Dawn T. Robinson, "Control theories in sociology"

CSGnet Archive CSG2007
1

[from Gary Cziko 2007.08.02 20:00 CDT]

Clark McPhail brought the following recently published article to my attention:

Control Theories in Sociology

Dawn T. Robinson

Annual Review of Sociology ,
August 2007, Vol. 33

···

,
Pages 157-174

(doi: 10.1146/annurev.soc.32.061604.123110)

I was pleased to see that Robinson devotes a section to Powers’s PCT with the following intro:

In the 1970s, a landmark work provided the modern inspiration for a new generation of sociological control theories. Powers’s (1973) book, Behavior: The Control of Perception, offered a comprehensive application of control system logic and mathematics to behavioral psychology.

She then goes on to discuss affect control theory and identity control theory, including empirical support for these theories.

While I haven’t finished the article or even yet fully digested what I have already read, I suspect that Robinson and much of the research she cites is “not quite PCT.” Little mention is made of controlled perceptual variables and no mention at all is made of the test for the controlled variable or the behavioral illusion. I suspect that the research she cites to support affect control theory and identify control theory is based on “casting nets” rather than "testing specimens (as Runkel would put it).

The publication of this article is leading me to consider revising my plans for writing an account of theories that are “not quite PCT.” Instead, I am considering checking out the work that she cites to see to what extent they are really PCT (which for some reason she calls "perception control theory) based on testing specimens to find controlled perceptions.

This advantage of this approach is that it gives me a more defined target and a greater chance of publication. The disadvantage (for me) is that I am not a sociologist and so I would be treading on a turf that was not my own.

I could envision such a piece as a prelude to a more comprehensive work that reviews all that PCT-inspired research has found about human (and maybe even animal behavior).

I have Robinson’s article in PDF format and can send it as an attachment to anyone interested in checking it out and giving me feedback on my plans to review it from “pure PCT” perspective.

–Gary

Gary Cziko
Professor
Educational Psychology & English as an International Language
University of Illinois
1310 S. Sixth Street
210F Education Building
Champaign, IL 61820-6990
USA

Telephone +1-217-333-8527
Fax: +1-217-244-7620
e-mail: g-cziko@uiuc.edu

Web: http://garycziko.net
Skype: garyjazz (http://www.skype.com)
Google Talk: gcziko
Amateur Radio: N9MJZ

Founder, Station Manager, Program Director, Chief Engineer and Announcer

The Latino Radio Service of La Casa Cultural Latina
1660 AM on the Urbana-Champaign campus of the University of Illinois
(http://latinoradioservice.org)

Founder and Major Contributor

ATALL Wikibook for
Autonomous Technology-Assisted Language Learning
(http://en.wikibooks.org/wiki/ATALL)

2

[from Dawn T. Robinson 2007.08.03 06:30 EDT]

Gary and All,

Yes, this paper does reviews control theories in sociology more broadly and, specifically, describes several research traditions that were directly inspired by Powers' B:CP. Most of the works reviewed (with McClelland's being one of the notable exceptions) are not that "PCT-like," but all involve some kind of control system model.

I thought about calling your attention to this review i n response to your earlier discussion of a review of "not quite PCT" theories in psychology (including those that were PCT-like but didn't know it and those that made erroneous claims of relatedness). It seemed awfully self-serving for my first post! But, now that you brought it up...

Excuse the formality below, but this seems to be the publisher's required language for me to share the paper:

I am pleased to provide you complimentary one-time access to my Annual Reviews article as a PDF file http://arjournals.annualreviews.org/eprint/mb4KixMGZquXuYVvwrwV/full/10.1146/annurev.soc.32.061604.123110

for your own personal use. Any further/multiple distribution, publication or commercial usage of this copyrighted material would require submission of a permission request addressed to the Annual Reviews Permissions Department, email permissions@AnnualReviews.org.

For any of you do not have institutional access to AR journals, Gary has already offered to send a PDF off-list. I'd be happy to do that as well.

Finally, many of the research traditions mentioned in this review are represented in an edited volume recommended earlier on this list:

McClelland K, Fararo TJ, eds. 2006. Purpose, Meaning and Action: Control Systems Theories in Sociology. New York: Palgrave Macmillan

If you decide to broaden and mention sociological control theories, that might provide some more helpful information for your paper.

Lastly, let me offer an apology for mistakingly calling PCT "perception control theory" in the review! I had seen both used before and thought that the "perceptual" version was the error. Perceptual control theory sounds like it would refer to using perceptual information to control something else; while perception control theory sounds like it refers to the control of perceptions. I think that and my greater familiarity with theories whose names are more parallel to the latter (affect control theory and identity control theory) probably led to my mistake. I clearly should have looked it up before writing the review. Sorry!!

