Behaviour

CSGnet Archive CSG2003
#1

I’d just like to ask a question regarding ‘behaviour’ in terms of PCT. In terms of behaviour I am referring to actions people do. It may be very basic to many

I am just having some trouble being sure were behaviour in this sense fits. I think my trouble may in an attempt not to fall into old traps, not tackling this point . The purpose of working this out is that some very good work (I think anyway) was done by Rodger Barker on Behaviour Setting in Enviro Psych field and I’m trying to interpret this from a PCT point of view. He got some really consistent results and valuable work was done. I’m pretty sure PCT will explain it well and provide a theory why it was so consistent rather than it being the result of the ‘mileau’. Anyway to the question.

I am aware and fundamentally agree that people control perceptions not behaviour in general. However it is true, is it not, that perceptions of behaviour as a controlled variable( the sort of actions someone from a behaviorist viewpoint would focus apon) is controlled at one level in the hierachy. I think it is level 9 relating to ‘programatic series of behaviour actions’. At this level people are controlling there perceptions of their behaviour (e.g. standing up out of a chair). True?

A perception relating to behaviour however may be being controlled due controlling an array of variables higher up the hierarchy. Such as respect and social for a important person, or disrespect for a teacher if it out of order. True?

And to control a perception relating to a ‘behaviour’ a number of other lower order perceptions need to be controlled. True?

However back to the first point - perceptions (controlled variable) relating to behaviour are being controlled at one point. True?

I hope this is not too ‘old hat’ but it would help my thesis if I can clarify it.

Thanks Rohan Lulham

···

-----Original Message-----
From: Control Systems Group Network (CSGnet) [mailto:CSGNET@listserv.uiuc.edu]On Behalf Of Bruce Nevin
Sent: Tuesday, January 07, 2003 8:07 AM
To:
CSGNET@listserv.uiuc.edu
Subject: Re: Control of Language

[From Bruce Nevin (2003.06.01 16:06 EST)]

Bill Powers (2003.01.06.0618 MST)–
At 08:29 AM 1/6/2003, Bill Powers wrote:

When you say you're controlling x, x is the name of a variable that can have many values: temperature, sweetness, position, rate of spin, pressure, income. Controlling *for* x requires x to be a specific state of a variable. Mixing the two usages sounds strange

There is a ready alternative to mixture: when you say you’re controlling x at reference r.

Why do I doubt that this will settle the matter?

Probably because such things are not settled by fiat. I say this having conceded the specific point to the extent that it is possible to, when I replied to Rick:

Bruce Nevin (2003.06.01 15:45 EST)–

You might be right. Time will tell.

      /Bruce N
#2

[From Rick Marken (2003.01.07.0940)]
Rohan Lulham wrote:

I’d just like to
ask a question regarding ‘behaviour’ in terms of PCT. In terms of behaviour
I am referring to actions people do. It may be very basic to many
I am just
having some trouble being sure were behaviour in this sense fits. I think
my trouble may in an attempt not to fall into old traps, not tackling this
point . The purpose of working this out is that some very good work (I
think anyway) was done by Rodger Barker on Behaviour Setting in Enviro
Psych field and I’m trying to interpret this from a PCT point of view.
He got some really consistent results and valuable work was done. I’m pretty
sure PCT will explain it well and provide a theory why it was so consistent
rather than it being the result of the ‘mileau’. Anyway to the question.
I am aware and fundamentally
agree that people control perceptions not behaviour in general. However
it is true, is it not, that perceptions of behaviour as a controlled variable(
the sort of actions someone from a behaviorist viewpoint would focus apon)
is controlled at one level in the hierachy. I think it is level 9 relating
to ‘programatic series of behaviour actions’. At this level people are
controlling there perceptions of their behaviour (e.g. standing up out
of a chair). True?
I think a better way to look at it is in terms
of seeing behavior from the perspective of the behaving system vs seeing
it from the perspective of the observer of the behaving system. Someone
from a behaviorist point of view looks at behavior ( a word that refers
to both actions and the results produced by actions) only from the observer’s
perspective. Someone from a PCT perspective tries to look at behavior from
both the observer’s and the behaving system’s perspective. This is basically
the issue I look at in my “Looking at Behavior through Control Theory Glasses”
paper (Review of General Psychology, 2002, Vol. 6, No. 3, 260-270).
I don’t recall what behavior Roger Barker
was dealing with but let’s assume it was something like the path a person
takes through some environment. The person’s behavior, from the observer’s
perspective, is the path: changes in the person’s 2 dimensional coordinates
over time. This behavior, from the person’s own perspective, consists
if the perceptions the person is controlling for. If the person is controlling
for a particular proximity to objects and for a particular target position
(as in the CROWD program) then the person’s behavior, from the person’s
own perspective, is not at all like the person’s behavior, from the observer’s
perspective. It’s possible, however, that the person is controlling for
a particular path. The person, for example, might be going to a list of
lat-lon positions using a compass (as my wife and I do occasionally when
we orienteer). In that case, the 2 D path seen by the observer does correspond
(so some extent) to the 2 D path controlled by the behaving system.

