Emergence, parapsychology

[From Rick Marken (920716.0830)]

I was going to let this one go; but my curiosity has gotten the best of me.

Allen Randall asked:

  How do emergent properties (a la connectionism) fit into this scheme?

And I replied:

Purposive behavior is an emergent property of the organization of the control
model.

And Martin Taylor (920715 15:45) said:

I have to disagree here. Purposive behaviour is built into the elementary
control system that is the basis of the structure. There are different
possibilities for emergence (a la connectionism). I think coordination may
be one (possible forthcoming thread, but not now, please). Purpose is not.

I guess my understanding of "emergence" differes from Martin's. I thought
"emergence" referred to a behavior or other characteristic of a system that
is qualitatively different than the behavior or charactteristics of its
constituents. One example of "emergence" that I learned was the salt crystal
that is qualitatively different than its constituents, Na and Cl. I think
that this same kind of "emergence" happens with a control loop. The
constituents of this loop are s-r (cause effect) components. This includes
the neurons, their synaptic connections and the physical entities that are
in the environment. The equations that describe the two basic components
of the control loop are s-r equations:

(1) o = f(p*-p) and
(2) p = g(o+d)

(1) decribes the behavior of the neural component that produces variations
in output as a function of input. (2) describes the behavior of the
environmental component that produces variations in input as a joint function
of outputs and independent enviromental events.

There is no purpose in either of these components; they behave according to
good old fashioned cause effect laws. Purpose emerges (to the considerable
surprise and satisfaction of those who first built these systems) when these
components are hooked up properly -- ie. so that there is negative feedback
and dynamic stability.

So I have to disagree with Martin's disagreement; I think the negative
feedback control system is a perfect example of emergence (as I understand
the term) -- where a phenomenon (purpose) emerges from a system of components
that, themselves, do not exhibit the phenomenon.

Bruce Nevin (Thu 920716 08:19:31) says:

though if you think funding for PCT is
precarious you should look at the history of parapsychology (interesting
summary in _Margins of Reality_, a recent book concerning issues of
consciousness).

I think this comparison would be more interesting if parapsychology had
data that were anywhere near the quality of the data we obtain regularly
in studies of control. To my knowledge, the parapsychological results are
even noisier than the results achieved in standard psychological
experiments. Since I consider the latter to be of no value, it is difficult
for me to characterize value of the former (negative, perhaps). The only
thing that is surprising to me is that funding agencies are willing
to pay for standard psychological research but not for parapsychological
research, which is of only imperceptibly poorer quality. I believe it must
be a religious thing -- the religion of standard psychology being the currently
dominant form. When psychololgy becomes a science (ie. when it stops turning
the statistical crank and just goes for the quality data -- something that
can only happen when it starts to focus its research efforts on trying to
understand how organisms control) such religious territorial squabbles should
diminish considerably.

Best regards

Rick

···

**************************************************************

Richard S. Marken USMail: 10459 Holman Ave
The Aerospace Corporation Los Angeles, CA 90024
E-mail: marken@aero.org
(310) 336-6214 (day)
(310) 474-0313 (evening)

[Martin Taylor 920716 18:00]
(Rick Marken 920716.0830)

We have a pure misunderstanding of words here. Easily fixed, I think.

Purposive behavior is an emergent property of the organization of the control
model.

And Martin Taylor (920715 15:45) said:

I have to disagree here. Purposive behaviour is built into the elementary
control system that is the basis of the structure. There are different
possibilities for emergence (a la connectionism). I think coordination may
be one (possible forthcoming thread, but not now, please). Purpose is not.

I guess my understanding of "emergence" differes from Martin's. I thought
"emergence" referred to a behavior or other characteristic of a system that
is qualitatively different than the behavior or charactteristics of its
constituents. ... Purpose emerges (to the considerable
surprise and satisfaction of those who first built these systems) when these
components are hooked up properly -- ie. so that there is negative feedback
and dynamic stability.

We agree about what "emergence" means. We had a misunderstanding about what
was the unit under discussion. I had taken Allan's question about organization
to be asking what emerged from the organization that is not inherent in
negative feedback control systems. I think my comment that you quoted made
that assumption explicit. You took the question to refer to the constituent
components of control systems, and the emergent to be control. I agree that
if we are talking about the s-r components of a feedback loop, then of course
purpose is an emergent from the organized structure.

I assumed that the unit was a complete ECS, and the question was what emerged
from their hierarchic organization. I proposed coordination, which seems to
me to be a collective function different from collaboration. I don't know
how strongly I would support that proposal if pressed.

Martin