Thank you Bill and Rick for your helpful (and patient!) replies and sorry for the
delay, I�ve been reading B:CP and thinking about it.
I�ve also been trying to integrate my understanding of PCT with the brand of
naturalistic psychology I�ve picked up as a psychology student over the last few
years. What I hope to do here is run my current attempts past you for your
feedback. After that, once you have a better idea of where I�m coming from, I�ll
reply to your replies of a few weeks back. Hopefully we�ll soon be agreeing (I
get the feeling we are making progress!) and I can get on with applying and
testing aspects of PCT to my Ph.D research.
At present, I find it helps if I divide PCT into two parts.
The first part is addressed most explicitly in chapter 4 of B:CP but is at the
same time evident throughout the entire book. I�ll call it psychological PCT for
clarity when I refer back to it later. Chapter 4 introduces an impressive means
of understanding complex human behavior with the concept of hierarchically nested
negative feedback loops. Previously disparate appearances (e.g., stimuli,
responses, consequences, etc) are integrated within a single descriptive (and,
dare I say, beautiful) framework. This is consistent with Lee�s (1988) view that
"the task of experimental science is to probe beyond appearances in order to
illuminate the patterns that underlie fields of particulars" (p. 32). In this
quotation, "beyond" and "underlie" denote general, abstract accounts (e.g., the
law of gravity) of specific, particular observations (e.g., this apple falling at
this velocity from this tree at this time). Further, because the abstract theory
is literally built out of the actual particulars (or more precisely, out of their
classes, properties, and relations), both exist in the one realm, or level of
discourse, which is physics in the example of gravity. They differ only in their
degree of abstraction.
As Bill and Rick taught me earlier in this thread, PCT is fundamentally concerned
with variables and the functions used to relate those variables. This links up
with the above picture of science if a) the variables are seen to correspond to
properties of particulars (e.g., position of spot, force of disturbance) or
relations between these properties (e.g., distance between position of spot and
position of target), and b) the functions (along with the organizational "glue"
holding them in place) are seen to correspond to the "underlying patterns." The
variables themselves are set up to take on the actual values which are
necessarily derived from particulars (i.e., actual occurrences). And what makes
PCT psychology as opposed to something else (e.g., physics) is that those
particulars are psychological. In earlier posts, I suggested that these
particulars are most literally conceptualized as the things people get done (and
relations between things done and other variables, such as disturbances). But we
needn�t agree on this (though I agree about the necessity of identifying the
controlled effect rather than just any old effect) for my present point should
stand either way. All that matters is that we agree that PCT is directly
concerned with a certain sub-class of observable events (it is also kind of
important that we agree that they are conceptually separable from organisms
accompanying them, so please bear with me on this and grant me this assumption
for the moment). These events inform the variables which the abstract functions
integrate.
In line with this interpretation, Bill closes chapter 4 with the suggestion that
"all these concepts apply to an objective analysis of behavior, and require no
model of the behaving system itself" (p. 55).
This leads into the second part of PCT (which I�ll call physiological PCT) I
wanted to mention, namely the "model of the behaving system" which gets
intermingled with psychological PCT throughout the book. In this second part,
physiology makes an entrance. The functions which were conceptualized above as
mathematical abstractions of psychological events are verbally inserted into the
central nervous systems of human beings. I�m uneasy about the assumptions which
seem to justify this step.
To try and communicate my difficulties with the step from psychological tp
physiological PCT, I�ll work with the justification that Bill gives early on in
B:CP (pp. 14-16) for presenting PCT as a model of brain organization. He draws
an analogy with the kinetic theory of heat. In this example, Bill tells the
story of how the kinetic theory arose not from merely observing the surface
appearances and generalizing from them, but instead from experimental evidence
for the existence of molecules moving within matter. This molecular information
provided the underlying explanation, or the deeper reality of the surface
appearances. So it will be with behavior as appearance and brain organization as
underlying reality, or so the story goes. Now I want to try and why I think this
analogy is misleading.
