Emotions

[Bruce_Gregory4]

[From Bruce Gregory (2004.01.02.0842)]

Marc Abrams (2004.01.01.2313)

When I look at the keyboard, what portion of my perception is
imagined?

All of it _AND_ none of it. I know it sounds crazy, BUT, I go back to
my
example of an art student trying to learn to draw real-life scenes and
my
own experience with it, plus a ton of research to this effect.

An interesting story, but not relevant to your claim, at least as I see
it. When you are looking at the paper you are perceiving the paper.
When you are looking at the object, you are perceiving the object. the
lesson, as I take it, is that your ability to draw improves when you
focus your attention on the object you are drawing and not the drawing.
This tells us nothing about the role of imagination in perceiving
either the object or the drawing, as far as I can tell.

Hardly, I 'see' things in my thoughts and ideas as well as my dreams.
Where
do you suppose they come from? When you 'hear' a voice, can you match
it to
perception?

All these things are perceptions. but that does not mean that they are
98% of all perceptions. If they are, I suspect one would be
institutionalized as severely psychotic.

Do you actually use all your senses in constructing what you
simply heard? We _love_ to make stories. Schank was and is right on.

I agree, but don't see the relevance.

I would say that a perception in some imagined or idealized state is
still a perception and not a reference condition. The error is minimal
when the perception matches the reference condition, but the two
remain
distinct.

OK, so what is a reference condition then?

I think Martin's post covers that quite well.

The first stop for the efferent signals in the cortex is the Primary
Sensory
Cortex. The Somatosensory cortex is believed to be located in the
parietal
lobe. The primary visual cortex is located in the occipital lobe and
the
primary auditory cortex is in the temporal lobe. At this point they are
still _not_ perceptions.

They are not perceptions or we are not aware of them?

From the primary cortex's they project to what is
called the unimodal sensory associative area's, located in the
parietal,
occipitotemporal, and temporal lobes respectively. It is here that
each of
three cortex's combine the different signals for each of the sensory
modalities into entities for each of the respective senses. They are
still
not considered perceptions. It is not until they project to the next
area,
The multi-modal sensory association area's where you can start calling
them
perceptions and not mere signals.

Of course this is not a distinction made in PCT.

So in my world 'perceptions' don't start until the 4th level.

O.K. So now we are dealing with MACT.

Now to answer Bruce's questions.

What do you mean by "more sensitive"? Higher gain?

Yes. If more amplification will make certain Level 4 aspects different
from
the rest. But I believe tha amplification or gain can actually be
added at
any point between transduction and Level 4.

In other words, no.

Can I translate your
statement as follows: We control certain perceptions at higher gains
than other perceptions. Normally higher gain leads to better control.
Is this what you mean?

Yes, with perceptions being defined by my organizational structure
above.

What does higher gain have to do with setting reference conditions
lower?

Nothing. The setting of reference conditions is a seperate issue.

And how do lowered reference conditions lead to higher sensitivity?

You don't lower a reference condition. You lower or raise the gain
_OF_ the
reference condition. I believe that emotions can cause internal
disturbances
that can effect the gain of both, perceptions and reference conditions.

In PCT gain is not associated with reference conditions but with ECU's,
so again we are talking MACT. Can you explain how this works in
principle? I realize that the details are unknown.

By the way, I do not find the notion of perception in PCT ambiguous.
We
can perceive many things, but that does not make the concept
ambiguous.

OK, I hope I clarified why I think they are, and still do.

No, but you have proposed an alternative way of describing perceptions.

Bruce Gregory

"Everything that needs to be said has already been said. But since no
one was listening, everything must be said again."

                                                                                Andre Gide

[Marc_Abrams1]

from [Marc Abrams (2004.01.02.0915)]

[From Bill Powers (2004.01.02,0616 MST)]

>Sort of and sometimes. The adrenal gland is 'operated' by our
>sympathetic nevrous system, that is, our spinal column. The innervation

of

>our spinal column by receptors can and does allow the adrenal gland to
>respond. It will also respond to commands from the brain (conjectured to

be

>the hypothalamus). So in effect, receptors will tell the spinal column

that

>adrenaline is required, it will signal the adrenal gland _and_ the brain,
>where the brain will 'supervise' the adrenaline through the spinal cord
>according to physiological and cognitive needs.

This doesn't answer my question, which is how the adrenal gland could
respond to an impending collision at the same time the cogitive part of

the

brain does.

It's truly amazing how selective we become when we want to read something we
want. I appreciate your quoting me above. Did you read the last sentence in
the above quote? _AND_ this part of my post?;

You asked:

Doesn't there have to be a
series of interpretations of the visual image before the adrenal gland
could figure out that a collision is imminent?

I answered:

Not necessarily, but in _this_ case it probably would. The adrenal gland

can

operate solely on the basis of some sensations picked up by mechanical,
chemical or thermal receptors, that go straight to the spinal column. In
effect the brain and adrenal gland are 'notified' at the same time by the
spinal column. Again, this does not preclude the brain from signaling the
adrenal for 'cognitive' stresses.

Again, please pay special attentioin to the first and last sentences in the
above paragraph

For this to happen, the adrenal gland must receive information
that there is a car ahead, that the distance to this car is decreasing
rapidly, and that if nothing is done there will be an event or

relationship

called a "collision" and all that such a thing implies. How does that
information get to the adrenal gland? You say

>There are two main sensory systems that provide information to the brain
>about the state of our internal organs. The afferents provided by the...

No. That was not my explanation for the adrenal gland. That was a later
explanation of the the two main sensory systems which had _nothing_ to do
with the collision example, which I gave _prior_ to this statement.

I don't see how pain receptors, receptors for touch, pressure,temperature
smell, taste, or pain could tell the adrenal gland that a collision is
about to happen.

They don't and didn't. As I said the first time.

