[From Bill Powers (941207.0905 MST)]
I've been having some trouble with uploads since our computer center put
in a new Kermit program. Bear with me, I'm trying to solve the problem
of strange line-endings. I'd love to hear from anyone who has solved
this problem on a VAX machine.
···
------------------------------------------------------------------------
Hans Blom (941205)--
RE: "good" and "bad"
What makes a reference level of zero so much different from a non-zero
level?
Nothing. It's just one setting within the possible range. But if a
reference level is set low for a given perception, we will act to
maintain that perception at a low level. In other words, we will avoid
experiencing it, and if it appears we will act to lessen it. This is
what we generally mean by a "bad" perception: it's one we don't want to
experience, like pain. On the other hand, if a reference level is set
high, then we act when possible to bring the perception to a high level.
In that case, we would call that perception "good," meaning that it's
one we seek rather than avoid.
------------------------------------
1. for an elementary control system "good" is a perception that matches
its reference level; "bad" is a perception not matching a reference
level.
I see the problem. In my usage I'm speaking of how we decide whether a
perception is considered a "good" type or a "bad" type. But you bring up
the subject of error signals -- whether a perception matches or doesn't
match its reference level. Pain is a bad type of perception, judging
from the fact that we almost always act to keep it at a low level.
However, if there is an error signal -- that is, if the pain signal is
greater than its reference signal, that is bad, too, but if the pain
signal is at or below its reference signal that is good.
Language was obviously not designed for precision in speaking of such
matters.
------------------------------------
I agree that the use of "level" is confusing. I try to be consistent
about it. A reference level is that level (amount, state) of a
perception which we try to maintain. A reference _signal_, on the other
hand, is a signal in a model of control organization which determines
the observable reference level of a perception. The reference level of a
perception is evidence of the existence of a reference signal. The
perception can occur at any level within its range; only one of those
levels is the reference level at a given time.
-------------------------------------
You keep insisting, Hans, that the actions of an organism correlate with
their consequences. That is true only under very special and limited
circumstances. But a model of behavior has to work under all the
circumstances where real behavior works. As long as you keep ignoring
disturbances and variable reference signals, you will keep missing this
critical point about PCT, and what makes it different from S-R
psychology.
--------------------------------
RE: routes
What I wanted to stress in this example is that this was PASSIVE dis-
covery: there was no control system or reorganizational system at work
that wanted to try to minimize some error.
Right, I switched examples. The goal in your example was simply to
follow another person, and in doing so you acquired knowledge about a
possible new route. What you say about accidentally discovering new
paths is reasonable. I agree that this is not reorganization.
-----------------------------------------------------------------------
Best to all,
Bill P.