I think that the most useful way to approach
conventional studies of behavior (like Barker’s) is to try to look at the
situation from the point of view of the participants in the study. Ask
yourself what you might be trying to do (perceive) if you were in the
same situation as the participant in the study. I think that’s what
it means to look at behavior through control theory glasses. You look at
the behavior of a rat in a Skinner box, for example, and try to put yourself
in the position of the (very) hungry rat. When you do this you realize
that what you are doing is acting to produce, as frequently as possible,
a perception of food. Or you look at the behavior of a gray lag goose
(as I did in the “PCT glasses” paper) and try to put yourself in the position
of the hen who has the egg taken away from her as she is dragging it towards
her nest. When you so this you realize that the hen is probably controlling
a perception of pressure on the back of her bill.

I think it is useful to try to understand
(and then model) behavior observed from a behaviorist perspective
in terms of PCT. I’d be interested in hearing about what Barker found
and what you make of it from a PCT perspective.

Best regards

Rick

···

–

Richard S. Marken, Ph.D.

Senior Behavioral Scientist

The RAND Corporation

PO Box 2138

1700 Main Street

Santa Monica, CA 90407-2138

Tel: 310-393-0411 x7971

Fax: 310-451-7018

E-mail: rmarken@rand.org

#3

[From Rick Marken (2003.01.07.1200)]

Fred Nickols (2003.01/07/1330 EST) --

>[From Rick Marken (2003.01.07.0940)]
>
>Someone from a behaviorist point of view looks at behavior ( a word that
>refers to both actions and the results produced by actions) only from the
>observer's perspective.

I am assuming that the statement above reflects your view of the
"behaviorist point of view" and that you do not consider yourself a
behaviorist. In any event, that's not the way I was trained in the
behaviorist viewpoint. I was taught that "behavior' refers to the
activities of the organism. Terms like "achievements" (Gilbert Ryle),
"accomplishments" (Thomas Gilbert) and "outcomes" (just about everyone)
were and are used to refer to the effects of those actions, in particular,
to the intended effects.

Yes. I agree. But behaviorists (and this includes cognitive psychologists and just
about every other kind of psychologist besides the PCT kind) describe these things
from _their_ point of view. That is, in terms of their own perceptions. From the
point of view of a behaviorist the gray lag goose accomplishes or achieves the
rolling of an egg into its nest. But from the point of view of a control theorist
the gray lag goose accomplishes or achieves the maintenance of a particular
feeling of pressure against the back of its bill. From the point of view of a
behaviorist a center fielder produces the outcome "run to catch the ball". From
the point of view of a control theorist the center fielder produces the outcome
"perception of constant vertical and horizontal optical velocity of ball".

Further, I suppose I could look at my own behavior from my own perspective
but is that not still the perspective of an observer?

Yes. But what you observe while you are behaving is quite different than what you
observe when others are doing the same behavior.

And, of course, I
don't see anyway I could actually see anyone else's behavior from his or
her perspective, although I can certainly try to be as empathetic as I can,
and that's what I am assuming you are saying, that someone operating from a
PCT point of view tries be empathetic.

