Making my transition easy

[From Rick Marken (940222.2200)]

Martin Taylor (940222 17:00) --

Fundamentally, the self-organization model is a control system with no
reference input. It is a negative feedback system that stabilizes some
structural feature of the environment. In physical terms, it depends on
nonlinearities in the response of the environment to some form of far
from equilibrium energy flow.

Thanks for making my move away from PCT to self-organization theory
so easy by explaining that a self-organization model is a control system
with no reference input. Well, then a self-organizing system must control
a perceptual variable relative to a fixed reference level (zero).
I don't understand why this "depends on non-linearities in the response
of the environment to some form of far from equilibrium energy flow" but
it sure sounds impressive so I can hardly wait to find out what it means.
Does it mean that this control system doesn't work unless there are
these non-linearities? If so, I wonder why? If a self-organizing system
is a control system then, from what I recall when I was captive of that
stubborn, shallow PCT cult, such systems work with or without
non-linearities in the various functions. I'm sure glad I got out of there
on time so that I could learn about this new property of control
systems (with no reference inputs). Could you elaborate a bit on this.

Boy am I excited about this self-organization stuff. Sounds even better
than PCT and it's probably even easier (don't have to worry about all
those pesky reference inputs -- although I am a tad conconcerned
about those non-linearities). One quick question -- just what is it
that we're trying to explain? Is self-organization control? If not,
what is it? and why is it explained by a control system?

It is relevant to understanding how groups of them interact, such
as in the development of languages, social conventions, organizational
and governmental structures, and the like.

Boy, am I gonna like being a self-organization theorist. Could you
show me how a control system with no reference input helps me
understand one of the simpler things on that list -- like the development
of organizational structures, for instance. That Powers fella has this
program that produced all these different organizational structures
using regular old PCT -- but I knew he didn't really understand how
those structures occurred, not in any deep sense anyway. Now I know
what he was missing -- control systems with no reference input AKA
self-organizing systems.

Me:

Could you give a description of how a self-organizing system operates?

Martin:

Like a control system, except for the absence of the reference signal.

Boy, I tell ya. This transition to self-organizing systems theory is
going to be a piece of cake!

The necessary constructs are some manner of diverting energy out of a
power flow into some small number of degrees of freedom, and some mechanism
whereby a small deviation from a particular state of those degrees of
freedom causes a substantial corresponding difference in the power flow to
those same degrees of freedom, which is the actual gain function.

But it acts like a control system, right? If you could just turn the
statement above into a set of program statements we could fire up the
model and show those shallow, narrow minded, intrasigent PCT weenies what
a self-organizing system can do.

As Bill P.
pointed out in the earlier discussion, this loop has to be asymmetric, or
there will be no gain, and as I point out, no self-organization.

You mean Bill P. already knows how self-organizing systems work?!?!
And he still does not include them in the PCT model. Boy, talk
about narrow minded. Am I glad I got out of that cult.

We do not have the Jan 1994 issue of the American Psychologist in our library
(or any issue since 1972), but we do have two people reputed to be APA
members, who might have it. If the article's description of self-organization
is as you say, I may have some harsh words of my own about it after I have
seen it.

Good for you! And I would really like to see your comments on the article,
being that I'm a new member of the self organizing systems group. I want
to see how self-organizing systems theory is properly applied. The
guy in American Psychologist was probably just trying to make the
theory sound acceptable to his audience. It didn't seem like he had
much to say about control systems with no reference input. I sure hope
you set him straight -- there's no telling how many American psychologists
have been led astray by that article. I'm glad that all you Canadian
psychologists were spared.

Best

Rick

<Martin Taylor 940223 12:00>

Rick Marken (940222.2200)

I recognize attempts at humour, but when so extended, they aren't really
funny. All the same, there are a couple of points worth continuing.

Thanks for making my move away from PCT to self-organization theory
so easy

I'm sorry you want to move away from PCT. It must be hard for a canon
to change his aim.

