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

Re: Auditory efferents and phase locking

At 10:09 AM +1000 8/30/11, Matt Flax wrote:
...  it suggests paradigms which are very
different to the classical active travelling wave theory (which
typically doesn't incorporate any form of efferent activity).

The travelling wave theory certainly should include MOC efferent effects, but often they are neglected for simplicity. I always include efferects in discussions of traveling waves, because they seem so critical to understanding how the dynamic range compression works by altering the dispersion relation that describes the wave mechanics.

See this great early description of efferent/afferent roles by Duck On Kim (I can provide a copy on request):
title={Functional roles of the inner-and outer-hair-cell subsystems in the cochlea and brainstem},
  author={Kim, DO},
  journal={Hearing Science},
  publisher={College-Hill Press, San Diego, CA}
This paper is not about traveling waves, but is implicitly in the context of traveling waves, as shows up clearly in various places, such as interpretation point 7 on p. 256, where he talks about the "the basal travel of the outer spiral fibers" to align same-CF fibers with the basalward region that amplifies the wave arriving at the related place.

Evans says "This efferent innervation probably acts to control the OHCs in their role as cellular amplifiers of the travelling wave (produced by sound) on the cochlea's basilar membrane."
title={Here today, hear tomorrow: a transient inhibitory synapse regulates spiking activity in developing inner hair cells through facilitation},
  author={Evans, M.G.},
  journal={The Journal of Physiology},
  publisher={Physiological Soc}

Here's a paper in Serbia that says "Activation of medial olivocochlear system (MOCS) alters the cochlear output decreasing the travelling wave within cochlea."
title={Contralateral acoustic suppression of transient evoked otoacoustic emissions: Activation of the medial olivocochlear system},
  author={Komazec, Z. and Filipovi{\'c}, D. and Milo{\v{s}}evi{\'c}, D.},
  journal={Medicinski pregled},

And of course, there are my papers; e.g. this oldie:

  title={{Automatic gain control in cochlear mechanics}},
  author={Lyon, R. F.},
  booktitle={The Mechanics and Biophysics of Hearing},
editor = {P Dallos and C. D. Geisler and J. W. Matthews and M. Ruggero and C. R. Steele},
  publisher = {Springer-Verlag},
  address={New York},
  pages = {395--420},
which says, "...we cover possible AGC mechanisms and mathematical modeling techniques. These involve active outer hair cells whose properties are controlled by the efferent system, resulting in variable-gain wave propagation in the cochlea."

This paper by Slaney and me does a good job describing a range of traveling wave models that incorporate efferent control:
  author =       {Slaney, M. and Lyon, R. F.},
title = {On the Importance of Time---A Temporal Representation of Sound},
  booktitle =    {Visual Representations of Speech Signals},
  publisher =    {John Wiley and Sons},
  address = {Sussex},
  editor =       {M. Cooke and S. Beet and M. Crawford},
  year =         {1993},
  pages =        {95--116},

A general method for mapping traveling wave models to cascade filterbanks with feedback (efferent) control is in this one:
  author = {Lyon, Richard F},
  title = {Filter cascades as analogs of the cochlea},
  editor = {Tor Sverre Lande},
  year = {1998},
  isbn = {0-7923-8158-0},
  publisher = {Kluwer Academic Publishers},
  address = {Norwell, Mass.},
  pages = {3--18},
  booktitle = {Neuromorphic Systems Engineering: Neural Networks in Silicon}
"The filter-cascade approach to modeling the cochlea is based on the obser- vation that small segments of the cochlea act as local filters on waves propa- gating through them. Thus, a cascade of filters can emulate the whole complex distributed hydrodynamic system. This modeling approach can include com- pressive and adaptive aspects of the peripheral auditory nervous system as well, using analogs of cochlear nonlinear distortion and efferent feedback."

The only way a cochlear model can deny the interaction of MOC efferents with the traveling wave is if the model doesn't have traveling waves. A model that neglects the efferents doesn't deny them; but I agree that is an all-too-common situation. It's not a defect of the "classic traveling wave theory", but perhaps of some embodiments of that theory.

Sometimes the details are not well worked out, but that doesn't mean the efferents don't have a role in controlling the "cochlear amplifier" that causes the traveling wave to pick up power as it travels.

Matt, in my MoH paper that you attended, I also presented a simple illustration of how the phase may be stable even as the efferent activity varies, so to first order the phase locking will be not much affected, as long as the level is within the range that drives the afferents. The system is amazingly robust in this sense. It is indeed tricky to ascribe a clear psychophysical or physiological effect to the action or inaction of this feedback path, since the time pattern delivered to the IHCs is so stable. Perhaps the gain-control role is primarily protective, as many have speculated?