Vicarious conditioning

Hi, Warren –

I seem to have misplaced the post, but I’ve been thinking abouty your
student’s question about how PCT deals with “vicarious
conditioning.” Questions like this often come up, but only now have
I seen what the answer should be. I’m copying this to CSGnet, too,
because it’s a question of general interest.

The basic problem is that “vicarious conditioning” isn’t a
phenomenon that needs an explanation. It’s an explanation. PCT doesn’t
explain explanations (at least not in the terms the questioner is
expecting).I would say that the person who uses the term imagines that
something called “conditioning” is taking place inside the
subject organism, and that it is “vicarious” because the actual
process of conditioning, whatever that is, is being applied to a
different organism that the subject organism is observing.

There is no term “conditioning” in PCT, not is the term
“vicarious” found in PCT. PCT is a different theoretical
intepretation of the observations to which those terms apply. The only
way to see what PCT might say is to strip away all the theoretical and
imaginary aspects of what is being alluded to, and simply describe the
observable phenomena with as little theoretical bias as possible, and
without whatever imaginary data are being asssumed.

So if your student would like to start over and ask about how PCT might
apply to the observable phenomena (and list them), I would be glad to
make a guess, although without doing an experiment of my own to check out
the reported observations I would not like to make any formal
proposals…

Best,

Bill P.

[From Rick Marken (2008.02.16.1800)]

Bill Powers write to Warren Mansell:

I seem to have misplaced the post, but I've been thinking abouty your
student's question about how PCT deals with "vicarious conditioning."
Questions like this often come up, but only now have I seen what the answer
should be. I'm copying this to CSGnet, too, because it's a question of
general interest.

It's certainly of interest to me. As I think you know, I am hoping to
eventually write a paper describing how PCT explains many of the
"classical" phenomena of psychology. I suppose it's something I should
do, being one of the few "official" psychologists involved in PCT.
Getting your take on a PCT explanation of phenomena like "vicarious
conditioning" would certainly help me with my paper and with my
seminar on "Personal Control".

The basic problem is that "vicarious conditioning" isn't a phenomenon that
needs an explanation. It's an explanation.

Yes, that seems to be true in this case. I think the phenomenon is
something like "imitation";
"virtual conditioning", from what I can tell from a brief web search,
is operant learning -- learning to control food input by controlling
some aspect of the environment, like a lever -- that goes faster when
the learner can see an other animal getting food by taking the
appropriate action. I would like to know exactly what goes on in these
"vicarious conditioning" studies.

The only way to see what PCT might say is to strip
away all the theoretical and imaginary aspects of what is being alluded to,
and simply describe the observable phenomena with as little theoretical bias
as possible, and without whatever imaginary data are being assumed.

Right.

I look forward to hearing the a theoretical description of the
"vicarious conditioning" phenomenon and your take on it form a PCT
perspective.

Best

Rick

···

--
Richard S. Marken PhD
rsmarken@gmail.com

It seems to be a widely understood term.
http://preview.tinyurl.com/2dj33j

[From Rick Marken (2008.02.16.2320)]

It seems to be a widely understood term.
http://preview.tinyurl.com/2dj33j

Yes, I had done a Google search myself on it. What I couldn't find
was a nice, clear description of an study in which "vicarious
conditioning" was observed. Hopefully, whoever asked about how PCT
deals with vicarious conditioning will give what Bill asked for -- and
what I'd like to see: a nice, theory-free description of what goes on
in a "vicarious conditioning" study.

Best

Rick

···

On Sat, Feb 16, 2008 at 8:21 PM, Bruce Nevin (bnevin) <bnevin@cisco.com> wrote:
--
Richard S. Marken PhD
rsmarken@gmail.com

It seems to be a widely
understood term.


http://preview.tinyurl.com/2dj33j

[From Bill Powers (2008.02.17.0730 MST)]
Bruce Nevin (1:21 PM 2/16/2008 -0500):
Here’s the first paragraph from the referenced item:

···

============================================================================
Basic Components
In classical (or Pavlovian) conditioning, there are four important
elements. In a normal conditioning experiment, these include the pairing
of a **conditioned stimulus (CS)**followed by the unconditioned
stimulus (UCS)
. The UCS innately elicits an unconditioned response
(UCR)
, which is known as the unconditioned reflex. After
CS-UCS pairings in training, the CS comes to elicit the conditioned
response (CR)
, which is evidence for the conditioned reflex.
The CR becomes stronger with more CS-UCS trials until it reaches
asymptotic level.

=============================================================================

I can’t figure out from this what (among other terms)
“conditioning” is, what is meant by a
“stimulus”, what “elicit” means, or what model
is implied by “reflex.” One is just supposed to know, I guess.
Or maybe nobody knows, but no one wants to be the first one to ask.

This is something like calling every action a “response,” or
discussing differences of “affinity” of various cannonballs for
Earth, or proposing a theory about the effect of
“dephlogisticated” air in creating “calxes.” Each
specialized term refers to an underlying model or theory or explanation
of what is going on.

We discussed classical conditioning when Wayne Hershberger was still on
the net. The first thing we talked about was what phenomenon is behind
the term “unconditioned response.” Apparently an unconditioned
response is an observable action that reliably takes place when some
variable in the environment is independently changed (that is a
theory-free description of the phenomenon). For example, when a pin is
thrust into someone’s hand, the person will reliable move the hand in the
direction away from the pin, rather than moving it so the pin goes deeper
into the hand or twists sideways.

The old idea, which is behind the concept of an unconditioned response,
was that the pin stimulated nerve endings, generating trains of neural
impulses which traveled into the spinal cord, reaching synapses with
other nerves which relayed the impulses onward. Eventually the impulses
were routed back to motor nerves in the spinal cord which actived just
those muscles that moved the hand away from the pin.

In PCT, we offer the theory that such actions are part of a control
process organized to maintain some variable in a reference state, with
the independently-manipulated variable acting as a disturbance of the
controlled variable. The controlled variable is affected in one way by
the independently-varied environmental variable, and in the opposite
direction by the action. We would say that the pin’s entry created a
perceptual signal in some control system having a reference level of zero
for that signal. The resulting error signal was routed and amplified to
operate the muscles that moved the hand in the way that reduced the
perceptul signal to zero.

Clearly, these are two different theories being used to explain the same
observations. The PCT model introduces the reference signal into the
picture, which determines the level of the input signal that the actions
of the system will create and maintain. It introduces the idea that
actions are generated for the purpose of controlling perceptual signals.
It explains how a teenager can actually seek the sensation of a pin
entering, say, the tongue, and direct the hand to force the pin entirely
through it, to be followed by a stud or ring. To make this process work,
the reference level for the controlled variable must be temporarily
raised from zero to some rather high value.

According to PCT, there is no wired-in connection directly from the
sensory input to contractions in specific muscles. The same input can
lead to different actions, depending on the setting of the reference
signal. The same actions, or different actions, can appear without any
prior change in the inputs, if the reference signal is changed, or if
other systems that use the same muscle act. So the manipulated
environmental variable may or may not be followed by the action that the
idea of an unconditioned stimulus would lead us to expect.