That explanation (the kinetic theory of heat) did not leave the bounds of
physics. In the end, physical phenomena were predicted from other (littler)
physical phenomena. This is fundamentally different to the PCT thesis, where
psychological phenomena are to be explained, it is proposed, in terms of
physiological phenomena. One nested slice of the natural world is to be
explained with recourse to another. Or, in other words, an abstraction (or
theory) about the subject matter of one science is forwarded as a realistic model
of the subject matter of another science. However, the empirical content of PCT
is psychological, and will remain psychological whether or not PCTers talk as if
it was physiological. For example, the instance of me picturing my grandmother
can never be comprehensively described, and therefore explained, almost solely in
terms of events occurring within my brain.
Previous psychologists have put this as follows. In 1941, Bentley wrote that
"�behaviors should be investigated where they are, that is, where observation of
them can be made�without limitation to where grammatical convenience guesses them
to be" (p. 11). Pronko (1984) said that "phenomena at each level are to be
explained in terms of factors observed at that particular level without resorting
to explanatory principles from lower levels of integration" (p. 173). J. R.
Kantor (e.g., 1983) long argued against reducing a complex psychological event to
a constituent (brain) of one of its constituents (body). For example, he said
"even when specific fields consist of interdependent and interlocked
physiological, general organic, mathematical, psychological and other components
none can be reduced to any other, nor can any type of component field be regarded
as the cause of any others" (p. 131).
So, to summarize, in earlier posts we talked about the basic units of
psychology. In this post, I talked about what a scientific account of these
units might consist of, and how this account might be interpreted. The two
topics go hand in hand, so hopefully you will now have a more complete picture of
what I�m on about and why I think these issues are important. To pull the two
threads together, I think that if you start out with the assumption that the
basic units are somehow to do with the organism more than anything else, when you
come to make abstractions (theories, models, etc.) you can end up in another
discipline (physiology). Whereas if you take your basic units as conceptually
separable from organisms at the outset, you can proceed to make abstractions that
remain separable from physiology and can have psychological theory with its own
place amongst the sciences.
Now for my replies to your last batch of comments:
Rick Marken (991013.1545)--
PCT (unlike the rest of psychology) begins with "control language" and treats
"things done"
(which I think of as referring to informally observed behavior) as a visible side
effect of the process of controlling perceptions.
Me:
What I�m suggesting is that PCT is more easily understood as including both a) a
mathematical analysis of the visible, and b) the postulation that the
mathematical model actually represents the mostly invisible physiological reality
of the brain. Rather than calling the visible stuff a "side effect" of the
hypothetical stuff, why not put more emphasis on PCTs organization of the visible
stuff first and the invisible stuff second. This would free up psychological
PCTers to pursue alternative hypotheses, and possibly to more swiftly identify
the laws that psychology as science desperately requires.
Rick:
PCT explains how control works. A crucial component of the
control process is a comparison of the state of the controlled
variable to a reference specification for that state. This
comparison must take place inside the controller.
Me:
I would be more cautious and say the comparison takes place in the model. From
here, you can choose to take the two additional steps of saying first that the
model is a model of real brain processes and second that this comparison actually
happens in the brain. I�m not saying these additional steps are wrong or
implausible, I�m just saying there are alternative interpretations. I think a
crucial factor here is how you choose to define "explain" and how you choose to
interpret "underlying reality." Perhaps you could tell me what you mean by these
terms.
Rick:
I have no idea how one would explain control in terms of the
"organization of things done". I don't even know what that
means. Are you suggesting that it's possible to build a model
of a controller (such as a driver controlling speed) that
ignores the organization of the system (person in this case)
doing the controlling?