A simple connection from a sensor to the gland would be
insufficient,

Right, becuase there is no connection between a sensor and a gland. There
is a connection between a receptor and either the spinal column or brain
stem. it is through either one of these two that the endocrine system works,
usually through the control of the cortex.

because it is necessary to detect shapes, locations,
velocities, events, and relationships, as well as projecting them into

the

near future, in order to apprehend that something bad is about to occur

(in

Gregory's example).

No. The _only_ thing the adrenal gland needs to 'know' is to secrete or not
to secrete. It has no idea about anything else and doesn't need to.

A single sensory nerve ending could not possibly detect such things.

Who said it did? Did you read my post to Bruce Gregory and the use of the
associative area's?

In fact, I don't see how the pending collision could be known
to any part of the organism without use of perceptual functions that we
would have to class as "cognitive."

Again, I would agree. Where did you think I didn't?

I see no way for the adrenal response
to occur before or even at the same time as the cognitive response. Please
explain how this could happen.

In this _particular_ case 'fear' as I have said in a previous post _is_
cognitive. ;

From [Marc Abrams (2004.01.01.1734)]

...For any. When you are cognitively going Uh! Oh!, your eyes are focusing

on

what is ahead of you, your heart starts beating a bit faster, your foot is
going for the gas or brake, you're tensing up. All of this is happening at
the same time. It's all happening physiologically, with something's you

have

no control over, some you have a little bit of control over and others you
have a great deal of sway over. For instance, How much you 'tense' up
depends on what kind of thoughts you have. The emotion of fear can, and
does, provide the trigger for the release of CRH (corticotropin-releasing
hormone) which underlies the autonomic responses (heart beat, sweating,
tightening of muscles, both smooth and skeletal)necessary to correct for
error. But a key ingredient in all of this is how much you will 'panic'.
"Panic" is something you have a good deal of control over and is cognitive.
This of course affects and is affected by our autonomic nervous system...

I hope I cleared up any misunderstandings.

Bill, a red flag went up with this post. I can see this thread might be
heading in a bad direction. I hope I'm wrong but we'll see. I sense a change
in your attitude, and when you misunderstand me so completely I wonder what
you have actually read. I'm not great, but I'm not this bad of a writer
either. I just don't know how you came up with what you did here. I'm not
asking you to justify anything here. I just want you to know that I sense a
change and not a good one. Somehow I feel like we might be going down a path
we have traveled before and I don't want to go there.

Marc

[Martin_Taylor2]

[Martin Taylor 2004.01.02.1016]

From [Marc Abrams (2004.01.01.1413)]

> [Martin Taylor 2004.01.01.2146]

>... A reference
> >condition in PCT is simply any one of these perceptions in some imagined
or

>idealized state. So you tell me how am I using it? I am suggesting that
>different aspects (intensity, relationships, events, etc) can each have

it's

>own reference levels and we pay attention to different 'aspects' of

larger

>combinations.

That really doesn't have much to do with the PCT meaning of "Reference

Signal".

Ok, lets take a look at your response. I assume you are talking about my
response to Bruce.

Just to clarify...I'm not commenting on your concepts in either of
your long messages. I'm trying to deal entirely with the PCT/HPCT
functional description of what is going on, so as to try to minimize
confusions when we look at what is in conflict (if anything) between
your ideas and those of HPCT. One place where confusion is likely is
in the use of the word "perception." You make this explicit in [Marc
Abrams (2004.01.01.2313)] where you say: "Of course I'm defining
perceptions the way 98% of the world does but I'll live with that
quirk."

Living with that quirk is almost guaranteed to ensure that there will
be huge misunderstandings when you start arguing that your view of
perception differs from the PCT view of perception. Of course it
does. You have defined it so.

>

An elementary control unit (ECU), myriads of which construct the
hypothesized hierarchy, consists of a loop that has two inputs from
outside the loop. One comes from what we conventionally call "below".
That's an input to a perceptual finction that produces a perceptual
signal (a perception), and as far as the ECU is concerned, it
represents a state of the environment (real or imagined). The other
comes from "above". It is the Reference Signal, and it tells the ECU
what value the (real or imagined) perception should take on. The
difference between these two is the error signal, and if the error is
not zero, it means the ECU should be changing its action on the
environment so as to influence the perception to come closer to its
reference signal.

Ok, so far so good. No conflicts here.
>
> It doesn't really matter whether the perception is an imagined

perception from an imagined environment that the imagined action
influences or a perception based on data from external and internal
sensors and the action works on a "real' world full of disturbances
so as to influence that "real" perception. The relation of reference
signal to perceptual signal is the same. You could think of the
reference signal as a command, if you want, but the perceptual signal
is an observation.

Sorry. My point was _not_ whether the environment was real or imagined or if
the refernce or perception were imagined, my point was and is that the one
disturbance that a cv passes through is in the external environment. I am
claiming that a perception passes through an 'internal' disturbance I like
to call emotions. It has nothing to do with whether it's real or imagined or
any combination of the two.

In that case, I guess when you said you agreed to the previous
paragraph, you didn't understand what it said. Let me reiterate, in
different words...

In standard PCT, an elementary control unit (ECU) consists of one
simple loop in which all signals are SCALAR variables. Going
backwards around the loop as compared to the way signal information
travels (because that's the way you have to analyze the "causes" of
signal levels in the loop), there is an output to the ECU's
environment (note _very_ carefully, NOT the organism's environment).
This output is the result of an output process operating on an error
signal within the ECU. The error signal is the difference between a
reference signal that comes from outside the ECU (we say
conventionally that its source is "above" the ECU) and a perceptual
signal. The perceptual signal is the result of a percpetual process
operating possibly many input signals from outside the ECU.

The output signal may be distributed to have effects on many, many,
aspects of the environment of the ECU. Likewise, the reference and
perceptual signals can have sources in many, many, places outside the
ECU. For most, perhaps all ECUs, ALL of those sources are inside the
organism. Why I say "most, perhaps all" is because I do not know the
extent to which the peripheral muscles and sensors have the
characteristics of ECUs. They are the only places where there is
contact between any ECU and the world outside the organism.

[Less standard PCT speculation follows.]

