Cells and People

[Marken_Richard_S1]

[From Rick Marken (930615.0900)]

Hans Blom (930615) and Oded Maler (931615) --

Let me see if I understand what your analogy where you compare
individual cells to individual people. I think you are pointing out
that both cells and people are control systems -- which I accept
as true. The behavior of collections of cells (such as the cells that
make up an organ, like the liver) is regulated by systems external
to the cells. So, for example, the collective behavior of the cells
in the liver is regulated via hormonal and neural control systems
in the brain. This shows (I think you are saying) that the collective
behavior of humans is regulated via control systems (Oded's "emergent
social, historical entities" and others' "social control systems") that
are outside of the collective. Is this what you guys are saying?

Assuming that it is what you are saying, let me try to explain where
I have problems with the analogy. I agree that the behavior of collections
of cells is regulated by other systems -- the collectives are part of
a higher level control system. But this higher level control system
must be using properties of these cells that the individual cells themselves
are not controlling. For example, the individual cells may be controlling
the differential concentration of electrolytes across the cell wall -- doing
this by varying the permeability of the cell wall. This means that the
concentration of electrolytes outside of each and every cell is part of
a variable controlled by each cell -- possibly relative to a slightly
different
reference specification. The fact that electrolyte concentration is
controlled by the individual cells means that a higher level control system
cannot "use" the collection of cells to control a variable that strongly
depends on elecrtolyte concentration. The higher level control system
can really only effectively use uncontrolled variable aspects of cell
behavior as part of its means of controlling whatever variable(s) it
controls. For example, it can use the cell's actions that control
electrolytes (the permeability changes) as part of its means of control
because the higher level system can make unresisted variations in this
variable by disturbing the variable being controlled by the cells -- cell
permeability is an uncontrolled action of the cell.

Suppose that changes in permeability required to control electrolyte
concentration cause changes in the size of the cell. Then the higher
level system can use changes in the size of the cell to control whatever
variable it is controlling -- like net blood flow through the collective of
cells (the liver). So the higher level system can vary the concentration
of electrolytes outside of the cell collective in order to influence the
net size of the cells in the collective and influence blood flow though the
liver -- which is the variable controlled by the higher level system.

Things could work this way with collectives of people; there could be a
higher level control system, outside of the collective of people, that
uses side effects of the control exerted by the individuals in the
collective as a means of controlling whatever variables are being con-
trolled by the higher level system. I'm perfectly willing to believe that
there are such higher level social control systems. I just don't see any
evidence of them. I see no evidence of side effects of collective action
being under control. The results of collective action that I have seen
controlled seem to be controlled by the individuals themselves. No
outside control system is using the side effects of people controlling
other variables to create a skyscraper or a symphony. In each case
each individual is controlling a variable specifically in order to produce
the intended collective result.

If you really think that there are higher level social control systems
that are using side effects of human control to produce the results
they (not the individual people) intend, then I would really like to see
some evidence of it.

Oded claims:

there is something in those cell assemblies (aka BP and RM)
which resists the idea that their level of organization is not the
ultimate one.

The reason we "resist" is because we have seen no evidence. If
you have some real strong evidence of control by a higher level
control system then please present it. The evidence for the
existence of higher level control would be a variable that is clearly
maintained against disturbance by effects produced by a collection
of individuals who have absolutely no interest in the state of that
variable; it is not controlled by the individuals, it is maintained
AGAINST DISTURBANCE by some other control system.

Happy hunting

Rick

[_Maler]

[From Oded Maler (930616)]

Sometimes I forget that the meaning of Control is very precise
among the hard-core participants, and this might cause misunderstanding.
So I should separate my "claim" into two parts:

1) There "are" *systems* at a higher levels of organization for
which humans, their artifacts, etc., play the role of cells or
molecules inside a human.

2) Those systems are control systems.

I was criticizing your refusal to accept (1). I was not claiming
explicitly that these systems are built exactly as control
systems in the sense of an individual person.

What is clear to me is that there are collective perceptions for such
entities, there are some collective reference, and there is some
mechanism that connects them thru actions. It is very hard for us to
imagine what are the perceptual varibles for such systems because they
may extend in time (thoushands of years) and space, and may be
realized by many generations of their mortal components. The whole
thing operates on a different scale and the relation between a
collective signal and its individual realization is maybe like the
relation between a perceptual signal within an individual to the
identity of ions that flow thru the relevant synapses.

I don't know whether these systems achieve their "goals" by
by influencing their environment or by constantly creating
new perceptual variables and updating old ones. Maybe the
time-scale for "normal" functioning is yeares, and for
reorganization it is decades. All this is very speculative,
I know. I don't intend to give any "model", and if you insist
I stick only to (1) and withdraw (2) alltogether.

--Oded

p.s.

I think there is some confusion between two different notions of
levels. The simple notion is the relation between higher and lower
signals. Although a signal for, say, a sensation, and a configuration
may have some different features concerning their update rate, etc.,
they are creatures of the same sort. On the other hand a biochemical
signal that "tells" a cell to synthesize this or that protein at
a given moment in the life-cycle is (a mon avis) qualitatively
different from the perceptual signal that this cell might be
implementing in case he or she happens to be a neuron.

···

--

Oded Maler, LGI-IMAG, Bat D, B.P. 53x, 38041 Grenoble, France
Phone: 76635846 Fax: 76446675 e-mail: maler@imag.fr

[Tom_BOURBON3]

From Tom Bourbon (930615.1402)

Holy smokes! Let your server go down for a day and look what happens on
this net! As a point to rejoin this discussion, I will use Rick's post.