Yes. Where by "empathy" I mean _imagining_ what you would be perceiving (and
controlling) if you were doing the observed behavior. The behaviorist sees Fred
Astaire dancing as what Fred is doing; the control theorist sees Fred Astaire's
dancing as extraordinary visual side effect Fred's control of his own perceptions
of intensities, sensations (of pressure), configurations (of pressures),
transitions (of pressure configurations) and so on.

I think it's fair to call PCT rigorously applied empathy.

Do I have that right?

Yes!

Best regards

Rick

···

--
Richard S. Marken, Ph.D.
Senior Behavioral Scientist
The RAND Corporation
PO Box 2138
1700 Main Street
Santa Monica, CA 90407-2138
Tel: 310-393-0411 x7971
Fax: 310-451-7018
E-mail: rmarken@rand.org

#4

[From Bill Powers (2003.01.07.0934 MSTZ)]

Rohan Lulham (2003.01.07);

I
am aware and fundamentally agree that people control perceptions not
behaviour in general. However it is true, is it not, that perceptions of
behaviour as a controlled variable( the sort of actions someone from a
behaviorist viewpoint would focus apon) is controlled at one level in the
hierachy. I think it is level 9 relating to ‘programatic series of
behaviour actions’. At this level people are controlling there
perceptions of their behaviour (e.g. standing up out of a chair).
True?

Standing up out of a chair is a relationship (body out of the chair,
level 6) achieved by an event (level 5) composed of transitions (level 4)
of configurations (level 3 – and so on). Control occurs at all these
levels. Representing this action symbolically involves representing many
such processes as examples of a single class of behavior, and creating a
structure of verbal symbols, “Standing up out of a chair,” with
which to refer to any example of this class (involving different people,
different chairs, different ways of standing up, different degrees of
“outness,” and different starting and ending configurations).
All the lower-order differences are discarded at and above the category
level, so that any instance of standing up out of a chair can be referred
to by using the same verbal description as any other instance of the
“same” behavior.

As soon as you recognize the hierarchical nature of perception and
control, you can see that what is a perception at one level
(straightening the legs) is the action by which a perception at the next
level is controlled (standing up). The perception of standing up then
becomes the action by means of which the next level of perception (out of
the chair) is achieved.

The term “behavior” can be and is used to refer to any change
in anything. We can speak of the behavior of a drop of rain sliding down
a window, the behavior of steam emerging from a valve, the behavior of a
rat pressing a lever to obtain food, the behavior of a lump of sugar
dissolving in a teacup, or the behavior of a kite in the wind. Behavior
is just whatever happens while you’re looking. It really has no technical
definition and is not specific to any one aspect of human, animal, or
nonliving activities.

In the attempt to tighten the usage of language here, I propose that we
speak of the variables in a typical control system explicitly (input
quantity, perceptual signal, reference signal, error signal, and output
signal). The eliminates any ambiguity, because these variables are always
defined for one and only one control system. To speak of the same
variables in another system of the same level, we would have to add
subscripts so we speak of p[1], p[2], or e[1], e[2], where the numbers
designate different control systems. We could add a second subscript to
designate the level of the system: p[1,1] is the perceptual signal in the
first system at the first level, and so forth.

We could now speak of the action or behavior produced by system S at
level L by referring to its output signal o[S,L]. This could never be
confused with the output of any other system, either at the same level or
a different level. All the ambiguities mentioned above would disappear.
If we use a term like action or behavior, it would be clear that we are
not being completely specific (which is all right in informal
conversation), so no conclusions should be drawn from any serious
implications of our word choice. If I want to be specific about what I
object to in someone’s “behavior” or “actions”, I
would have to say something like “I don’t like the angle at which he
wears his hat,” or “…his constantly uttering
obscenities.” Neither “action” nor “behavior” is
specific enough to point to exactly what I mean.

I haven’t specifically answered your question or commented on your
propositions, but perhaps now you can work out what I would have
said.

Best,

Bill P.

#5

[From Fred Nickols (2003.01/07/1330 EST)] --

[From Rick Marken (2003.01.07.0940)]

Someone from a behaviorist point of view looks at behavior ( a word that
refers to both actions and the results produced by actions) only from the
observer's perspective.

