[From Rick Marken (990114.1200)]
Martin Taylor (990114 14:22) --
It seems to me that surprise when perception differs from belief
is evidence __against__ the belief being a reference value for the
perception.
So why do you think the person acts surprised?
Best
Rick
--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken
[From Bruce Gregory (990114.1512 EST)]
Rick Marken (990114.1200)
Martin Taylor (990114 14:22) --
> It seems to me that surprise when perception differs from belief
> is evidence __against__ the belief being a reference value for the
> perception.So why do you think the person acts surprised?
Because what the person perceives differs from what the person anticipated
he or she would perceive. A thermostat has no anticipations, so it is never
surprised.
Bruce Gregory
[From Rick Marken (990114.1320)]
Me:
So why do you think the person acts surprised?
Bruce Gregory (990114.1512 EST)--
Because what the person perceives differs from what the person
anticipated he or she would perceive.
Right. This is basically all I'm saying. What the person
anticipated is represented by the setting of the reference
signal. When the reference signal is used to produce an
imagined perception (via the imagination loop) then the
imagined perception is the "anticipated perception"; it's
what I "picture in my mind" when I think about what I
_will see_ when the cylinders are placed in the water.
When the cylinders are actually placed in the water then
what I see is surprising because it is being compared to
the reference signal that _I had used_ to produce the
imagined future perception.
I am trying to propose a _mechanism_ that would explain the
phenomenon of surprise. I am saying that the mechanism is
comparison of the actual perception to the reference specification
for that perception. This comparison does not need to result in
actions that control the perception. We can have references for
perceptions without necessarily having control systems in place
to make those perceptions match the reference. That is, there
is nothing in PCT that says that a reference signal has to
be the input to a control loop. We probably have a reference
for the transition "sun moves slowly across the sky"; if the
sun stopped for some period of time I bet many people would
be _very_ surprised; but there wouldn't be much the person could
do about it; we haven't built up systems that control for the
perception "sun moves slowly across the sky" even though we
(I, anyway) have a reference for that perception.
I guess my larger point is that we are always evaluating
our perceptions relative to references for thoses perceptions
_whether we are controlling those perceptions or not_. In
HPCT, perceptions are just perceptions, neither "good" nor
"bad", "right" nor "wrong". I am saying that surprise, outrage,
disdain, satisfaction and other indications of judgement
regarding what we are perceiving are based on a comparison of
what we are perceiving to what we think we _should be_ perceiving.
The difference, for example, between people who are outraged
by the impeachment trial and those who are satisfied by it is a
difference in _reference signals_ (what they want to perception),
not perceptual signals (what they actually perceive).
Best
Rick
--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken
[From Rick Marken (990114.1420)]
Bruce Gregory (990114.1648 EST)--
This use of the term "reference signal" is new to me and
possibly to others.
Nope. Same ol' reference signal.
I have always assumed that a neural signal can be labeled a
"reference signal" because it enters the comparator of an
ECU and is subtracted from (or added to) another signal from
lower in the hierarchy that is labeled a "perceptual signal".
Yep, that's the reference signal I'm talking about. All I'm
saying is that the output of the comparison need not be part
of a control loop. It's just an error signal -- usually zero (as
in the case of "sun moves across sky"). This error signal can go
_anywhere_ in the control hierarchy; it could go lower level
systems that move limbs or make faces. Where the error signal
goes doesn't matter since we're talking about references for
perceptions that can't really be controlled.
I suspect that the result of error in systems that don't really
control is physiological preparation for responses that can
have no effect on the perception. This preparation is experienced
as an aspect of emotion (see Bill's chapter on "Emotion" in LCS II).
So if we have a reference for a perception (like the perception of
the status of the impeachment trial) that we cannot possibly control,
any difference between reference and perception (as there is for me)
results in an error that is experienced as an emotion (that's
why I'm feeling pissed off).
Best
Rick
--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken
[From Bruce Gregory (990114.1648 EST)]
Rick Marken (990114.1320)
I am trying to propose a _mechanism_ that would explain the
phenomenon of surprise. I am saying that the mechanism is
comparison of the actual perception to the reference specification
for that perception. This comparison does not need to result in
actions that control the perception. We can have references for
perceptions without necessarily having control systems in place
to make those perceptions match the reference.
This use of the term "reference signal" is new to me and possibly to others.
