IUBio Biosequences .. Software .. Molbio soft .. Network News .. FTP

More than you wanted to know about hearing and hearing aids

Jerry Larson jlarson at op.net
Tue May 20 09:00:28 EST 1997

Following the recent discussion about how hearing aids work, which was
really about how hearing works, I talked to a friend who is an ENT
physician and auditory physiologist, and the king of oto-acoustic
admissions, and I found out some interesting stuff.

First of all, while what I was saying was basically correct (hearing aids
make use of residual hearing, and there is residual hearing for two
reasons: there are a lot of hair cells, and their frequency ranges
overlap), I underestimated the maximum conductive loss, which is more like
50 0r 60 dB.  That would be with a completely ankylosed ossicular chain.  A
perforated eardrum could cause the 35dB loss I suggested was the maximum. 
However, as I said, there is good surgery for this kind of problem, and a
stapedectomy could bring that 60dB loss down to 5 or 10dB, so in a
developed country you won't see a lot of significant pure conductive

As for the population of hair cells, there are about 16000 of them,
arranged in four rows, like this:
.  ...
.  ...
.  ...
.  ...
.  ...
.  ...
.  ...

Three rows of outer hair cells and one row of inner hair cells.  The inner
ones actually transduce the acoustic signal and pass the message on to the
nervous system; the outer ones are highly motile, and actively tune the
resonance of the basilar membrane.  They are responsible for amplification
and gain control.  They are much more vulnerable to most kinds of insult,
so they usually go first, leaving hearing sensitivity diminished by about
40dB, and also leaving the person without gain control, and hence with
abnormal loudness growth and sensitivity to noise.  That's why hearing aids
have automatic gain control, compression, etc.

This amplification by the outer hair cells is responsible for a great deal
of the frequency-specific response of the basilar membrane.  A depopulated
membrane has a very broad frequency-response curve, but the outer hair
cells make it much sharper.  

I don't have an exact figure on the frequency range of a given hair cell,
but I would guess there would be about a 6dB/octave rolloff, maybe 3, maybe
12.  In other words, a cell (group of cells, actually) that responds best
at 500 Hz might respond with halved  intensity to a 250Hz sound.  You
wouldn't be able to measure the difference in frequency response between
two adjacent groups of cells, and there are 4000 such groups for about a
20Khz bandwidth.  

I don't know whether an individual hair cell can function partially or
intermittently, or whether it's all or nothing.  But I imagine there would
also be interactive effects, such that a significant loss of outer hair
cells in one region, by diminishing the amplification and tuning of the
basilar membrane as a whole, would produce functional deficits in other
regions, or in the entire cochlea.  You have to look at the functioning of
the cochlea as a whole, not just individual cells.  So basically, as I
said, a damaged ear has diminished sensitivity, as well as reduced dynamic
range and gain control, distorted frequency response, etc.  Amplifying the
sound overcomes the problem of diminished sensitivity, but may make some of
the other problems worse, so you can try to design the aid to deal with
those problems too; and that's all I know about how a hearing aid helps.

jerry at neuromon.com

More information about the Neur-sci mailing list

Send comments to us at biosci-help [At] net.bio.net