Standard PCT has an extension that is less well accepted or agreed.
Let's say it's speculative. That is the so-called "imagination loop."
One concept of the imagination loop (I don't know if it's exactly
Bill P's concept) is that INSIDE the ECU there is a connection
between the output process and the perceptual process that goes
through what might be called a "simulated environment" process
(a.k.a. "imagination"). The simulated environment is presumably an
aspect of memory, and if the simulated environment affects the output
signal in the same way as does the real environment, then the
perceptual signal will be the same as would the perceptual signal be
were the output signal to be distributed into the real environment.

If the "imagination loop" exists in some or all of the ECUs that
compose the hierarchy, then there must be a means of merging or
switching between the perceptual input from outside the ECU (but
within the organism) and the perceptual input through the imagination
loop. At high levels, when the signals to and from the world outside
the ECU are switched out, we call the process "planning"

> In PCT, a reference signal is not a perceptual signal. It's like a

command to make a perceptual signal take on some value.

_THAT_ has been my point. I guess when I said that a reference condition is
an idealized perception that some thought I was saying they are the same
thing.

I thought so.

That was not my intent. Your explanation is much less ambiguous and I
will adopt it. Thanks. btw, forget about the example I asked for above
think I figured out where it came from.

Thanks. I hope this message makes it yet clearer.

My main point here (actually two main points).

(1) If you want to have a productive discussion relating your
concepts to PCT, make a distinction between the everyday use of the
word "perception" and its technical use in HPCT. For example, call
one of them "E-perception" and the other "PCT-perception" or
something like that.

(2) Remember that outside an ECU is the rest of the organism, and
emotions and the like come from outside the ECU, just as do signals
from the world outside the organism (unless, of course, they are
generated in one of the ECU's processes, meaning that the ECU's job
is to create an emotion corresponding to a reference value for that
emotion).

I think your contributions are worthwhile and bear pursuit, but the
pursuit will lead through bramble thickets if we don't keep these
distinctions (as well as the usual "functional" vs. "implementation")
clear.

Martin

[Bill_Powers]

[From Bill Powers (2004.01.02.0846 MST)]

Marc Abrams (2004.01.02.0915)--

It's truly amazing how selective we become when we want to read something we
want. I appreciate your quoting me above. Did you read the last sentence in
the above quote? _AND_ this part of my post?;

Yes, and you keep going back and forth about this. After carefully pointing
out that the collision example would NOT involve the adrenals reacting
before the cognitive systems had recognized the danger, you conclude by saying

When you are cognitively going Uh! Oh!, your eyes are focusing
on what is ahead of you, your heart starts beating a bit faster, your foot is
going for the gas or brake, you're tensing up. All of this is happening at
the same time.

This, of course, denies what you have just been saying. Why does your heart
beat faster and your foot move toward the brake? Because of an error signal
in a higher-level control system, according to my model. According to
yours, the adrenals get "notified" at the same time as the cognitive
systems, even though you agree that in this example, the cognitive systems
have to work first before any need to act can even be perceived. But having
said that, you go right back to describing the emotional and cognitive
processes as operating in parallel. This is very much like the ideas I read
50 years ago that led me to reject pretty much everyhing in the
conventional literature on this subject. It didn't hang together then any
more than your current description of it does now.

Bill, a red flag went up with this post. I can see this thread might be
heading in a bad direction. I hope I'm wrong but we'll see. I sense a change
in your attitude, and when you misunderstand me so completely I wonder what
you have actually read. I'm not great, but I'm not this bad of a writer
either.

When it comes to explaining theory, Marc, you are a terrible writer. Your
ideas are disorganized and inconsistent, and you seem to skim the surface
without seeing the crucial details you are leaving out, or seeing how one
paragraph contradicts another. Your ideas about brain organization are
lifted out of textbooks which simply echo proposals that have been around
for a century, which I knew about when I wrote B:CP, and which I rejected
as implausible and much too sketchy to constitute any sort of theory of
behavior. The fact that you are learning about these ideas for the first
time, and reading about them in books with recent publication dates, does
not mean that they are new to everyone in this discussion, or even very
many of them. And the fact that they are printed in books does not make
them correct or even "modern."

All the same connections you speak of are in the HPCT model (with the
restoration of the emotion chapter), except for the autonomic nervous
system. I have simply organized the facts in a more systematic way, partly
through analyzing perception and control in a way that no neurologists ever
have ever done. As I have explained many times on CSGnet, it is perfectly
possible for lower-order control systems to act against disturbances,
readjusting the physiological systems as they do so, before higher-level
systems can act. I have explained why this leads to an illusion of the
emotions arising before the conscious control systems realize anything is
going on, so they conclude, erroneously, that the emotions came first. The
"direct" connections from spinal cord to adrenals that you imagine are in
my model already -- but they are connections through low-order control
systems, not simple S-R links. And neuroanatomy fully supports these
connections, though the texts never put all the observed linkages together
into a coherent kind of system, a control system.

I just don't know how you came up with what you did here. I'm not
asking you to justify anything here. I just want you to know that I sense a
change and not a good one. Somehow I feel like we might be going down a path
we have traveled before and I don't want to go there.

What you are noticing, Marc, is that I am holding you to the same standards
I set for myself in constructing theories. You are being hasty, even
sloppy, and your renditions of control theory verge on delusory. You seem
to have almost no conception of what a signal is, an input function, a
reference level/signal, gain, or any of the fundamental terms of PCT, as
others have been more or less gently hinting (while you ignore the hints).
What you have constructed in your mind using my words as raw materials
bears almost no resemblance to control theory, system analysis, PCT, or
anything else I have proposed.

While you have no obligation to adopt PCT or HPCT or control theory in
general, I should think that you would want to avoid ridicule by at least
learning what the theory you are trying to improve on says. Rather than
doing that, I should think you would fare better simply by focusing on your
own theory (which Bruce G. refers to as MACT, trying to hint that you are
pretty far from PCT or any other kind of CT).

