[From Rick Marken (960729.2230)]
Hans Blom (960729) --
Would your answers have been different if you had evaluated the
merits of MCT independently?
I was not evaluating the merits of MCT. I was comparing MCT to PCT.
Me:
My goal is simply to show that the implications of MCT and PCT
regarding the nature of human nature are quite different.
Hans:
Good! Then they have different perspectives to contribute!
Yes. Then we test these "perspectives" against observation to see which one
does best. This is called doing "science"; happily maintaining contradictory,
untested "perspectives" is called liberal theology -- one of the great
oxymorons of the 20th century;-)
How does PCT explain that different people prefer different models?
Does the term "reference signal" have a familiar ring?
Me:
... The "internal models" in MCT are models of the laws of physics,
not perceptual representations of external variables.
Hans:
I'm sorry that you got this impression. What you say is, of course,
absurd. Instead, the "internal models" in MCT are models of HOW WE
PERCEIVE the laws of physics.
The internal models are models of the environmental laws that relate system
output to perceptual input. They are not models of "how" we perceive
anything. A perception is the system's represention of variables in its
environment. The only perception in your "model based" controller was
the variable you called x (as I recall); x was the result of multipying
the environmental variable (the scalar "plant" output) by k. So the
perceptual function in your controller was a multiplier of k applied to
the only variable in the system's environment..
MCT says that the perceptual function changes over time.
No. It says that the system's model of the environmental relationship
between its own output and its perceptual input changes over time. This
model is _not_ a perception because it is _not_ a representation of an
environmental variable; it is just a set of equations that are changed
based on the differnece between actual and predicted perceptual input.
MCT says that if the internal model is the same, the same perceptual aspects
of the environment will be controlled.
It is not obvious to me that this is the case. Perhaps you could show me
the proof? You can use equations only, if you like; no diagrams necessary;-)
What do you think a model is?
In your model based controller, the "model" is an equation that is a
hypothetical representation of the environmental law relating system output to
system input; the Kalman filter algorithm changed the coefficients of this
equation based on the difference between predicted and actual perception.
So the "model" in the controller was a model in the scientific sense of a
mathematical representation of the environmental laws responsible for what
we observe; it is not a "model" in the sense of being a representation
_of_ what is realy "out there" to be observed (as a model boat is a "model"
of the real ocean liner).
Tell me about the outcomes of your Tests.
People control the one and two dimensional position of lines; they control
frequency of tones, the area of rectangles, etc etc.
Have you discovered any goals that persons control for _at all times_, the
things that are _really_ important?
Sure. I discovered your goal (showing that MCT has some explanatory value);
you've discovered mine (showing that it doesn't;-)
How does PCT explain exploration? Or is there no such thing in your opinion?
Of course exploration exists. "Exploration" is a word that describes one
person's perception of the external apprearance of another person's
controlling.
The term "optimal behavior" must be confusing from a PCT perspective,
Well, "confusing" is the nice way of putting it;-)
In MCT models, "therapy" can be achieved by simply changing some parameter
values which have to do with how reliable the model thinks its knowledge
(other parameter values) is.
Could you give a concrete example of how this is done. It sounds like a
high tech metaphor to me.
But what invariably shows up is a period of "crisis", very bad control
behavior; the old model is being relinquished but the new one is not
well tuned yet...
Can a PCT model explain this phenomenon?
Sounds like reorganizaton to me; or conflict resolution. So, yes, PCT
can explain it.
A central heating thermostat is a control system as well. Can't it be
controlled?
It's actions can be controlled the way human actions can be controlled; by
disturbing a controlled variable (sensed temperature). But humans control
many variables, some of which are likely to be disturbed when you try to
control certain actions arbitrarily. If you persist in your efforts to
control the human, you will simply intensify a conflict, possibly leading
to violence.
A high [similarity] score [between MCT and PCT] would have told you that
MCT has nothing additional to offer... A low score makes it incorrect
for you, I presume.
High similarity would have meant that we had essentially the same theory with
two different names. A low score is good becauase it means we have two
different explanations of the same phenomenon (behavior) so we should be
able to come up with experiments that allow us to chose between the two
theories. I don't evaluate a theory in terms of its similarity to another
theory; I evaluate a theory in terms of the "similarity" of it's behavior
to that of a living control system.
The thing is: MCT can "explain" some of what we see happening around us,
which PCT cannot (yet).
Saying it doesn;t make it so, Hans. You have yet to show one (count them,
1.0) example behavior that can be explained (quantitavely and precisely)
by MCT but not by PCT. I REALLY would like to see an example of such
behavior, but thus far, all I get is a lot of hot bits;-)
Best
Rick