induction

[Ken Easlon]

Let's take a look at role played by induction in the perception
of physical reality.

Here I'll define induction as the mental process of filling in
the blanks between sensory observations. I see before me the
appearance of a brick wall. I go to one end and feel the
outlines of a few bricks, I go to the other end and do the same.
The perception of a solid brick wall is induced in my mind, a
wall which I believe will yield a similar tactile experience no
matter which part of it I touch.

If I've had previous experience with holograms or painted or
virtual brick walls, then the mental process of induction may be
tempered with a counter process which reminds me that the
perception of solidity is a mental process.

If all my experience with brick walls is self reenforcing, and I
have no counter experience of phony walls, then the mental
process of induction becomes a hardened habit, producing a solid
world view which defies attempts to reduce it to a perceptual
phenomenon.

A hardened world view is not necessarily a bad thing,
particularly for engineers, so long as the view accurately
predicts the kind of response one can expect from interacting
with the extra-mental environment.

In my view however, if we are engaged in the business of mental
engineering (the design or maintenance of mental systems in
either biological or silicon brains), we need to distinguish
between the mental phenomenon of induction, and the mental
phenomenon of sensation. We need to be able to distinguish
between the raw information we collect from the extra-mental
environment, and the way the information arranges itself in our
mind.

For example, we repeatedly hear about mental models which refer
to real features in an external physical world. A more accurate
way of describing mental models would talk about intellectual
ideas which refer to the induced perception of a physical world
resulting from sensory and cultural experience.

When it comes to designing an artificial mind, we need to
remember that we don't simply teach this mind about the external
world, be we must actually construct the entire "physical" world
inside the artificial mind.

This is the point where we need to pay particular attention
to the process of induction, because we don't want to create
a "physical" world down to the last quark, but simply an
inductive process which will create the illusion of a
physical world made of quarks (and such).

Comments invited.

--
Ken Easlon
ke...@delphi.com