I am assuming that the statement above reflects your view of the
"behaviorist point of view" and that you do not consider yourself a
behaviorist. In any event, that's not the way I was trained in the
behaviorist viewpoint. I was taught that "behavior' refers to the
activities of the organism. Terms like "achievements" (Gilbert Ryle),
"accomplishments" (Thomas Gilbert) and "outcomes" (just about everyone)
were and are used to refer to the effects of those actions, in particular,
to the intended effects.

Further, I suppose I could look at my own behavior from my own perspective
but is that not still the perspective of an observer? And, of course, I
don't see anyway I could actually see anyone else's behavior from his or
her perspective, although I can certainly try to be as empathetic as I can,
and that's what I am assuming you are saying, that someone operating from a
PCT point of view tries be empathetic.

Do I have that right?

Fred Nickols
nickols@safe-t.net

#6

Rohan,

I for one would like to know more about your thesis. The Barker work was
influential in the development of ACT
(http://www.indiana.edu/~socpsy/ACT/),
an offspring of PCT, yet there are still interesting questions about how
and why named behaviors occur with predictable variations in
frequency in different parts of the environment.

Dave Heise

···

At 11:02 PM 1/7/2003 +1100, you wrote:

I am just
having some trouble being sure were behaviour in this sense fits. I think
my trouble may in an attempt not to fall into old traps, not tackling
this point . The purpose of working this out is that some very good work
(I think anyway) was done by Rodger Barker on Behaviour Setting in Enviro
Psych field and I’m trying to interpret this from a PCT point of view. He
got some really consistent results and valuable work was done. I’m pretty
sure PCT will explain it well and provide a theory why it was so
consistent rather than it being the result of the ‘mileau’. Anyway to the
question.

#7

Fred thanks for the reply,

Is the key here from a PCT view that you need to look at behaviour or output
signals in terms of the perceptions ( reference signals) a person is
controlling (the viewpoint of the person). Looking at an output signal
(behaviour) without knowing the reference signal of the person ( just
observing) can lead to a myriad of problems with ecological validity (what
you observe to be happening is not really the case and thus interventions
may miss the mark based on the observation).

You can't get the viewpoint of the observer exactly, but trying to find out
what reference signal they are controlling really helps.

Regards Rohan

···

-----Original Message-----
From: Control Systems Group Network (CSGnet)
[mailto:CSGNET@listserv.uiuc.edu]On Behalf Of Fred Nickols
Sent: Wednesday, January 08, 2003 5:37 AM
To: CSGNET@listserv.uiuc.edu
Subject: Re: Behaviour

[From Fred Nickols (2003.01/07/1330 EST)] --

[From Rick Marken (2003.01.07.0940)]

Someone from a behaviorist point of view looks at behavior ( a word that
refers to both actions and the results produced by actions) only from the
observer's perspective.

I am assuming that the statement above reflects your view of the
"behaviorist point of view" and that you do not consider yourself a
behaviorist. In any event, that's not the way I was trained in the
behaviorist viewpoint. I was taught that "behavior' refers to the
activities of the organism. Terms like "achievements" (Gilbert Ryle),
"accomplishments" (Thomas Gilbert) and "outcomes" (just about everyone)
were and are used to refer to the effects of those actions, in particular,
to the intended effects.

Further, I suppose I could look at my own behavior from my own perspective
but is that not still the perspective of an observer? And, of course, I
don't see anyway I could actually see anyone else's behavior from his or
her perspective, although I can certainly try to be as empathetic as I can,
and that's what I am assuming you are saying, that someone operating from a
PCT point of view tries be empathetic.

Do I have that right?

Fred Nickols
nickols@safe-t.net

#8

Thanks for the reply Bill and David,

If I could just have another go of explaining it in PCT language it would be much appreciated. Your response has put it in a better perspective, but also I may need to describe Barkers work very quickly and a very basic (possible) PCT explanation.

Basically what Barker found that in many established setting - he called them behaviour settings - there were very consistent ‘standard patterns of behaviour’. A behaviour setting is a defined space occuppied by people with consistent roles where certain behaviours patterns occur such as a classroom, a shop counter, a church. He found in each setting a number of very consistent behaviour patterns nearly all people did- (noteablely from ‘observation’ which PCT may have a problem with). In fact by definition a ‘behaviour pattern’ is a ‘behaviour’ (o(n) or r(N-1) ) that is displayed by all people with a degree of consitency above 85 percent between people (e.g. 85 % of people consitently do it). Studies were done on whole towns, schools, churches, suburbs etc.