I have always assumed that a neural signal can be labeled a "reference
signal" because it enters the comparator of an ECU and is subtracted from
(or added to) another signal from lower in the hierarchy that is labeled a
"perceptual signal". You want to broaden the use of the term "reference
signal" in a way that seems likely to prove confusing. Perhaps it deserves a
different name and a clear operational definition that allows us to
incorporate it into PCT models.
Bruce Gregory
[From Bruce Gregory (990114.1750 EST)]
Rick Marken (990114.1420)
I suspect that the result of error in systems that don't really
control is physiological preparation for responses that can
have no effect on the perception. This preparation is experienced
as an aspect of emotion (see Bill's chapter on "Emotion" in LCS II).
So if we have a reference for a perception (like the perception of
the status of the impeachment trial) that we cannot possibly control,
any difference between reference and perception (as there is for me)
results in an error that is experienced as an emotion (that's
why I'm feeling pissed off).
An interesting idea. I can't think of any way to test it however. Can you? I
suspect that emotions such as frustration arise from perceptions that we
believe we _ought_ to be able control, when we are not succeeding.
Bruce Gregory
[From Rick Marken (90114.1830)]
Me:
I suspect that the result of error in systems that don't really
control is physiological preparation for responses that can
have no effect on the perception.
Bruce Gregory (990114.1750 EST)--
An interesting idea. I can't think of any way to test it however.
Can you?
I think so. The idea would be to take some physiological measures
(like heart rate, GSR, etc) while the person is watching, say,
a videotape of some events that the person has previously been
found to like, not care about or dislike. The physiological measures
should vary according to what is happening on the videotape.
I suspect that emotions such as frustration arise from perceptions
that we believe we _ought_ to be able control, when we are not
succeeding.
I disagree. I don't think, for example, that anyone believes that
they ought to be able to control the course of the sun as it moves
across the sky. But I bet you would find people experiencing intense
emotions if this uncontrollable perception deviated even a teensy
bit from their reference level for it.
Best
Rick
--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken/
[From Bruce Gregory (990115.0705 EST)]
Rick Marken (90114.1830)
> An interesting idea. I can't think of any way to test it however.
> Can you?I think so. The idea would be to take some physiological measures
(like heart rate, GSR, etc) while the person is watching, say,
a videotape of some events that the person has previously been
found to like, not care about or dislike. The physiological measures
should vary according to what is happening on the videotape.
Do you think any S-R psychologist would be surprised by the results?
> I suspect that emotions such as frustration arise from perceptions
> that we believe we _ought_ to be able control, when we are not
> succeeding.I disagree. I don't think, for example, that anyone believes that
they ought to be able to control the course of the sun as it moves
across the sky. But I bet you would find people experiencing intense
emotions if this uncontrollable perception deviated even a teensy
bit from their reference level for it.
I agree, but somehow I doubt that frustration would be among the emotions
experienced.
Bruce Gregory
[From Rick Marken (990115.0730)]
Bruce Gregory (990115.0705 EST)--
Do you think any S-R psychologist would be surprised by the
results?
No. But S-R psychologists are not surprised by the results
of Bill's "Nature of Control" demo (high correlation between
unsensed disturbance and response, low correlation between
visible cursor and both disturbance and response) or my
"Mind Reading" demo (an observer can't tell what a person
is doing by looking at what they are doing without testing
for controlled variables) either. You have to have a reference
for a perception in order to be surprised by a perception that
doesn't match that reference. Conventional psychologists have
no clear references for perceptions of behavior; whatever
behavior they observed is no surprise to them (they can explain
_anything_). That's one reason why PCT is a hard sell; there
is nothing surprising about it for conventional psychologists.
I think I was an unusual conventional psychology because for
some reason I really had a reference for seeing causality in
behavior. For some reason I took the basic assumption of
conventional research methodology -- the assumption that input
casues output -- to heart. So the results of what I now call
the "Nature of Control" task, which I first read about in one
of Bill Powers' Byte articles, were a _big_ surprise to me;
as big a surprise as I would have had if the sun had stopped
in its tracks one day. This surprise led me to perform
experiments (the ones described in Chapter 3 of Mind Readings
and the one called "S-R and Control" at
http://home.earthlink.net/~rmarken/demos.html )
which were part of a reorganization process that lead to
the development of references that allowed me to fit the
results of the "Nature of Control" demo into my cognitive
map; I became a perceptual control theorist.
Best
Rick
---
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken
[From Kenny Kitzke (990115.1120 EST)]
<Rick Marken (990114.1320)>
<I am saying that surprise, outrage,
disdain, satisfaction and other indications of judgement
regarding what we are perceiving are based on a comparison of
what we are perceiving to what we think we _should be_ perceiving.
