High-frequency hearing

[Albert Bregman <IN09@xxxxxxxxxxxxxxxx>]

Dear list,

I would imagine that if the sound is delivered over loudspeakers,
the exact position of the loudspeakers and listeners in the room
would introduce enough variation as to make the argument about
sampling rate moot.

Another factor is the deterioration of high-frequency sensitivity
with aging.  I'd be surprised if I could hear anything above
16kHz.

So if you're delivering sounds over headphones to a very young
human being (or a middle-aged dog), the sampling rate might have
to be higher than 44.1kHz.

As for our fine sensitivity to the interaural delays involved in
perceiving spatial location, again the room acoustics would be
primary if loudspeakers were used.  Remember also that when we
publish a threshold, it is the average taken over a large number
of trials that we give.  On any single trial there could be a lot
of error.

I can see how higher sampling rates might be necessary for
specialized efforts, such as research in AI or psychoacoustics.
I don't know exactly how the proposed standard would affect these
efforts.

Al

---------------------------------------------------------------------
Albert S. Bregman, Professor,  Dept of Psychology,  McGill University
1205 Docteur Penfield  Avenue,  Montreal,  Quebec,  Canada   H3A 1B1.
Phone: +1 514-398-6103 Fax: -4896 Email: bregman@hebb.psych.mcgill.ca
---------------------------------------------------------------------