Subject: Re: Physiological models of cochlea activity - alternatives to the travelling wave From: Matt Flax <flatmax@xxxxxxxx> Date: Tue, 2 Oct 2007 12:57:59 +1000 List-Archive:<http://lists.mcgill.ca/scripts/wa.exe?LIST=AUDITORY>Hi aja and the list, You have brought up feedback ... thank goodness as we are now back onto the original topic - physiological modelling. Indeed feedback is powerful. Current physiologically based models incorporate feedback. Alas a majority of them are dependent on the traditional travelling wave process. This is not a problem however, I can use the travelling wave as a passive mode of input if the research community suggests it is the way .... Perhaps the biggest problem with feedback models is that they don't specify the physiological correlation between the feedback and the hair cell, Organ of Corti nor Cochlea. Here are some feedback models with no physiological basis for the feedback : a] Zwicker's fantastic model - which begins with the travelling wave [1] b] Hopf bifurcation models - which augment the travelling wave. [2] for example. The Hopf bifurcation models at the moment are on shaky ground as they were dreamt up from pure mathematics and later sought a physiological explanation. This can work sometimes, however creates problems in their current form. By the way, they produce some fantastic results. They are also fundamentally flawed as they can't produce other results, such as frequency dispersion. Does anyone else have PHYSIOLOGICALLY BASED models to mention out there? Matt [1] @xxxxxxxx{Zwicker:1986a, author = {Zwicker, E.}, title = {A hardware cochlear nonlinear preprocessing model with active feedback}, journal = {Journal of the Acoustical Society of America}, year = {1986}, volume = {80}, pages = {146-153}, number = {1}, month = {July} } [2] @xxxxxxxx{duke:2003, author = {Duke, T. and J{\"u}licher, F.}, title = {{Active Traveling Wave in the Cochlea}}, journal = {Physical Review Letters}, year = {2003}, volume = {90}, pages = {158101}, number = {15}, publisher = {APS} } On Mon, Oct 01, 2007 at 06:12:11PM -0400, A.J. Aranyosi wrote: > I think the point Martin is trying to make is a bit more complicated than > that, if I'm interpreting it correctly. I suspect most of us would agree > that the amplitude of BM displacement at 0 dB SPL for a "dead" or > compromised cochlea is not functionally significant. However, Martin's > claim seems to be that in a cochlea with normal sensitivity, this passive > BM displacement is what drives OHC amplification because the OHCs are > dependent sources (i.e., their hair bundles must be displaced in order for > them to amplify). In Martin's view, if this passive motion is too small to > cause significant OHC bundle deflection, then the cochlea cannot possibly > be driven by BM deflection. > > I disagree with this conclusion because it ignores the phenomenon of > feedback. If the OHCs are part of a positive feedback loop, this problem > goes away because the feedback insures that the motion of the BM is never > as small as it is in the passive case. Imagine if the BM moves one > picometer; this motion deflects hair bundles by a tiny amount, causing OHCs > to increase BM motion, leading to more bundle deflection, etc. When you > try to reason it through step-by-step like this it still sounds like the > OHCs have to detect the passive vibration, which technically is true. But > the key to feedback is that an input that causes an insignificant (and > probably undetectable) response on a single pass through the feedback > system can grow rapidly as it passes repeatedly through this loop. > > Having said all of this, there are problems with positive feedback models > as well (for example, they are extremely sensitive to the amount of gain, > which is usually not well controlled in biological systems). I > wholeheartedly agree with Martin's statement that we need more measurements > before we can truly separate the good from the bad models. > > Erik Larsen wrote: >> I wouldn't want to insult anybody's intelligence by posting the >> calculation of that number online. Its not a meaningful thing to do anyway >> for a cochlea with dysfunctional outer hair cells. Its like asking how >> long it would take someone to run a marathon with a broken leg. >> >> Specifically for the issue of motion sensitivity, you appear to assume >> that the cochlea somehow behaves as a laser doppler vibrometer: you take >> the limitations of this piece of equipment to somehow imply that the >> cochlea would have the same limitations under similar circumstances. If >> you could describe to the readers of this thread how you arrived at that >> conclusion, I'm sure we will tremendously enjoy reading about it. >> >> Martin Braun wrote: >>> Erik Larsen wrote: >>> >>>> Perhaps Martin meant that nobody has been able to measure BM motion >>>> below 60 dB with dysfunctional outer hair cells? But that wouldn't be a >>>> proof there is no motion. In fact, it would be difficult to understand >>>> why there wouldn't be any motion below 60 dB, however small. >>> >>> For many years now, the measurement sensitivity for BM excursions has >>> been well below 1 nanometer. Now, if BM motion is < 1 nm at 60 dB, how >>> big then can it be at 0 dB? >>> >>> Please do the maths, and then come back. You can be sure that all readers >>> of this thread will tremendously enjoy that figure. >>> >>> Martin >>> >>> --------------------------------------------------------------------- >>> Martin Braun >>> Neuroscience of Music >>> S-671 95 Klässbol >>> Sweden >>> web site: http://w1.570.telia.com/~u57011259/index.htm >> -- http://www.flatmaxstudios.com/ http://www.flatmax.org Public Projects : http://sourceforge.net/search/?type_of_search=soft&words=mffm http://www.psysound.org ----- End forwarded message ----- -- http://www.flatmaxstudios.com/ http://www.flatmax.org Public Projects : http://sourceforge.net/search/?type_of_search=soft&words=mffm http://www.psysound.org