Bell, Book and Science

CSGnet Archive CSG1995
1

[From Rick Marken (950214.1540)

Martin Taylor (950214 12:10) --

My procedure in both cases was what Bill P required--to present what ought
to be incontrovertible demonstrations.

They ought to have been "incontrovertible", perhaps, but they were not; Bill
and I controverted them.

Here's the claims: (1) It is through the perceptual function that the
information needed for control is obtained.

What is information? How can we tell that it is coming through the perceptual
function? Who "obtains" it? How do we know it is obtained? What is done with
it once it has been obtained?

Your inability to answer these questions made it clear that your only
interest in making claims about "the information needed for control" was to
preserve, in your own mind, the relevence of an old, beloved theory
(information theory) to perceptual control theory. Preserve away.

(2) There are more perceptual functions involving controllable perceptions
than can be controlled at any moment,

Even if this were true, so what?

and there is a need for some mechanism to shift what is controlled when
some currently uncontrolled perception is disturbed sufficiently to matter.

This doesn't follow from the first part of the statement. Moreover, the
mechanism that does this shifting MUST be a control system -- it must have a
reference for the maximum amount of error in another control system and it
must be able to act on this other control system so that, when the perceived
error exceeds the reference (is "sufficiently" large) the error is brought
back below the reference. So an "alerting mechanism" is a control system --
nothing new.

Perhaps I should ask you to take the other position, and show, using only
the core PCT assumptions, how it is possible to avoid the necessity for
the existence of some alerting mechanism, given the larger number of
potentially controllable perceptions than available outputs.

Can I start with an easier problem, like "starting with the core assumptions
of christianisty, show that it is possible to avoid the necessity of
purchasing a pin when you've already got a large number of angels dancing on
the head of a gold club":wink:

Looking at the world for data, to show that they are so, happened earlier
in the case of the alerting system

So alerting systems have already been discovered, eh? Just like reflexes and
reinforcers? I guess now that these have all been discovered we know that
they really exist? No more looking at the world is necessary, eh?

Martin Taylor (950214 11:10) --

There's no "goal" to reorganization. Systems just stay the way they are when
reorganization is not happening much, and that is when perceptual control is
generally good. That's true within the individual, and therefore true
within social structure composed of individuals.

Is this another one of the "truths" you are bringing to PCT. Do you really
think reorganization has no goal?

Very interesting;-)

Best

Rick

2

[Martin Taylor 950215 11:10]

Rick Marken (950214.1540)

Martin Taylor (950214 12:10) --

My procedure in both cases was what Bill P required--to present what ought
to be incontrovertible demonstrations.

They ought to have been "incontrovertible", perhaps, but they were not; Bill
and I controverted them.

No you didn't. You sidestepped them without dealing with the issues.

What is information? How can we tell that it is coming through the perceptual
function? Who "obtains" it? How do we know it is obtained? What is done with
it once it has been obtained?

Your inability to answer these questions ...

All of these questions were answered in detail, from several different
perspectives. You chose (and choose) simply to assert that they were
not answered, because each answer leads to the same conclusion, that
something you took to be true is not.

(2) There are more perceptual functions involving controllable perceptions
than can be controlled at any moment,

Even if this were true, so what?

Yep. So a person can't follow a moving target with a cursor unless the
person contains/is a control system. So what? Ask any psychologist.

There are implications in simple, obvious facts. You may choose not to
see them, if you want. In the UK, Nelson's telescope is the example
normally used to illustrate this simple, obvious fact.

and there is a need for some mechanism to shift what is controlled when
some currently uncontrolled perception is disturbed sufficiently to matter.

This doesn't follow from the first part of the statement.

OK. Provide an alternative. That way you can show that it doesn't follow.
Don't just assert.

Moreover, the
mechanism that does this shifting MUST be a control system -- it must have a
reference for the maximum amount of error in another control system and it
must be able to act on this other control system so that, when the perceived
error exceeds the reference (is "sufficiently" large) the error is brought
back below the reference.

Yes. But if there is no such mechanism (see your previous sentence), how can
this nonexistent mechanism be a control system? Of course it's a control
system, probably many.

So an "alerting mechanism" is a control system -- nothing new.

If you remember, one of the points I tried over and over to get across was
exactly this--that alerting is nothing new. It's just a necessary consequence
of core PCT, and uses only core PCT.

Perhaps I should ask you to take the other position, and show, using only
the core PCT assumptions, how it is possible to avoid the necessity for
the existence of some alerting mechanism, given the larger number of
potentially controllable perceptions than available outputs.

Can I start with an easier problem, like "starting with the core assumptions
of christianisty, show that it is possible to avoid the necessity of
purchasing a pin when you've already got a large number of angels dancing on
the head of a gold club":wink:

Sidestepping again, eh? Of course your proposal presents an easier problem,
because I think the one I proposed to you is one you cannot solve. You can't
get away from the fact that there has to be some mechanism for shifting
the set of controlled perceptions, and that unpredictable disturbances from
the outer world are often the reason. Control is control against disturbance
as well as control for changing reference levels.

Looking at the world for data, to show that they are so, happened earlier
in the case of the alerting system

So alerting systems have already been discovered, eh? Just like reflexes and
reinforcers? I guess now that these have all been discovered we know that
they really exist? No more looking at the world is necessary, eh?

You have a beautiful way of contradicting yourself, both within and between
postings. It's most elegant. Don't look for theory until you have a
phenomenon. Don't take any notice of phenomena observed before you have
a theory. Stasis. No need to think.

Do you really think reorganization has no goal?

In the sense that there is a desirable organization toward which
reorganization forces the hierarchy, no, no goal. Reinforcement is
a mechanism of control, the action output of a control system. That
control system has a goal, but it is only the maintenance of intrinsic
variables near their reference levels. It isn't to produce some particular
structure of control systems.

Is this another one of the "truths" you are bringing to PCT.

Not my truth. One I learned at the feet of the master :slight_smile:

Very interesting;-)

Would that I believed you found it so.

Oh-oh. I must have rediscovered that lost goal of getting Rick to see
the truth. Alerting systems set off--reflex--prodded by stimulus, response
is automatic--AAARRRGH. Must stop myself in future....not now.

Martin