Collective Control

[From Fred Nickols (2016.09.28.1208 ET)]

I’m interested in reading up on “collective control” – how groups of people might cooperate, collaborate and communicate in ways through which they manage to control the value of a variable none of them could control on their own. Where should I look?

Regards,

Fred Nickols, Knowledge Worker

My Objective is to Help You Achieve Yours

DISTANCE CONSULTING LLC

“Assistance at a Distance”SM

At everything Kent has written…

···

On Wed, Sep 28, 2016 at 5:21 PM, Fred Nickols fred@nickols.us wrote:

[From Fred Nickols (2016.09.28.1208 ET)]

Â

I’m interested in reading up on “collective controlâ€? – how grroups of people might cooperate, collaborate and communicate in ways through which they manage to control the value of a variable none of them could control on their own. Where should I look?

Â

Regards,

Â

Fred Nickols, Knowledge Worker

My Objective is to Help You Achieve Yours

DISTANCE CONSULTING LLC

“Assistance at a Distanceâ€?SM

Â

Dr Warren Mansell
Reader in Clinical Psychology

School of Health Sciences
2nd Floor Zochonis Building
University of Manchester
Oxford Road
Manchester M13 9PL
Email: warren.mansell@manchester.ac.uk
Â
Tel: +44 (0) 161 275 8589
Â
Website: http://www.psych-sci.manchester.ac.uk/staff/131406
Â
Advanced notice of a new transdiagnostic therapy manual, authored by Carey, Mansell & Tai - Principles-Based Counselling and Psychotherapy: A Method of Levels Approach

Available Now

Check www.pctweb.org for further information on Perceptual Control Theory

[From Rick Marken (2016.09.28.1610)]

···

Fred Nickols (2016.09.28.1208 ET)]

Â

FN: I’m interested in reading up on “collective controlâ€? – how grouups of people might cooperate, collaborate and communicate in ways through which they manage to control the value of a variable none of them could control on their own. Where should I look?

RM: Tom Bourbon did some nice work on collaborative control; it’s described in his “Dancer and the Dance” paper in the American Behavioral Science issue dedicated to PCT, Â volume 34/number 1, Sept/Oct 1990. He also described some nice work on cooperative control at a CSG meeting some time ago but I don’t believe it was published anywhere.Â

BestÂ

Rick

Â

Â

Regards,

Â

Fred Nickols, Knowledge Worker

My Objective is to Help You Achieve Yours

DISTANCE CONSULTING LLC

“Assistance at a Distanceâ€?SM

Â


Richard S. MarkenÂ

“The childhood of the human race is far from over. We
have a long way to go before most people will understand that what they do for
others is just as important to their well-being as what they do for
themselves.” – William T. Powers

[From Kent McClelland (2016.09.29.1400)]

Fred Nickols (2016.09.28.1208 ET)

I’m interested in reading up on “collective controlâ€? – how groups of people might cooperate, collaborate and communicate in ways through which they manage to control the value of a variable none of them could control on their own. Where should I look?

Hi Fred,

“Collective control� is a term that I coined about 20 years ago in my effort to apply PCT to sociology. In my definition, collective control refers situations in which two or more control systems control the same or highly similar perceptions
in a common environment, where the control efforts of one affect control by the others, and vice versa.

Collective control can be either collaborative, when the control systems have the same or highly similar references for the perceptions they control, or conflictive, when the control systems try to control their similar perceptions using different
references. Conflictive control typically leads to rapid conflict escalation if the control systems have high loop gain for the contested perceptions they are controlling. I have argued that most human social interactions involve a mixture of collaboration
and conflict, with the result of the collective control being a greater stabilization of the environment in which it takes place than any single control system could achieve, but with simultaneous occurrence of some degree of conflict, as well.

in my view, collective control is the source of the social stabilities that sociologists have described as "social structure,� including languages and cultures. I’ve explained that view at some length in my chapter for the LCS IV volume to be
published by Bill’s sister, Alice McElhone. which I hope will be available soon. Martin Taylor’s long chapter for that volume, which I’ve had a chance to see in draft, also has some very interesting things to say about collective control.

I understand from a CSG post a few years ago that Rick Marken has come up with his own definition for the term collective control, a definition different than mine. I’ll leave him to explain his definition to you, rather than trying to paraphrase
it myself.

Here are references to my publications about collective control:

McClelland, Kent. 2004. “Collective Control of Perception: Constructing Order from Conflict.� International Journal of Human-Computer Studies 60: 65-99.

McClelland, Kent A. 2006. “Understanding Collective Control Processes.� Pp. 31-56 in Purpose, Meaning, and Action: Control Systems Theories in Sociology, edited by Kent A. McClelland and Thomas J. Fararo. New York: Palgrave Macmillan.