Dick Robertson, some time ago, had some useful comments to make about the
eye-blink response to a puff of air. Perhaps he would repeat them for us
here. The variations in this and other similar phenomena are simply
ignored under the customs of psychology, as statistical fluctuations of
no importance. Most people have been left with the false impression that
every time there is a puff of air delivered to the cornea, there is a
blink of the eye. That’s not the case.

Notice that PCT has nothing to say about classical conditioning in
relation to actions, just as modern chemical theory has nothing to say
about the properties of phlogiston in relation to combustion. These
theories are basically different attempts to explain and predict
observations that are (or should be) described without reference to any
proposed theory. They are not about each other. To use theoretical terms
in a description of a phenomenon is to “beg the question” – to
assume the truth of a proposition in offering a proof of its truth (look
up “beg” in a dictionary). To call an action a response is to
assume that there must have been some stimulus causing it; but the
occurrance of an action is not proof that there was a stimulus.

To compare theories, we must reduce the descriptions of phenomena as
nearly as possible to theory-free terms, or at least terms that don’t
depend on adopting any particular theory being discussed.

Best,

Bill P.

No virus found in this incoming message.

Checked by AVG Free Edition.

Version: 7.5.516 / Virus Database: 269.20.6/1282 - Release Date:
2/15/2008 7:08 PM

No virus found in this incoming message.

Checked by AVG Free Edition.

Version: 7.5.516 / Virus Database: 269.20.6/1282 - Release Date:
2/15/2008 7:08 PM

[From Rick Marken (2008.02.17.1140)]

Bill Powers (2008.02.17.0730 MST)--

Beautiful. Thanks

Best

Rick

···

--
Richard S. Marken PhD
rsmarken@gmail.com

[From Erling Jorgensen (2008.02.18 1145 EST)]

Bill Powers (2008.02.17.0730 MST)

[building on an earlier post --]

Bill Powers (Sat, 16 Feb 2008 04:00:11 -0700)

[responding to Warren Mansell]

...I've been thinking abouty your student's question about
how PCT deals with "vicarious conditioning." Questions like
this often come up, but only now have I seen what the answer
should be. I'm copying this to CSGnet, too, because it's a
question of general interest.

I'm going to take a couple swings at this as well. (I figure
if I limit it to two swings, then by definition I can't
strike out! :-> )

To the extent there is a reliable phenomenon involved in
"vicarious conditioning," there may be something for PCT to
explain in terms of its constructs. Reliability always
suggests the possibility of control, since in a world of
disturbances, reliable reproducable phenomena usually require
control systems to bring them about.

Here's my take on so-called vicarious conditioning. First
the conditioning part. Bill's comments (in both his posts)
are important here, because it is necessary to look behind
theory-laden descriptions to the underlying phenomena.

For those using the conditioning map, I'm looking first at
the pairing of a so-called "Conditioned Stimulus" with an
"Unconditioned Stimulus," which to me speaks of some kind
of signalling. I think that resolves into learning to
perceive and/or control a series of perceptions, which ascends
up the PCT Hierarchy.

Let's say a tone is paired with the pin (mentioned in Bill's
second post), which when thrust into a hand has reliably led
to pulling the hand away. Pair the tone and the pin often
enough, and the hand tends to move when first the tone is
heard. I believe a "signal" is a relatively theory-free
description of a key part of this phenomenon.

So, where might PCT take this? (Capitals, below, serve as
allusions to Hierarchical PCT.)

Obviously, sounds of various kinds are heard all the time,
most of them via passive observation. A tone Configuration
may first come to be perceived as a distinct Event having
potential significance. 'Significance' is an indicator of
a reference state, leading to the action of continued
attention, which controls the Event by having it noticed
again.

Having singled out the tone as an Event, the tone-with-pin
Relationship would then be learned, soon followed by the
tone-then-pin Sequence. Those learned perceptions would then
make possible the controlled Program, if-tone-then-move-hand,
which the other theory wants to call a "Conditioned Response."

Brief aside about a fine-point. Above, I described this part
of the phenomenon as "the hand tends to move." 'Tends to'
may indicate that a statistically constrained tone-with-
pin-on-average association gets learned. This is a shift
from a single-instance Relationship, to correlation over time,
and so it certainly seems to be at a different logical level,
although I am not sure what Hierarchical descriptor to use.
In any event, that statistical frequency of occurrence may
then be reproduced in the frequency of carrying out the
if-tone-then-move-hand Program. To take the aside one step
further, this sounds like a contribution to the 'reference
function' for the Program, so I don't know if this says
anything about how to model a Principle. Other reflections
have led me to wonder about the on-average quality of
controlling for Principles, but perhaps I'm taking my third
strike already here. Anyway, back to the main thrust of
this post.

When it comes to the vicarious part of (the theory-laden
descriptor) "vicarious conditioning," it seems the underlying
phenomenon could be called learning-by-observation, or
learning-by-imitation. In this case, I believe a similar
ascending series of learned perceptions would hold true for
the (so-called) "Unconditioned Response" and "Unconditioned
Stimulus," now being observed in another, before the above
series involving the tone would come into play.

I would list this learned series as follws. First, the
observed Transition of (another's) hand-being-moved would be
noted as an Event of interest. 'Interest' here implies a
reference state, where control consists of looking for a
repeat occurrence. Next, the pin-with-hand-being-moved
Relationship and the pin-then-hand-being-moved Sequence would
be observed and learned. 'Learned', here, may be a theory-
laden term, but for PCT I believe it may signify that the
perception is now capable of being addressed as a reference
standard for a subsequent controlled occurrence of that
perception.

There may also be other concommitant observations interpreted
as pain-for-the-other, which combine with the above listed
perceptions involving the pin and the other's hand being
moved. But with those learned perceptions on the palette,
a comparable ascending series of perceptions involving the
tone could commence, as listed above, albeit being learned
at a distance via observation.

If it similarly results in a learned Program, if-tone-then-
move-hand, that learned perception could then be addressed
as the reference standard for one's own Program to carry
out, perhaps serving a Principle such as keep-pain-at-zero.

This is the kind of overlay I would make, between the map
of PCT, and the phenomenological territory of learning-by-
imitation-as-signalled-by-a particular-event.

On some maps, that's called "vicarious conditioning." But
I believe our PCT map captures the contours of the land much
more thoroughly. And as Bill suggests, such a map -- with
just a 'change of elevation' for a reference signal -- can
even explain seemingly contradictory situations such as a
teenager deliberately seeking a pin-through-tongue series of
perceptions, in the service of displaying body jewelry for
some System Concept perception of self.

That seems to me a pretty versatile map.

All the best,
Erling

[From Bruce Nevin (2008.02.18.2305 EST)]

Vicarious conditioning is given a simple definition at
http://www.webref.org/psychology/v/vicarious_conditioning.htm:
"Learning by observing the reactions of others to stimuli or by
listening to what they say."

The issue, then, has two parts.
1. What is conditioning?
2. How is this accomplished "vicariously" by observing what others are
doing or saying?

Notice the use of the word "learning" in that definition.

My working definition of conditioning: any learning that the observer
believes he has brought about by manipulating the environment of the
learner. The term is also used for "strengthening or weakening of an
operant response" that has already been established.

