[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Reverse traveling wave does not exist
The design of Ren's experiment has a serious flaw. If you wish to look
for evidence of a reverse traveling wave using DPOAEs, the best way to do
this is to choose primaries that are going to peak at a location well
apical of your measurement location. This means that the distortion
product will be at a frequency that is significantly below the best
frequency of the measurement location. This in turn means that the
displacements will be small even though the pressures will be significant.
We often forget that a DPOAE is a pressure measurement, not a displacement
measurement. Ren's instrumentation does not have enough sensitivity to
do this sort of experiment without putting a reflector on the basilar
membrane (BM), something that he has not chosen to do because placing
reflectors on the BM can affect the sensitivity at that location. Data
taken with primaries around the measurement site can not address the
reverse traveling wave issue.
Ren did try to move his primaries to a location a bit apical of his
measurement site, but he used a very small f2/f1 ratio. Under this
condition, there is considerable phase cancellation and not much of an
emission comes out. It is also likely that with these small f2/ft ratios
that the emission measured in the ear canal largely originates from a
region basal to the best place for the primaries.
There is an experiment that can be done at the BM level that should be
able to detect a reverse traveling wave. This is to keep the primaries
well away from the measurement site and to use a reflector to improve the
sensitivity of the measurement. Even if the active process is damaged by
the placement of the reflector, it is likely that the active, nonlinearity
at the best place for the primaries will not be damaged and this can be
confirmed by simulataneously measuring the DPOAE in the ear canal.
The Ren paper also ignores much of the data that supports a reverse
traveling wave. There are many papers that present phase curves or group
delays for otoacoustic emisisons that support the reverse traveling wave
hypothesis. For example, we have published evidence that
electrically-evoked emissions elicited by current injection into scala
media propagate via reverse traveling waves:
Nakajima, H.H., Olson, E.S., Mountain, D.C. and Hubbard, A.E. (1994)
Electrically-evoked otoacoustic emissions from the apical turns of the
gerbil cochlea. J. Acoust. Soc. Am. 96 : 786-794
Through a comparison of electrically-evoked basilar membrane motion with
electrically-evoked otoacoustic emissions, we have also found
direct evidence for a reverse traveling wave:
Xue, S., Mountain, D.C., and Hubbard, A.E. (1996) Electrically-Evoked
Otoacoustic Emissions: Direct Comparisons with Basilar Membrane Motion.
Aud. Neurosci. 2:301-308.
And finally, through comparison of DPOAEs, cochlear microphonic, and
electrically-evoked emissions, we have further evidence for reverse
traveling waves. The forward travel time measured with the CM is
approximately equal to the reverse travel time measured with the
electrically-evoke emission and the round trip travel time for the cubic
distortion product is equal to twice the reverse travel time measured with
the electrically-evoked emission.
Mountain, D.C. Nakajima, H.H., Rafee, S., and Hubbard, A.E. (2000)
Forward and reverse traveling waves in the gerbil cochlea. In: Symposium
on Recent Developments in Auditory Mechanics, H. Wada, T. Takasaka, K.
Ikeda, K. Ohyama, T. Koike eds. World Scientific Publishing, Singapore,
All standard models of cochlear mechanics support forward and reverse
traveling waves in cochlear regions where the characteristic frequencies
are equal to or higher than the frequency of the wave. This is just
physics, and I think we all agree that ultimately, the cochlea is just a
physical system, all be it a complicated one.
For those list subscribers who are familar with MATLAB, I suggest you play
with a simple model for otoacoustic emissions that we have put up on the
When you run this model, you will see that although there is a significant
pressure signal with phase characteristics of a reverse traveling wave,
the BM displacement basal to the generation site is very small. Note to
elicit such a wave, the stimulus frequency must be below the
characteristic frequency of the generation site, but then this is exactly
the situation for the 2f1-f2 distortion product.
David C. Mountain, Ph.D.
Professor of Biomedical Engineering
44 Cummington St.
Boston, MA 02215
Phone: (617) 353-4343
FAX: (617) 353-6766
Office: ERB 413
On Tue, 30 Mar 2004, Andrew Bell wrote:
> List members will no doubt be interested to note a recent paper by Tianying
> Ren [Nature Neuroscience, 21 March 2004] which supplies a convincing
> demonstration that the reverse traveling wave in the cochlea does not exist.
> Instead, Ren could find only a fast backward compressional (pressure) wave.
> The reverse traveling wave is a key entity required by conventional
> cochlear mechanics to explain acoustic emissions of all kinds, including, in
> the case under observation, distortion product emissions (DPOAEs). The
> 2f1-f2 distortion product, for example, originates at a location on the
> partition through non-linear interaction of two primary tones (e.g., 17000
> Hz and 15455 kHz in one gerbil experiment reported) giving a strong
> vibration at 13910 Hz; this tone is presumed to travel backwards, via a
> traveling wave of displacement, to the stapes and the ear canal, where it is
> To Ren's surprise, he found with his scanning laser interferometer that he
> could detect no displacement of the basilar membrane at 13910 Hz until
> _after_ the stapes had started vibrating at this frequency. That is, the
> stapes vibrated some 50 us _before_ the basilar membrane did. There must
> have been a fast (nearly instantaneous) compressional wave at 13910 Hz which
> originated at some distance along the partition and excited movement of the
> stapes; only later did the basilar membrane join in.
> This paper will cause us to reexamine the basics of how the cochlea works.
> For example, how do we explain spontaneous emissions in terms of 2-way
> (forward and backward) travel times in a reverberating cochlea? A factor of
> 2 is hard to hide. A couple of years back discussion on this list centred
> around the question of whether the (forward-) traveling wave is itself an
> epiphenomenon, arising only as a consequence of outer hair cells reacting to
> a fast pressure wave entering the cochlea. This latest work gives strong
> support to this notion. If outer hair cells can produce a fast pressure
> wave, it seems natural to suppose that, by reciprocity, the fast pressure
> wave is the key stimulus in the cochlea.
> Andrew Bell
> Research School of Biological Sciences
> Institute of Advanced Studies
> Australian National University
> Canberra, ACT 0200, Australia
> phone +61 2 6125 9634
> fax +61 2 6125 3808