Me:
No. That�s not what I�m saying. What I�m saying is that PCT doesn�t build a
model of a controller. Now, I better qualify that statement quickly before you
write me off as a nutcase. PCT models variables which take on values from
various measures of, among other things, the effects people have (or whatever you
want to call the things your data literally represent). Try this example. Lets
say 180 different people cooperate on a control task where the idea is to keep a
dot in the middle of a circle. Each person controls the position of the dot
along a straight line, but each line is rotated by 1 degree. So each individual
can only control the dot�s position along that one line. We could then model
successful control (as far as I can tell) just as we would if one person had
total control. Either way, we�d need just one model, because the model quite
literally deals with visible particulars that are conceptually separable from the
individual organisms. But PCT would seem to say that because 180 brains are
involved, we�d need 180 different models, and thus 180 x N reference signals,
comparators etc etc.
Rick:
Perhaps I could understand your point of view better if you described the
mechanism of your "organization of things done" model of how a driver controls
the speed of a car.
Me:
No problem. I�ll start with some research I�ve been involved with then apply it
to controlling the speed of a car. I�ve been involved with some experiments
where participants were required to earn a high score by pressing a mouse
button. Sometimes a press was always followed by a score increase, other times a
delay of 2.5 s was required. Anyway, we recorded as many effects (and their
properties) as we could think of (such as time-stamped button depressions,
releases, score increases, reprimand messages, and so on). We analyzed the data
with the hypothesis that participants were controlling for a high score increase
rate. On detailed analysis, all sorts of "hidden" details were found. For
example, when the 2.5 s delay conditions kicked in, participants were maintaining
a high score increase rate by very reliably waiting longer after fails. Here we
had a pattern of remarkable order, and had literally "uncovered" some of the
"reality" underlying the "surface appearances" of the data, which looked kind of
messy. If anyone would like, I can send you further examples. We actually used a
hierarchical model to map out the recorded events, in a manner very consistent
with PCT. It�s preliminary stuff, but I believe it demonstrates how psychology
can proceed from particulars to abstract accounts without recourse to physiology
or any other science.
So applying this kind of analysis to controlling the speed of a car, we could
proceed in the same way. Once data of appropriate complexity were in, we could
organize them hierarchically (perhaps with position of the speedometer at the top
below which components of lower levels of order are arranged) and then search for
hidden patterns, which might include systematic relations between properties of
classes of effects (such as changing the visual field to include the speedo and
adjusting the position of the accelerator pedal) classes of properties of
effects, classes of classes, and so on. This is how Quine characterized the
operation of science. It is hard to picture such a "things done" model ahead of
actually getting in there for a look, but hopefully this will give you an idea of
how a different interpretation of psychological events is at least conceivable.
Rick:
Yes. If the person kept moving relative to the speedometer
needle. But this was not the point of the example. I was
picturing a driver who is stationary in his new position
relative to the speedometer. Let's assume that the driver is
in a position such that the needle appears to be over the
"100" when it would appear to be over the "90" if the driver
were viewing the speedometer head on. When the driver keeps the
needle, as perceived, over the "100", he is actually keeping
the car speed at 90. So the driver is controlling a perception
of speed (indicated by the needle _as seen_), not the actual
speed.
Me:
I think we are agreeing here on something like the fact that an organism can only
control what it can see or perceive. We are just disagreeing about whether the
term "a perception" is the most useful way of saying this (I find that nouns that
have been converted from psychological verbs tend to end up making trouble). I
guess it comes down to a matter of personal preference. As Bill says in the
final sentence of chapter four "it is not necessary to talk of perceptions,
however convenient."
Rick:
The "facts" of the situation don't determine what aspect of the situation is
controlled.
Me:
Agreed. This is why there is such a huge literature in psychology about figuring
out what an organism is actually getting done (or attempting to get done), or
controlling, or bringing about etc despite appearances. And why PCT�s "test for
the controlled variable" is so handy. We can only make guesses and then pit
those guesses against the data.
Rick:
Each control system controls a different _perceptual_ representation of the
_same_ external state of affairs.