Classically, conditioning is brought about by conflicting with the
learner's control of some variable that is important to them, typically
food. The experimentor then relaxes resistance to the learner's control
of that variable (e.g. by making a bit of food available) whenever the
learner does something of the sort that the experimenter wants. The
learner begins to control some secondary variable as means of
controlling the primary variable, which he cannot otherwise control (due
to the experimenter's manipulations). The learner's means of controlling
that secondary variable are what the experimenter observes as "the
behaviors" that are "conditioned".

The essence of it is that an organism changes when it cannot control
some variable that it must control to survive, and keeps changing until
it can control that variable. The changes may be of the sort that we
call reorganization, or they may involve application some existing
control capability to resolve the difficulty, e.g. program or sequence
control of some other variable in a manner that includes "the behaviors"
that the experimenter is looking for.

We look upon coercive manipulation with distaste, and we note that there
are other, non-coercive ways in which learning takes place, but we're
not about to deny the phenomenon.

There are loose places in that description so that a hostile reader
could pick holes in it, but I am only trying to indicate a way to
understand what is really going on with conditioning. There is also a
lot of looseness in the range of things that are all considered to be
"conditioning". This may be essential as we move on to "vicarious."

"In the case of Vicarious Conditioning, behavior is the product of
observed consequences. In other words, we do what we do because of what
we see happen after we see someone else do it. For example, your
neighbor buys a certain type of lawn mower and you see that it works out
well for him, so you buy one just like it. Or you don't go to a certain
part of town because you heard something bad happened to someone else
when they went there."

Yes, it is certainly possible to learn from the example of others.
Here's an example from a practice test for psych students:

"Recently, a suggestion box appeared at work. The only person who filled
out one of the cards to make a suggestion was Doris. A couple of days
later, the boss came into the office, told Doris that he was pleased
that she had made a suggestion, and handed her a check for $100. The
next day, the suggestion box was full."

According to one textbook, however, what you observe happening to the
other guy has to be frightening, arousing, titillating, or the like. It
has to evoke some emotion of fear or desire:

"Vicarious conditioning requires that the observer be at least somewhat
emotionally aroused while observing the CS-UCS [conditioned
stimulus-unconditioned stimulus] pairing in the other person."


chapter_outline.html

Torture someone where others can see or hear, with the message "this
will happen to you unless you talk."

There are also cases of learning by example that don't fit the
definition of vicarious conditioning. You're driving through some flood
water. The car ahead of you suddenly dips the bumper under water on the
right side. You steer to the left to avoid the pothole hidden under the
water. You learned from his example. No conditioning involved. Nor would
this one-off "behavior" count as a conditioned response. But there's a
simple kind of learning by example.

Psychopaths are relatively unaffected by vicarious conditioning.
http://www.ncbi.nlm.nih.gov/pubmed/422732 and
http://preview.tinyurl.com/yskqe2
(Duh.)

The whole subject veers into "social learning theory."

Whether this is useful or not probably hinges on whether or not you
accept the idea that conditioning is manipulative influencing of
reorganization so as to "shape" the means by which an organism learns to
bring some coercively conflicted variable under control.

  /Bruce Nevin

[From Bruce Nevin
(2008.02.18.2305 EST)]

Vicarious conditioning is given a simple definition at


http://www.webref.org/psychology/v/vicarious_conditioning.htm
:

"Learning by observing the reactions of others to stimuli or by

listening to what they say."

The issue, then, has two parts.

  1. What is conditioning?

  2. How is this accomplished “vicariously” by observing what
    others are

doing or saying?
[From Bill Powers (2008.02.19.0828 MST)]

A useful and interesting post, as most of yours are, but it mistakes my
position. I don’t deny that people can learn to do things on the basis of
watching what other people do in various cirumstances, or by observing
what happens to the others when they behave. My objection is to the terms
“vicarious” and “conditioning,” which bring in other
theories and unspoken assumptions about the way behavior works. I’ll just
focus on the second term here. With regard to “vicarious” I’ll
just assert my theoretical position: people experience only their own
perceptions.

As you point out in a number of places, the assumption behind
“conditioning” is that behavior is changed by what is done to
an organism. “Conditioning” occurs when something is done to
“condition” behavior, where “to condition” is
understood as a transitive verb with behavior as its object. I’ve also
seen a usage (perhaps the original one) to the effect that conditioning
procedures make behavior “conditional on” the occurrance of
prior events, which is just S-R theory.

In PCT, changing the organization of behavior in relation to the
environment is the result of a capacity to reorganize, and the reason for
the changes is that variables important to the organism are not being
controlled sufficiently well. In other words, PCT proposes that the
changes that we can observe are brought about by processes in the
organism, not by external agencies. You can’t “condition” an
animal to produce a certain behavior by giving it food if it has a way of
providing for itself as much as it wants of the same food whenever it’s
hungry (you say this indirectly in your post). The conventional
interpretations of these phenomena hinge on the “establishing
conditions,” which are rarely taken into account or even noticed. In
PCT apparent effects of establishing conditions are among the phenomena
that need an explanation. Why does an organism have to be hungry before
it can be rewarded with food?

The difference from the conventional interpretation is similar to the
difference of PCT from stimulus-response theory. It looks as if the
actions of organisms are often caused by stimulation of their sensory
nerves by events in the environment, so conventionally one imagines some
sort of wired connection between the sensory nerves at the input and the
muscles at the output. But in the PCT model, there is a controlled
variable between the stimulus event and the behavioral response, with the
behavior being such as to oppose the effects of the stimulus event on the
controlled variable. A conventional scientist and a PCT scientist may
observe exactly the same externally-visible processes, but they imagine
very different processes (with very different implications about
causality) going on inside the organism.

You will notice that in the definition above, the phrase “…the
reactions of others to stimuli” appears as a description of what one
person observes about another. This phrase is not a description of an
observation. It is an assertion of a theory, the same theory that
underlies the idea of conditioning. It proposes an imaginary connection
between a purported cause and its effect. A PCTer, indulging in the same
sort of question-begging, would say that we observe “… the actions
of an organism that oppose disturbances caused by stimulus events.”
That phrase is not an observation, either – it asserts that inside the
organism there is a control system. We do not observe that control
systems exist inside an organism (not in the year 2008, or for a lot of
years to come). We observe the same processes that a conventional
psychologists observes from outside the organism. But we propose a
different explanation of how these processes are generated.

A pure observation simply reports what is actually observed and nothing
that is only imagined (as nearly as possible and relevant). What I have
been asking for is that descriptions of the sort with which you opened
your post be recast so that they do not assert any causal theory,
explictly or through connotation or common usage. I expect PCTers to do
this just as much as I wish conventional scientists would do it. It is
this very strong preference that has pushed me into using demonstrations
as a way of communicating control theory. A demonstration shows you what
happens without offering any interpretation, explanation, or theory about
it. Any theory has to account for what actually occurs during a
demonstration. Believing in one theory or another one has no effect on
what is observed during the demonstration – although beliefs have a lot
of effect on what one imagines is going on.

I’m leaving the rest of Bruce Nevin’s post appended for Warren Mansell’s
sake (he broached this subject and started this thread).

Best,

Bill P.

[continued from quote at the top]

···

Notice the
use of the word “learning” in that definition.