McClelland Kent. 2014. “Cycles of Conflict: A Computational Modeling Alternative to Collins’s Theory of Conflict Escalation.� Sociological Theory 32: 100-127. DOI: 10.1177/0735275114536387

If you’re interested in seeing these publications, but can’t easily get access to them, let me know by direct email (not through CSGnet), and I’ll send you a copy. They’re also available on ResearchGate. Alice has asked us not to distribute the
draft chapters for LCS IV, so I can’t give you a copy of that.

Best,

Kent

···

Fred Nickols (2016.09.28.1208 ET)]

FN: I’m interested in reading up on “collective controlâ€? – how groups of people might cooperrate, collaborate and communicate in ways through which they manage to control the value of a variable none of them could control on their own.
Where should I look?

RM: Tom Bourbon did some nice work on collaborative control; it’s described in his “Dancer and the Dance” paper in the American Behavioral Science issue dedicated to PCT, volume 34/number 1, Sept/Oct 1990. He also described some nice work on
cooperative control at a CSG meeting some time ago but I don’t believe it was published anywhere.

Best

Rick

Regards,

Fred Nickols, Knowledge Worker

My Objective is to Help You Achieve Yours

DISTANCE
CONSULTING LLC

“Assistance at a Distance�SM


Richard S. Marken

“The childhood of the human race is far from over. We have a long way to go before most people will understand that what they do for others is just as important to their well-being as what they do for themselves.” – William T. Powers

[From Fred Nickols (2016.09.29.1601 ET)]

Thanks, Kent. I’ll contact you off the list.

Does anyone know when LCS IV is slated to be released?

Fred Nickols

···

From: McClelland, Kent [mailto:MCCLEL@Grinnell.EDU]
Sent: Thursday, September 29, 2016 3:49 PM
To: csgnet@lists.illinois.edu
Subject: Re: Collective Control

[From Kent McClelland (2016.09.29.1400)]

Fred Nickols (2016.09.28.1208 ET)

I’m interested in reading up on “collective controlâ€? – howw groups of people might cooperate, collaborate and communicate in ways through which they manage to control the value of a variable none of them could control on their own. Where should I look?

Hi Fred,

“Collective control� is a term that I coined about 20 years ago in my effort to apply PCT to sociology. In my definition, collective control refers situations in which two or more control systems control the same or highly similar perceptions in a common environment, where the control efforts of one affect control by the others, and vice versa.

Collective control can be either collaborative, when the control systems have the same or highly similar references for the perceptions they control, or conflictive, when the control systems try to control their similar perceptions using different references. Conflictive control typically leads to rapid conflict escalation if the control systems have high loop gain for the contested perceptions they are controlling. I have argued that most human social interactions involve a mixture of collaboration and conflict, with the result of the collective control being a greater stabilization of the environment in which it takes place than any single control system could achieve, but with simultaneous occurrence of some degree of conflict, as well.

in my view, collective control is the source of the social stabilities that sociologists have described as "social structure,� including languages and cultures. I’ve explained that view at some length in my chapter for the LCS IV volume to be published by Bill’s sister, Alice McElhone. which I hope will be available soon. Martin Taylor’s long chapter for that volume, which I’ve had a chance to see in draft, also has some very interesting things to say about collective control.

I understand from a CSG post a few years ago that Rick Marken has come up with his own definition for the term collective control, a definition different than mine. I’ll leave him to explain his definition to you, rather than trying to paraphrase it myself.

Here are references to my publications about collective control:

McClelland, Kent. 2004. “Collective Control of Perception: Constructing Order from Conflict.� International Journal of Human-Computer Studies 60: 65-99.

McClelland, Kent A. 2006. “Understanding Collective Control Processes.� Pp. 31-56 in Purpose, Meaning, and Action: Control Systems Theories in Sociology, edited by Kent A. McClelland and Thomas J. Fararo. New York: Palgrave Macmillan.

McClelland Kent. 2014. “Cycles of Conflict: A Computational Modeling Alternative to Collins’s Theory of Conflict Escalation.� Sociological Theory 32: 100-127. DOI: 10.1177/0735275114536387

If you’re interested in seeing these publications, but can’t easily get access to them, let me know by direct email (not through CSGnet), and I’ll send you a copy. They’re also available on ResearchGate. Alice has asked us not to distribute the draft chapters for LCS IV, so I can’t give you a copy of that.

Best,

Kent

On Sep 28, 2016, at 6:07 PM, Richard Marken rsmarken@gmail.com wrote:

[From Rick Marken (2016.09.28.1610)]

Fred Nickols (2016.09.28.1208 ET)]

FN: I’m interested in reading up on “collective controlâ€? – how groups of people might cooperate, collaborate and communicate in ways through which they manage to control the value of a variable none of them could control on their own. Where should I look?

