NetNews Usenet Archive 1992 #30/NN_1992_30.iso/spool/sci/virtual/4160

Internet Message Format | 1992-12-21 | 3.9 KB

Path: sparky!uunet!noc.near.net!hri.com!spool.mu.edu!uwm.edu!ogicse!news.u.washington.edu!stein.u.washington.edu!hlab
From: [e-mail] (Dave Stampe)
Newsgroups: sci.virtual-worlds
Subject: Re: TECH: Neural Interfacing
Message-ID: <[e-mail]>
Date: 19 Dec 92 00:22:32 GMT
References: <[e-mail]>
Sender: [e-mail] (USENET News System)
Organization: Department of Psychology, University of Toronto
Lines: 64
Approved: [e-mail]
Originator: [e-mail]


[e-mail] (Devon Copley) writes:
>
>Couldn't biofeedback be the crucial part of developing a really powerful input
>device?  I'm going to make a couple of assumptions; if they're true (and I
>really have no proof that they are, though if anyone has some, please offer
>it) perhaps "thinking at your computer" isn't impossible.
>
>1: We "learn" to use much of our nervous system through painstakingly
>repeating and practicing actions, strengthening the neural connections
>and pathways associated with them.  And I'm not just talking about learning
>how to play the piano, either: simple things like moving your index finger
>reliably are a result of long years of refining that output, by basically
>firing a few motor neurons and watching what happens.
>
>2: A reciever can be implanted in the brain, sensing the output of a large
>number of neurons.
>
>Admittedly, now, a non-destructive (2) is pretty far from reality, and
>the only concrete examples I can come up with of (1) are in the motor
>realm (although that in itself might not be a problem).
>
>My point is, though, that to some degree it MAY NOT MATTER which neurons
>the computer senses, because if an appropriate feedback mechanism is
>provided, the subject will "figure it out" for himself, and with
>practice could conceivably develop the same precise control over the
>computer as I have over my right hand.
>
>Am I way off base here?  Are there many neurons in the cortex whose output
>we have no control over no matter how much feedback we get?
>
My own opinion here (but seems quite feasible from a neurological point of
view):
Most neurons in the brain learn how to connect themselves by feedback for
example, from recurrent connections from other cortical areas where they
send their own output to.  Now if you have a fairly fine electrode array
that has both sensing and exictation capabilities, you can train the local
neurons to get input/output in a useful fashion from a neural area.

Possible problems: You might have to implant this stuff fairly early in
childhood, or somehow create medically the "critical period" neural
plasticity needed for such training.  On the other hand, it is well known
that areas of the primary sensory/motor cortex seem to retain at least some
of their plasticity.  The type of plasticity must be that for which maps
of inputs are formed, which usually occurs before o a month or so after
birth, not (usually) during adult life.

The key is: local neural feedback, and plasticity.  I don't really think that
global (wide-area or bulk) neural stuff is going to be as useful, mainly
because that's not how sensory data in the brain is represented.  I think
that the best you could do with that is fairly global emotional/arousal
types of signals.  "Interested/Tired/Extroverted/Tense"  seems like a
rather difficult kind of signalling code to use.

The problem is that this kind of stuff needs lots of experimantal work,
of which nearly zilch has been done so far.


--------------------------------------------------------------------------
| My life is Hardware,                    |         Dave Stampe          |
| my destiny is Software,                 | [e-mail]    |
| my CPU is Wetware...                    | [e-mail]   |
| Am I a techno-psychologist, or just an engineer dabbling in psychology?|
---------------------------------------------------------------------------