wrote in message ...
Harry Lavo wrote:
"Don Pearce" wrote in message
...

The only problem was that the cables I was changing were not in
fact
the ones carrying the signal. The real cables remained unchanged
throughout the test.

So without the pressure of blind testing, the participants were all
relaxed enough to hear the differences easily. Shame really that
there
were no differences to hear.


Yep...the first conclusion you can draw from this is that people
don't
expect to be lied to.

The second is that people can easily imagine differences where none
exist
*IF* they have a reason to believe such differences should exist.

And isn't this exactly the problem with all sighted evaluations? Any
visual difference, or any known difference between the units under
test, establishes a reason to believe [audible] differences would
exist. You can't reject Don's experiment on this logic without
rejecting every sighted comparison you have ever done.


Yes, except there is a big difference when evaluating, lets say, a known
expensive and well-regarded piece of equipment vs. a cheapo, versus trying
to ascertain differences between two different pieces of gear that are
roughly of same value and manufacturing quality. There may be a bias
towards hearing differences where none exist, but that bias is not likely to
be so strong as to override an ability to hear real differences, if that is
the goal.

As a specific example, I recently bought a used power amp by a manufacturer
based on my satisfaction with another piece of gear from the same range by
that same manufactur. It was hopefully to replace a piece of gear that I
had been relatively happy with, but I felt was slightly lacking in a
specific regard. When I got the unit, I pulled out a few "test" disks and
substitued the units back and forth, playing and replaying sections from the
disks. My overall evaluation was that the units sounded essentially alike
in frequency response and speaker control, and the new unit had the
characteristic I had been looking for (also good). So I was predisposed to
keep/like the unit. I put it in the system and used it as I worked at the
computer for a week...but I noticed that I became tired of listening and
slightly irritated after several hours..that had never happened with the old
unit. Switched it back in, went another week, no problem. Put the "new"
unit back in, another few days...same irritation problem. Back in went the
old...no problem..and it is staying there and I am selling the new unit. If
anything my expectation bias was that I would like the new unit, and the
comparative testing tended to support this. But clearly long term there is
a problem and it is a piece of gear I cannot live with. If I had "blind"
tested, I might have concluded the units sounded the same (they were that
close) or that the new unit sounded slightly different (and probably better)
since it clearly has the characteristic I was after. It clearly would not
have shown anything about the long term irritation that has caused me to
reject the unit.

None of this proves that people can't hear real differences sighted.

Why would we need to prove this?


Because there is a tendency here to assume that anybody who hears
differences is kidding themselves, and that most likely there is no
difference unless we are talking about phono cartridges or loudspeakers.
That "standard" is applied to turntables, tonearms, tuners, CD players,
amplifiers almost without discrimination.


Nor does it prove that blinding does not remove some real differences
that
are perceived under different listening conditions.

Since when is it anyone else's job to disprove your baseless
hypothesis?


Well, you haven't proved it baseless. But you don't have to prove it, and I
don't have to accept your preference for blind testing.


This is why blind testing really works, and why the results it
gives
can be trusted.

Yep, seems logical on the face. But not proven via controlled
testing,

Tell that to the folks in the Psychology Department of your local
university. They could use a good laugh.


Come on. We've gone over this before. Nothing proven that specifically
applies to some of the esoterica of audio component evaluation.

especially as regards to something as slippery as open ended
evaluation of
audio components.

The reason it's slippery is that it involves not just sound, but also
the mental state of the listener. The whole point of blind testing is
to eliminate--to the extent possible--mental states as a factor.
Designers of audio equipment should want to do this, because they
cannot engineer for mental states.

Nice if you ignore the fact that the test itself changes the mental state.