The behaviours patterns from a PCT view would be I think be reference signals at the program level (r[9]) . As such they would be the output signal for level ten- principles (o[10]).

It is also noted that the control of the reference signal at nine(r[9]) requires an output signal that sets the reference signal to the lower level eight (r[8]) and so forth. However where I think the answer to explaining the consitency of these behaviours is up the hierachy.

(1) What I think the PCT explanation for the consitency of the reference signals at the ninth level (r[9]) - is a common reference signal controlled at a high level of gain at the 10th level (r[10]). As the refernce level at 9th is the output at level 10. People are controlling a principle level (10) reference signal (r[10]), which is common to many people, at a high level of gain.

(2) Or, as it is just comming to me, they may be controlling different refernce signals at level 10 r[10] that required the use of the same outputs signals(o[10]). And as they use the same output (o[10]), the same reference signal at level 9 (r[9]) is controlled with consitency. ----- Arrh I liked the first one but know see the second is the more accurate and what you guys would go with.

I need to do a bit of postualating but the reference signal controlled at Level 10 (r[10]) may be something like ‘getting on with others’ or ‘fitting in with others’. My field would currently explain it as the result of something primarily fundamental of the environment, both cultural and physical, which is ‘mysterious’. The cause is often put solely outside of the person.

I’m not sure if I explained it quite correctly. I hope I have hit the mark, but quite possibly as it is in with this theory I may have not. In relation to what Bill once said about people learning PCT, ‘people think they know it in a week but it actually takes 2 years’- I think I’m at about the 18 -20 month mark.

Thanks for your time. If this sort of post is not appropriate for the list (e.g. too basic) please let me know.

Rohan.

···

-----Original Message-----
From: Control Systems Group Network (CSGnet) [mailto:CSGNET@listserv.uiuc.edu]On Behalf Of Bill Powers
Sent: Wednesday, January 08, 2003 4:21 AM
To:
CSGNET@listserv.uiuc.edu
Subject: Re: Behaviour

[From Bill Powers (2003.01.07.0934 MSTZ)]

Rohan Lulham (2003.01.07);

I am aware and fundamentally agree that people control perceptions not behaviour in general. However it is true, is it not, that perceptions of behaviour as a controlled variable( the sort of actions someone from a behaviorist viewpoint would focus apon) is controlled at one level in the hierachy. I think it is level 9 relating to 'programatic series of behaviour actions'. At this level people are controlling there perceptions of their behaviour (e.g. standing up out of a chair). True?

Standing up out of a chair is a relationship (body out of the chair, level 6) achieved by an event (level 5) composed of transitions (level 4) of configurations (level 3 – and so on). Control occurs at all these levels. Representing this action symbolically involves representing many such processes as examples of a single class of behavior, and creating a structure of verbal symbols, “Standing up out of a chair,” with which to refer to any example of this class (involving different people, different chairs, different ways of standing up, different degrees of “outness,” and different starting and ending configurations). All the lower-order differences are discarded at and above the category level, so that any instance of standing up out of a chair can be referred to by using the same verbal description as any other instance of the “same” behavior.

As soon as you recognize the hierarchical nature of perception and control, you can see that what is a perception at one level (straightening the legs) is the action by which a perception at the next level is controlled (standing up). The perception of standing up then becomes the action by means of which the next level of perception (out of the chair) is achieved.

The term “behavior” can be and is used to refer to any change in anything. We can speak of the behavior of a drop of rain sliding down a window, the behavior of steam emerging from a valve, the behavior of a rat pressing a lever to obtain food, the behavior of a lump of sugar dissolving in a teacup, or the behavior of a kite in the wind. Behavior is just whatever happens while you’re looking. It really has no technical definition and is not specific to any one aspect of human, animal, or nonliving activities.