My working definition of conditioning: any learning that the
observer

believes he has brought about by manipulating the environment of the

learner. The term is also used for "strengthening or weakening of
an

operant response" that has already been established.

Classically, conditioning is brought about by conflicting with the

learner’s control of some variable that is important to them,
typically

food. The experimentor then relaxes resistance to the learner’s
control

of that variable (e.g. by making a bit of food available) whenever
the

learner does something of the sort that the experimenter wants. The

learner begins to control some secondary variable as means of

controlling the primary variable, which he cannot otherwise control
(due

to the experimenter’s manipulations). The learner’s means of
controlling

that secondary variable are what the experimenter observes as
"the

behaviors" that are “conditioned”.

The essence of it is that an organism changes when it cannot control

some variable that it must control to survive, and keeps changing
until

it can control that variable. The changes may be of the sort that we

call reorganization, or they may involve application some existing

control capability to resolve the difficulty, e.g. program or
sequence

control of some other variable in a manner that includes “the
behaviors”

that the experimenter is looking for.

We look upon coercive manipulation with distaste, and we note that
there

are other, non-coercive ways in which learning takes place, but
we’re

not about to deny the phenomenon.

There are loose places in that description so that a hostile reader

could pick holes in it, but I am only trying to indicate a way to

understand what is really going on with conditioning. There is also
a

lot of looseness in the range of things that are all considered to
be

“conditioning”. This may be essential as we move on to
“vicarious.”

"In the case of Vicarious Conditioning, behavior is the product
of

observed consequences. In other words, we do what we do because of
what

we see happen after we see someone else do it. For example, your

neighbor buys a certain type of lawn mower and you see that it works
out

well for him, so you buy one just like it. Or you don’t go to a
certain

part of town because you heard something bad happened to someone
else

when they went there."


http://www.barkingdogs.net/bmod.shtml

Yes, it is certainly possible to learn from the example of others.

Here’s an example from a practice test for psych students:

"Recently, a suggestion box appeared at work. The only person who
filled

out one of the cards to make a suggestion was Doris. A couple of
days

later, the boss came into the office, told Doris that he was pleased

that she had made a suggestion, and handed her a check for $100. The

next day, the suggestion box was full."


http://preview.tinyurl.com/2d4tfs

According to one textbook, however, what you observe happening to
the

other guy has to be frightening, arousing, titillating, or the like.
It

has to evoke some emotion of fear or desire:

"Vicarious conditioning requires that the observer be at least
somewhat

emotionally aroused while observing the CS-UCS [conditioned

stimulus-unconditioned stimulus] pairing in the other person."


http://highered.mcgraw-hill.com/sites/0072490462/student_view0/chapter3/

chapter_outline.html

Torture someone where others can see or hear, with the message
"this

will happen to you unless you talk."

There are also cases of learning by example that don’t fit the

definition of vicarious conditioning. You’re driving through some
flood

water. The car ahead of you suddenly dips the bumper under water on
the

right side. You steer to the left to avoid the pothole hidden under
the

water. You learned from his example. No conditioning involved. Nor
would

this one-off “behavior” count as a conditioned response. But
there’s a

simple kind of learning by example.

Psychopaths are relatively unaffected by vicarious conditioning.


http://www.ncbi.nlm.nih.gov/pubmed/422732
and


http://preview.tinyurl.com/yskqe2

(Duh.)

The whole subject veers into “social learning theory.”


http://en.wikipedia.org/wiki/Social_learning_theory

Whether this is useful or not probably hinges on whether or not you

accept the idea that conditioning is manipulative influencing of

reorganization so as to “shape” the means by which an organism
learns to

bring some coercively conflicted variable under control.

    /Bruce

Nevin

No virus found in this incoming message.

Checked by AVG Free Edition.

Version: 7.5.516 / Virus Database: 269.20.7/1285 - Release Date:
2/18/2008 5:50 AM

No virus found in this incoming message.

Checked by AVG Free Edition.

Version: 7.5.516 / Virus Database: 269.20.7/1285 - Release Date:
2/18/2008 5:50 AM

[From Rick Marken (2008.02.19.1040)]

Bill Powers (2008.02.19.0828 MST)--

A pure observation simply reports what is actually observed and nothing
that is only imagined (as nearly as possible and relevant). What I have been
asking for is that descriptions of the sort with which you opened your post
be recast so that they do not assert any causal theory, explictly or through
connotation or common usage. I expect PCTers to do this just as much as I
wish conventional scientists would do it. It is this very strong preference
that has pushed me into using demonstrations as a way of communicating
control theory. A demonstration shows you what happens without offering any
interpretation, explanation, or theory about it. Any theory has to account
for what actually occurs during a demonstration. Believing in one theory or
another one has no effect on what is observed during the demonstration --
although beliefs have a lot of effect on what one imagines is going on.

I'm leaving the rest of Bruce Nevin's post appended for Warren Mansell's
sake (he broached this subject and started this thread).

I would sure like to see a description of the observations that have
been made and interpreted as resulting from "vicarious conditioning".
Wouldn't all this be easier if Warren and his student joined CSGNet?
Perhaps you could ask them to do so. It's really easy. Just have them
go to

http://listserv.uiuc.edu/wa.cgi?SUBED1=csgnet&A=1

and follow the instructions.

I think everyone who is truly interested in PCT should be on CSGNet.
It is the _only_ place I know of where PCT is discussed intelligently.

Best

Rick

···

--
Richard S. Marken PhD
rsmarken@gmail.com

[From Bruce Nevin (2008.02.19.1503 EST)]

I had no intention of imputing any position to you, just trying to identify what’s going on when people talk about conditioning and vicarious conditioning.

I don’t deny that people can learn to do things on the basis of watching

what other people do in various cirumstances, or by observing what

happens to the others when they behave.

What are they perceiving and controlling when they do this watching or observing, and how does it bring about this learning? These are the interesting unanswered questions. For “conditioning” we can point to reorganization because of coercive interference with their control of some vitally important variable. Can imagining the inability to control under duress also bring about reorganization? Is there control for avoiding such coercive situations?

With regard to “vicarious” I’ll just assert my theoretical position: people

experience only their own perceptions.

Yes, of course. They perceive what is going on with another person. These are their own perceptions. Perhaps they then imagine this going on with themselves. These imagined perceptions are their own perceptions. Hidden under these questions is the interesting difference between a normally empathetic person and person with diminished capacity for empathy (psychopath, sociopath). What is empathy?

I agree with you of course about the inspecificity of “conditioning” as a term and as a concept. They are being as specific as they know how. It’s up to us to be more specific.

/Bruce
···

From: Control Systems Group Network (CSGnet) [mailto:CSGNET@LISTSERV.UIUC.EDU] On Behalf Of Bill Powers
Sent: Tuesday, February 19, 2008 11:42 AM
To:
CSGNET@LISTSERV.UIUC.EDU
Subject: Re: Vicarious conditioning

[From Bruce Nevin (2008.02.18.2305 EST)]

Vicarious conditioning is given a simple definition at

http://www.webref.org/psychology/v/vicarious_conditioning.htm:
“Learning by observing the reactions of others to stimuli or by
listening to what they say.”

