Prove it

FROM: mclaren
TO: The new Alternative Wanking List
SUBJECT: Prove it

In message 854, Paul "All Claims, No Proof" Erlich once
again made an unsubstantiated statement when he
claimed:

--- In crazy_music@y..., "Robert Walker" <robertwalker@n...> wrote:
> Hi Paul,
>
> N.b. why was it you thought notes should sound
> higher in pitch with 1st as strongest partial?
>
> Can you explain a bit more?
>
"That's not exactly what I said . . . but when the cochlea is exposed
to more than one frequency at once, the pitches are
subjectively `stretched apart' compared to how they sound as isolated
sine waves. For a timbre without a second partial, the first partial
won't be affected as much by this `stretching.' So I would expect
that a timbre without a second partial would sound a little higher
(maybe a few cents, subjectively rather than physically speaking)
than a timbre with a second partial." -- Paul Erlich

Prove it.
C'mon, Paul Erlich.
Show us proof.
Provide us with 3 detailed citations from the psychoacoustic
and/or neuroauditory scientific literature which show that what
you are saying is anything but idle speculation.
C'mon, Erlich, show us the hard evidence.
Give 3 quotes from the scientific literature that provide hard
proof what you're saying is more than unsubstantiated buls**t.
-------------
--mclaren

I wrote,

>> "That's not exactly what I said . . . but when the cochlea is
exposed
>> to more than one frequency at once, the pitches are
>> subjectively `stretched apart' compared to how they sound as
isolated
>> sine waves.

The ever-polite Brian McLaren wrote,

> Prove it.
> C'mon, Paul Erlich.
> Show us proof.
> Provide us with 3 detailed citations from the psychoacoustic
> and/or neuroauditory scientific literature which show that what
> you are saying is anything but idle speculation.
> C'mon, Erlich, show us the hard evidence.
> Give 3 quotes from the scientific literature that provide hard
> proof what you're saying is more than unsubstantiated buls**t.
> -------------
> --mclaren

Terhardt, E. (1971). Die Tonhöhe Harmonischer Klänge und das
Oktavintervall. Acustica 24, 126-136

The frequencies of a sinusoidal tone and of a complex tone with the
same pitch are slightly different. The investigations show that
usually the pitch of a complex tone is lower than the pitch of
sinusoidal tone of the same (fundamental) frequency. The frequency
ratio corresponding to the subjectively correct pitch interval of a
musical octave usually differs slightly from the value 2. This
phenomenon was investigated with low pure tones and with complex
tones. The results for complex tones are explained by the octave
intervals that were found with simple tones and the pitch differences
between simple and complex tones.

These conclusions were revisited here:

Terhardt, E., Grubert, A. (1987). Factors affecting pitch judgments
as a function of spectral composition. Percept. Psychophys. 42, 511-
514

Stimulated by a recent paper by Platt & Racine [Perception &
Psychophysics 38, 543-553 (1985)], we discuss the factors that
probably are involved in certain inconsistencies observed in pitch
judgments of tones with different spectral composition. Typically,
discrepancies reported in the literature are of the order of 10 cents
in magnitude. We point out that measurement of such small pitch
effects is heavily dependent on systematic individual differences;
and, when individual differences are averaged out (as is essentially
the case in Platt & Racine's experiments), verification of the actual
auditory stimulus SPL within a few dB is necessary. Utilizing the
virtual-pitch theory, we evaluate the effects of frequency, SPL, and
earphone frequency response. Further, we present experimental and
theoretical data on pitch of piano tones relevant to the problem. The
study elucidates that, taking into account the factors mentioned,
agreement between the various data considered, as well as theoretical
understanding, actually is much better than may have become apparent
on first sight.

In Terhardt, E. (1998). Akustische Kommunikation - Grundlagen mit
Hörbeispielen. Springer, Berlin/Heidelberg. 505 S. ISBN 3-540-63408-
8, pp. 353-356 describe how, in many cases, the pitch of the harmonic
complex tone turns out to be slightly lower than that of the pure
tone.

Webster, J.C., Miller, P.H., Thompson, P.O., Davenport, E.W. (1952).
The masking and pitch shifts of pure tones near abrupt changes in a
thermal noise spectrum J. Acoust. Soc. Am. 24, 147-152

When a pure tone is mixed with a noise of uniform spectrum, its
threshold is sraised. At levels above threshold, its loudness and
pitch are changed by the presence of the noise. Introducing abrupt
changes in the slope of the noise spectrum by filtering out
(rejecting) one octave changes these effects in the vicinity of this
gap. The masked threshold for a pure tone varies from the value for
unfiltered noise at the edges of the gap to a value approximately 25
dB lower at the middle. This indicates that gappedi-noise may be used
to mask out sounds outside of the gap without unduly raising the
threshold of sounds in the gap. Presence of white noise generally
raises the pitch of a pure tone whose frequency is between 500 and
4000 cps. Presence of noise with the gap does not raise the pitch of
a pure tone located in the upper half of the gap. For a tone located
in the lower half of the gap, the pitch is raised more than it would
be in the presence of unfiltered noise. The changes in the judged
loudnessof pure tones partially masked by a gapped-noise reaffirm the
importance of the tails of the excitation pattern in their effect on
loudness.

Webster, J.C., Schubert, E.D. (1954). Pitch shifts accompanying
certain auditory threshold shifts. J. Acoust. Soc. Am. 26, 754-758

A method was developed for getting a continuous and running record of
pitch matches between the two ears ...

Ten music students made these pitch matches (90 matches per octave)
from 150 to 9600 cps. In the ear in which the standard tone was heard
various types of hearing loss were simulated by masking noises. The
pitch tended to shift away from a region of hearing loss to a region
of no, or less hearing loss. Downward shifts were observed but were
never as marked as the upward shifts.

Allanson, J.T., Schenkel, K.D. (1965). The effect of band-limited
noise ond the pitch of pure tones. J. Sound Vib. 2, 402-408

An investigation has been made of the effect of a band of noise, one-
third of an octave wide, on the perceived pitch of a pure tone. In
general, the pitch was found to "move away" from the interfering
noise. However, in contrast with the results of erlier workers, the
shifts in pitch were found to be quite small and it is suggested that
this may be due to the difference in experimental procedures