I don't understand why this "depends on non-linearities in the response
of the environment to some form of far from equilibrium energy flow" but
it sure sounds impressive so I can hardly wait to find out what it means.
Does it mean that this control system doesn't work unless there are
these non-linearities?

What doesn't work without nonlinearities is the amplifier that provides
the gain, without which there is no effective negative feedback. The
non-linearities are not necessarily in the feedback loop itself. As with
any control loop, the feedback loop might be linear over some portion
of its range. That's a separate issue. The non-linearities allow the
diversion of energy into the degree(s) of freedom that are maintained
at their stable value(s) by the feedback loop. That powers the gain
function.

Well, then a self-organizing system must control
a perceptual variable relative to a fixed reference level (zero).

If you care to identify a perceptual variable in the loop, you can certainly
say that it controls to a fixed level, but I don't see how you can say
that the fixed level is zero, unless you do it be definition. The
controlled degree of freedom has some sustained value, for sure. What
that value is labelled is up to am outside observer.

(Aside) It seems to me that to identify a specific perceptual variable
distinct from the environmental variable, there has to be a transformation
through a perceptual function. In a self-organized system, there need
not be any such stage of transformation, and one might identify the
CEV itself with the perceptual signal. In other words, there's no
obviously necessary "inside" and "outside," to which parts of the loop
can be assigned. (End Aside)

One quick question -- just what is it
that we're trying to explain? Is self-organization control? If not,
what is it? and why is it explained by a control system?

Who we, LA-man? You said self-organization was a dormitive principle,
explaining itself. What I tried to explain to you was that self-organized
structures appear through feedback loops dependant on non-linearities
in far-from-equilibrium energy flows. They almost always appear in such
flows, and that is the phenomenon for which the feedback process provides
an explanation. In your language, is that control? I'm not proposing
to play word-games here; just dealing with generic processes that occur
in those systems that have adequate power sources and appropriate
possibilities for feedback.

Boy, am I gonna like being a self-organization theorist. Could you
show me how a control system with no reference input helps me
understand one of the simpler things on that list -- like the development
of organizational structures, for instance. That Powers fella has this
program that produced all these different organizational structures
using regular old PCT -- but I knew he didn't really understand how
those structures occurred, not in any deep sense anyway. Now I know
what he was missing -- control systems with no reference input AKA
self-organizing systems.

Sorry, I haven't seen the demos of autocracy, bureaucracy, group dialect
formation, ... I have seen the development of localized rings of people
in crowds, trains of followers, and so forth. The reason that the self-
organizing principles might well be applicable to all these sitations is that
the elements are control systems that provided disturbances to each others'
perceptions in a reciprocal way. There are all sorts of powered loops that
pass through more than one of the individual control systems. In Durango,
I followed this process through in respect of the development of stable
language conventions, but it applies also to other stable conventions
that are not dictated by the physical environment. That includes the
conventions of organizations. Those organizations that are structured
in such a way that stable conventions support the restoration of the
structure after disturbance will be the organizations that we see surviving.
I'm not sure crowd rings qualify as self-organized structures (but I'm
not sure they don't, either). The reason for my doubt is that the effect
seems to be static, in that you don't get some members leaving the ring
as others enter. In other words, the ring seems to be more like the
return of a ball to the bottom of a bowl than it is to a self-organized
structure. It looks more like a minimum energy configuration that
would stay there if all the elements just switched themselves off. Now
the reason why I think the rings just might qualify as self-organized
structures is that when there are too many active "people" to fit in the
ring, new ones do seem to displace old, while the ring itself maintains
some kind of integrity. So I'm not sure about crowd rings.

Leave PCT for a study of self-organization if you want. We will regret
the loss, but it's your choice. And I don't think you will be able to
understand how the self-organization of social structures happens if you
drop PCT.

We do not have the Jan 1994 issue of the American Psychologist in our library
(or any issue since 1972), but we do have two people reputed to be APA
members, who might have it.

Good for you! And I would really like to see your comments on the article,

They turned out not to be members of APA. But I have been told that another
person is. Unfortunately, he's off sick. So no comments yet.

Martin