The issue, then, has two parts.
1. What is conditioning?
2. How is this accomplished “vicariously” by observing what others are
doing or saying?
[From Bill Powers (2008.02.19.0828 MST)]

A useful and interesting post, as most of yours are, but it mistakes my position. I don’t deny that people can learn to do things on the basis of watching what other people do in various cirumstances, or by observing what happens to the others when they behave. My objection is to the terms “vicarious” and “conditioning,” which bring in other theories and unspoken assumptions about the way behavior works. I’ll just focus on the second term here. With regard to “vicarious” I’ll just assert my theoretical position: people experience only their own perceptions.

As you point out in a number of places, the assumption behind “conditioning” is that behavior is changed by what is done to an organism. “Conditioning” occurs when something is done to “condition” behavior, where “to condition” is understood as a transitive verb with behavior as its object. I’ve also seen a usage (perhaps the original one) to the effect that conditioning procedures make behavior “conditional on” the occurrance of prior events, which is just S-R theory.

In PCT, changing the organization of behavior in relation to the environment is the result of a capacity to reorganize, and the reason for the changes is that variables important to the organism are not being controlled sufficiently well. In other words, PCT proposes that the changes that we can observe are brought about by processes in the organism, not by external agencies. You can’t “condition” an animal to produce a certain behavior by giving it food if it has a way of providing for itself as much as it wants of the same food whenever it’s hungry (you say this indirectly in your post). The conventional interpretations of these phenomena hinge on the “establishing conditions,” which are rarely taken into account or even noticed. In PCT apparent effects of establishing conditions are among the phenomena that need an explanation. Why does an organism have to be hungry before it can be rewarded with food?

The difference from the conventional interpretation is similar to the difference of PCT from stimulus-response theory. It looks as if the actions of organisms are often caused by stimulation of their sensory nerves by events in the environment, so conventionally one imagines some sort of wired connection between the sensory nerves at the input and the muscles at the output. But in the PCT model, there is a controlled variable between the stimulus event and the behavioral response, with the behavior being such as to oppose the effects of the stimulus event on the controlled variable. A conventional scientist and a PCT scientist may observe exactly the same externally-visible processes, but they imagine very different processes (with very different implications about causality) going on inside the organism.

You will notice that in the definition above, the phrase “…the reactions of others to stimuli” appears as a description of what one person observes about another. This phrase is not a description of an observation. It is an assertion of a theory, the same theory that underlies the idea of conditioning. It proposes an imaginary connection between a purported cause and its effect. A PCTer, indulging in the same sort of question-begging, would say that we observe “… the actions of an organism that oppose disturbances caused by stimulus events.” That phrase is not an observation, either – it asserts that inside the organism there is a control system. We do not observe that control systems exist inside an organism (not in the year 2008, or for a lot of years to come). We observe the same processes that a conventional psychologists observes from outside the organism. But we propose a different explanation of how these processes are generated.

A pure observation simply reports what is actually observed and nothing that is only imagined (as nearly as possible and relevant). What I have been asking for is that descriptions of the sort with which you opened your post be recast so that they do not assert any causal theory, explictly or through connotation or common usage. I expect PCTers to do this just as much as I wish conventional scientists would do it. It is this very strong preference that has pushed me into using demonstrations as a way of communicating control theory. A demonstration shows you what happens without offering any interpretation, explanation, or theory about it. Any theory has to account for what actually occurs during a demonstration. Believing in one theory or another one has no effect on what is observed during the demonstration – although beliefs have a lot of effect on what one imagines is going on.

I’m leaving the rest of Bruce Nevin’s post appended for Warren Mansell’s sake (he broached this subject and started this thread).

Best,

Bill P.

[continued from quote at the top]

Notice the  use of the word "learning" in that definition.

My working definition of conditioning: any learning that the observer
believes he has brought about by manipulating the environment of the
learner. The term is also used for "strengthening or weakening of an
operant response" that has already been established.

Classically, conditioning is brought about by conflicting with the
learner's control of some variable that is important to them, typically
food. The experimentor then relaxes resistance to the learner's control
of that variable (e.g. by making a bit of food available) whenever the
learner does something of the sort that the experimenter wants. The
learner begins to control some secondary variable as means of
controlling the primary variable, which he cannot otherwise control (due
to the experimenter's manipulations). The learner's means of controlling
that secondary variable are what the experimenter observes as "the
behaviors" that are "conditioned".

The essence of it is that an organism changes when it cannot control
some variable that it must control to survive, and keeps changing until
it can control that variable. The changes may be of the sort that we
call reorganization, or they may involve application some existing
control capability to resolve the difficulty, e.g. program or sequence
control of some other variable in a manner that includes "the behaviors"

that the experimenter is looking for.

We look upon coercive manipulation with distaste, and we note that there
are other, non-coercive ways in which learning takes place, but we're
not about to deny the phenomenon.

There are loose places in that description so that a hostile reader
could pick holes in it, but I am only trying to indicate a way to
understand what is really going on with conditioning. There is also a
lot of looseness in the range of things that are all considered to be
"conditioning". This may be essential as we move on to "vicarious."


"In the case of Vicarious Conditioning, behavior is the product of
observed consequences. In other words, we do what we do because of what
we see happen after we see someone else do it. For example, your
neighbor buys a certain type of lawn mower and you see that it works out
well for him, so you buy one just like it. Or you don't go to a certain
part of town because you heard something bad happened to someone else

when they went there."
http://www.barkingdogs.net/bmod.shtml

Yes, it is certainly possible to learn from the example of others.
Here's an example from a practice test for psych students:

"Recently, a suggestion box appeared at work. The only person who filled
out one of the cards to make a suggestion was Doris. A couple of days
later, the boss came into the office, told Doris that he was pleased
that she had made a suggestion, and handed her a check for $100. The
next day, the suggestion box was full."

http://preview.tinyurl.com/2d4tfs

According to one textbook, however, what you observe happening to the
other guy has to be frightening, arousing, titillating, or the like. It
has to evoke some emotion of fear or desire:

"Vicarious conditioning requires that the observer be at least somewhat
emotionally aroused while observing the CS-UCS [conditioned
stimulus-unconditioned stimulus] pairing in the other person."

http://highered.mcgraw-hill.com/sites/0072490462/student_view0/chapter3/

chapter_outline.html

Torture someone where others can see or hear, with the message "this
will happen to you unless you talk." 

There are also cases of learning by example that don't fit the
definition of vicarious conditioning. You're driving through some flood
water. The car ahead of you suddenly dips the bumper under water on the
right side. You steer to the left to avoid the pothole hidden under the
water. You learned from his example. No conditioning involved. Nor would
this one-off "behavior" count as a conditioned response. But there's a

simple kind of learning by example.

Psychopaths are relatively unaffected by vicarious conditioning.

http://www.ncbi.nlm.nih.gov/pubmed/422732 and
http://preview.tinyurl.com/yskqe2
(Duh.)

The whole subject veers into "social learning theory."

http://en.wikipedia.org/wiki/Social_learning_theory

Whether this is useful or not probably hinges on whether or not you
accept the idea that conditioning is manipulative influencing of
reorganization so as to "shape" the means by which an organism learns to
bring some coercively conflicted variable under control.