In the attempt to tighten the usage of language here, I propose that we speak of the variables in a typical control system explicitly (input quantity, perceptual signal, reference signal, error signal, and output signal). The eliminates any ambiguity, because these variables are always defined for one and only one control system. To speak of the same variables in another system of the same level, we would have to add subscripts so we speak of p[1], p[2], or e[1], e[2], where the numbers designate different control systems. We could add a second subscript to designate the level of the system: p[1,1] is the perceptual signal in the first system at the first level, and so forth.

We could now speak of the action or behavior produced by system S at level L by referring to its output signal o[S,L]. This could never be confused with the output of any other system, either at the same level or a different level. All the ambiguities mentioned above would disappear. If we use a term like action or behavior, it would be clear that we are not being completely specific (which is all right in informal conversation), so no conclusions should be drawn from any serious implications of our word choice. If I want to be specific about what I object to in someone’s “behavior” or “actions”, I would have to say something like “I don’t like the angle at which he wears his hat,” or “…his constantly uttering obscenities.” Neither “action” nor “behavior” is specific enough to point to exactly what I mean.

I haven’t specifically answered your question or commented on your propositions, but perhaps now you can work out what I would have said.

Best,

Bill P.

#9

[From Rick Marken (2003.01.11.1100)]

Rohan Lulham wrote:

Fred thanks for the reply,

Is the key here from a PCT view that you need to look at behaviour
or output

signals in terms of the perceptions ( reference signals) a person
is

controlling (the viewpoint of the person).

Well, no one asked me, but…

I would say that the key to the PCT view is that the behavior we see
is a hierarchical arrangement of actions and results controlled by those
actions. Actions at one level of the hierarchy are themselves results that
are controlled by actions lower in the hierarchy “Signals” are theoretical
entities that explain the hierarchical behavioral relationships that we
observe. We don’t really see signals, be they perceptual, reference or
output.

Basically what Barker
found that in many established setting - he called them behaviour settings

  • there were very consistent ‘standard patterns of behaviour’.
    This is similar to one of the fundamental observations of PCT: that people
    (and all organisms) produce consistent behavioral results, such as the
    one’s you mention: going to class, paying for products over the counter
    and worshipping in church. What is interesting about this, from a PCT perspective,
    is that this consistency is produced in ever-changing environmental circumstances:
    the route to class, for example, always has different obstacles. Different
    pews are occupied each time you show up at church.
    People manage to produce consistent results under changing circumstances
    by varying the means (actions) they use to produce these results. Perhaps
    Barker noticed this. If so, then he had his nose right up against the door
    to PCT. What he also might have noticed is that people produce consistent
    result using different means (actions) every time. Kids consistently show
    up for class, for example, but the path they take to class is different
    each time.

Consistent results are produced by variable means. Skinner noticed
this fact about behavior and explained it away with he concept of the “operant”.
What Skinner failed to notice is that the variation in the means used to
produce a consistent result (such as a bar press) is exactly what is required
in order to compensate for the variations in the circumstances obtaining
each time the result is produced. Indeed, people must vary the means
they use to produce consistent results or those results will not be consistent.
This is the observation that is basis of PCT. Indeed, PCT exists as an
explanation of how this works, that is, how people are able to vary their
actions appropriately so that they are able to produce consistent results.
The PCT explanation is that consistent results are consistently produced
perceptions. Reference signals inside the person specify the intended state
of these perceptions and actions are driven (in a closed loop) by any discrepancy
between reference signals and perceptions. The consistent results people
produce are called “controlled results” and the process of producing these
consistent results is called “control”.