Me:
I read this as something like "a person can control multiple effects
simultaneously" with the addition of some superfluous and misleading traditional
pronouncements.
Rick:
Another (probably better) way to see the importance of _perception_ in control is
to read (and do) my "Control of Perception" demo
Me:
Nice demo! Thanks. However it made me wonder why you program such impressive
and complex examples of control then mask this complexity with a vocabulary
derived from sources foreign to the occurring events.
Me (In my last post, dated 991013.1200)
2) Before going any further, I am convinced that is is essential to clarify what
we mean (or what we specify) by our action language. That is, what units do we
point to with statements like "Bob driving at 100?" One interpretation is that
we point to Bob. This is the interpretation adopted by both behaviorism and
cognitivism, though differing in the part of Bob they zoom in on (either his
bodily responses or the contents of his head/mind). I think
this is fundamentally the wrong interpretation. Interpreted literally, action
language refers to the results of bodily efforts (Guthrie) or the things that
people get done (Lee, e.g., 1995).
Bill Powers (991014.0959 MDT)--
This is part of the story but not enough of it. I agree that we must speak mainly
of what behavior accomplishes rather than the means by which behavior is
produced, but this leaves open the question of how to define what is
accomplished.
Me:
Excellent point. I appreciate the fundamental importance of this question and
the difficulty of establishing an answer. I agree that taking on the doer�s
perspective is one way of minimizing the biases that might otherwise sneak in.
The "Test for the Controlled Variable" is a valuable tool.
Bill:
Speaking of changes, if you mean this literally, is appropriate only if
change is the intended effect. For example, I can move a coin on a tabletop
an inch to the left of its present position. Since the absolute ending
position relative to other objects is not specified, I can speak of
controlling a _change_ in position (as I see it), where the change is
independent of the location in space of the starting position.
However, what I control may be a position relative to something else: move
the coin until it covers a spot on the tablecloth. Since no starting
position is specified, the amount of change of position is unspecified. But
I can still achieve the end-relationship quite easily, using position
control rather than change control.
Me:
Interesting point. I hadn�t realized how ambiguous "changes" are. What I mean
by them, loosely speaking, is any change from one state of the world to another,
where the change a) has a person as a constituent, and b) that state change was a
"thing done," where "thing done" is contrasted with any old effect (i.e., what
you would call the controlled variable). It doesn�t matter where the coin was,
only that it is now covering the spot. That state of affairs has been brought
about. I would be better to emphasize the criteria that must be satisfied in
order for a particular thing to have been done (e.g., the coin is over the mark).
Me (in my last post):
the only reason I can see for focusing on this particular constituent to the
neglect of others is that there is some sort of residual attachment to the
organocentric interpretation of action language as described above.
Bill:
You seem to be using "organocentric" as a pejorative term, here. If I refer to
the operation of a car by describing how its engine, gears, and wheels work,
would you dismiss this as "mechanocentric?" Would a description of the thyroid
gland's operation in a loop involving thyroid-stimulating hormone and circulating
thyroxin be dismissed as "chemocentric?" In the pressing of a mouse-button, the
important components involve, as you say, the physical properties of the mouse
(according to our physics-models), the neural properties of the brain and sensors
(according to neurological models), and the physiological properties of the
muscles (according to the models of physiology). Is it organocentric to include
the brain and muscles in the description of mouse-pressing? Would it not be
physico-centric to
leave them out?
Me:
I used the term "organocentric" to voice my concerns that PCT not only includes
the brain and muscles as any comprehensive psychology should, but that it seems
to elevate them from constituents of psychological events to the whole thing
(excepting the section of the first-order loop outside of the skin). The
argument I have adapted from others is that the change in the state of the mouse
button from up to down, the event that is recorded at the instant that button
becomes "depressed," is properly a basic psychological unit, and that it can be
understood as a psychological unit. Of course, the other sciences have things to
say about the various constituents of that event, one of which is a nervous
system, but by definition have nothing to say about that event as a psychological
event. My interest is in psychological explanations of psychological events.
