ideas. In the beginning
I had some of the same reservations. The difference between us is that I've put
allof them to thje test. Time, switching....et al; all of them and yet I've not
found a circumstance where any subject (even vociferiously ardent ones) who has
been able to reliably identify wires (speaker cables, interconnects) or
amplifiers (power amps/pre-amplifiers) of nominal competence (able to deliver a
level/frequency matched signal to the speaker terminals) with even the most
MODEST of bias-controls (opague cloth over terminals) in their personal
reference systems with their personally selected progam materials with no time
limits.
If you say YOU can do this I'm all ears, as it were. So why don't you just
prove this once and for all? You don't need to make any investments, no need to
hire a large bunch of subjects; no equipment required; no laboratory facilities
.... just do it and we're finished.
The only pre-nup agreements are that all results/details will be made publicly
known on this newsgroup and I (and you) reserve the right to publish same under
our own copyright. For starters I submit that I will prepare a paper on the
experiment, submit for presentation at an AES Convention and then submit for
publication in the JAES.