Re: DBT in audio - a protocol

On 13 Jan 2006 07:26:20 -0800, "andy" <andy19191@xxxxxxxxxx> wrote:

>Don Pearce wrote:
>> So what do you think?
>
>I think you may be putting the cart before the horse. Before designing
>an experiment one needs to define precisely what observation is being
>investigated. For example, "X and Y sound different/same" is too
>imprecise an observation to design an experiment.
>
I will be explicit in what is being investigated. Somebody says that
two interconnects (or whatever) sound totally different, and the
circumstances of the claim lead one to believe that they should not
sound different. Clearly he has made the assertion on the basis of
unplugging one set and substituting another, so that is how the test
works. The basic precept is that the circumstances which prompted the
assertion in the first place should be replicated as closely as
possible.

>The perception of a difference when switching quickly between A and B
>is not the same as the perception of a difference when there is a pause
>between hearing A and B. You have plumped for the latter but this can
>be criticised for being less sensitive than the former. A solution
>would be more precision about the observation: "A and B sound
>different/same when compared with a 5 minute pause between switching".
>

As I say, I am replicating the circumstance which gave rise to the
assertion - with the slight difference of not being able to see which
is connected. This is not intended to be a general purpose protocol.

>The perception of a difference between simple tones and music is also
>different as is the perception between different types of music: "A and
>B sound different/same when compared with a 5 minute pause between
>switching using music X".
>
>Do you want the perception to include or exclude the possible effects
>of suggestion and/or visual stimulus on the brains interpretation? Most
>studies want to exclude this and adopt some form of blind procedure as
>you suggest, making the observation: "A and B sound different/same when
>compared with a 5 minute pause between switching using music X under
>blind conditions".
>
Provided the subject can't see which cable is connected, he can run
the test any way he likes. If five minute breaks help him, he can have
them. He is likewise free to make his own choice of material, although
I think it is reasonable to exclude test tones - just music or speech
allowed. Instant switching is not on, because it involves switch
boxes, which will be invoked as a confounding factor when the subject
fails to identify.

>I have almost certainly missed a few other factors of relevance which
>will add to the qualifications and pin down exactly what form of
>perception is being investigated. Nonetheless it is clear where
>following scientific reasoning is going and I would suggest an
>audiophiles response to the results of such an experiment is going to
>be the same as similar ones performed in the past.
>

Quite so. The person taking the test, when he fails, will say he was
stressed. Everybody else will still say that they could do it with no
problem, but they can't be bothered.

>To be more positive, the component of an experiment that appears to
>generate the strongest response from an audiophile is blind testing. I
>would suggest that the results from an experiment involving sited
>testing and not blind testing will be more persuasive. This is going to
>be difficult to design if you wish to quantify the effect of suggestion
>and/or visual stimulation since this will have to be independent of A
>and B. Nonetheless it might be fun having a go and the results are
>likely to be more interesting than simply repeating the same type of
>experiment performed in the past.

Read my post yesterday describing a sighted test which yielded
differences, even though nothing was changed. Sighted testing is of no
value whatever.

d

Pearce Consulting
http://www.pearce.uk.com
.

Relevant Pages

  • Re: DBT in audio - a protocol
    ... imprecise an observation to design an experiment. ... The perception of a difference when switching quickly between A and B ... different/same when compared with a 5 minute pause between switching". ... generate the strongest response from an audiophile is blind testing. ...
    (uk.rec.audio)
  • Re: Charles Murray "The Plan" ends poverty
    ... The early days of electronic design automation are indeed interesting and if someone is going to write a history they had better hurry up. ... Donald Durr of RCA developed the first automated placement and wire routing program for LSI circuits in '65, ... I've had some experience with large systems in which the wrong circuit bounce in a switch or relay could set some relays to chatter and the system to wildly respond. ... Capacitors slowed down the switching, and helped the bouncing, but nowadays there are processors polling relays without debouncing circuits to determine whether they have switched. ...
    (soc.retirement)
  • Re: Charles Murray "The Plan" ends poverty
    ... The early days of electronic design automation are indeed interesting and if someone is going to write a history they had better hurry up. ... Donald Durr of RCA developed the first automated placement and wire routing program for LSI circuits in '65, ... I've had some experience with large systems in which the wrong circuit bounce in a switch or relay could set some relays to chatter and the system to wildly respond. ... Capacitors slowed down the switching, and helped the bouncing, but nowadays there are processors polling relays without debouncing circuits to determine whether they have switched. ...
    (soc.retirement)
  • Re: Evolution of Personality
    ... on autopilot? ... "design" to mean in this context. ... Some seem able to see and feel God everywhere while ... Perception is the only way i can support it. ...
    (talk.origins)
  • Re: Evolution of Personality
    ... just a wad of bilogical and chemical processes on autopilot. ... "design" to mean in this context. ... Some seem able to see and feel God everywhere while some cannot feel or see anything. ... Perception is the only way i can support it. ...
    (talk.origins)