Me(In my last post):
5) Now I can state the problem bluntly. If PCT, like all psychology, properly
begins with action language, and interprets this action language in terms of
things done (as it seems empirically to do), then why does it couch its
explanation in terms of brain or person organization rather than in terms of the
organization of things done?
Bill:
But PCT forces us to acknowledge that what an
organism is "doing" is far from obvious, for all we can see from the
outside is how the organism physically affects the environment, both
directly and indirectly. And without control theory to organize our
investigations, we have no way to find out which of the effects _we_
consider important is also important to the organism.
Me:
Yes! This is the psychological PCT I�m getting excited about. It�s great stuff
and I am very pleased that you brought it into the world. I�m also honored that
you are prepared to discuss it with me.
Bill:
Because without postulating brain processes, there is no other way to
explain how all these environmental interactions produce the result we see.
A proper model must do justice to the physical effects, but that is not
enough to explain the observations.
Me:
We seem to be using "explain" differently here. I understand an explanation as
an abstract description of the events of interest that uncovers their hidden
patterns and makes predictions about what we would observe if we were to look in
certain places. How does your conception of "explain" differ from this?
Bill:
When we ask how Bob can possibly produce just the foot-pressure that is
needed, when there are so many other variable forces acting, mostly unknown
to Bob, we pose a problem that can ONLY be answered by the theory of
negative feedback control.
Me:
I think I agree.
Bill:
While there are many ways to embody a given negative feedback control system, we
can at least speak of a canonical system to which all other systems that behave
the same way are equivalent. We know that the state of the variable being
controlled has to be sensed somehow; we know that the reference state has to be
specified somehow; we know that the action must vary in a systematic relationship
to the difference between the sensed state and the specified reference state of
the controlled variable. Given those requirements, it is easy (if you understand
control systems) to construct a picture of a simple neuromuscular control loop in
which the required relationships are specified as functional blocks, if not
actual neural connections. And in some cases, we can establish the parallels
between the canonical system's interconnected functional blocks and the anatomy
of the nervous system and musculature.
Without a model like this, we can only accept the empirical findings and marvel
at them.
Me:
I believe that we can do a lot more than this. We need a model, in the sense of
an abstract theory, but so far as I can make out, there�s no requirement that we
put it in the head.
Me (in my last post):
I only disagree (or at least am curious to clarify) about where the referents of
those functions are supposed to reside (and I think this _is_ an important
issue).
Bill:
They reside in the physical form of neural signals in the brain of the
actor -- that is the basic PCT model. HPCT further proposes that they are
the outputs of higher control organizations in the brain which control more
abstract and aggregate variables. The highest level must of course be dealt
with in a different way.
Me:
This is purely what I called physiological PCT. I argued that psychological PCT
logically comes first (i.e., we need observables, and a model of those
observables, before we can put the model in the brain). So I can�t understand
why you say this is the "basic" PCT model. For me, the functions reside as
abstract, conceptual components in a model. They are verbal constructions, not
things. I�d like to think that I can apply PCT to my advantage in the world
while keeping the brain well out of the spotlight.
I apologize profusely for the ridiculous length of this post, and sincerely hope
that some of you made it this far and will respond with your comments.
Henceforth I�ll endeavor to be succinct.
Cheers,
Dan Palmer
Bentley, A. F. (1941). The human skin: Philosophy's last line of defence.
Philosophy of Science, 8, 1-19
Kantor, J. R. Psychological Comments and Queries (1984, Principia Press).
Lee, V. L. (1988). Beyond behaviorism. Hillsdale, NJ: Erlbaum.
Pronko, N. H. (1988). From AI to Zeitgeist: A philosophical guide for the
skeptical psychologist. New York: Greenwood Press.