        /Bruce Nevin


No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.5.516 / Virus Database: 269.20.7/1285 - Release Date: 2/18/2008 5:50 AM


No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.5.516 / Virus Database: 269.20.7/1285 - Release Date: 2/18/2008 5:50 AM

[From Rick Marken (2008.02.20.0815)]

Bruce Nevin (2008.02.19.1503 EST)

What are they perceiving and controlling when they do this watching or
observing, and how does it bring about this learning? These are the
interesting unanswered questions.

I think the first questions to be answered (possibly not quite as
interesting as yours) are ones pertaining to what exactly happened in
the "vicarious conditioning" experiments. What kind of animal was
used? Were they starved to a particular body weight level? What did
they have to do to get food (if that's what they were supposed to do)?
What previous experience had they had with the situation? Where was
the "model" relative to the animal who acted to get the food (if
that's what was involved)? What were the data that suggested that
that "model" was relevant to the results? etc etc.

I agree with you of course about the inspecificity of "conditioning" as a
term and as a concept. They are being as specific as they know how. It's up
to us to be more specific.

I don't much care about their terminology or how specific it is. What
I want to know is what is the observation to be explained. If that is
what you mean by being more specific about the meaning of
"conditioning" then I'm with you. I would like to know what exactly
happens in an experiment that people call an example of "vicarious
conditioning". I hope Mansell and/or his student will provide such a
description soon. Until then, all this talk about how PCT might
explain "vicarious conditioning" seems like guessing about the number
of angels that can dance on the head of a pin (my guess is << 1, by
the way;-)).

Best

Rick

···

--
Richard S. Marken PhD
rsmarken@gmail.com

[From Bill Powers (2008.02.20.1002 MST)]

Bruce
Nevin (2008.02.19.1503 EST) –

I had no
intention of imputing any position to you, just trying to identify what’s
going on when people talk about conditioning and vicarious
conditioning.

I don’t deny that people can learn to do things on the basis of
watching

what other people do in various cirumstances, or by observing what

happens to the others when they behave.

What are
they perceiving and controlling when they do this watching or observing,
and how does it bring about this learning?

Those are good questions, but I still want to know what the experimenter
is seeing (as opposed to imagining) when he or she reports that vicarious
conditioning is going on. Just one example would be helpful. The
questions you ask are about the theory we would use to explain the
observations, or the data we would assume as a premise in making
hypotheses. The experimenter can’t (just by looking) see what aspect of
the scene is being perceived by the person doing the watching. At best,
the experimenter sees the watcher looking in the direction of some other
person doing something, and then behaving in a way that the experimenter
sees as similar in some regard. The experimenter can record that, and
record what the experimenter notices about what the observed person is
doing. There could be some questioning of the observing person about what
is being noticed. There could be tests to see what the observing person
considers important to control. Reports about all these things would be
helpful in describing what the experimenter is observing and trying to
explain.

Is what I’m asking about something that everyone else but me knows about
and considers too obvious to discuss? It seems extraordinarily hard to
get anyone with knowledge in this area simply to describe what an
experimenter watching vicarious conditioning taking place would actually
see happening. It seems hard to get psychologists just to report what
happened without putting some kind of theoretical spin on it, like
calling every action a “response.”

I agree with you
of course about the inspecificity of “conditioning” as a term
and as a concept. They are being as specific as they know how. It’s up to
us to be more specific.

About what? All I have right now are some words referring to a phenomenon
and a few very sketchy generalizations about what goes on in
“vicarious conditioning.” I would like to get more specific,
but never having seen an experiment with vicarious conditioning, or an
adequate description of one, I don’t know what it is I’m supposed to be
more specific about. I haven’t heard or read enough even to decide for
myself whether this phenomenon ever actually occurs. I’m hoping that
someone more familiar with the literature can sum up the things I’m
asking about – partly because I’m overloaded with programming and other
things and can’t do it myself right now.

Maybe this will just have to go into my Unfinished Business
file.

Best,

Bill P.

[From Bruce Nevin (2008.02.20.1435)]

(BTW, do we still need this CSG-unique manually typed datestamp? Years ago someone craved it for archiving. Every email message comes with a datestamp and the name of the sender in the email header.)

It sounds to me like the people who talk knowledgeably about conditioning infer causation from association, and they attribute “association” to statistically meager correlations. By that I refer to Dick’s conversation with the Experimental guy down the hall, who was unperturbed at throwing out failed cases of conditioning. In other words, “Conditioned association is a real phenomenon, it just doesn’t always happen.” The causation is assumed, and any contrary data is noise to be ignored. And I think the reason behind that is that without conditioning they don’t know any other explanatory principle for learning.

The example from the psych exam seemed to be straight from a rather old fashioned management “how to motivate your people” playbook.

  1. Put out suggestion box.

  2. Only one person puts a suggestion in it.

  3. Publicly reward that person with a check for $100.

  4. Next day the box is full of suggestions.

Omitted:

  1. Give each suggestion-maker $100? How long do you keep this up?

    Or give $100 just to those whose suggestions you like? Is that “shaping”?

    Or only give the $100 to the first person and watch the suggestions drop off? (That’s all we need, thanks.)

[The advantage of money is that it abstracts from the particularities of barter to a fungible medium of exchange. The difficulty with money is that this very abstraction disconnects the “reward” from the real motivation (the CV and its reference value). But that’s another topic entirely.]

I don’t think you’re going to get anywhere expecting examples from the literature of conditioning to include more detail. The explanatory principle “conditioning” does not require more detail. I think it’s necessary to conduct new experiments that qualify as conditioning, but in which more information is gathered. Dick’s excerpt from Intro to Modern Psych **points the way.

Email datestamp - Mon 2/18/2008 8:33 PM

CSG stamp - from Dick Robertson,2008.02.18.1920CST

In that excerpt, the point of the first part of the discussion is that labelling things differently has prevented seeing what those things have in common.

“Association” between perception and “reflex” = Pavlov’s “conditioning”

“Association” between two perceptions = “association of ideas”

In the first case, the reflex is control of a perceptual variable (wetness of eyeball, Rick surmised) which is usually not attended to consciously. The experiment by Dick’s students suggests that “conditioning” occurs when this “reflexive” control remains out of awareness, but that “conditioning” does not occur when the “reflex” is brought to awareness. Dick’s text (the quoted part, anyway) does not propose any mechanism for the association when it does occur. Presumably the same mechanism, whatever it is, applies to the second case, “association of ideas”.

Dick’s example experiment involves the “blink reflex” when a puff of air is directed at the eyeball. The perception of the tone or click comes “just before” the puff of air. The principle seems to be like the traffic truism, “Red means stop, green means go, yellow means go like hell 'cause red’s a-coming.”

Unanswered question: do they blink sooner, that is, “just before” they would have without the warning tone?

Unanswered question: Is the CV “dryness of the eyeball”? If so, then isn’t the puff of air also associated with the real CV? Or is the CV any physical disturbance to the eyelashes? It takes quite a while holding my eyes open for a sensation to arise that I interpret as dryness of the eyeball, about 3 seconds in my current conditions of temperature and humidity. It is possible that the puff of air speeds evaporation from the surface of the eye and the blink is to relubricate the eyeball. On the other hand, if I am walking through bushes and a leaf brushes against my eyelid I blink. If a strand of spiderweb drifts across my eyelid I blink. Are eyelashes like cats’ whiskers, sensors warning of approaching objects before they reach the cornea?

