modulation audiogram?

Eckard Blumschein blumschein at et.uni-magdeburg.de
Fri Jan 4 05:16:21 EST 2002


Steve, you wrote:

>I personally find infrasound quite disturbing. Years
>ago, when I was in the pro-sound business, it was
>well known that ventilation noise is a big problem.
>I also note that folks raised in a city are better
>at ignoring infrasound.

Well, I was also told that women are more prone
to develop strong sensitivity against sound of very
large modulation periods. Maybe, fearful listening
is to blame for an exceptionally mistrained  brain.

>I think 20 Hz is the limit where we perceive
>sound as a continuous wave, and begin sensing
>individual waves.

I am not sure about that. Isn’t 20 Hz just the 70 dB
limit of audibility for a sinusoidal tone?
I imagine an essential change in the mechanism
of perception to happen already at higher frequency.
Zwicker/Fastl wrote that the thresholds for FM and
AM coincide at a modulation frequency between
64 and 128 Hz. Warren observed at about 100 Hz
a change from spectral perception to perception
with apparently doubled repetition rate. A French
musician told me that musical range ends at about
70 Hz. When I was a little boy, I wondered why
tones on the left end of the partition of a piano
sounded very strangely. Today, I suspect that
spectral hearing gradually ends at about 70 Hz.
In other words, perception of lower frequency
must be based on the same mechanism that is
also responsible for perception of modulation.
As shown in a JASA paper by Steinschneider
et al. in 1999,  this mechanism of hearing does
not rely on tonotopy but the perception of the
lowest tones is largely mediated  by auditory
nerve fibers belonging to high best frequencies.

>Myself, and others are quite sensitive to
>infrasound, but it may not be our ears that
>we use to sense it. Infrasonic weapons
>are based on the fact that various parts of the
>body resonate at infrasonic frequencies.

I agree that infrasound has various effects and
it may even cause VAD. However, what about
its inaudibility per definition, the mentioned
temporal mechanism implies its strong
dependency on SPL, and there seem to be
large differences among individuals.

>My impression was that infrasound produces a
>modulation of our perception like a audio system
>that clips from a low frequency signal that is later
>filtered out, leaving only the modulated higher
>frequencies.

Well. Let’s agree that average people may
still hear for instance sinusoidal tones of 15 Hz
at 140 dB while fluctuation strength has its
largest values around a modulation frequency
of 4 Hz. I am going a step further and conclude:
Audition must not be interpreted in terms of
an outdated mechanical model. In general, ears
do not simply perform frequency analysis.
This discrepancy between mechanistic theory
and a more complex reality is in particular
obvious to me below the limit of spectral perception.

>Note how our hearing is impaired by a
>pressure change, then consider doing that ~2-10
>times a second.  Perhaps some people perceived the
>effect as the same as external modulation.
>If so, the presence of unmodulated pink noise
>should negate the effect, since real infrasound
>would modulate everything the individual hears.

I substituted affect by effect and presents by
presence in order to possibly better understand
your ideas. Was this acceptable?
If you were correct, some people could actually
hear infrasound via modulation of any uniform
sound.
Admittedly, I imagine different mechanisms.
When I suggested a modulation audiogram,
I rather ascribed the role of a missing fundamental
to an infrasonic component. It does not matter
much whether or not it is already missing in the
physical spectrum of sound or in the cochlear
spectral analysis. The modulation audiogram
could indicate the modulation transfer function
of the cochlea as a whole together with the
mental ability for  reconstruction and
recognition of periods.

>In any case, the natural sources of infrasound
>are all things to avoid, so we are programmed
>to get very nervous when we sense it.

Why not also considering a contribution by
early learning?

Eckard

> > http://iesk.et.uni-magdeburg.de/~blumsche/




More information about the Audiolog mailing list