You may think that all these remarks are uncalled for and that you have
made a very different impression on others listening in or participating in
this discussion. If you are right and I am wrong, I want to know about it,
and if it's the other way around I suppose you would want to know about it,
too. So I ask all onlookers and participants to render their verdicts here.
Is my opinion of Marc's recent communications off the track or inaccurate?
Do any of you see a coherent theory of emotion or behavior or experience
arising out of what Marc has been proposing? I will apologize and subside
if I have misconstrued what is going on. But I can't just let it go on
without voicing my protest.

Best,

Bill P.

[Marc_Abrams1]

From [Marc Abrams (2004.01.02.1013)]

[From Bruce Gregory (2004.01.02.0842)]

An interesting story, but not relevant to your claim, at least as I see
it. When you are looking at the paper you are perceiving the paper.
When you are looking at the object, you are perceiving the object.

No. When you are looking at the paper you are perceiving what you are drawing on the paper, and what you are drawing is coming from your imagination

the lesson, as I take it, is that your ability to draw improves when you
focus your attention on the object you are drawing and not the drawing.

Yes, but it’s becuase of what you are paying attention to that matters.

This tells us nothing about the role of imagination in perceiving
either the object or the drawing, as far as I can tell.

I see it differently.

All these things are perceptions. but that does not mean that they are
98% of all perceptions. If they are, I suspect one would be
institutionalized as severely psychotic.

This goes back to what and how you define a ‘perception’. It’s not only that I simply don’t care for the current PCT definition and will not use it in my work. It becomes really problematic in communicating PCT to others. On the surface it represents no real problems because a perception that most people understand to be a perception is one in PCT as well. It’s the extended version based on the hierarchy that gets people confused. In fact, I know of no other body of work who define perceptions as you do. All I will say is that when PCTer’s talk about perceptions, it’s not of generally accepted variety.

So in re-stating my comments according to your defintion of a perception, I would agree that most perceptions are not imagined. I will leave it up to empirical research to figure out just what the percentages actually are.

I would say that a perception in some imagined or idealized state is
still a perception and not a reference condition. The error is minimal
when the perception matches the reference condition, but the two remain
distinct.

OK, so what is a reference condition then?

I think Martin’s post covers that quite well.

Yes, and you quoted yourself here. The triple bracketed lines are yours, not mine, and Martin was addressing your definition of a reference being a perception. I explained in my reply to Martin that my use of the term ‘idealized perception’ was misunderstood.

They are not perceptions or we are not aware of them?

Again, depends on how and what you define as a perception. To me it’s real simple. A perception is a_cognitive_ construct. It is a conscious interpretation of the environment. So according to my definition (and most others) a perception would only be a perception after the projection into the primary multi-modal associative area in the frontal lobe. You may disagree with this, and that’s fine. What data do you have to dispute this claim and back-up yours?

Of course this is not a distinction made in PCT.

Yep, So? Are we trying to be accurate here or idealogical?

So in my world ‘perceptions’ don’t start until the 4th level.

O.K. So now we are dealing with MACT.

You can call it what you want. I call it PCT. My PCT model has everything the model in B:CP & the chapter on emotions in LCS II has. What is different is how I have organized and structured my information. If PCT is defined by the specific organization and structure Bill has set up in B:CP exclusively than I guess I am doing MACT. But I think your wrong.

Bill why don’t you settle this once and for all. Am I doing PCT or MACT? If you feel I’m doing MACT I’m outta here, and in a flash.

And how do lowered reference conditions lead to higher sensitivity?

You don’t lower a reference condition. You lower or raise the gain
OF the
reference condition. I believe that emotions can cause internal
disturbances
that can effect the gain of both, perceptions and reference conditions.

In PCT gain is not associated with reference conditions but with ECU’s,

I assume you are referring to Martin Taylor’s use of the term ECU, yes? If so, an ECU is, from a Martin Taylor paper titled; Perceptual Control and Layered Protocols in Interface Design

"…The above discussion leads to a canonical description of a control loop, shown in Figure 2a. The loop can be considered as having two parts, one that senses some aspect of its environment, compares the sensed value (its “perceptual signal value”) with a reference value to generate an “error signal value”, and produces an output. We call this part of a control loop an Elementary Control Unit (ECU). The other part of the loop consists of the effect that the output, acting through the external environment, has on the sensing system. This external part is the “environmental feedback function.”

What am I missing here Bruce? As far as I can determine the reference condition is part of an ECU. If you’re suggesting that gain is part of the ‘entire’ ECU and not part of one of it’s parts, I have a difficult time understanding that logic.

so again we are talking MACT. Can you explain how this works in
principle? I realize that the details are unknown.

Sure, if you could explain exactly what part MACT it is your asking about

No, but you have proposed an alternative way of describing perceptions.

Good. That was my intent. So I did indeed clarify my position.

Marc

[Richard_Marken2]

[From Rick Marken (2004.01.02.1300)]

Bill Powers (2004.01.02.0846 MST)--

So I ask all onlookers and participants to render their verdicts here.
Is my opinion of Marc's recent communications off the track or
inaccurate?
Do any of you see a coherent theory of emotion or behavior or
experience
arising out of what Marc has been proposing?

No.

Happy New Year

Rick

···

---
Richard S. Marken
marken@mindreadings.com
Home 310 474-0313
Cell 310 729-1400

[Marc_Abrams1]

from [Marc Abrams (2004.01.02.1309)]

[Martin Taylor 2004.01.02.1016]

Just to clarify...I'm not commenting on your concepts in either of
your long messages. I'm trying to deal entirely with the PCT/HPCT
functional description of what is going on, so as to try to minimize
confusions when we look at what is in conflict (if anything) between
your ideas and those of HPCT. One place where confusion is likely is
in the use of the word "perception." You make this explicit in [Marc
Abrams (2004.01.01.2313)] where you say: "Of course I'm defining
perceptions the way 98% of the world does but I'll live with that
quirk."

