[From Bruce Gregory (980305.1500 EST)]
In a e-mail to me, Marc Abrams said:
Could you elaborate a bit more. Do you have any pet theories
about how perceptions are "learned" and control established.
I'm responding on the CSGnet in case anyone else is interested.
Learning a new skill involves extending the domain of control.
The first step in this process seems to be organizing a new
perception, because until the perceptual organization exists,
there is nothing to control. Looking over a variety of skills,
it seems to me that directing attention is the first step in
creating a new perceptual possibility, e.g., recognizing a
solder connection. Once you "fix" this perceptual
organization, you can begin to solder. But at this point, you
probably produce both good solder joints and bad or "cold"
solder joints. Improving your performance now requires a new
perceptual organization -- one that embodies the difference
between a "good" solder flow and a "poor" solder flow. Once you
can literally _see_ this distinction, it becomes possible for
you to make "good solder connections". "Good solder connection"
seems to embody both a perception to be controlled and a
reference value for that perception. These seem to come as a
package when we learn a new skill. You might consider that the
perceptual to be controlled is "solder" joint formation and that
the appearance of the solder as it flows is the reference value.
Indeed, this is the way that we would probably model the
phenomenon. I prefer the perceptual reorganization description,
however, because I seem to have no direct access to many of my
reference levels (I can't adjust my desire for chocolate!) but I
_can_ direct my attention. when I do this, I am able to create
new "possibilities of perception" or perceptual organizations.
This suggests to me that what we call reorganization associated
with learning largely comes about through perceptual
reorganization rather than through altering reference levels or
links between perceptual inputs and outputs. The latter changes
can very well occur, but the learner seems to have no access to
them. (If you want to change the way someone acts, you have to
do it by directing their attention to a new perception).
If there is any merit to this analysis, it suggests that the
key question to ask when it comes to learning something new is,
"What do I have to perceive to makes this possible." Teaching a
new skill involves the ability to provide a helpful answer to
this question.
Bruce