RM: Tom Bourbon did some nice work on collaborative control; it’s described in his “Dancer and the Dance” paper in the American Behavioral Science issue dedicated to PCT, volume 34/number 1, Sept/Oct 1990. He also described some nice work on cooperative control at a CSG meeting some time ago but I don’t believe it was published anywhere.

Best

Rick

Regards,

Fred Nickols, Knowledge Worker

My Objective is to Help You Achieve Yours

DISTANCE CONSULTING LLC

“Assistance at a Distance�SM

Richard S. Marken

“The childhood of the human race is far from over. We have a long way to go before most people will understand that what they do for others is just as important to their well-being as what they do for themselves.” – William T. Powers

[From Fred Nickols (2016.09.28.1208 ET)]

      I’m interested in reading up on “collective

control” – how groups of people might cooperate, collaborate
and communicate in ways through which they manage to control
the value of a variable none of them could control on their
own. Where should I look?

3.1.1a_CollectiveTug.jpg

3.1.1b_CollectiveTug_2.jpg

ctrl3.logo.png

[From Fred Nickols (2016.09.30.0703 ET)]

Thanks for the long and thoughtful response, Martin. I think I understand it. I’m taken by the distinction between “controlled” and “controllable.” I’ve much pondering to do.

Fred Nickols

image00249.jpg

image00419.jpg

···

From: Martin Taylor [mailto:mmt-csg@mmtaylor.net]
Sent: Thursday, September 29, 2016 11:26 PM
To: csgnet@lists.illinois.edu
Subject: Re: Collective Control

[Martin Taylor 2016.09.29.10.14]

[From Fred Nickols (2016.09.28.1208 ET)]

I’m interested in reading up on “collective control” – how groups of people might cooperate, collaborate and communicate in ways through which they manage to control the value of a variable none of them could control on their own. Where should I look?

I’d start by watching Kent’s presentation to CSG '93, which he allowed me to put on my web site:
http://www.mmtaylor.net/PCT/Movie/McClelland_CSG93.mp4

Kent called it “Conflictive control” at the time, but it developed into collective control, and the presentation gives the main point, which is that when more than one controller influences some environmental variable, neither controller is able to bring the variable to their reference value for it (Conflict), but to an outside observer the variable seems to be controlled by a controller with a gain that is the sum of the two gains, at a reference value that is between the reference values of the two controllers in proportion to their gains.

The same thing can happen with many controllers influencing the same variable. It doesn’t matter whether they even know of each others’ existence (in fact, they can’t, since they perceive only the value of their perceptual signal), the result is as if there exists what we (Kent and I) have called a “Giant Virtual Controller” that has a very high gain and a reference value that is a weighted average of the reference values of the different controllers.

After you watch the movie, you could look up Kent’s work on Google scholar (or ask him directly for suitable reading material).

The diagrams show cartoons of the situation in 2D, as though the controllers were tugging on rubber bands trying to get the box to where each one of them wants it to be. If one of them tugs a bit more strongly, as suggested in the right-hand figure, it will move the box a little, but not very far from the reference position that is the average of the reference positions of the other controllers. The shaded controller is effectively in conflict with a Giant Virtual Controller consisting of all the other controllers. (So is any one of the unshaded ones, but in the diagram the box is near their reference location for it, so we treat them as a group opposed to the shaded one).

Which brings us to your core question, of how such conflicted controllers could collaborate. But before dealing with that, I want to address your other question, posed in a message entitled “The Concept of Controlled Variable” [From Fred Nickols (2016.09.28.1211 ET)]. The answer brings up a point that is not neglected in B:CP, but that is often overlooked.

In B:CP (both editions), Bill didn’t define “controlled variable”; instead, he defined “controlled quantity” (which I assume is the same thing. His definition was as follows:

“An environmental variable corresponding to the perceptual signal in a control system; a physical quantity (or a function of several physical quantities) that is affected and controlled by the outputs from a control system’s output function.”

Bear with me while I tease out the criteria that something must meet in order to qualify as a controlled variable, one that I might try to control.

  1. Its value must be capable of varying; if it doesn’t, it’s not a variable at all

  2. I must have some desired value in mind

  3. I must be able to perceive its current value

  4. I must be able to affect its value

  5. I must want to control its value

Do I have these right?

Over the decades on CSGnet and its predecessors, there have been sporadic outbreaks of disagreement as to whether the controlled variable is on the one hand a perception, a unitary signal value that is compared to a reference value inside the organism (which would not fit your criteria) or on the other hand a complex structure of variables in the environment, perceptible to an outside observer or experimenter. The very names of PCT “Perceptual Control Theory” and of its founding book “Behavior: The Control of Perception” argue for the former, and even engineering control theory requires that the controlled variable be directly compared to a reference value.

No reference value for anything can exist in the environment, though a reference value does specify a state of the environment. A reference value is a property of an individual controller inside an organism, not a property of the environment that anyone else could observe and use. So, technically, no environmental variable can be controlled. All that can be controlled is some representation of it inside the organism, a representation we call a “perceptual signal” or simply a “perception”.