Living with that quirk is almost guaranteed to ensure that there will
be huge misunderstandings when you start arguing that your view of
perception differs from the PCT view of perception. Of course it
does. You have defined it so.

Yes, and I went into some detail yesterday with my response to Bruce Gregory
in;

from [Marc Abrams (2004.01.01.2313)]

"...Anyway, depending on the kind of receptor, the transduced signal moves
on to
either the spinal cord or brain stem...."

In that case, I guess when you said you agreed to the previous
paragraph, you didn't understand what it said. Let me reiterate, in
different words...

Ok.

In standard PCT, an elementary control unit (ECU) consists of one
simple loop in which all signals are SCALAR variables.

OK. I assume by 'standard' you mean B:CP exclusively, correct?

The output signal may be distributed to have effects on many, many,
aspects of the environment of the ECU. Likewise, the reference and
perceptual signals can have sources in many, many, places outside the
ECU. For most, perhaps all ECUs, ALL of those sources are inside the
organism. Why I say "most, perhaps all" is because I do not know the
extent to which the peripheral muscles and sensors have the
characteristics of ECUs. They are the only places where there is
contact between any ECU and the world outside the organism.

Martin, lets focus on the ECU for a moment. I'm afraid I'm a bit confused.
In studying PCT all these years I felt the environment was _always_
considered _outside_ the body, and disturbances from the _external_
environment were part of the model. No one ever tried to disuade me or
anyone else from this notion. In fact in B:CP, LCS I, and Mind Readings, any
number of diagrams are shown with the control unit inside the organism and
the environment outside, clearly delineated as such. Your notion of the ECU
and its environment makes perfect sense. A disturbance to an ECU is a
disturbance, it doesn't make a difference where the disturbance actually
comes from, its _external_ to the ECU. But I do not believe that this is the
picture that is given in the PCT literature. I say this because in showing
HPCT, in Mind Readings for example pg. 137 is a diagram of a three level
hierarchy. In it the ECU's pass through the 'environment' only at the first
level and are exposed to disturbances only at that point. They are _not_
exposed to 'internal' disturbances on the way down the hierarchy from level
to level.

I am _not_ advocating the need for disturbances here. I believe the effects
of attention and emotion can be handled in the input function. This of
course is subject to change I like the notion Martin presents on the
ECU's and toward that end I have a question or two. How _do_ we distinguish
between an organism and its external environment when using ECU's? I thought
a reference condition was endogenous to a PCT control system. That is the
reference condition is set by internal rather than external means. How do we
distinguish this with ECU's?

[Less standard PCT speculation follows.]

Standard PCT has an extension that is less well accepted or agreed.
Let's say it's speculative. That is the so-called "imagination loop."
One concept of the imagination loop (I don't know if it's exactly
Bill P's concept) is that INSIDE the ECU there is a connection
between the output process and the perceptual process that goes
through what might be called a "simulated environment" process
(a.k.a. "imagination"). The simulated environment is presumably an
aspect of memory, and if the simulated environment affects the output
signal in the same way as does the real environment, then the
perceptual signal will be the same as would the perceptual signal be
were the output signal to be distributed into the real environment.

If the "imagination loop" exists in some or all of the ECUs that
compose the hierarchy, then there must be a means of merging or
switching between the perceptual input from outside the ECU (but
within the organism) and the perceptual input through the imagination
loop. At high levels, when the signals to and from the world outside
the ECU are switched out, we call the process "planning"

Why the need for switching? If we understand how perceptions are constructed
(or can come close in theorizing about them _functionally_) I believe that
will provide us with some key insights into how memory functions. (at least
as far as the PCT model is concerned) I am much less concerned with the
actual mechanisms of how memory works than I am in how it functions and is
integrated into our ability to make perceptions, plan, etc.

Thanks. I hope this message makes it yet clearer.

Yes, and raises some questions as well. Thank you

My main point here (actually two main points).

(1) If you want to have a productive discussion relating your
concepts to PCT, make a distinction between the everyday use of the
word "perception" and its technical use in HPCT. For example, call
one of them "E-perception" and the other "PCT-perception" or
something like that.

Yes, this is a problem I will follow your suggestion and simply call _my_
definition of perception an M-perception. Have I made my self clear on the
distinctions between my views on perceptions and those of Bill?

(2) Remember that outside an ECU is the rest of the organism, and
emotions and the like come from outside the ECU, just as do signals
from the world outside the organism (unless, of course, they are
generated in one of the ECU's processes, meaning that the ECU's job
is to create an emotion corresponding to a reference value for that
emotion).

Not to nit here Martin but 'signals' don't exist in the external
environment. Various energy stimuli do, and I think the distinction here is
important for a number of reasons. Our sensory organs (our 5 main senses)
transduce the _external_ energy stimuli into internal 'signals'. It is these
internal signals and not the external energy that is 'controlled' in our
bodies. Second, emotions, I believe, are a side-effects of what ECU's
produce and do not 'come' from the outside. How about an ECU's job of
creating an emotion based on an error signal?

I think your contributions are worthwhile and bear pursuit, but the
pursuit will lead through bramble thickets if we don't keep these
distinctions (as well as the usual "functional" vs. "implementation")
clear.

Thanks, I agree and I'll try my best.

Marc

[Martin_Taylor2]

[Martin Taylor 2004.01.02.1621]

From [Marc Abrams (2004.01.02.1309)]

> [Martin Taylor 2004.01.02.1016]

In standard PCT, an elementary control unit (ECU) consists of one
simple loop in which all signals are SCALAR variables.

OK. I assume by 'standard' you mean B:CP exclusively, correct?

B:CP and the generally accepted results of discussions on CSGnet over
the dozen years I've been reading it.

The output signal may be distributed to have effects on many, many,
aspects of the environment of the ECU. Likewise, the reference and
perceptual signals can have sources in many, many, places outside the
ECU. For most, perhaps all ECUs, ALL of those sources are inside the
organism. Why I say "most, perhaps all" is because I do not know the
extent to which the peripheral muscles and sensors have the
characteristics of ECUs. They are the only places where there is
contact between any ECU and the world outside the organism.

