Re: cochlear discussions

[Jont Allen <jba@xxxxxxxxxxxxxxxx>]

Andrew Bell wrote:
>
> Dear Neil and List:
>
> I strongly agree that the combination of tectorial membrane (TM) and
> OHC provides the basis for frequency analysis in the mammalian ear.

I think there is strong evidence (see my papers at my web site for
published references) that the TM modifies cochlear tuning. However,
this has been in strong debate for more than 15 years. I dont think
you can end this debate simply by pontification.

> Yes, auditory science has been "labouring under the possibly false belief
> that the frequency selective properties of mamallian ear are due to the
> mechanical properties of the basilar membrane."
>
> Let me see if I can emulate your succinct style, and provide 5 good reasons
> why the basilar membrane is not an appropriate structure for frequency
> analysis.
>
> 1. Its combined mass and stiffness do not vary by the required factor of
> 10^6 (to provide a response over 3 decades of frequency).

Fletcher (JASA, 1930; also see his book, recently reprinted
by ASA "Speech and Hearing in communication") accounts for the variation
in stiffness and mass by the required amount. However there are other
ways to argue this question, and I dont think it has been proven that
the variation in Stiffness/Mass is insufficient to account for cochlear
tuning. It HAS been argued, by Naidu (1997), "Diversity in Auditory Mechanics"
World Scientific, Ed. Lewis et al. page 542-8.

Again, you are being dogmatic, to the extreme, IMO.

> 2. In some bats, the mass and stiffness are not even tonotopically
> organised, so that there is a so-called "paradoxical change in stiffness"
> (Vater, 1988) in which the stiffness of the 70-111 kHz region is much
> _lower_ than that in the adjacent 60-70 kHz region.

Actually Vatter and Kossl's (1995) story is wonderful for those of us who
believe in BM/TM tuning. There are two cochlear maps, that join, but have
different slopes, and two sets of tunings, that correspond to these two
regions. One might also conclude from their data that the TM plays a major
role in that tuning. But it is widely believed that the BM is also involved.
By direct measurements, the BM is tuning, after all.

>
> 3. The BM has interfering structures, such as blood vessels, running along
> its length.

I dont think this is relevant to the argument.

>
> 4. It is sometimes found, along with a completely formed (and presumably
> functional) cochlear partition, sitting on bone.
>
> 5. In birds and crocodiles, it is found with holes penetrating it,
> short-circuiting the differential pressure that is supposedly required to
> generate a traveling wave.

It is widely accepted (but perhaps arguable) that birds and crocodiles do
not have traveling waves, or mammalian-like auditory systems. So this
does not argue the case either, IMO.

>
> Nevertheless, it does possess graded mass and stiffness. Braun (Hear. Res.
> 78 [1994], 98-114) suggested that the BM was designed to absorb excess
> cochlear vibration (at high SPLs), which is a good suggestion.

Stiff objects do not "absorb" anything. Matched transmission lines and resistors
(dash pots) absorb vibration. Intensity has nothing to do with absorption,
rather impedance matching is the key to good absorption. What is
"excess cochlear vibration" anyway? My question is posed rhetorically.

>
> Andrew.

Jont Allen

--
Jont B. Allen
AT&T Labs-Research, Shannon Laboratory, E161
180 Park Ave., Florham Park NJ, 07932-0971
973/360-8545voice, x7111fax, http://www.research.att.com/~jba

"You can't hope to win unless you know how to lose."

Three things must be in harmony on the internet:
        Law, economics, and technology.