[sci.virtual-worlds] VR Directions of Growth: VII, VIII

williamb@milton.u.washington.edu (William Bricken) (01/03/91)

Virtual Reality:  Directions of Growth
    Notes from the SIGGRAPH '90 Panel 
    
    Copyright (C) 1990  All Rights Reserved by William Bricken

        William Bricken
        Human Interface Technology Laboratory
        University of Washington, FU-20
        Seattle, WA  98125
        9/10/90
        william@hitl.vrnet.washington.edu



VII.  EDUCATIONAL APPROACHES

VR provides an exciting educational medium for exploring worlds and
for exploring ourselves.  It provides a training environment that is
rich, replicable, and responsive.  It permits direct evaluation of
educational theory.  The central educational issue for VR is one of
transfer of experience.  Do skills and habits learned in VR transfer
to the physical world?  Here are some educational issues:

        Constructivism
                "Human knowledge is essentially active."  Piaget
        Natural Semantics
                non-symbolic, preoperational interaction
        Programmable Participation
                conducive and responsive environments
        Cognitive Presence
                modifiable self-concepts, learning by becoming
        Social Reality
                unique concurrent worlds

Educational psychologists have long known that people actively
construct their experience of reality.  In VR, students will construct
their knowledge, then dwell within it, exploring their understanding.

Natural semantics means that the computational environment hides
symbolism in favor of displaying information in an innately
recognizable form.  The two-year-old criterion: if a kid recognizes
it, its natural.  The three Rs, all symbolic, will become the three
ACTs: enact, interact, and abstract.

VR provides the potential for completely customized, individualized
learning.  Educational environments will uniquely respond to the
participant-learner, in terms of both needs and preferences.  A
student model will not be necessary, instead the teacher and student
will modify the environment in support of student behavior.

We also have a tool for affective education, for sharing perspectives,
for mapping perspectives into broader contexts, for changing
self-image, for remapping capabilities.

Education is inherently social.  Explicitly shared worlds and multiple
concurrent agreements provide the opportunity for groupwork, social
consensus, and the construction of functional, multiparticipant
environments.

In general, everything we do to educate with words and pictures can be
provided as virtual experience.



VIII.  LESSONS LEARNED

Tom Furness, the Director of HITL, has over twenty years experience in
VR.  He pioneered most of the hardware interface devices we use today,
in the extremely demanding environment of military aircraft.
Personally, I have worked on VR related projects for six years,
beginning at Atari Research Labs in 1984.  Meredith Bricken designed
Autodesk's worlds, built Virtual Seattle for CHI'90, and has pioneered
research into the design of comfortable virtual environments.  Over
the years we have learned some lessons:

        Psychology is the Physics of VR.
        Our body is our interface.
        Knowledge is in experience.
        Data is in the environment.
        Scale and time are explorable dimensions.
        One experience is worth a trillion bits.
        Realism is not necessary.

A major theme of VR research is that Psychology, in the broad sense of
behavior, perception, cognition and intention, provides the rules and
the constraints of virtual worlds.  Psychology is the Physics of VR.

This may come as a shock, it is one of those truths that is obvious
after it is said, but elusive before it is stated explicitly.  Our
body is our interface.  Interface is not something that is out there,
in some machine.  Interface is a boundary which both connects and
separates, interface takes place at the surface of our skin.  From the
perspective of VR, interface is physiology, interaction is natural
behavior.  We simply want to use the power of computers to make
computation invisible.

Knowledge is in experience, it is not in some abstract, symbolic
representation.  Data is in the environment, it is not stored away in
some memory array.  These observations serve to remind us that we are
not the computer.  To understand computation, we should participate
within it, rather than writing programs to dominate it.  Humans have a
great skill for projecting outward, for becoming the tool we are
handling.  We need reminding that we are creatures who dwell inside an
environment.

VR is inherently multidimensional.  As well as freedom of translation
and rotation, in VR we can travel in scale and in time.  Think of
scale as simply another direction; when we traverse scale, size
instead of location changes.  We can also travel through time using
any of the techniques of film editing, including slow-motion, fast
forward, and temporal discontinuity.

There is a tremendous compression ratio between digital information
and human experience.  Very approximately, it takes a hundred million
polygons to simulate what we see in one scene.  Add duration,
multisensory channels, and interaction, and you get a lot of digital
information being transacted with each moment of consciousness.
Computation will not come close to this bandwidth for a long time.
Fortunately, virtual world experience does not require the information
density of physical reality.

Because our minds provide such tremendous flexibility in interpreting
what is outside of us, realism in VR is simply not necessary.  Our
cognitive plasticity permits even simple cartoon worlds of 500
polygons to be experientially satisfying.  We must design worlds that
respect our physiological needs.  For example, we conceptualize
perspective in physical space as having six degrees of freedom, three
in translation and three in rotation.  But our bodies have roughly
four and one-twelth degrees of freedom.  We move easily in all
directions on a plane, forward and to the side, but not up, off the
surface we stand on.  We rotate freely around the vertical by turning,
but our natural rotation forward, around our waist by leaning, is at
best 270 degrees (3/4 of a full 360 degree rotation).  And our ability
to bend side to side is only about 120 degrees, one-third of a full
rotation.  This adds up to a little more than four degrees of
physiological freedom.  Input devices which permit complete freedom of
translation and rotation usually get people lost in space.  The
dimensionality of our abstract perspective does not match that of our
physical construction.  We must also differentiate that which is
innate from that which is learned.  Pilots, for example, have learned
to fly in all six degrees of freedom.  Realism is both physiological
necessity and cognitive interpretation.  In VR, world design that
conforms to physiological necessity frees our minds to furnish the
rest of our reality.