Re: Invitation to criticize dissertation idea.

[David Spondike <Dspondike@xxxxxxx>]

Dear Dr. Bregman,

    Thank you for sending the paper in html.
I was able to get the text from my brother, but not the images/tables.

    Just a thought about band limited noise: If narrow bands of noise that
had harmonically related center frequencies are simultaneously sounded, would
that increase the pitch strength at the center frequency that is the
"fundamental" of the harmonic relationship?

    Terhardt's "pitch weight" does seem very similar to your "pitch
strength":  a measurement of the perceived salience of "pitch" derived from a
particular complex sound.  According to Terhardt's algorithm, strongly
harmonic complexes (that have virtual pitch heights in the heart or the human
auditory sensitivity range) have the strongest pitch weights. Complex tones
exhibit higher pitch weights than pure sine waves due the reinforcement of
the fundamental by the harmonically related partials.

    I am debating with myself on what course to take with my theoretical
dissertation. My professors would like me to keep it very focused. Therefore,
my original idea of developing a music theory curriculum of "musical scene
analysis" that could be taught along side traditional (Schenkerian) theory
could best be left for future work. Much "timbre" research seems to be in the
auditory (as opposed to musical) domain and seems to be focused on auditory
mechanisms -- very "psycho-acoustic."
    From my correspondence with you and others on the auditory list, it would
seem that exploring the idea of crossing musical voices would be best done as
research in orchestration, as opposed to timbral research. What are musically
plausible crossing conditions and what are the perceptual results? Of course,
one can look for actual musical examples that exploit these conditions.
Perhaps this may be the best tack to take. I suppose that I should be careful
to apply psycho-acoustics to music and not do a dissertation in
psycho-acoustics.

    I guess the timbral question I was trying to explore is: what are some of
the timbral cues that cause stream segregation? I figured that any difference
of timbre that could overcome the pitch-proximity grouping cue in the scale
crossing experiment could be considered a significant difference, even though
smaller differences may cause segregation in a different experimental
condition. For instance, "brightness" or "sharpness" is generally considered
an aspect of timbre. Yet, I am not sure how much difference in brightness
will allow the scales to be heard as crossing. I can explain the relevance of
the question musically. Does it hold any value from an "auditory" perspective?

    Thank you again for taking the time to resend your paper.
David

Dspondike@aol.com