Cheers,
Dawn

Gary Cziko wrote:

···

[from Gary Cziko 2007.08.02 20:00 CDT]

Clark McPhail brought the following recently published article to my attention:

Control Theories in Sociology
Dawn T. Robinson
Annual Review of Sociology, August 2007, Vol. 33, Pages 157-174
(doi: 10.1146/annurev.soc.32.061604.123110)

I was pleased to see that Robinson devotes a section to Powers's PCT with the following intro:

    In the 1970s, a landmark work provided the modern inspiration for
    a new generation of sociological control theories. Powers's (1973)
    book, Behavior: The Control of Perception, offered a comprehensive
    application of control system logic and mathematics to behavioral
    psychology.

She then goes on to discuss affect control theory and identity control theory, including empirical support for these theories.

While I haven't finished the article or even yet fully digested what I have already read, I suspect that Robinson and much of the research she cites is "not quite PCT." Little mention is made of controlled perceptual variables and no mention at all is made of the test for the controlled variable or the behavioral illusion. I suspect that the research she cites to support affect control theory and identify control theory is based on "casting nets" rather than "testing specimens (as Runkel would put it).

The publication of this article is leading me to consider revising my plans for writing an account of theories that are "not quite PCT." Instead, I am considering checking out the work that she cites to see to what extent they are really PCT (which for some reason she calls "perception control theory) based on testing specimens to find controlled perceptions.

This advantage of this approach is that it gives me a more defined target and a greater chance of publication. The disadvantage (for me) is that I am not a sociologist and so I would be treading on a turf that was not my own.

I could envision such a piece as a prelude to a more comprehensive work that reviews all that PCT-inspired research has found about human (and maybe even animal behavior).

I have Robinson's article in PDF format and can send it as an attachment to anyone interested in checking it out and giving me feedback on my plans to review it from "pure PCT" perspective.

--Gary

---------------------------------------------------
Gary Cziko Professor Educational Psychology & English as an International Language
University of Illinois 1310 S. Sixth Street 210F Education Building Champaign, IL 61820-6990
USA

Telephone +1-217-333-8527
Fax: +1-217-244-7620
e-mail: g-cziko@uiuc.edu <mailto:g-cziko@uiuc.edu>
Web: http://garycziko.net
Skype: garyjazz (http://www.skype.com)
Google Talk: gcziko
Amateur Radio: N9MJZ

Founder, Station Manager, Program Director, Chief Engineer and Announcer
The Latino Radio Service of La Casa Cultural Latina
1660 AM on the Urbana-Champaign campus of the University of Illinois
(http://latinoradioservice.org)

Founder and Major Contributor
ATALL Wikibook for
Autonomous Technology-Assisted Language Learning
(Autonomous Technology-Assisted Language Learning - Wikibooks, open books for an open world)

--
Dawn T Robinson
Associate Professor of Sociology
University of Georgia
Athens, GA 30602-1611
phone: 706-542-8948
fax: 706-542-4320

3

[from Gary Cziko 2007.08.03 07:30 CDT]

Dawn:

Welcome to CSGnet!

Thank you for making your presence known on CSGnet (as one
of the listowners of CSGnet, I suppose I should have already known you
were here) and for making your review chapter available for CSGnet
participants.

[from Dawn T. Robinson 2007.08.03 06:30 EDT]

Yes, this paper does reviews control theories in sociology more broadly
and, specifically, describes several research traditions that were
directly inspired by Powers’ B:CP. Most of the works reviewed (with

McClelland’s being one of the notable exceptions) are not that
“PCT-like,” but all involve some kind of control system model.

As I mentioned earlier, I have not yet fully read and digested your paper, but it does appear to be a very useful and concise introduction to PCT-inspired theory and research in sociology.

Most of the works reviewed (with
McClelland’s being one of the notable exceptions) are not that

“PCT-like,” but all involve some kind of control system model.

There are two principal ways in which PCT has been used for behavioral research. The first is doing the Test of the Controlled Variable in which disturbances are applied to a suspected controlled perceptual variable and the behavior of a living system is observed to see if it acts to resist the disturbance. The other is to create a working model of perceptual control, either simple or hierarchical (usually simulated by a computer program) and the behavior of the working model is compared with the behavior of a living system. I’d be very interested to know the extent to which either of these two approaches for PCT research are employed in the research that you review in your chapter.

Lastly, let me offer an apology for mistakingly calling PCT "perception

control theory" in the review! I had seen both used before and thought
that the “perceptual” version was the error.

I had never seen (or at least noticed) the descriptor “perception control theory” before reading your chapter. But a Google search reveals:

“perception control theory” = 301 instances
“perceptual control theory” = 14,800 instances

So perception control theory is indeed out there, but dwarfed by perceptual control theory.