In short, I don’t think we can expect associationists to do PCT research, we have to do that ourselves. Instead of demanding that an associationist experiment qualify as a PCT experiment, we can do a PCT experiment that associationists perceive as an associationist experiment with a bunch of extraneous data that they don’t use.

My interest, as you know, is in how the simpler phenomena are extended socially, hence my interest in “vicarious conditioning.” I have some trouble trying to imagine how an observer of a subject in the eyeblink experiment would be emotionally so aroused as to themselves associate the tone with the blink. However, observers at a boxing match sometimes flinch in sympathy, and spectators at other sports put their own “body english” into the efforts of players on the field. Somewhere I have a photo of myself in high school taken by the photographer for the local small town paper. I was watching the high jumpers. He was amused that I moved my knee and shoulder in sympathetic synchrony with each jumper’s effort.

On the social side, a key ingredient is perceiving oneself from the imagined point of view of another person. We have perceptions of our appearance and actions; we imagine how our appearance and actions are perceived by another person. We also perceive the other’s facial expressions, gestures, and other actions (and lack of expected actions) as indicating how they perceive us. (“You look doubtful. What part of that was unclear?”)

It is not a far step thence to extend this to environmental perceptions (perceptions other than of yourself), and to imagine how one’s own perceptions of the environment appear from another person’s point of view. This socalled “theory of mind” (ToM) develops normally at around age 4 in humans. http://en.wikipedia.org/wiki/Theory_of_mind I reported some time ago on some experimental work that my daughter helped with. There is a certain consensus that on the Aspergers-Autism spectrum development of ToM is incomplete or lacking. Here’s one person’s somewhat poignant recollection of growing up autistic:

http://djdialogue143.blogspot.com/2007/04/just-silly-one-about-theory-of-mind.html

(NT = “neurologically typical, autism related”).

People considered psychopaths or sociopaths, on the other hand, are typically manipulative, and you simply can’t be manipulative without ToM. (For example, you can’t exercise counter-control without perceiving that the other is controlling some variable that you can influence.) What people call sociopathy and psychopathy may be a problem with input functions. For example, people with amygdala deficit associated with psychopathic behavior seem not to recognize fear in the eyes of others, but simply asking them to focus on other people’s eyes helps correct this, at least temporarily

http://bjp.rcpsych.org/cgi/content/full/189/3/280

In addition, their perception of degrees or intensity of emotional expression may be miscalibrated

http://cjb.sagepub.com/cgi/content/abstract/34/4/531

Cephalopods probably are doing something much more rudimentary, though no one yet knows how they do these amazing matches of self to environment.

http://www.nytimes.com/2008/02/19/science/19camo.html?scp=1&sq=cuttlefish&st=nyt

/Bruce

OK, I’ll try omitting the time and date stamp.

First, thanks Bruce, for the references. I enjoyed the autist’s blog and loved the article about the cephalopods. Realist-artists must envy their abilities.

Your expansion of what we could have done with the eyeblink experiment was enlightening. I confess I never thought of the greater number of possible interpretations we could have tried to explore. I hope somebody will. Are you listening, Rick?

Science, even when we have a good grounding in theory like PCT, just lurches ahead a step at a time mostly, doesn’t it?

Best,

Dick R.

[From Rick Marken (2008.02.21.1900)]

I like the time stamp;-)

Dick Robertson said (sans time stamp):

Your expansion of what we could have done with the eyeblink experiment was enlightening. I confess I never thought of the greater number of possible interpretations we could have tried to explore. I hope somebody will. Are you listening, Rick?

The main thing I got from reading about your eye blink study was that there was not always a blink to a puff of air. So categorizing all the responses to a puff as “eye blink” misses the point, which is that the eye blink is part of a control loop – the output part – which is controlling a variable to which the air puff is a disturbance. If the CV is moistness of the eyeball, then assuming that the blink spreads lubrication over the eye, the blink is only necessary when the moistness level gets below the reference. Since the puff doesn’t drive this CV below the reference every time it occurs you wouldn’t expect a blink to every puff – which is what you found.

A possible expansion of your work would be to monitor the hypothetical controlled variable (moistness of the eye) and see if it is maintained at some level that would not be expected given air puff disturbances. I don’t know how you would monitor the moistness of the cornea. Maybe you could do the equivalent by varying the intensity of the puff, knowing the expected evaporation that would occur, and see if the blink rate mirrors the puff intensity in the way expected if moistness were under control.

I think your observation about the eyeblink reflex was very useful. It certainly tells us that a control process is very likely involved.

Best

Rick

···

Best,

Dick R.


Richard S. Marken PhD
rsmarken@gmail.com

[From Bill Powers (2008.02.22.1810 MST)]

Bruce
Nevin (2008.02.20.1435) –

(BTW, do
we still need this CSG-unique manually typed datestamp? Years ago someone
craved it for archiving. Every email message comes with a datestamp and
the name of the sender in the email
header.)

Right. Here is what the header of your post looks like in a Eudora
mailbox file (separator lines of == added):

···

=============================================================================

Delivered-To: powers_w@frontier.net

X-ASG-Debug-ID: 1203635099-53b302f50000-07bNlb

X-Barracuda-URL:
[
http://spamfilter.brainstorminternet.net:8000/cgi-bin/mark.cgi

](http://spamfilter.brainstorminternet.net:8000/cgi-bin/mark.cgi)X-MS-Has-Attach:

X-MS-TNEF-Correlator:

Thread-Topic: Vicarious conditioning

Thread-Index: Achz50skQi3SKFBLQHaA53C3RuyYvgAEAWug

X-OriginalArrivalTime: 21 Feb 2008 23:01:22.0801 (UTC)

FILETIME=[AD87C610:01C874DD]

X-Spam-Score: 0

X-Spam-Details: rule=no-quarantine_notspam policy=no-quarantine
score=0

spamscore=0 ipscore=0 phishscore=0 bulkscore=0 adultscore=0

classifier=spam adjust=0 reason=mlx engine=3.1.0-0801230000

definitions=main-0802210131

X-Spam-OrigSender: bnevin@cisco.com

X-Spam-Bar:

Date: Thu, 21 Feb 2008 18:01:21 -0500

Reply-To: “Control Systems Group Network (CSGnet)”
CSGNET@LISTSERV.UIUC.EDU

Sender: “Control Systems Group Network (CSGnet)”
CSGNET@LISTSERV.UIUC.EDU

From: “Bruce Nevin (bnevin)” bnevin@CISCO.COM

X-ASG-Orig-Subj: Re: Vicarious conditioning

Subject: Re: Vicarious conditioning

To: CSGNET@LISTSERV.UIUC.EDU

X-Barracuda-Connect: vacuum.cso.uiuc.edu[128.174.5.113]