The behaviours patterns
from a PCT view would be I think be reference signals at the program level
(r[9]) . As such they would be the output signal for level ten- principles
(o[10]).
I would say that consistent behavior patterns correspond to controlled
perceptions in PCT. Reference signals (presumably) specify the states of
these behavior patterns (perceptions), which can be specified for perceptions
at any level in the hierarchy. People consistently produce sensations (of
warmth, for example), transitions (speed), configurations (a pen), etc.
on up the hierarchy (all at the same time, of course).
It is also noted
that the control of the reference signal at nine(r[9]) requires an
output signal that sets the reference signal to the lower level eight (r[8])
and so forth.
I think you mean control of the perceptual signal. Reference signals are
not really controlled. At higher levels of the hierarchy they are varied
as necessary as the means of keeping higher level perceptions under control.
However where I
think the answer to explaining the consitency of these behaviours is up
the hierachy.
I think the consistency of behavior is explained by the control loop (as
I described it above). Consistent behavior, like consistently sitting in
church, occurs (according to PCT) because people set references for perceiving
particular results such as “sitting in church”) and vary their actions,
as necessary, to produce this perception (when it is intended).
The fact that a person sets a particular reference consistently may
be what you are talking about. PCT explains this consistency in terms of
hierarchical control. Going to church is presumably an action that is produced
as the means of controlling a higher level goal, such as “being part of
a group”. As long as there are not disturbances to the perception
of “being part of the group” that require a change of action, the same
action (going to church) will be consistently produced as a lower level
means of controlling for “being part of the group”. If, however, the group
starts going bowling instead of to church, the “going to church” result
will no longer keep “being part of the group” under control. So the reference
for “going to church” will be changed to “go bowling”, which will now be
produced consistently instead of “going to church”.

  1. What I think
    the PCT explanation for the consitency of the reference signals at the
    ninth level (r[9]) - is a common reference signal controlled at a high
    level of gain at the 10th level (r[10]). As the refernce level at 9th is
    the output at level 10. People are controlling a principle level (10) reference
    signal (r[10]), which is common to many people, at a high level of gain.
    (2) Or, as it is just comming to me, they may be controlling different
    refernce signals at level 10 r[10] that required the use of the same outputs
    signals(o[10]). And as they use the same output (o[10]), the same reference
    signal at level 9 (r[9]) is controlled with consitency. ----- Arrh I liked
    the first one but know see the second is the more accurate and what you
    guys would go with.
    Perhaps you can see that the PCT explanation for the consistency of reference
    signals is neither 1) nor 2). The explanation (to put it as simply as possible)
    is: there will be a consistent lower level reference when there is no disturbance
    to the higher level perception controlled by variation in this reference.
    It’s basically the same as the explanation for why there is sometimes no
    change in action during a tracking task. If there is is disturbance to
    cursor position (the controlled perception) there is no need to change
    the action that is keeping this perception under control.
    Hope this helps.

Best regards

Rick

···

Richard S. Marken

MindReadings.com

marken@mindreadings.com

310 474-0313

#10

Thanks for comments I thinked they may have nailed a number of aspects of it.

Rohan Lulham wrote:

···

Fred thanks for the reply,

Is the key here from a PCT view that you need to look at behaviour or output
signals in terms of the perceptions ( reference signals) a person is
controlling (the viewpoint of the person). <

Well, no one asked me, but… — <

I thought by recieving e-mails from this list you are in sense being asked what you think.

I would say that the key to the PCT view is that the behavior we see is a hierarchical arrangement of actions and results controlled by those actions. <

Are not the results controlled by those actions the reference signal? And isn’t it important to investigate behaviour in relation to the results being controlled.

Actions at one level of the hierarchy are themselves results that are controlled by actions > lower in the hierarchy .
“Signals” are theoretical entities that explain the > hierarchical behavioral relationships that we observe. We don’t really see signals, be they perceptual, reference or output.

I am not sure what you are getting at but possibly it was my intial use of the words ‘look at’. Poor choice of words and probably should have used ‘investigate’- you don’t miss much. I realise you can not see perceptual signals. We can’t see them but we can investigate them and ‘test’ for them??

<People manage to produce consistent results under changing circumstances by varying the means (actions) they use to produce these results. Perhaps Barker noticed this. If so, then he had his nose right up against the door to PCT. What he also might have noticed is that people produce consistent result using different means (actions) every time. Kids consistently show up for class, for example, but the path they take to class is different each time. >

Thanks for making that comparison and insights. Yes I think he did and its explained a bit later. In addittion to the fact individuals always manage to control the for the achieving the same results- even though the disturabmce - do you think it is also important and relevant to PCT that different people in similar roles appear to be controlling for the same results. Why are they controlling similar actions patterns(output)?-

1) What I think the PCT explanation for the consitency of the reference signals at the ninth level (r[9]) - is a common reference signal controlled at a high level of gain at the 10th level (r[10]). As the refernce level at 9th is the output at level 10. People are controlling a principle level (10) reference signal (r[10]), which is common to many people, at a high level of gain. (2) Or, as it is just comming to me, they may be controlling different refernce signals at level 10 r[10] that required the use of the same outputs signals(o[10]). And as they use the same output (o[10]), the same reference signal at level 9 (r[9]) is controlled with consitency. ----- Arrh I liked the first one but know see the second is the more accurate and what you guys would go with.

