Re: mechanical cochlear model

["Richard F. Lyon" <DickLyon@xxxxxxx>]

Martin,

Maybe I misinterpreted what was behind your statements, and I agree 
it was impolitic (but maybe short of slander) to say what I thought.

> Just one example. In the 1950s it was well established and widely 
> discussed that temporary threshold shift (TTS) due to high-level 
> sound exposure does not occur at the place of the characteristic 
> frequency (CF) of the stimulus (say, 2 kHz) but about half an octave 
> towards the base of the cochlea (i.e., at the CF place of 3 kHz). 
> These often repeated observations were, and still are, evidence for 
> the view that the basilar membrane traveling wave peaks about half 
> an octave basalwards of the neural CF. Yet, Bekesy and his followers 
> succeeded in treating this evidence as if it did not exist.

The half-octave shift was hard to interpret back when the view was 
that the cochlea was pretty much linear, but has since been 
incorporated as an inevitable side effect of variable-peak-gain 
nonlinear models that match the observed BM mechanics.  It's easily 
incorporated in nonlinear traveling wave models.

Nonlinear models, like the real cochlea, have a maximum response that 
occurs at places basalward of the place that has the corresponding 
CF.  You are correct in saying that there is good "evidence for the 
view that the basilar membrane traveling wave peaks about half an 
octave basalwards of the neural CF", for high enough levels.  Only at 
very low levels does the wave, or the neural response, peak near the 
CF place.  This is well known and accepted, like the traveling wave 
concept itself, and is true independent of whether the model is based 
on the traveling wave being the main mechanism of energy transport. 
The half-octave basalward shift therefore doesn't provide much 
support one way or the other for what I think this argument might be 
about.  Or maybe I'm wrong on what it's about, since I seem to be 
unable to fathom the motivations...

The half-octave shift of the response maximum with level (or at least 
part of it) can also be induced by stimulating the efferents, or by a 
loud contralateral sound that stimulates the efferents.  Anything 
that "turns down the gain" of the "cochlear amplifier" will make the 
wave peak earlier (more basally) and damp out sooner.  Noise-induced 
TTS is just one extreme example, and it certainly hasn't been 
ignored.  Other such effects are widely investigated.  Here's a 
recent paper available online that talks about some: 
http://lib.ioa.ac.cn/ScienceDB/JASA/JASA1997/PDFS/VOL_102/ISS_3/1719_1.PDF 
(I don't know what this site in China is about, but they seem to have 
a lot of JASA papers freely available -- is that legit?).

I didn't know von Bekesy, but probably some of "his followers" are 
among the people I know who have incorporated nonlinearity into our 
understanding of hearing over the last four decades.  Why do you say 
they treat the evidence as if it does not exist?  Evidence that's 
hard to explain is always good to examine; that's where new ideas and 
findings come from.  So bring on some more.

As an example of evidence that's hard to explain, we have the 
blocked-round-window experiment of Perez 2009.  It needs to be 
investigated further, so we can see if it can be replicated, and if 
so what it means.  If occlusion by super-glue is really stiff enough 
to prevent the stapes initiating a traveling wave, we should be able 
to see that by greatly reduced stapes motion, or reflected energy in 
the ear canal, or impedance change at the eardrum or some such 
measurement.  If the threshold of hearing is really unchanged when 
the stapes isn't moving, that's one thing.  If the stapes is moving 
in spite of super-glue on the round window, that's something else.

The comment on Martin's site says
"Immobilization of the round window reduces, or even abolishes, a 
local motion of the basilar membrane (BM) toward the scala timpani 
upon sound exposure. The reason is that a volume shift in the scala 
timpani is impeded, or even made impossible, if the round window 
membrane can no longer bulge outwards. Therefore, a sound induced BM 
traveling wave along the cochlea must necessarily be greatly reduced 
under this condition.
The findings that a reduction of the BM traveling wave leaves hearing 
thresholds unchanged, but increases vulnerability against acoustic 
overload, constitute a direct experimental confirmation of a 
completely new theory of cochlear functions that was first published 
16 years ago. (Comment Martin Braun)"

This seems to be way beyond what the authors of the paper could dare 
to conclude.  If it turns out to be true, then we would all have to 
admit that the traveling wave isn't the effective stimulus for 
hearing; but we're not there yet.

I'm not convinced that Reinhart's explanation solves the problem.  If 
the stapes moves and initiates a traveling wave, the fluid has to 
have some place to go; that is, the basal region is in the long-wave 
region, not just a wave confined to the around the membrane.

If it's not the round window, then maybe there's another possible 
path, as Dave mentioned, "There is some very good evidence that there 
is a third window in the cochlea from H. Nakajima et al that was 
presented at this year's ARO meeting."  They considered various 
possibilities.  For those who were not there, here's their abstract:

Is There a Physiological Third Window? Measurements of Human 
Cadaveric Intracochlear
Differential Pressures for Round Window Stimulation with Fixed Stapes
Hideko Nakajima, Saumil N. Merchant, John Rosowski
Abstract:
It has been hypothesized that the inner ear is a rigid structure 
filled with incompressible fluid, with only two flexible windows to 
produce differential pressure across the partition for transduction 
of sound. This hypothesis has been supported with experiments in the 
past for the case of a normal inner ear with forward stimulation 
through the oval window. However, this evidence does not rule out the 
existence of a normal physiological third window that is small in 
conductance compared to the oval and round windows, such that the 
small third window influence is insignificant under normal 
conditions. In the case of reverse round window (RW) cochlear 
stimulation, fixation of the stapes should prevent cochlear 
transduction if the hypothesis held true. However, it has been 
reported that RW stimulation has improved hearing in patients with 
fixed stapes (Beltrame et al 2009). RW stimulation has potential as a 
treatment for stapes fixation when stapedectomy is contraindicated. 
Our experimental results in cadaveric temporal bone preparations show 
that reverse RW stimulation after fixation of the stapes resulted in 
higher pressures in both scalae, which were similar (within 1 dB) in 
magnitude. However, the calculation of the complex differential 
pressure revealed significant pressure difference, indicating 
achievable hearing. These results suggest that a clinically-relevant 
physiological third window likely exists in the scala vestibuli 
compartment. Possible fluid release mechanisms include the 
endolymphatic duct, neural and vascular channels, as well as the 
possibility that the bone surrounding the inner ear is not infinitely 
rigid throughout, and/or that the combination of having unequal 
volumes in scala vestibuli and scala tympani with slight fluid 
compressibility results in a physiological third window.

I think this is pretty good evidence that people are working on 
understanding the data, not ignoring it.

Martin, since I don't have the Perez paper handy, can you tell us 
what the statistical support was for their observation that "In the 
four control ears, there was no change in ABR threshold 24 hours 
after the round window was occluded."?  Would a hypothesis of a few 
dB of threshold rise be equally compatible with their data?  That's 
what the "third physiological window" idea would suggest.

Dick