[From Bill Powers (941007.0755 MDT)]
Hans Blom (941006) --
Rick reminds me of something I should have commented on:
...assume, just for a minute, that there is no higher-level system and
that the highest-level reference level is fixed. Are the lower-level
reference levels now fixed as well? If not, might we not say that it is
the _perceptions_ that adjust the lower goals?
If there is no higher-level system, then there is no higher-level
reference signal either. To what would such a reference signal connect?
A reference signal is an input to a comparator. If there's no
comparator, what possible function could a reference signal have? I
simply can't imagine what your mental model looks like.
The same goes for your model of a goal. How in the world can a
perceptual signal adjust a goal? A perceptual signal indicates the state
of the world as it is right now, not the state to which it is to be
brought by actions. To specify a goal, we need a second signal, set to
the value that the perceptual signal is to attain. Only then can we
compare the actual state of the world represented by the perceptual
signal with the desired state represented by the reference signal, to
get an error signal to drive the output.
Perhaps you are asking what sets the highest level of reference signal.
I remind you that the absence of a reference signal is equivalent to
presence of a reference signal set to zero. Zero is as valid a value as
any other; it can be matched by bringing the perceptual signal to zero.
A zero perceptual signal can represent a specific state of the world,
such as an arm at one limit of its range of positions. If you
decerebrate a cat (ugh), all the limbs extend fully. That is apparently
the meaning of a zero reference signal reaching the highest systems that
remain.
Of course there are other possible sources for the highest reference
signals; they might come from memories of past perceptual signals, for
example. Whatever the answer, there is some source needed to account for
them, other than the perceptual signal that is to be compared with them.
Maybe I should ask you: what is a goal, that it can create the physical
actions which achieve it in the environment of the moment?
···
----------------------------------------------------------------------
Oded Maler (941007)--:
Suppose a researcher R wants to test the controlled variable of a
subject S. Suppose his conclusion is X. How is X expressed? Being a
perception it cannot be described in physical terms (maybe except for
simple sensations such as muscle tension).
Who says it can't be described in physical terms? I didn't say that. I
said (or certainly meant to say) only that it can't be described in
_objective_ physical terms -- that is, terms that do not rely on
perceptions for their definitions. All the terms of physics, chemistry,
engineering, etc., rest ultimately on perceptions for their
definitions, don't they?.
If I propose that a person in a tracking experiment is controlling the
position of a cursor on a screen, I measure the cursor position in
pixels from some reference point on the screen. In my experience, what I
see as a certain "number of pixels" on a "screen" is what another person
would tell me he or she sees, too. So I can measure the proposed
controlled variable in a way that others (at least other adult human
beings, if not my cat) would agree is correct, and do the test for the
controlled variable using that measure of the variable. Another person,
using what that person believes to be the same perception I am using to
define the controlled variable, would most likely come to the same
conclusion, that it is under control. Luckily, this seems to work quite
well.
What's the problem, Oded? Do you want to say that we can know the actual
nature of the world outside us by some method other than looking at our
own perceptual signals? If so, I'd like to hear what that method is.
-----------------------------------------------------------------------
Greg Williams (direct) --
I hope to get the next issue out ca. 10-1.
Seeing that this is 10-7, I presume you meant 11-1.
-----------------------------------------------------------------------
Best to all,
Bill P.