TRT and PCT (from Mary)

[from Mary Powers 941212]

Bruce Abbott says:

     Behavior varies (reorganization is in progress), affecting
     the environment, changing the state of a perceptual
     variable, changing the error between that variable and its
     reference level. If the error is reduced, that behavior,
     which a higher-level system has 'selected', remains selected
     and thus becomes part of the output function of the lower-
     level system controlling that perceptual variable...
     [This description] is PCT, if I've described it correctly...

But I don't think you have described it correctly. The giveaway
is that parenthetic (reorganization is in process). Behavior
always varies, whether or not reorganization is going on. Even
the most 'repetitive' or 'stereotyped' behavior. A repeated
pattern or sequence is a perception, brought about by variable
means. If a ballerina doing 30 fouettes in a row 'selected' the
same behavior for the 30th as the first, she'd fall down -
because of fatigue she must use her muscles in an entirely
different way to repeat the pattern successfully.

In order to learn to do the turns, she has to learn to balance on
point, to whip her other leg so that she turns, to spot on each
turn so she won't get dizzy, etc. But this is all learning
perceptions, not behaviors. She learns it looking in a mirror, by
voice instruction from her teacher, and primarily
kinesthetically. The outcome, to the audience, is repetitive
behavior. How it's done is by controlling perceptions. That is
not just the mechanistic explanation, but also the functional one
- at least from the PCT point of view. To say

     Those responses in the situation that are followed by a
     satisfying state of affairs become more strongly connected
     to the situation so that, when the situation recurs, the
     response is more likely to occur

is a functional explanation seems wide of the mark to me. This is
a conclusion drawn from appearances, and begs the question of how
connecting responses to a situation can do anything successfully
when both the responses and the situation vary.

The problem is that what appears to be going on is not what is
happening. It's like watching the sun come up, cross the sky, and
go down. Lots of 'functional' explanations for that appearance,
but the only explanation that we accept today is that the earth
is turning. PCT is an 'earth is turning' theory, in a pre-
Copernican world of psychology.

Mary P.

<[Bill Leach 941213.18:20 EST(EDT)]

[Mary Powers 941212]

It seems to me that there is an additional problem in this discussion of
"Law of Effect". The example of repetitive operations is a good one and
while I'm sure that most of us have not attempted Ballet, I am equally
sure that virtually all can think of similar examples.

I wonder though, if the term "Consequences" itself is not a source of
misunderstanding? I know that for most of the discussions here, it is
recognized that the term refers to "occurances in the environment"
perceived BY THE OBSERVER. However, whenever one talks about
consequences with respect to one's own actions, one is almost always
referring to one's OWN perception of consequence which most generally
would actually BE a controlled perception.

I realize that PCT has no use for a term that is ambigious but if others
are using the term; believing that they are using the term as "a
description of events independent of the subject" when in reality the
term personally means perception... well it seems to me that it would be
damned difficult to "shake the concept of consequences "selecting"
behaviour for such a person until the person saw for themselves that
their meaning for the term was different than their stated meaning.

I have an awful lot of "trouble" with a great deal of this "selection" of
behaviour by higher level systems also. It seems that the "sequencing"
level of HPCT might itself be multilevel. It also seems "reasonable" to
me that much of the "selecting" that is professed as necessary... is NOT!

My thinking there is that experience that already exists in PCT
experimental data concerning our own inability to predict what
perceptions are actually controlled for a particular task suggests that
many operations "with surprise disturbances" are corrected without more
than reference level changes. Such things as slipping might involve a
sequence change but I suspect that slew rates and loop delays explain the
control failure and new reference levels explain the successful control.

I recognize that the example of walking on an icy surface might not be
one of the better choices for an example but even with this one, I doubt
that significant control system changes (compared to "normal" walking)
are present.


Those responses in the situation that are followed by a satisfying state
of affairs become more strongly connected to the situation so that, when
the situation recurs, the response is more likely to occur


... This is a conclusion drawn from appearances, and begs the question
of how connecting responses to a situation can do anything successfully
when both the responses and the situation vary.

This is the point. I don't think that any PCTers are trying to say that
there is anything "wrong" with Bruce's statement above from a
"non-rigorous" view. It is generally true that people will "likely"
repeat "things" that are "satisfying" and "make them happy." It is
however, lousey science.

I think that this whole discussion has raised a couple of important

1) PCT is not interested in "likely occurrances", PCT is an exacting

2) PCT recognizes that "multiple responses" or "repeated behaviour"
    usually isn't and PCT views this factual difference as significant.

Just as the Alchemists methods for transmuting lead into gold is
irreconcilable with modern physics, non-PCT methods of behaviour studies
are irreconcilable with PCT to the extent that they deviate from
fundamental PCT principles.