Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!samsung!think.com!mintaka!ogicse!milton!hlab
From: shebs@Apple.COM (Stan Shebs)
Newsgroups: sci.virtual-worlds
Subject: Re: Virtual Reality, Helmets, and Gloves (+ Safety Issue)
Message-ID: <1991Apr5.185934.15316@milton.u.washington.edu>
Date: 5 Apr 91 18:11:31 GMT
References: <1991Apr4.171457.17426@milton.u.washington.edu>
Sender: hlab@milton.u.washington.edu (Human Int. Technology Lab)
Organization: Apple Computer Inc., Cupertino, CA
Lines: 41
Approved: cyberoid@milton.u.washington.edu



In article <1991Apr4.171457.17426@milton.u.washington.edu> dowen@BBN.COM 
(Dan Owen) writes:

>[...]  The number of senses provided with substitute
>sensory data and the overall fidelity of the virtual reality which is
>provided need to pass a threshold to be accepted.  This threshold may be
>defined as that level of fidelity to a plausible reality which allows
>the virtual participant to temporarily suspend disbelief in the
>fictitious nature of the experience.  Once this important act of
>perceptive cognition has occurred then for a time the virtual reality is
>accepted as actual reality.  Numerous anecdotes of the fear, anger and
>nausea experienced by pilots in flight simulators and tank crews in
>networked simulation training environments attest to this fundamental
>requirement of virtual reality.

I like this definition.  The "suspension of disbelief" is a good subjective
way to avoid insisting on particular hardware.  For instance, when I play
NetHack, even with its character graphics and non-real-time behavior, I find
myself experiencing fear (trapped in a corridor), relief (escaped), and
anger/depression (died).  So, if your model is sufficiently compelling, then
the nature of the display is of lesser importance.  Perhaps the attraction
of the full sensory replacement is that it works well with both good and
bad software - without any alternative input, the brain has to make do with
what is being provided.

This brings to mind another issue that I've been concerned about.  Accepting
a synthetic reality as actual reality is potentially hazardous - the nausea
attendant upon the eyes getting information that conflicts with that coming
from the inner ear, for instance, is just the tip of the iceberg.  What if
the software convinced you that your virtual leg had been cut off?  Would
your real leg become paralyzed?  Could you actually be killed by suggestion
alone?  Can we set some sort of limits on what can be simulated?  One solution
for nausea would be to restrict virtual acceleration to under the sensitivity
of the inner ear - if you want to move fast, it should happen instantaneously.
Does anybody have, or is developing, such a set of safety principles?

                                                Stan Shebs
                                                Apple ATG System Software
                                                shebs@apple.com