[From Rick Marken (930615.0900)]

Who replied to:

Hans Blom (930615) and Oded Maler (931615) --

Let me see if I understand what your analogy where you compare
individual cells to individual people. I think you are pointing out
that both cells and people are control systems -- which I accept
as true. The behavior of collections of cells (such as the cells that
make up an organ, like the liver) is regulated by systems external
to the cells. So, for example, the collective behavior of the cells
in the liver is regulated via hormonal and neural control systems
in the brain. This shows (I think you are saying) that the collective
behavior of humans is regulated via control systems (Oded's "emergent
social, historical entities" and others' "social control systems") that
are outside of the collective. Is this what you guys are saying?

Is that what you guys are saying? Bill Powers (930615.0900) also posted
along the same line as Rick. I think he might also ask you this question.

..

If you really think that there are higher level social control systems
that are using side effects of human control to produce the results
they (not the individual people) intend, then I would really like to see
some evidence of it.

As would I.

Oded claims:

there is something in those cell assemblies (aka BP and RM)
which resists the idea that their level of organization is not the
ultimate one.

Will anyone ever catch on to the fact that I also "resist" those claims?

The reason we "resist" is because we have seen no evidence. If
you have some real strong evidence of control by a higher level
control system then please present it. The evidence for the
existence of higher level control would be a variable that is clearly
maintained against disturbance by effects produced by a collection
of individuals who have absolutely no interest in the state of that
variable; it is not controlled by the individuals, it is maintained
AGAINST DISTURBANCE by some other control system.

Exactly!

Hans and Oded, in lieu of working models, why don't you cast your ideas in
the form of a few simple PCT diagrams that map out the interactions you
think occur among cells in a person, and among people in a group, so
that we can *see* why the two of you think they are the same? As an example
of what I mean, here is a re-post of part of Tom Bourbon (930611.0750) in
which I offerred my idea of a comparison between people as control systems
and social groups as "control systems."

···

===========================

Subject: Re: BBS4

From Tom Bourbon (930611.0750)

..

MIS D. "Social systems (sometimes the topic is social
hierarchies) are really PCT systems, with a different person
playing the role of each function in a PCT loop." ...

My reply.
++++++++++++++++++++++++++++++++++++++++++++++
CASE 1: simplest possible PCT model for a person keeping a
disturbed cursor even with a target.

                 p* (see: c - t = 0)
                 >
      __________\|/__________
     > > > >
     > i -p-> c -e-> o |
     >_______|_______|_______|
     /|\/|\ |
      > > > qo: accelerates arm
      > c <---- arm ----|
      > /|\
      > >
      t d

+++++++++++++++++++++++++++++++++++++++++++++++
CASE 2: Minimal two-level hierarchical PCT model for the same
pseson modeled in CASE 1. Output from the top level (visual
relationship) IS the reference signal for the bottom level (arm
movement).

                          p* (see: c - t = 0)
                          >
               __________\|/__________
              > > > >
LEVEL 2 | i -p-> c -e-> o |
              >_______|_______|_______|
              /|\/|\ |
               > > >___ qo2 = k * e
               > > >
               > > >
               > > p* (accelerate arm by qo2)
               > > >
               > > __________\|/__________
               > > > > > >
LEVEL 1 | | | i -p-> c -e-> o |
               > > >_______|_______|_______|
               > > /|\ |
               > > > > qo1: accelerate
               > > > > arm
               > c <------- arm -------------|
               > /|\
               > >
               t d

+++++++++++++++++++++++++++++++++++++++++++++++++++++
CASE 3: Approximation of minimal connections between independent
PCT models that would satisfy the popular belief that social
hierarchies are really PCT systems, with a different person
playing the role of each function in a PCT loop, which in this
example is the minimal PCT model shown in CASE 1. A separate
system represents each of p*, c, i, and o. Output from each
system must be turned into words spoken to the appropriate other
person; unlike the situation in CASE 2, no person has direct
access to the reference signal of any other person. (A similar
diagram for CASE 2 would be even more complex, with each function
in each model in CASE 2 represented by a different PCT model, each
of which should be hierarchical.)

                      p* (hear myself tell below,
                      > "c - t should = 0")
                      >
           __________\|/__________
          > > > >
          > i -p-> c -e-> o |
          >_______|_______|_______|
             /|\ |
              > >
        ______|_______________| spoken words "c - t should = 0"
      >
      > p* (compare what I hear from above with
      > > what I hear from below left and tell
      > > below right how to move)
      > >
      > __________\|/__________
      > > > > >
      > > i -p-> c -e-> o |
      > >_______|_______|_______|
      > /|\/|\ |
      >_______| | |
                 > > words about
  words about | | how to move
  what is seen ->|___________ |
                             > >
            p* (watch c & t | | p* (listen for
            > and say what | | | how to move;
            > I see) | | | feel it
            > > > > happen)
__________\|/__________ | | __________\|/__________

      > > > > > > > > >
  i -p-> c -e-> o | | | | i -p-> c -e-> o |
_______|_______|_______| | | |_______|_______|_______|

  /|\/|\ | | | /|\/|\ |
   > > >________| |____| | | qo
   > > > >
   > c <------------------------------ arm -------------|
   > /|\
   > >
   t d

=====================================

Hans and Oded, if each of you were to cast your ideas into similar diagrams,
I might more easily see what you mean when you equate people in social
groups with cells in a person. When I use PCT to model individuals in
groups, I think of an organization like that shown in CASE 3. When I think
of where a cell fits in the body, it is more like one element in the output
box at the lower level of CASE 2. How do you envisage them?

Until later,
   Tom Bourbon