From: hal@finney.org
Date: Sun Aug 15 1999 - 14:02:08 MDT
Robert J. Bradbury, <bradbury@www.aeiveos.com>, writes:
> They may not actually reside "in" the neurons, but sit in the synapses.
> Nanobots, I beleive, don't have the power to interfere with the electrical
> signals, but can wrestle with the chemical messages.
As I recall, a neural impulse (technically called an action potential)
is an electrochemical process. I had to study this (and model it) in
detail for a college class, but that was 20+ years ago. The idea is
that there is normally a voltage difference between the inside and the
outside of the neuron. This is an artificial difference maintained by
constantly pumping ions so that the concentration outside is different
from that inside, kind of like a bilge pump in a leaky ship.
As the neural impulse approaches a particular region of the neural
membrane, the electrical potential gets reversed. This causes certain
channels in the membrane to open, and ions flood through (icebergs
puncturing the hull of the Titanic and water rushing in). This will
maintain the disturbance in the potential and propagate it farther down
the neural membrane. After a relaxation period the channels close,
calcium and other ions are pumped back to their starting concentrations,
and the neuron is ready to fire again.
The process is initiated by chemical effects, neurotransmitters released
by other neurons causing channels to open and the flooding process
to begin. (Sensory neurons use other chemical or mechanical signals
to initiate the process.) Once any segment of the neural membrane is
caused to fire like this, it propagates all around the entire membrane.
Typically this involves travelling down an axon and eventually causing
other neurotransmitters to be released when the signal arrives.
The upshot is that it might be possible for a "nanite" (hate that word)
to sit inside the neuron and release chemicals internally which would
open the channels and initiate a firing impulse. More invasively,
it might be possible to create artificial channels which would reside
in the neural membrane and pump ions to stimulate or counteract the
neural firing. I don't have a clear picture of how much room there is
for all this machinery in the cell or its membrane, though. Maybe it
would do better to be between the cells if there's more room there.
As far as VR using this mechanism, you not only have to stimulate the
sensory neurons (and prevent them from responding to physical stimuli
that aren't part of the simulation), you also have to inhibit the
motor neurons, while detecting their firing and feeding that into the
external-world simulator. So the communication requirements are two-way.
There is also the issue of proprioceptive and vestibular senses, which
tell you how your body is positioned and oriented. Those too would need
to be faked with the VR sim.
Hal
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:04:46 MST