Audiogram defects (was Re: Cochlea -- quality of filtering)

susan moreland smorelan at MAGNUS.ACS.OHIO-STATE.EDU
Sat Sep 23 11:53:18 EST 1995


>On 22 Sep 1995, susan moreland wrote:
>> a widening of critical bands is often experienced by the person as reduced
>> speech discrimination in noise.  as the bands expand, it becomes more and
>> more likely that the noise will mask the signal.  there's a fairly lengthy
>> explanation behind that phenomeonon and it was, in fact, another of my comps
>> questions.
>Noise had been view as a double edge sword.  Reducing perception and
>increasing perception.  Neuro firing are assumed to generate noise, but
>the noise can be used to amplify a weak signal.  How does increase in
>width in bands change the advantage created by noise for signal detection
>in your opinion?  Ron Blue
>


The most important point is that there exists an internal auditory bandpass
filter that can be centered ideally at the signal frequency.  This filter
reflects peripheral activity on the basilar membrane, independent of central
mechanisms.  The filter, via resonance, removes the noise at frequencies
below fl and above fu outside the filter and passes any signal between fu
and fl.  Signals that are enveloped by two different critical bands will be
perceived as being two separate signals, but signals that are close to each
other (in the same critical band) can not be distinguished from each other.
Therefore, the critical band is offered as the basis for our ability to
discriminate between different frequencies. 

Critical bands are the reason why we enjoy the degree of frequency
selectivity that we do.  When outer hair cells are damaged due to aging or
other causes, there occurs not only a loss of sensitivity, but also
selectivity.  This happens most often in the basal parts of the cochlea so
the resulting selectivity problems are usually visible in the higher
frequencies, at least in the beginning.  When the hair cells are destroyed
in a particular location, the critical bands in that area widen.  The
narrower a filter is, the better the frequency resolution, so if the filters
become abnormally large the listener will have reduced capacity to
distinguish one sound from another if the two sounds are relatively close
together in frequency.   In turn, this affects the person's ability to
understand speech (the signal) in the presence of noise (the masker).
Specifically, according to Sommers and Humes (1993), increased filter widths
are "hypothesized to decrease the signal-to-noise ratio at the output of the
auditory filters and therefore make identification of speech signals more
difficult."


>
susan 

we stroll together silently,
just the two of us -- my wolf and me...




More information about the Audiolog mailing list