[From From Rick Marken (2000.12.20.0850)]
Me:
So I don't understand why my comments about teachers controlling
for order in class constitute a notable "oversimplification" or
an "issue" any more than do my comments (on the net and in a soon
to be published article) about the fact that there is evidence
(data) indicating that fielders control for vertical and lateral
>optical velocity.
Bruce Gregory (2000.1219.1935)--
When the fielder is controlling for catching the ball, all other
goals are presumably subordinate to that one.
HPCT doesn't presume that "all goals" are subordinate to any
particular controlled variable. According to HPCT, control of
nearly all perceptual variables (except those at the top of the
hierarchy) is subordinate to control of other variables. Control
of vertical and lateral optical velocity, for example, is
subordinate to the goal of catching the ball; control of catching
the ball is subordinate to the goal of playing baseball; control
of playing baseball is subordinate to the goal of making money,
and so on. There are many goals subordinate to catching a ball
(goals regarding visual variables, muscle tensions, body
orientation, rates of limb movement, etc); but there also many
goals to which control of catching the ball is subordinate
itself (like the goals of winning the game, being a Dodger,
making the cut, impressing the fans, etc).
We have no evidence that this [all other goals being subordinate]
is true in the case of the classroom teacher controlling for
order in the class.
We have no evidence that it's true in the case of the fielder
either. In fact, according to HPCT, it is not true in the case
of the fielder (controlling optical velocity) or the teacher
(controlling classroom order). That's the nice thing about HPCT;
the same model explains such superficially different behaviors
as catching a ball and maintaining order in a classroom; it's
all (hierarchical) control.
So I'm not convinced that I "oversimplified" in the case of the
teacher but not the fielder. Indeed, controlling for order in
class is probably at about the same level in the teacher's control
hierarchy (and, thus, has as many goals sub- and superordinate
to it) as is controlling for catching the ball in the fielder's
control hierarchy.
Bruce Nevin (2000.12.19 14:26 EST)--
Catching a ball is a pretty simple task (albeit requiring
practiced skill), with no feedback loops through the
environment including other control systems.
How do you know that catching a ball is simpler that keeping a
class orderly? Given the number of fielders compared to the
number of teachers it seems to me that it may actually be
the opposite; nearly anyone can apparently keep a class orderly
but only a few, highly skilled people can catch a fly to deep
center. That may be why fielders make the big bucks while
teachers make bupkis.
And what does the fact that a controlled variable involves
other control systems (students, say) have to do with "over-
simplification"? In the CROWD program each person controls
variables (like proximity) that involve other control systems.
Is it an oversimplification to say that the individuals in the
CROWD program control proximity? I don't think so. Whether
the state of a variable (such as order in class or optical
velocity) involves the behavior of another control system (as
does order in the classroom) or not (as does the optical
velocity of a baseball) is irrelevant to whether or not the
variable is controlled. If the fielder caught flies instead
of fly balls he would still do so by controlling optical
velocity (of the optical image of the fly, in this case,
instead of the ball). The control system nature of the fly
is irrelevant to the fact that the fielder is controlling
the optical velocity of the fly.
Body temperature is not relevant to catching a ball. Nor is
center of gravity directly relevant
Actually, center of gravity is. But are you now saying that my
"oversimplification" was my failure to mention all variables
that are relevant to (influenced by the actions used to control)
a controlled variable, such as order in the class? If so, then
it is an oversimplification to talk about _any_ controlled
variable without mentioning all the other variables that are
relevant to that variable (in the sense that control of those
other variables is influenced by control of the controlled
variable). So even in the case of controlling the position of
a cursor on a computer monitor it would be wrong to say that
the subject is controlling cursor position because the actions
used to control the cursor influence control of a large number
of other variables too; mouse position, hand position, eye
position, torque on the wrist, the state of the activity in
progress before taking time out to do the experiment, etc, etc).
But I think it's quite unnecessary to go to these ludicrous
extremes. There is absolutely nothing wrong with talking about
(and testing for) controlled variables one at a time. So we
can talk about the fact that a person is controlling cursor
position, optical velocity or class order without being required
to mention all the other controlled variables whose states are
likely to be disturbed by the actions used to control these
variables; the operation of the control hierarchy makes these
disturbances irrelevant..
Or perhaps you are saying that the teacher is only controlling
disruption and that is quite distinct from teaching...
Not at all. I am saying that _one_ of the variables a teacher
controls is classroom order. The teacher controls _many_ variables
(just like any organism), some as the means of controlling order
in the classroom; and classroom order is itself controlled as the
means of controlling other variables, like maintaining an
appropriate learning environment.
On both counts, a period of actual RTP classroom observation
would be perhaps a learning experience.
What would observation alone tell me about what the people in
the class are controlling? Remember, you can't tell what
people are doing (controlling) by just observing what they
are doing (overt behavior). You have to test for controlled
variables. I thought the PCT researchers involved in RTP
would report the results of such tests. But, so far, the
only RTP research I've seen has reported the discovery
that I am an enemy of RTP.
I think a period of actual modeling of hierarchical control
systems and testing for controlled variables would be a valuable
learning experience for you. I do hope you also take yourself
up on your suggestion that you do some more work on the coin
game to get a better idea of what it means to identify a
controlled variable. I also suggest that you do some more work
with the spreadsheet hierarchy; you can use this model to see
how it is possible to correctly identify a single controlled
variable even though it is controlled in the context of many
other controlled variables, nearly all of which are "relevant"
to control of that variable.
Best
Rick
ยทยทยท
---
Richard S. Marken Phone or Fax: 310 474-0313
MindReadings.com mailto: marken@mindreadings.com
www.mindreadings.com