🚀 go-pugleaf

On Wed, 27 Sep 2006 13:10:28 -0500, "Mike Rieves" <mriev@hotspam.com>
wrote:

<snip>
>
>The Neurophone however, is a totally different animal. 
>I notice that at www.neurophone.com they are talking about ultrasonic 
>transmission, rather than RF transmission, which I think may be plausible, 
>though I'm far from an expert on the subject. 
>

It doesn't matter whether it's ultrasound, RF, microwave, magnetic, or
electric fields.  The problem is not getting a signal into the head,
its getting *many* signals into the head in just the right places to
represent the frequency content of each auditory channel.  Even
if you drilled a hole and placed an electrode at any site you chose,
and pumped the audio signal in directly, you would just get a series
of buzzes in time with the signal peaks.

Once again, the auditory system is not even remotely like radio
or telephones.  The cochlea breaks down the signal into separate
frequency components, and each goes to the brain on its own
nerve.  If you look at the signal on any one of those nerves, it is
a spike train where the spikes per second is proportional to the
signal *strength*... not frequency.  The spike rates are in the
low 100s of spikes per second... definitely not adequate to 
encode anything like audio.  If you stuck your electrode or
neurophone signal directly into the neuron and applied audio,
only the peaks that were large enough to exceed the neuron's
threshold would produce spikes.  After the neuron fires its
spike, it goes into a "refractory period" where it has to recover
before it can fire another spike.  So, no possible way to send
audio over a nerve.  All the nerve carries is information about
the envelope of its single frequency component, sort of like
one column of LEDs on one of those home stereo spectrum
displays, where there are different columns of LEDs for each
band.  It's the full array of thousands of these that make up
the experience of hearing.

Now since the Neurophone has no way single out individual
neurons and send just the right spike rate to each one, the
best it can do is fire a whole bunch of neurons together.

Think of the task as playing the piano: Normal hearing has
way more than 88 "keys", but that will do for an example.
To get a melody, you have to press down different keys
at different times.  But the Neurophone can only press
*all* the keys or no keys, it can't single out specific keys
at specific times.  So lay a board across the piano
keys and try hitting it in the center with a hammer.
You can convey *information* this way, but not a melody.
In the same way, a single-channel stimulation system can 
not convey speech or music to the brain.

Best regards,





Bob Masta
dqatechATdaqartaDOTcom
 
            D A Q A R T A
Data AcQuisition And Real-Time Analysis
           www.daqarta.com
Home of DaqGen, the FREEWARE signal generator