But here’s where things get a bit messy, and where “collective control” mixes in. In B:CP, Powers argues that while there may be some loss of precision by combining the neural firings in some bundle into a “neural current” and treating that current as a signal, the loss usually won’t be as bad as 10%, and that’s OK because we aren’t trying to be exact, so much as explain what’s happening, and it’s much easier to deal with a single neural current than with a mix of nerve firings of nerves that don’t exactly respond the same way to sensory input, but nearly do.

The perceptual “neural current” is a virtual signal, that exists nowhere in the brain. This is true also of a reference “neural current”. Each is as virtual as are the perception and reference of the collective “Giant Virtual Controller”. We cannot know anything of the environment except what we perceive, and what we perceive is virtual anyway. So where does that leave “control”? More to the point, where does it leave “Social Control” or “Collective Control”? Does Fred have his criteria for a controllable (not controlled) environmental variable correct, even though technically an environmental variable cannot be a controlled variable?

There are two places where the distributed “neural current” signals converge in to a single value. One is in a conscious perception, which is ordinarily unitary and smoothly varying, while the other is at an environmental variable that corresponds to a virtual perception that is brought to its virtual reference value. This is true both inside the head and in “Collective Control” (except that presumably in collective control there is nothing analogous to a conscious perception). The virtual perception is not necessarily made conscious, and in most cases, especially at lower levels, it is hard to make it conscious. So we are left with a sole point of convergence, the place where the influence of the controller’s output meets the influence of the disturbance, where it creates a value that looks to an outside observer/experimenter doing “The Test for the Controlled Variable” as if it corresponds to a controlled perception. The external variable looks “controlled”.

Maybe we should relax the definition of “controlled variable” a little, if we can nowhere find a controlled perceptual value with a reference value that corresponds to a value of the environmental variable. Maybe we should do as Rick has long advocated, and call the environmental variable a “controlled quantity”, as it is in the B:CP quote with which Fred precedes his list of criteria. We know that technically it isn’t controlled, but for all practical purposes its influence on its environment (such as an observer) is the same as it would be if it were actually controlled.

If we allow this relaxation, then Fred’s criteria correctly define a controllable variable, but not a controlled variable. For that, criteria 3 and 4 should replace “be able to” by actually doing (perceiving, affecting the value). Then criterion 5 may be unnecessary or be folded into criterion 2.

I realize this is likely to be unnecessarily complicated, and that “neural currents” as “perceptual values” and “reference values” work perfectly well in most situations. However, when we discuss Collective Control, we ought not to make it seem so very different from ordinary control. Both have virtual perceptual and reference values, but act on a single environmental variable. So the stage is set to address Fred’s main question:

…how groups of people might cooperate, collaborate and communicate in ways through which they manage to control the value of a variable none of them could control on their own.

Kent’s 1993 presentation answers the last part of the question. To answer the rest, it may help to distinguish different types of collective control. I identify six, of which three seem relevant to this question:

  1. Conflicted Control: Several people push on a rock, possibly all wanting it in a different place. The participants have individually derived reference values for perceptions of the same environmental variable. The environmental variable responds to disturbances as if it corresponds to a controlled perception, but the outputs of the individual controllers tend to increase as in any conflict.

  2. Collaborative Control: Several people push on a rock trying to move it to a place on which they have agreed by some negotiation. The participants control a higher level set of perceptions of belonging and being seen to belong to “the group”. Supporting their control of “belonging to the group”, they control for having their reference values for the common environmental variable to have a common value, eliminating the conflict while maintaining strong control.

  3. Coordinated Control: Several people push on a rock trying to get it to a place designated by one of them. All members (who control for perceiving themselves and being perceived as belonging to the group) accept reference values provided by an agreed leader.

The latter two types of collective control both involved control of a higher-level variable, which has as input (at least in part) a perception of the relevant goals of the others. These inputs may well be verbal, but they need not be, as each individual has the opportunity to perform the Test for the Controlled Variable on the others. But it’s much easier if you just ask “Can I help” when you see someone or a few people trying to move a rock that is too big for them.

My other three types of collective control involve the negotiation of different roles for some of the controllers. For example, a “Can I help” question might be answered with “Sure. Hold this post for me while I hammer it in”. That’s different in kind from two people trying to hammer in a post that keeps falling over after each blow. I won’t go into the others because they need a bit of preliminary material, but I will point you to an argument I made at CSG '05 that true social control units can and do exist: http://www.mmtaylor.net/PCT/CSG2005/CSG2005cSocialControl.ppt.

I doubt that this long discussion of collective control inside and outside the brain really answers your question in the way you wanted, but I doubt that a simple answer is likely to be correct or that a correct answer can be simple.

Martin