X-Barracuda-Start-Time: 1203635100

X-Barracuda-Virus-Scanned: by Barracuda Spam Firewall at
brainstorminternet.net

X-Barracuda-Spam-Score: 0.20

X-Barracuda-Spam-Status: No, SCORE=0.20 using per-user scores of
TAG_LEVEL=1000.0 QUARANTINE_LEVEL=5.0 KILL_LEVEL=1000.0
tests=BSF_SC7_SA015e, HTML_MESSAGE

X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.42787

    Rule

breakdown below

    pts rule

name
description

    ----


    0.00

HTML_MESSAGE
BODY: HTML included in message

    0.20

BSF_SC7_SA015e Custom
Rule SA015e

X-Antivirus: AVG for E-mail 7.5.516 [269.20.9/1291]

X-Spambayes-Classification: ham

X-Spambayes-MailId: 1203641256

========================================================================================

Ah, there it is, in the 20th line. Doesn’t exactly jump out and grab you,
does it? The time and date are several lines prior to that. When I do a
reply, the header doesn’t show, and the top line in the reply
reads

“At 06:01 PM 2/21/2008 -0500, you wrote:”

Which doesn’t identify who “you” is. The “To” field
in my reply has the Sender in it, not the From item, so it reads

"Control Systems Group Network
(CSGnet)"CSGNET@LISTSERV.UIUC.EDU.

Therefore I have to open the original post to see who it is from if I
forget whose name was in the “who” field of the summary
listing.

And the time is not 6:01 PM minus five hours. That notation means the
sender’s local time is 6:05 PM, and the -0500 indicates that the zone is
GMT - 5 hours.

I find it handy when replying to delete the “[From” and
“]” and add a “–” to indicate which post is the
subject of my comments to follow. Then the name of the sender and the
time are shown all in one line at the start of the post you’re
reading.

I can accommodate to anyone using a different way of indicating the
originator of the post, but the way we’ve been doing it helps my fading
short-term memory, and saves a lot of searching for information.

Best,

Bill P.

[From Bruce Nevin (2008.02.22.4744 EST)]

Rick Marken (2008.02.21.1900)

···

The main thing I got from reading about your eye blink study

was that there was not always a blink to a puff of air.

There was a blink when the subject was not paying attention to the puff of air and was concentrating on the assigned task while ignoring distractions such as puffs of air.

There was no blink when the subject was paying attention to the puff of air on the assumption that it was a purposeful part of the experiment, and controlling to not let it affect him.

PCT is able to make this distinction and identify its source in control processes. A “conditioning” experimenter throws it away as bad data, noise, because he knows what’s supposed to happen.

/BN

[From Rick Marken (2008.02.24.0030)]

Bruce Nevin (2008.02.22.4744 EST)--

>Rick Marken (2008.02.21.1900) --

>The main thing I got from reading about your eye blink study
>was that there was not always a blink to a puff of air.

There was a blink when the subject was not paying attention to the puff of
air and was concentrating on the assigned task while ignoring distractions
such as puffs of air.

Yes, now I remember. I got the idea about the blinks not being the
same on each trial from a comment Bill Powers made when he asked Dick
about his study. I bet that Bill is right about this (the
categorization of responses); the "blinks" that occur each time there
is a puff are probably not the same each time. But the study Dick
described didn't mention this. What Dick described was a study that
showed that some people will blink fairly regularly to a puff of air
(the one's who seem to be concentrating on something else) and some
who don't (those who seem to be purposefully trying not to blink.

There was no blink when the subject was paying attention to the puff of air
on the assumption that it was a purposeful part of the experiment, and
controlling to not let it affect him.

Right. I would interpret this observation as being a reference setting
phenomenon rather than an attention phenomenon. The person who was
presumably "not paying attention" to the puff had a different
reference for whatever perception is controlled by blinking to the
puff than the person who was "paying attention". If what is controlled
is moisture on the eye then the "not paying attention" person has a
higher reference for moisture -- and thus produces more blinking when
the puffs bring the moisture below the reference-- than does the
"paying attention" person". By setting a much lower reference for
moisture on the eye the "paying attention" person will blink less,
which satisfies his other goal of trying not to blink.

PCT is able to make this distinction and identify its source in control
processes. A "conditioning" experimenter throws it away as bad data, noise,
because he knows what's supposed to happen.

I think the "theory free" observations in this experiment are that 1)
people react quite differently to the same puffs on the eye and 2)
people report having noticed different things about the puffs; one
person reports not noticing the puffs and the other reports noticing
the puffs and thinking that his goal is to avoid blinking. Saying that
one person "attended" to the puffs and that the other didn't is
already introducing theory; you can't see attention; nor can you see
concentration or ignoring of distractions. You can only see what
people do when the puffs occur and you can hear people's reports of
what they thought. This kind of "theory free" description of the eye
blink response is probably available in the descriptions of the
eyeblink response given by a "conditioning" experimenter (though I
agree that you would probably have to filter through a lot of theory
laden parts of the description) as much as they would be in the
descriptions by a PCT experimenter. It's this kind of "theory free"
description that I'm hoping to get of the "vicarious conditioning"
experiment.

Best

Rick

···

--
Richard S. Marken PhD
rsmarken@gmail.com

[From Bruce Nevin (2008.02.22.1814 EST)]

Back on the thread of vicarious conditioning,

Changing the subject heading back, as indicated.

I came across a photograph in which a vicarious something-or-other is clearly going on. […] I think this is a nice example of how, when one imagines carrying out a task, some of the imagined reference signals seem to leak through to the lower levels, which actually start to do what one is imagining doing.

Yes, this is a nice example, caught on film. (And a nice picture. Thanks for sharing it!) She observes what you’re doing. She’s doing the “same thing” with an imagined knot in imagined yarn on an imagined handle. Or maybe it’s not exactly the same thing, but what she would do if she were moving your fingers aside and showing you, “No, pull the other one that way.” In either case, she’s controlling her perception of what you’re controlling, her perception of your reference values for those CVs, and her perception of your control actions as well. Pretty certainly she’s not controlling a perception of her own control actions.

There isn’t any “conditioning” here, but “vicarious conditioning” appears to involve controlling in imagination what another is controlling, like this.

I don’t know what this “conditioning” nonsense is supposed to represent.

“Conditioning” is learning under duress. The duress is calculated manipulation of some variable, the control of which is especially important to the learner, e.g. food. The learning is the establishment of reference values and/or control systems. The learner may be “taught” or “conditioned” to control some variable that has nothing to do with the original CV (such as a heard click or tone) as means of controlling that CV.

None of this is particularly controversial for us, I think. What is bogus is the presumption that the mere statement that the coercion with the learning constitutes an explanation of behavior. Since those who talk of “conditioning” are satisfied with that “explanation” they don’t need and are not interested in the additional data that is required for a PCT experiment. However, a PCT experiment providing that additional data will be perceived by them as an instance of a “conditioning” experiment. They may be puzzled why we are interested in the additional data, but that won’t disqualify the experiment by their criteria.

Probably just another aspect of the general idea that the environment (and especially the behavioral scientists in the environment) control the behavior of us poor fumbling organisms. Actually, that’s an attractive theory because it means that other people were really responsible for all the wrong things I have done. It’s the “Don’t Blame Me” theory of behavior. Naturally I retain the credit for all the right things I’ve done, like conditioning other people for their own good.

I suppose.

/BN
···