[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Alexandra Hettergot wrote:
>I don't believe in the CBW concept's being essentially wrong (though I'd
think other aspects
>(traditionally) apparent in music, as, e.g., timbral difference (i.e., due
to both spectral
>and temporal structure), spatial distance, multichannel treatment, etc.,
worth to being considered
>in that respect, too).
The latter is exactly what I would like to suggest. There are people like
me who consider virtual pitch a plausible result of neural principles
rather than a "gestalt" phenomenon. The same reasoning provides a
functional understanding of the neural basis for sensations like
consonance/dissonance on the most basic level, and for more emotional
judgements like pleasantness at a higher level.
>>You would like to refer to something like spectral acoustic energy over
What about acoustic energy, I am aware of many serious reasons to avoid
this misleading expression.
>Yes, and for several reasons, one being -- note that here I was imagining
sound sequences other
>than of purely "melodic" appearance/harmonic origin -- that of pitch
>in dependence on the spectral bandwidth of the sound stimulus (e.g., that
of a pure tone vs that
>of a complex tone, a narrow-band noise, a band-pass, a low-pass, a
high-pass noise, an AM noise,
>a comb-filtered noise..., respectively, cf. e.g., Zwicker, Fastl p.125) ;
and another one being
>the temporal aspect of which I am well aware as to pitch sensation in
relation to stimulus
>(temporal) structure and duration.
>Btw, as has been discussed on this list concerning the spectral-centroid
concept recently I do
>share well some of the somewhat critical opinions as to non-stationary
(transient) sounds, as,
>e.g., in human speech, and the impact of temporal envelopes (one might
consider the example of
>a reversed percussion sound/impulse), or furthermore examples of
manipulated sounds in
>electroacoustic music with rapidly changing spectral envelope/pitch
sensation, etc. -- and
>this is why I would like to have it (the spectral centroid) time-running,
as has been proposed
>recently, too (while I am really finding it a very useful concept, just
with regard to (the analysis
>of) contemporary music, *yet* with more strongly taking the temporal
aspect into account). --
>As you might derive from this I am not against temporal aspects in the
sensation of sound complexes
>*at all* ; I do, however, see a light (logical) contradiction in what you
are writing below points
>#33 "based ... simultaneously on temporal coincidence, too" and #34
[being] "the only plausible
>explanation" : while I would agree with #33 ("not just -- but too"), I
wouldn't accept #34 as to
>it being the "only" (plausible) explanation ; (to me) it is clear that
"tonal listening" is, as
>you were writing above, "simultaneously" tonotopically *and* temporally
I apologize for putting you boldly together with the majority into the same
drawer with Zwicker/Fastl 1990. Nonetheless, even Fastl's 1999 edition does
not reflect basic doubts of mine whether the traditional theory is really
correct. Some textbooks do so, at least in part. I recommend Auditory
Perception by R. Warren, Cambridge Univ. Press (1999) and chapter 4 of
Encyclopedia of Acoustics by Malcolm J. Crocker, Wiley (1997). On the web
you might find further food for reflection at
www.uni-leipzig.de/~psycho/kaernbach/pitch.html, and www.leonhard.dk.
Leonhard was prompted to his somewhat provocative conclusions by the
obvious failure of traditional theory in practice. The same is true for me
and also for J. Manger, the founder of Manger Products, who managed to
derive contributions to understanding and improvement of loudspeakers from
his realization. The importance of time was stressed by Malcolm Slaney. Roy
Patterson also contributed a lot, etc.
What is the point, if you are not denying temporal aspects? You did find it
out yourself. When I wrote "the only plausible explanation" this implied
that the theory by Terhard is not plausible to me. I just cannot imagine
that neurons communicate with each other within frequency domain, and I
share this view with Roederer (1977). He has been quoted in
Neurowissenschaft by Dudel/Menzel/Schmidt, Springer (1996) as follows: Die
Wahrnehmung von Tonhöhe und musikalischen Intervallen ist nur durch diese
Kodierung von Frequenz im Zeitbereich möglich. (Perception of pitch and
musical intervals is only possible by this coding of frequency within time
domain.) You may substitute the term frequency by periodicity. With other
words, if you agree that tonal perception is dually based on both tonotopic
and temporal coding at a time and in the same neurons, neurons certainly
interact within time domain. Suggested coincidence detection does not just
provide a simple elucidation to harmony. Other observations can also only
be understood on temporal basis, e.g. the discrepancy between doubled
frequency and the perceived, slightly enlarged octave.
Hopefully, you are not confused by my addition that the tonal perception is
not the only one but there is also a so called temporal perception in the
sense of modulation transfer, being diffusely spread across tonotopy.
>btw, what will "local resonances" cause if not spectral pitches (due to
spectral peaks) ?
Well, local resonance actually performs a coarse spectral analysis.
However, look at the neural pattern drawn by Secker-Walker and Searle
(1990). Temporal period seems to be much more precise than what Vercoe
named spectral "blockvoting". Furthermore, spectral resolution within
cochlea cannot account for the astonishing frequency resolution of hearing.
This also indicates that frequency discrimination by ear is based on
perception of period rather than frequency.
BTW, I have to correct myself. The paper by Shamma is based on the same
original data by Miller and Sachs. So it would be highly desirable to have
at least a second set of such data.
>>What about dyads, I just guess, larger numbers were too demanding
>Yes, this was a little my impression, too -- which yet doesn't prevent me
>this investigation, as I was writing, quite "instructive" (among others I
am aware of).
Why not trying and translate this quality into the time domain?