Perceptual control theory
sounds like it would refer to using perceptual information to control

something else; while perception control theory sounds like it refers to

the control of perceptions.

As I understand PCT, both perception and “something else” are controlled. The “something else” is some part of the physical world outside of the living control system which is represented in some way by the perception (
e.g., the position of the actual knot on the actual rubber band is controlled by controlling one’s perception of it).

I think that and my greater familiarity
with theories whose names are more parallel to the latter (affect
control theory and identity control theory) . . .

Nonetheless, “perception control theory” makes sense to me, too. Many things that are controlled don’t seem to have an adjectival form (e.g., speed control, tobacco control, firearm control), but “perception” does have “perceptual.”

If I recall correctly, it was Kent McClelland who first used the term
“perceptual control theory” to refer to control theory applied to living
organisms, so blame it on a sociologist! Although I believe it was me
who suggested the widespread adoption of this term to describe
Powers-type control theory, so I share some of the blame, too!

–Gary

P.S. I have added “Perception control theory” to the Wikipedia with a redirect to “Perceptual control theory.”

···

On 8/3/07, Dawn T. Robinson sodawn@uga.edu wrote:

4

[From Fred Nickols (2007.08.08.0902 ET)]

Dawn Robinson provided a link to the article Gary mentions and so I downloaded and read her paper. I think Gary's intuitive take is consistent with mine: a lot of what is cited is probably not-quite-PCT and relies on casting nets instead of testing specimens. There may be some exceptions (e.g., Kent McClelland's work).

I encountered a couple of puzzlers in reading her paper.

On page 162, as part of her review of what she calls "perception control theory," she indicates that "The reference signal itself is exogenous to the system." I don't understand that. I thought reference signals were part of the behaving person and thus endogenous. Oh well, it's probably just the case that I don't understand big, fancy words.

Also on page 162, her review of the eleven levels of the hierarchy omits the 8th level. I'm not sure what that's about either.

On page 169 she writes "...McClelland focuses primarily on what happens when two or more social actors strive to control a signal variable..." My confusion here ties to the use of "signal." I don't understand that at all and wonder if the word should have been "single" instead of signal.

Overall, the paper was well worth the read, pointing to many more published papers and research efforts tied to PCT than I knew about before reading it, some of which I will check out because they tie to my own interests.

So, thanks to Gary for bringing it up and thanks to Dawn (and her publisher) for making the paper available.

···

--
Regards,

Fred Nickols
Managing Principal
Distance Consulting
nickols@att.net
www.nickols.us

"Assistance at A Distance"
      
-------------- Original message ----------------------
From: Gary Cziko <g-cziko@UIUC.EDU>

[from Gary Cziko 2007.08.02 20:00 CDT]

Clark McPhail brought the following recently published article to my
attention:

Control Theories in Sociology
Dawn T. Robinson
Annual Review of Sociology, August 2007, Vol. 33, Pages 157-174
(doi: 10.1146/annurev.soc.32.061604.123110)

I was pleased to see that Robinson devotes a section to Powers's PCT with
the following intro:

In the 1970s, a landmark work provided the modern inspiration for a new
> generation of sociological control theories. Powers's (1973) book, Behavior:
> The Control of Perception, offered a comprehensive application of control
> system logic and mathematics to behavioral psychology.
>

She then goes on to discuss affect control theory and identity control
theory, including empirical support for these theories.

While I haven't finished the article or even yet fully digested what I have
already read, I suspect that Robinson and much of the research she cites is
"not quite PCT." Little mention is made of controlled perceptual variables
and no mention at all is made of the test for the controlled variable or the
behavioral illusion. I suspect that the research she cites to support affect
control theory and identify control theory is based on "casting nets" rather
than "testing specimens (as Runkel would put it).

The publication of this article is leading me to consider revising my plans
for writing an account of theories that are "not quite PCT." Instead, I am
considering checking out the work that she cites to see to what extent they
are really PCT (which for some reason she calls "perception control theory)
based on testing specimens to find controlled perceptions.

This advantage of this approach is that it gives me a more defined target
and a greater chance of publication. The disadvantage (for me) is that I am
not a sociologist and so I would be treading on a turf that was not my own.

I could envision such a piece as a prelude to a more comprehensive work that
reviews all that PCT-inspired research has found about human (and maybe even
animal behavior).

I have Robinson's article in PDF format and can send it as an attachment to
anyone interested in checking it out and giving me feedback on my plans to
review it from "pure PCT" perspective.