Martin, lets focus on the ECU for a moment. I'm afraid I'm a bit confused.
In studying PCT all these years I felt the environment was _always_
considered _outside_ the body, and disturbances from the _external_
environment were part of the model.

I can see where your confusion comes from (I think). It's true that
most discussion talks about "a control system" without distinguishing
between the entire complex (of which the hierarchy is the
hypothesised implementation) or an ECU (which is the hypothesised
atomic unit of the complex). When "a control system" means an
organism that is influencing its perceptions (very plural, and at
many different levels of abstraction) through its "environmental
feedback path", then indeed the environmental feedback path is
entirely outside the organism.

However, when one is dealing with ANY control system, whether it is
an ECU, a complex network of ECUs, a level in the hypothetical
hierarchy, or anything else, what is in the environment is everything
affected by the output(s) of that control system and everything that
influences its input(s). So, if the boundary of the control system
under consideration is not the skin of the organism, then some of the
environment is inside the organism.

No one ever tried to disuade me or
anyone else from this notion.

Probably either because it was not clear that you (or others) might
not recognize the distinction between control system and organism, or
because in most experimental cases, what an experimenter can observe
lies usually outside the organism, so the control system in question
is asserted to have its "skin" coincident with the organism's skin.

In fact in B:CP, LCS I, and Mind Readings, any
number of diagrams are shown with the control unit inside the organism and
the environment outside, clearly delineated as such.

True, I suspect for the second reason above. But you must recognize
that any diagram of a hierarchy shows ECUs at levels above the
peripheral sensors and muscles. All such ECUs have an environmental
feedback path partly inside the organism. In fact, if you think about
it, the environment of any ECU includes everything outside of it, and
that usually includes the source of the reference signal as well as
the ECU's environmental feedback path.

Your notion of the ECU
and its environment makes perfect sense. A disturbance to an ECU is a
disturbance, it doesn't make a difference where the disturbance actually
comes from, its _external_ to the ECU. But I do not believe that this is the
picture that is given in the PCT literature. I say this because in showing
HPCT, in Mind Readings for example pg. 137 is a diagram of a three level
hierarchy. In it the ECU's pass through the 'environment' only at the first
level and are exposed to disturbances only at that point. They are _not_
exposed to 'internal' disturbances on the way down the hierarchy from level
to level.

True, again. But the idea there is simply to show that multi-level
hierarchies can work in the face of external disturbances. I don't
think it would be too hard for Rick to extend his spreadsheet model ,
as a proof of concept, to incorporate disturbances that enter into
the second or third level perceptions (not to be used as a model of
actual human processing). Rick??

I am _not_ advocating the need for disturbances here. I believe the effects
of attention and emotion can be handled in the input function. This of
course is subject to change I like the notion Martin presents on the
ECU's and toward that end I have a question or two. How _do_ we distinguish
between an organism and its external environment when using ECU's?

I think that's easy, once you define what sensory inputs are
available to the (possibly complex) control system, and what
effectors it has. Let's get physiological for a moment (a terrain
with which I am barely familiar :-). It is (reasonably) clear that an
entire human under normal conditions has input sensors for certain
electromagnetic waves, whereas other EM waves have effects that are
not sensed but do affect internal states of the organism (think gamma
ray damage to DNA, for example). Both varieties affect the behaviour
of the organism, but only the sensed ones do so through the immediate
action of the hypothetical hierarchy. The others do so by affecting
the hierarchy itself, either through direct damage (in the case of
Gamma radiation) or through reorganization caused by the departure of
intrinsic variables from their reference values.

Define the "skin" of your control system (complex) and you have
defined the external environment. It's everything not inside that
"skin".

I thought
a reference condition was endogenous to a PCT control system. That is the
reference condition is set by internal rather than external means. How do we
distinguish this with ECU's?

A reference condition is the input to some process that creates a
reference signal. That process has been much less studied or
discussed than has the perceptual input process. Nevertheless, in the
hypothetical hierarchy, there must be some process that allows the
outputs of many higher-level ECUs to combine into the scalar that is
the reference signal for one lower-level ECU.

What this means is that there is an easy confusion between the terms
"reference condition" and "reference signal." I don't think that
distinction has been articulated on CSGnet as far as I can remember,
but it should be. The "reference condition" is a good way to describe
the vector of signals that combine through a "reference input
process" to form a scalar "reference signal." Normally, when we talk
about the behaviour of an ECU, we are concerned only with the
reference signal, not the reference condition.

> If the "imagination loop" exists in some or all of the ECUs that

compose the hierarchy, then there must be a means of merging or
switching between the perceptual input from outside the ECU (but
within the organism) and the perceptual input through the imagination
loop. At high levels, when the signals to and from the world outside
the ECU are switched out, we call the process "planning"

Why the need for switching?

I said "merging or switching". The reason is that the perceptual
signal is a scalar. At best, the input from outside and the input
from imagination form a 2-vector.

If we understand how perceptions are constructed
(or can come close in theorizing about them _functionally_) I believe that
will provide us with some key insights into how memory functions. (at least
as far as the PCT model is concerned) I am much less concerned with the
actual mechanisms of how memory works than I am in how it functions and is
integrated into our ability to make perceptions, plan, etc.

No problems here. It's a worthy enterprise. But to deal with it in
terms that can be accommodated in a discussion of PCT, it's a good
idea to be clear about just what construct is under consideration.
There are lots of possible ways imagination and memory may enter the
model, and (Gods of PCT forgive me) everyday experience and much
clinical and experimental psychological evidence tells us that there
are several different kinds of memory (kinds, according to how they
manifest in experience and in behaviour). There's the "know how to
ride a bike" kind, the "I went to the beach last summer" kind, just
to name two.

> Thanks. I hope this message makes it yet clearer.

Yes, and raises some questions as well. Thank you

Good questions are always welcome.