The difference, for example, between people who are outraged
by the impeachment trial and those who are satisfied by it is a
difference in _reference signals_ (what they want to perception),
not perceptual signals (what they actually perceive).>
We agree here. Then why do you want to judge other people and call them
filthy names for having different references than the *always right* Ricky?
Can't you just let other people perceive what they want without feeling the
need to defend or justify their references with you? Do you have any
evidence that apes have a reference for perjury even though they don't
appear to be able to talk to humans? Are any apes interested in Mr. Bill's
impeachment trial? I wonder if science or PCT can explain their
disinterest?
Kenny
[From Rick Marken (990115.0950)]
Me:
I am saying that surprise, outrage, disdain, satisfaction and
other indications of judgement regarding what we are perceiving
are based on a comparison of what we are perceiving to what we
think we _should be_ perceiving.
Kenny Kitzke (990115.1120 EST) --
We agree here.
Great!
Then why do you want to judge other people
Why would you ask this if you agree with me? I don't _want_ to
judge people; I can't help it, nor can you (notice how you are
judging me and "Mr. Bill", for example). What we perceive is
continuously being compared to references for what those perceptions
should be. We judge (compare perceptions to references) whether we
want to or not.
and call them filthy names for having different references than
the *always right* Ricky?
I don't recall calling anyone a filthy name for having references
different than mine. I have certainly expressed disgust at some
behavior I see but I don't recall calling anyone a filthy name.
And I don't think I am always right. Just recently, for example,
I was wrong in the "control of behavior" debate with Bruce Abbott.
I admitted I was wrong, learned from my error and even wrote a
demo (now at my web site:
http://home.earthlink.net/~rmarken/ControlDemo/Coercion.html) showing
how control by disturbance
works. But, tell me, Kenny, when you make these critical comments
about me are you always thinking you are right? Or are you just
thinking you are right sometimes? Or are you making these comments
and thinging you are wrong?
Can't you just let other people perceive what they want without
feeling the need to defend or justify their references with you?
I can't do anything about what other people perceive. But this is
a scientific discussion group and in science ideas that are
proposed publicly will be challenged and tested. I could care
less what you want to believe (as long as it doesn't hurt other
people). But if you make statements on this net about what you
believe to be true then those statements are likely to be
challenged. If this feels like being attacked for what you
believe then I suggest that you just not make your beliefs
public.
Do you have any evidence that apes have a reference for perjury
even though they don't appear to be able to talk to humans?
Since non-hominid apes can't talk I think it's unlikely that they
can intend to do purjury. But I think there is evidence that such
apes (chimps, for example) can (and will) intentionally lie; I seem
to recall research where a chimp intentionally misled a researcher
about the location of some object in order to keep the object for
himself.
Are any apes interested in Mr. Bill's impeachment trial?
Yes. It looks like a large group of hominid apes are interested
in it.
I wonder if science or PCT can explain their disinterest?
I should hope so.
Best
Rick
--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken
[From Rick Marken (990115.0955)]
Bruce Gregory (990115.1200 EST)--
I phrased my question badly. What is it about the experiments you
propose that supports your model as opposed to any other model?
Good question. I think I would add a look at individual differences
to the study. For example, show the same tape (of the impeachment
trial, say) to a person who wants to perceive self-righteous
hypocrisy and to another who doesn't; the same "stimulus" will
produce very different physiological responses in each. Of course,
this could be because these people perceive the same situation
differently; so we would have to test to make sure that both
people heard the same words, got the same meaning from the words,
etc. Not easy.
Best
Rick
--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken
[From Bruce Gregory (990115.1200 EST)]
Rick Marken (990115.0730)
Bruce Gregory (990115.0705 EST)--
> Do you think any S-R psychologist would be surprised by the
> results?No. But S-R psychologists are not surprised by the results
of Bill's "Nature of Control" demo (high correlation between
unsensed disturbance and response, low correlation between
visible cursor and both disturbance and response) or my
"Mind Reading" demo (an observer can't tell what a person
is doing by looking at what they are doing without testing
for controlled variables) either. You have to have a reference
for a perception in order to be surprised by a perception that
doesn't match that reference. Conventional psychologists have
no clear references for perceptions of behavior; whatever
behavior they observed is no surprise to them (they can explain
_anything_). That's one reason why PCT is a hard sell; there
is nothing surprising about it for conventional psychologists.
I phrased my question badly. What is it about the experiments you
propose that supports your model as opposed to any other model?
Bruce Gregory