Perhaps you can see that the PCT explanation for the consistency of reference signals is neither 1) nor 2). The explanation (to put it as simply as possible) is: there will be a consistent lower level reference when there is no disturbance to the higher level perception controlled by variation in this reference. It’s basically the same as the explanation for why there is sometimes no change in action during a tracking task. If there is is disturbance to cursor position (the controlled perception) there is no need to change the action that is keeping this perception under control.

  Thanks for this insight. I think its really important. I am going to need to think about it more but I think at one level you may have nailed an importnat part of explaining some of Barkers work. The findings of his and follwers work (most of it done in the sisteies and seventies) moved on to look at the staffing of behaviour setting- e.g what happened when there was more or less people than usual in a setting e.g. shop, church, football game. The approach basically found that in situations in which there were a lot more people or a lot less people had to change there actions to make the setting work- e.g control for the same results ( they called it adaptive mechanisms). They proposed settings had a mainatence miminum( least number of people) and capacity (most number of people) in which the setting would still work ( people could control their results).

Thanks greatly for these ideas it helps. As evident, I need to work on these ideas in more detail.

If it is Ok in a month or so I’ll post a again on another theory in Environmental psychology which I’d like to look at from PCT view. Its a theory by Altman(1975) on Privacy, Personal Space and Territoriality. It basically suggests how people control for privacy and social interaction through manipulating territorial and and personl space(e.g. controlling results related to social interaction through controlling terriorality and personal space . I think PCT will have a lot it can say about it.

It is also interesting that both these theories have somewhat died in the field although recognised as some of the best work- and I think partly it is because how they work psychologically has not been explained.

Thanks greatly,

Rohan Lulham

#11

[From Rick Marken (2003.01.13.0900)]

Rohan Lulham wrote:

Thanks for comments
I thinked they may have nailed a number of aspects of it.
Great. You’re welcome.

Well, no one asked me, but…
— <
I thought by recieving e-mails from
this list you are in sense being asked what you think.

My experience is that you can’t be too careful in this medium (e-mail).
I’ve gotten in big trouble for what I thought were far less egregious violations
of civility than speaking out of turn.

Are
not the results controlled by those actions the reference signal?

Not really. The results of actions are states of variables, like
the position of a cursor or the level of taxation of dividend income. The
reference signal is a theoretical entity that explains the fact that
a controller maintains a result in one state (“on target”, “high tax rate”)
rather than another (“1 cm to the left of the target”, “low tax rate”).

And
isn’t it important to investigate behaviour in relation to the results
being controlled.
Yes. That is probably PCT’s most important message to conventional psychology.

do
you think it is also important and relevant to PCT that different people
in similar roles appear to be controlling for the same results. Why
are they controlling similar actions patterns(output)?-
Yes. I think this is what PCT sociology is about. Groups of people
obviously control for the same perceptions (and often produce the same
action patterns when producing those perceptions – when there are no significant
disturbances to the higher level perception). Religious worship and football
games are good examples of this. I think that in these cases we are
seeing many individuals controlling for the same (or very similar) system
level perceptions; indeed, the fact that these perceptions are controlled
collectively is probably the reason they are controlled individually: because
it lets each person perceive themselves as part of a group.

If
it is Ok in a month or so I’ll post a again on another theory in Environmental
psychology which I’d like to look at from PCT view.
I look forward to seeing it.

Thanks
greatly, Rohan Lulham

Thank you.

Best regards

Rick

···

–

Richard S. Marken, Ph.D.

Senior Behavioral Scientist

The RAND Corporation

PO Box 2138

1700 Main Street

Santa Monica, CA 90407-2138

Tel: 310-393-0411 x7971

Fax: 310-451-7018

E-mail: rmarken@rand.org