[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: place pitch and temporal pitch

Dear Christian,
At 08:38 19.03.2004 +0100 you wrote:

>there is clear-cut evidence for both, for place pitch and for temporal pitch

> I conclude that pitch is best if the brain can exploit both cues,
> as is usually the case with most natural signals.
> Is there anything wrong with this view?

To some extent yes, I think so. Pitch is obviously a unitary perception.
Furthermore, merely a faint pitch remains when you excluded the spectral
code. Not by chance, place code is clearly the dominant one.
I see the mistake already in the traditional distinction between an
allegedly musical spectral cue and temporal envelope. Of course, the
natural spectrogram has a time axis as well as a frequency axis resembling
place. However, nobody could convincing explain to me harmony in terms of
interrelating place templates. As psychophysics shows, space code rather
than temporal envelope provides extremely high temporal precision. Given,
the ultimate goal is temporal signal processing in cortex, and spectral
code can be retranslated into intervals, then many open questions can be
resolved. Apparent weakness of  temporal code turns out to just reflect
missing spectral pre-analysis. For a while, you might share widespread
distrust against the idea that mammal hearing is largely based on the same
two-stage principle as is so called IFFT method in sonar technology.
However, there is overwhelming evidence for that. Neither immediate
measurement of periodicity nor frequency analysis could come close to the
same accuracy as can be reached by means of applying FCT twice. In so far,
it doesn't come a surprise that the spectral code is the main one.
Psychophysical indications for a nearly equally important role of
autocorrelation do not contradict to this finding since, according to
Wiener-Chintchine theorem,  the two stages together largely correspond to
autocorrelation and cepstral analysis.

> In K&D 1998 (see below) we presented evidence against AC models
> based on AC of the raw sound waveform.

Understanding your reasoning quite well, I can nonetheless not confirm
that. Your filters excluded the spectral code. Consequently, the admittedly
hidden two-stage autocorrelation of the raw sound waveform was also
excluded. This caused the discrepancy between your correct conclusion that
there is no autocorrelation and Peter Cariani's claim that autocorrelation
plays an important role. Do not be disappointed. You discovered what
remains without spectrally based autocorrelation.

>  Nobody would expect an AC model to explain all phenomena in pitch

Nobody would in particular me expect to explain anything.

> E.g., our difficulty to hear second-order periodicities clearly rules out
all-order statistics.

because the spectral basis of joint two-stage autocorrelation is missing.

Best regards and my sincere apology if you will not appreciate my effort to
reconcile Peter and you,

>Kaernbach, C., Bering, C. (2001). Exploring the temporal mechanism
>involved in the pitch of unresolved harmonics, Journal of the Acoustical
>Society of America 110, 1039-1048.
>Kaernbach, C., Demany, L. (1998). Psychophysical evidence against the
>autocorrelation theory of auditory temporal processing, Journal of the
>Acoustical Society of America 104, 2298-2306.