> (2) Remember that outside an ECU is the rest of the organism, and

emotions and the like come from outside the ECU, just as do signals
from the world outside the organism (unless, of course, they are
generated in one of the ECU's processes, meaning that the ECU's job
is to create an emotion corresponding to a reference value for that
emotion).

Not to nit here Martin but 'signals' don't exist in the external
environment.

I didn't say they did. I referred to "signals from outside the
organism" which doesn't mean that the signals themselves exist
outside the organism. Remember Bill's attempt to define a signal?
It's an information-carrying device, and there's no information
without a receiver of that information. The signal is a signal only
when received.

Various energy stimuli do, and I think the distinction here is
important for a number of reasons. Our sensory organs (our 5 main senses)
transduce the _external_ energy stimuli into internal 'signals'. It is these
internal signals and not the external energy that is 'controlled' in our
bodies.

I don't think you'll hear a peep of disagreement from me, Bill P., or
Rick, about that. Or at least, I hope not

Second, emotions, I believe, are a side-effects of what ECU's
produce and do not 'come' from the outside. How about an ECU's job of
creating an emotion based on an error signal?

A whole nother domain of enquiry. It's well worth pursuing.

For what it is worth, I personally don't believe that emotion is a
general consequence of error in just any ECU. But then again, I could
easily be wrong. For all the evidence I have at hand, your guess is
as good as mine. But to talk about it, we need an agreed substrate of
understood material, and on CSGnet, that usually is an acceptance of
the idea of an ECU and of ECUs being organized into a hierachy. When
that basis is not accepted, the deviation has to be spelled out (as I
have done on one or two occasions in the past, and no doubt will
again in the future).

Martin

[David_Goldstein1]

[From David Goldstein (2004.01.02.1812 EST)]
[About Bill Powers (2004.01.02.0846 MST)]

No, I don't see a coherent theory of emotion or behavior in Marc's
communications.

David
David M. Goldstein, Ph.D.

[Richard_Marken2]

[From Rick Marken (2004.01.02.1515)]

Martin Taylor (2004.01.02.1621)--

Marc Abrams (2004.01.02.1309)--

Your notion of the ECU
and its environment makes perfect sense. A disturbance to an ECU is a
disturbance, it doesn't make a difference where the disturbance
actually
comes from, its _external_ to the ECU. But I do not believe that this
is the
picture that is given in the PCT literature. I say this because in
showing
HPCT, in Mind Readings for example pg. 137 is a diagram of a three
level
hierarchy. In it the ECU's pass through the 'environment' only at the
first
level and are exposed to disturbances only at that point. They are
_not_
exposed to 'internal' disturbances on the way down the hierarchy from
level
to level.

True, again. But the idea there is simply to show that multi-level
hierarchies can work in the face of external disturbances. I don't
think it would be too hard for Rick to extend his spreadsheet model ,
as a proof of concept, to incorporate disturbances that enter into
the second or third level perceptions (not to be used as a model of
actual human processing). Rick??

Yes, it's easy to incorporate "internal" disturbances into my
spreadsheet hierarchy. But it's really not necessary to do this since
these internal disturbances already exist in the spreadsheet in the
form of the effects of systems at the same level on the perceptions
controlled by other system at that level.

For example, consider the perception controlled by the first control
system on the left at level 2 (in the spreadsheet, which can be
downloaded from Redirect by those who
don't have it handy). The perception controlled by this system is
computed as follows:

p12 = a1*p11+a2*p21...a6*p61.

where p12 is the perception controlled by system 1, level 2, p11 is the
perception controlled by system 1 level 1, p21 is the perception
controlled by system 2 level 1 and so on. That is, the first subscript
is the system identification and the second is the level of the system
in the hierarchy. The output of this system -- system 1, level 2 --
affects only p11 so variations in p21, p31...p61 are independent of
variations in the output of this system. Thus, by definition, p21...p61
are disturbance variables with respect to p12, the variable controlled
by system 1, level 2. System 1, level 2 controls p12 by varying it's
output -- the reference for p11 -- so that that p11 varies in a way
that compensates for variations in the weighted sum of p21...p61 values.

So the hierarchical control models already handles "internal"
disturbances to controlled perceptions. If controlled perceptions at
level 2 and up in the hierarchy of control systems were based, in part,
on perceptions of emotions -- if, for example, one or more of p21...p61
were emotional perceptions -- then each control system would be
successfully controlling its own perceptions whiles compensating for
the independent effects of emotional disturbances to these perceptions.

Best regards

Rick

···

---
Richard S. Marken
marken@mindreadings.com
Home 310 474-0313
Cell 310 729-1400

[Marc_Abrams1]

from [Marc Abrams (2004.01.02.1829)]

[From David Goldstein (2004.01.02.1812 EST)]

No, I don't see a coherent theory of emotion or behavior in Marc's
communications.

Gee, what a surprise. ha, Ha, Ha, ha, ha, ha. I thought you weren't going to
post again on this. I guess you just couldn't pass up this opportunity
HA,HA, HA, HA, HA, HA, HA

Lets see, I got 3 private communications asking to be put on my list, Bill
got 2 reply's _so far_ on CSGnet. I'm winning 3 to 2. C'mon folks, lets see
who else has the class of an ape besides these two jerks and actually
'votes' on this nonsense.

Marc

[Bruce_Gregory4]

[From Bruce Gregory (2004.01.01.1840)]

Rick Marken (2004.01.02.1515)

So the hierarchical control models already handles "internal"
disturbances to controlled perceptions. If controlled perceptions at
level 2 and up in the hierarchy of control systems were based, in part,
on perceptions of emotions -- if, for example, one or more of p21...p61
were emotional perceptions -- then each control system would be
successfully controlling its own perceptions whiles compensating for
the independent effects of emotional disturbances to these perceptions.

I find this very helpful, but a bit difficult to reconcile with your
conviction and Bill's that that there is nothing valuable in Marc's
suggestions.

Bruce Gregory

"Everything that needs to be said has already been said. But since no
one was listening, everything must be said again."

