Learning proposal; Tracy's quote

[From Bruce Abbott (961130.1355 EST)]

Rick Marken (961129.1420) --

Bruce Abbott (961127.1840 EST)

These seem to be serious gaps in what was supposed to be an HPCT-based
explanation, if I understood you correctly...I dare say that it hardly seems
less speculative than the little story I offered . . . Or did I miss
something?

The speculativeness of your story wasn't the problem; it was the
plausibility of your story, given what we know about how behavior
(control) works.

I don't think you've even _understood_ my story (probably because I haven't
been clear enough in telling it), and therefore are in no position to judge
its plausibility.

Re what we _know_ about how behavior works: Do you _know_ that all
control-learning depends on random reorganization, that there are no more
efficient processes available to rat or human? If so, please provide the
proof: I have only seen demonstrations of principle, i.e., that it _can_
work in a reasonable time if given few enough degrees of freedom while
varying transfer function properties. I strongly suspect that this process
by itself is not sufficient to account for the speedy repetition and
refinement of behavioral acts observed during acquisition of the required
operant in the operant chamber, or elsewhere, although it may be a basis for
fine-tuning of existing control systems (gain adjustment, etc.).

What empirical observations DO you see as offering difficulties for HPCT
as now constructed? Surely after 14 or so years you've run into _some_.

Are you asking "what empirical observations _have already been_ made that
offer difficulties for HPCT as now constructed?" If so, then obviously the
answer is "none".

Why do you suppose that is?

You start by asserting that "control systems can't control by repeating
acts," but then show how they can control _to_ repeat acts.

Your problem here is that you are ignoring the hierarchical nature of
control.

No fair -- it was my point that this is precisely what _you_ were doing.
Find your own criticism. (;->

In a hierarchical control organization, a control system at
any level above the lowest in the hierarchy must vary its _actions_ (the
reference for lower order "acts") in order to produce an "act" ( the
intended state of its own controlled perceptual variable). Your model says
that a control system repeats a particular action (the reference for a lower
order "act", such as pressing a lever) when a reward occurs.

Actually, it says that the higher-level _learns_ to raise a particular
reference (or set of them) as the _means_ of reducing the error in its own
controlled perception.

Again,
I suggest that you build the model and see what happens. I don't believe
that such a model could possibly learn to control.

So a mechanism that uses the perceptual signals available to it to determine
which reference(s) to vary could not possibly learn to control, but a
mechanism that merely chooses them at random (i.e., e-coli reorganization)
can? Tell me how the latter can work but the former cannot possibly do so
-- what is the crucial difference between guided and random
choice-mechanisms that makes this so.

The rest of your post was nicely answered by Tracy Harms(1996;11,27.19:12 MST)
and Bill Powers (961128.2200 MST).

So you wish to believe. But my replies to both (which you ignore) "nicely
answered" why both their answers missed the mark.

Re: Tracy's quote

Bruce seems to think that "discarding forever the principle of controlling
each other's behavior" is an admonition to "stop all controlling" or, in other
words, to "be a door mat". Of course, we can't stop controlling, whether
we want to or not. But we can stop some of our controlling -- the controlling
that involves control of other living control systems.

No, I'm not suggesting that this is a an admonition to stop all controlling.
We're control systems. That's what control systems do. Control systems
control.

I'm not addressing the issue of whether giving up attempting to control
others is the right thing to do -- I think it's a fine idea, and control
theory shows why. I'm addressing whether it is practical to do so, given
the world as it is.

If you don't want your neighbor to kill you, then there are basically
three ways to deal with the problem; 1) control him 2) negotiate with
him or 3) defend against him. PCT suggests that option 1) always
leads to problems -- ergo, Bill's statement above. Option 3) is what you
do, whether you want to or not, when all else fails (or is impossible); if
you are controlling for not getting hurt then you do what you can to resist
disturbances to this variable. Option 2) is the "best" option; PCT shows
how to negotiate effectively; the goal of negotiation is to help _both_
parties go "up a level".

There are only three ways to implement Option 3: (1) Make yourself
unavailable (hide, or lock yourself in a fortress) (2) get him before he
gets you, and (3) control him. The first is unacceptable, the second
illegal and against pacifist principles, and the third identical to Option
1. So if you want to live a normal life it comes down to two options:
control or negotiate. I'm all in favor of the latter, but before it can
happen, my neighbor has to be willing to listen. If he won't, then there is
nothing left to do but control.

As I recall, negotiation was the option Nevil Chamberlin chose. When that
failed, he was forced into Option 1.

Like "love thy neighbor," Bill's suggestion is an excellent one that should
be followed whenever possible. It is not always possible.

Regards,

Bruce