--Gary

---------------------------------------------------
Gary Cziko
Professor
Educational Psychology & English as an International Language
University of Illinois
1310 S. Sixth Street
210F Education Building
Champaign, IL 61820-6990
USA

Telephone +1-217-333-8527
Fax: +1-217-244-7620
e-mail: g-cziko@uiuc.edu
Web: http://garycziko.net
Skype: garyjazz (http://www.skype.com)
Google Talk: gcziko
Amateur Radio: N9MJZ

Founder, Station Manager, Program Director, Chief Engineer and Announcer
The Latino Radio Service of La Casa Cultural Latina
1660 AM on the Urbana-Champaign campus of the University of Illinois
(http://latinoradioservice.org)

Founder and Major Contributor
ATALL Wikibook for
Autonomous Technology-Assisted Language Learning
(Autonomous Technology-Assisted Language Learning - Wikibooks, open books for an open world)

5

[From Bill Powers (2007.08.03.0740 MDT)]

Still awkwardly replying via webmail on a dial-up connection.

Dawn T. Robinson 2007.08.03 06:30 EDT --

Welcome aboard as a speaking member!

I'll read your review when my system is back to normal.

RE perception control theory:

This is the way Clark McPhail has always referred to PCT, and I haven't objected because he is probably right: B is the control of perception, not the control of perceptual. Perceptual control unfortunately comes too close to meaning control BY perception instead of control OF perception, though the latter usage does exist in the language. I was trying for ambiguity in the title of that book, for reasons that are no longer completely clear to me except that maybe I thought the shock of discovering the intended meaning would help fix the concept in the reader's mind. I have had to outgrow these overly complicated trains of thought.

Until later,

Best,

Bill P.

6

[From Bill Powers (2007.08.04.1005 MDT)]

Dawn T. Robinson 2007.08.03 06:30 EDT–

What a pleasure it is to read your article! You can imagine what it means
to me to see my work well-described, endorsed, and gainfully employed. A
fine job.

Best regards,

Bill Powers

7

[From Dawn Robinson (2007.08.04 12:09 ET)]

Thanks, Bill for your kind words and welcome. Thanks, also to Jeff for the helpful references and to Fred and Gary for pointing out some of the errors in the paper. Goes to show why *responsible* authors send their works to the relevant experts *before* they get published. Below are some more specific responses.

nickols@att.net wrote:

[From Fred Nickols (2007.08.08.0902 ET)]

Dawn Robinson provided a link to the article Gary mentions and so I downloaded and read her paper. I think Gary's intuitive take is consistent with mine: a lot of what is cited is probably not-quite-PCT and relies on casting nets instead of testing specimens. There may be some exceptions (e.g., Kent McClelland's work).

You're right, Fred. By and large, the traditions I mention do rely on group comparisons in some fashion or another. And, most of the theories I mention are not at all PCT. But, most of the contemporary ones benefited directly from Powers' ideas. You might want to see them as friendly offspring (or wayward children, if you must).

[From Fred Nickols (2007.08.08.0902 ET)]

On page 162, as part of her review of what she calls "perception control theory," she indicates that "The reference signal itself is exogenous to the system." I don't understand that. I thought reference signals were part of the behaving person and thus endogenous.
  [From Ted Cloak (2007.08.03.1000 MST)]
Doesn't Dawn mean here "exogenous to the _control_ system", not referring
loosely to the behaving organism as "the system"?

Ted is right. I was referring to the control system rather than a person when I said "the system". That distinction becomes fairly important in affect control theory, even though it is less important for pct and ict.

[From Ted Cloak (2007.08.03.1000 MST)]
[Cloak quoting Powers]
"One change in the model covers all cases, including many we have not yet
considered. The change is this: We will assume from now on that _all_
_reference_ _signals_ _are_ _retrieved_ _recordings_ _of_ _past_
_perceptual_ _signals_ [emphasis in original]. This requires giving the
outputs from higher-order systems the function of address signals, whereas
formerly they were reference signals. The address signals select from
lower-order memory those past values of perceptual signals that are to be
recreated in present time."
[back to Cloak]
Hence reference signals are _always_ endogenous, not only to the organism
but also to the control system.

Hm. That still to me sounds like they are exogenous to the control system of interest. Nonetheless, I shall take your word for it. I will no longer refer to reference signals as coming from outside the control system when discussing pct.

Also on page 162, her review of the eleven levels of the hierarchy omits the 8th level. I'm not sure what that's about either.

On page 169 she writes "...McClelland focuses primarily on what happens when two or more social actors strive to control a signal variable..." My confusion here ties to the use of "signal." I don't understand that at all and wonder if the word should have been "single" instead of signal.

Yikes! That is indeed a typo - should have been "single". It certainly is a confusing one, though. Hopefully, folks will go straight to Kent's work to get it straightened out.

Thanks again for the warm welcomes and the helpful feedback.

Cheers,
Dawn

···

--
Dawn T Robinson
Associate Professor of Sociology
University of Georgia
Athens, GA 30602
706-542-8948
sodawn@uga.edu