                                                                                Andre Gide

[Marc_Abrams1]

from [Marc Abrams (2004.01.02.1833)]

[From Rick Marken (2004.01.02.1515)]

OK Bryan, Ken K. you might want to hit your delete keys now because what I
am about to say is not meant for your ears.

Ok, if you are reading this, don't complain to me about it afterward. I gave
you fair warning.

Marken you're a prick, a coward and an asshole. You sure know how to jump on
someone when their down, but you don't have the balls or guts to face
someone and tell them what you think right to their face. You are a
_LOWLIFE_ of the worst kind, a snake and you are right, You and Bill were
MADE for each other. You guys are like Siamese twins.

Have a happy new year lapdog

Marc

[Marc_Abrams1]

from [Marc Abrams (2004.01.02.1844)]

[From Bruce Gregory (2004.01.01.1840)]

I find this very helpful, but a bit difficult to reconcile with your
conviction and Bill's that that there is nothing valuable in Marc's
suggestions.

Thanks Bruce, but don't waste you breath on this moron. I'm outta here.

Marc

[Bill_Williams1]

[From Bill Williams 2 January 2003 6:30 PM CST]

I regard to the recent scientific and dispationate discussion of emotion, I think it is time to repeat, and for us to relfect upon what Thorstein Veblen once said,

   "All this is simple enough. It is so simple and [it] is
    so obvious that is difficult to see it until it has been
    pointed out, and after it has been pointed out it seems
    to have been unnecessary to speak of it." p. 177.

       "Kant's Critique of Judgment" 1884 reprinted in

          _Essays in Our Changing order_

Bill Williams

···

-----Original Message-----
From: Control Systems Group Network (CSGnet) on behalf of Marc Abrams
Sent: Fri 1/2/2004 5:43 PM
To: CSGNET@listserv.uiuc.edu
Subject: Re: Emotions

from [Marc Abrams (2004.01.02.1833)]

[From Rick Marken (2004.01.02.1515)]

OK Bryan, Ken K. you might want to hit your delete keys now because what I
am about to say is not meant for your ears.

Ok, if you are reading this, don't complain to me about it afterward. I gave
you fair warning.

Marken you're a prick, a coward and an asshole. You sure know how to jump on
someone when their down, but you don't have the balls or guts to face
someone and tell them what you think right to their face. You are a
_LOWLIFE_ of the worst kind, a snake and you are right, You and Bill were
MADE for each other. You guys are like Siamese twins.

Have a happy new year lapdog

Marc

[Bruce_Gregory4]

[From Bruce Gregory (2004.01.02.2021)]

[Marc Abrams (2004.01.02.1833)

I think that upon reflection you will find this post intemperate and
that an apology is called for.

Bruce Gregory

"Everything that needs to be said has already been said. But since no
one was listening, everything must be said again."

                                                                                Andre Gide

[Richard_Marken2]

[From RIck Marken (2004.01.02.1800)]

Bruce Gregory (2004.01.01.1840)--

Rick Marken (2004.01.02.1515)

So the hierarchical control models already handles "internal"
disturbances to controlled perceptions.

I find this very helpful,

Thank you.

but a bit difficult to reconcile with your
conviction and Bill's that that there is nothing valuable in Marc's
suggestions.

I didn't say that there is nothing valuable in Marc's suggestions. I
just said "No" in reply to Bill's question about whether anyone sees in
Marc's recent communications any coherent theory of emotion, behavior
or experience. I think there is much of value in Marc's suggestions,
if only because they stimulate what I often find to be very interesting
replies.

Best regards

Rick

···

---
Richard S. Marken
marken@mindreadings.com
Home 310 474-0313
Cell 310 729-1400

[Bill_Williams1]

[From Bill Willliams 2 January 2004 8:40 PM CST]

I thought Marc's argument all along had been that PCT, and HPCT didn't provide an adaquate treatment of emotions. And, that he was interested in pursuing the development of a more inclusive treatment of emotions, not that he _had_ or was ready to expound the principles of such a system.

My impression has been that PCT or HPCT has a conceptually adaquate treatment of emotions, but there hasn't been that much attention given to developing an exposition of what is already there. However, Franz Plooji's _Why Babies Cry_ might be considered in some sense to contain a lot _about emotions_ even if it isn't directly a _theory_ of emotions, let alone a comprehensive treatment.

Whether, Marc is in a position to make progress with his program is, I think, uncertain-- as I think Marc might admit. I think what Marc could claim without much risk of contradictions is that developing a systematic account of emotions remains project that hasn't yet been carried out with any comprehensiveness.

How the oath with seven letters comes into the question has me puzzled.

Bill Williams

[Bill_Powers]

[From Bill Powers (2004.01.02.2041 MST)]

Bill Willliams 2 January 2004 8:40 PM CST--

Whether Marc is in a position to make progress with his program is, I
think, uncertain-- as I think Marc might admit. I think what Marc could
claim without much risk of contradictions is that developing a systematic
account of emotions remains project that hasn't yet been carried out with
any comprehensiveness.

You are right on all counts. I have no objection to Marc's developing his
own proposals for a theory of emotions, perception, control, hierarchies,
and so forth. While I am acutely aware of the many shortcomings of my own
attempts to construct a complete theory, I would not like to see us
retrogress by wasting our time on even less-developed conjectures and
assertions. The questions I raised were legitimate and important. The
answers were inadequate. That's fine -- all our answers are inadequate. But
some are less inadequate than others, and those are the ones I prefer.

Best,

Bill P.

[Bruce_Gregory4]

[From Bruce Gregory (2004.01.03.1051)]

As Rick reminded us, the hierarchical model is capable of accommodating
memory, imagination, and emotion, both as controlled perceptions and as
disturbances to controlled perceptions. Is it fair to say that the
absence of emphasis on these topics in HPCT is therefore a reflection
of the lack of data?

Bruce Gregory

"Everything that needs to be said has already been said. But since no
one was listening, everything must be said again."

                                                                                Andre Gide