[From Rick Marken (940402.1200)]
Scott Clair (940401) --
I am pleased to announce that Robin Vallacher & Andrzej Nowak will
be speaking at this years University of Houston Psychology Symposium.
They make a point of emphasizing how to apply Dynamical
Systems concepts METHODOLOGICALLY in various areas of social psychology.
Scott, even if Tom can't make it (he may be laughed-out after he reads
my last post) it would sure be nice if you could post a synopsis of their
presentation. It would be nice to hear what some real, card-carrying
dynamical systems theorists REALLY have to say about the application of
Dynamical Systems concepts to human behavior -- especially social
Bill Leach (940402.01:10 EST) --
The only addition that I would recommend is that there is a second known
cause of conflict and that is when the disturbance exceeds the control
An "insuperable" disturbance (the kind you describe here) is NOT a cause
of conflict. This is a very common mistake -- it seems to come from the
equation of "conflict" with "error". When there is an insuperable disturbance,
the control system is not able to move the controlled variable to it's
reference level so there is a large, chronic error in the control loop. This
is not conflict -- it is failure to control. Conflict exists when the
distrubance to a controlled variable is the output of another control system
that is trying to keep the same (or a similar) controlled variable in a
different reference state. This kind of disturbance is not necessarily
insuperable (if the gain of the opposing control system is low) but it
is always proportionately opposed to your output; the harder you "push"
to oppose the disturbance, the harder the opposing control system's
output (the disturbance) pushes back.
Bill Leach (940402.10:15 EST) --
Oh Rick... you had me in stitches. Particularly with your completely
unexpected "Al Gore memorial information highway overpass".
Don't get excited; my politics are slightly to the left of "whoppee"
(that's from an old movie called "A thousand clowns"; not an allusion
to Whoopi Goldberg). Actually, I can't tell if I'm left or right any-
more (though I think I'm always right). I just think that all controllers
control best when all controllers can control well. How this is achieved?
The Zen of PCT -- of course.
I don't particularly want to clutter up and already busy net
No clutter. And this ain't busy at all. It's busy when I can't
find time to yell at the help ;-).
there are only three things that can cause
any change in output and those are a change in disturbance that results
in a change in a controlled perception
This is what Martin is arguing; this is what is wrong. The disturbance
does cause the output, but it's via the action of the loop, not via its
effect on the controlled perceptual varible. This is counter-intuitive;
but it's the way things work in a closed loop. Your statement above implies
that the distrubance caused change in perception is the cause of output;
that's what I am saying is wrong; the controlled perceptual variable is
NOT the cause of changes in the output of a control system. One reason
this is true is that the change in the output is ITSELF a cause of changes
in the controlled perceptual variable -- simultaenously with the disturbance.
So the output is part of the cause of itself. Chuck Tucker can send you
data showing evidence that there is nothing about temporal variations in
the value of the controlled perceptual variable that could possibly be
called the cause of change in the output of the control system.
"Choosing to perceive" anything is the act of setting a reference to
control a perception (in this case reseting the referenece for the
control of the perception of the "manager").
And here we get to what I think is the heart of the problem in pop
presentations of PCT. It turns on the distintion I made a while back
between perceptions changing within and across dimensions. It seems to
me that your client (and his attorney) still does not understand the
difference. Setting a reference selects a value WITHIN a dimension
of a perception; this is ordinary control. It is what is done when you
change the setting of the thermostat from 68 to 72; the system will
change its perception (within the dimension of temperature) keeping it
at a new value. I suggest that this be referred to as what it is --
changing the VALUE of a perceptual variable. In a human hierarchy
of control systems it is normal to regularly change the value at which
a perceptual variable is controlled in order to satisfy other
systems that requires that that perception be at various values;
so we vary the amount of upward force we perceive in order to lift a
particular suitcase; the force perception changes value in response to
changes in reference from the higher level systems that want to perceive a
lifted suitcase. A higher level example is changing the college you
attend in order to pursue a particular curriculum; the dimension of
perception (type of college) is still the same; you are changing the value at
which you hold this variable -- technical vs liberal arts -- in order to
achieve the higher level curriculum goal.
In the examples of "changing perceptions" I read about in pop PCT,
the recommended changes are not within a dimension (a good thing too
because there is no recommending to be done; within dimension changes
occur automatically as a means of achieving higher level goals). The pop
PCTers recommend across dimension perceptual change: perceive your wife in
a new way, your employees in a new way, etc. This is the kind of change you
DO need when control is chronically unsuccessful. This is like changing
the sensor on the thermostat from a temperature sensor to a humidity
sensor. The perceptual signal now "means" something different; I still
can control it at different levels but when I do so I am controlling
(if successful) different values of humidity rather than temperature.
This kind of across-dimension change is what may be needed in a chronically
dysfunctional control system. It is the kind of change that seems to
go on in various kinds of learning and therapy. It can result in improved
control -- but it also might not. It MIGHT result from doing exercises like
"quality time"; it might result from forcing oneself to do new things.
But this is pretty fundemental change when it happens. For this kind
of change, however, I would reserve the phrase "change perception" -- becuase
you are changing the dimensions of perception -- not the values at which
you control perpeptions on existing dimensions.
I see nothing "wrong" with trying to help people perceive the world in
a new way -- heck, that's what I'm trying to do here when I discuss
PCT. But when it comes to daily human problems, I think that more
often than not these problems can be solved by "going up a level" and
looking at the world from the perspective of the control systems that are
selecting the conflicting goals. This change in perspective could ALSO
be called "changing perceptions" -- but that would be confusing because
you are not changing any perceptions when you do this process; the
perceptions are always there -- you just move your awareness (an aspect
of your consciousness) to these perceptions. The "on" relationship between
my monitor and my processor/drive is always there to be perceived, I'm
just not usually aware of it. So I prefer to call the kind of "change in
perception" involved in "going up a level" a "change in awareness".
Some of the problems I have with pop PCT may be based on idiosyncratic
differences in how we talk about these things. It would really help
if we could use a common set of terms to refer to this stuff. It seems
to be that BCP went a long way to providing the basic glossary; I suggest
we try to use BCP terminology when we talk about control.
Happy post April fools day