Levels of control and consciousness, etc.

CSGnet Archive CSG1996
1

[From Bill Powers (960430.1700 MDT)]

Bruce Gregory (960430.1215 EDT) --

     Bill and Rick are of the "Soft" school. PCT describes are
     hierarchical control system, but an autonomous "force"
     (consciousness) can intervene and redirect elements of the system.
     Peter belongs to the "Hard" school which maintains that it is "PCT
     all the way down" (more accurately "PCT all the way up"). In the
     Hard school, there is no need to invoke a consciousness or
     awareness out side the hierarchical system -- consciousness and
     awareness are part and parcel of the hierarchical system.

Nice distinction, but you got it backward. Rick and I are the Hard
school, Peter is describing the Soft school.

Rick, Peter, and I, I presume, begin with the same observations: there
seems to be an observer and an observed, with the observer being some
sort of receiver of information from the observed. Rick and I do not
postulate any kind of receiver; we just say there is one. It might
consist of neural circuits (a higher-level system), or ectoplasm, or a
fourth-dimensional pseudopod of a Cosmic Panjandrum. Who cares? We don't
know what it is, so the best we can do is try to characterize what it
does.

The view Peter puts forth, on the other hand, attempts to explain the
nature of this receiver. It is said to consist of "reverberating
reentrant (or recursive) signals." It is "the whole system." It is
"coherence." It is "patterns." This, I submit, is the Soft, or Arm-
Waving, terminology. These terms don't actually represent descriptions
of mechanisms; they're just suggestive words unconnected to any specific
functions. They create the impression that one is talking about a
mechanism, but somehow the mechanism itself is not described, nor is it
explained how such a mechanism could result in the observed phenomena.
That is the essence of the Soft approach.

The Hard approach doesn't fingo any hypotheticorum. If we don't know how
it works, we say so. We try to think of experiments that would at least
shed some light on the function of the apparent phenomenon. We don't
say, "Well, like, it's just the Whole Thing, man, it's all this
Coherence and Reverberation, you know, all that Reentrant Self-Reflexive
snake-up-its-own-ass stuff that you have to just sort of feel because
definitions are like restrictive, man. But it's all, you know, neurons."

Sorry, Peter -- got carried away a bit with memories of annual meetings
of the ASC ;-)>

I can say this with confidence: I don't know what awareness is or how it
works, and neither does anyone else.

···

-----------------------------------------------------------------------
Richard Thurman (960430.0815) --

     So I ask, once again, is awareness necessarily the same as
     attention and/or consciousness?

Here are the theoretical distinctions I make.

1. Perception. The existence of a signal in an afferent pathway at any
level in the hierarchy. Perceptions are necessarily present when control
is taking place, because they are what is controlled.

2. Awareness. Receiving information about the activitiies of some system
in the hierarchy (by what, I refuse to try to say). It seems that what
awareness receives are copies of perceptual signals, but maybe other
signals in the hierarchy are accessible, too. Awareness receives
_selectively_ -- that is, at one time it may be receiving copies of one
set of perceptual signals, and at another time copies of another set.
This mobility of the locus of awareness is the main import of the term
"attention." These sets of perceptions may come from anywhere in the
hierarchy, bottom to top level. We can be aware of intensities, and of
system concepts, and all types of perceptions between. We are seldom
aware of more than one or two levels of information at a time, and then
only some of the existing perceptual signals are connected to awareness.
It is also possible for awareness to be recieving no signals. This is
the state of "awareness without object" that has been discussed by
numerous people. The receiver, as it were, can be tuned between
stations.

3. Consciousness. The state that consists of awareness coupled to
specific segments of the hierarchy. The _content_ of consciousness is
the activity in this subset of the learned systems, including
interpretations of experience, plus goals, plans, analyses, theories,
imagination, and everything else that the hierarchy can do. When
awareness is not connected to a given segment of the hierarchy, the
control systems involved may or may not go on working the same way as
when awareness is connected. This is something that requires
experimental investigation based on our ability to measure the
properties of operating control processes.

I think that Rick and I and others agree that the center from which one
observes the world seems to stay with the center of awareness -- the
receiver -- rather than with any particuilar objects of awareness (such
as thoughts).

If you want to say that this awareness thing is a neural network, go
ahead, but you'll have to explain how it makes the world appear to us as
it does. If you can't explain that, you might as well propose that it's
a Soul, or a Universal Oneness of Being, or a Cosmic Synchronicity, or
any other words that make you feel good. But it's easiest to answer such
questions as Mark Twain answered the lady who asked him a very important
and difficult question. "Madame," he said (more or less), "I am
delighted to say that I can answer your question immediately. I don't
know."
-----------------------------------------------------------------------
Stefan Balke (960430.1530 CET) --

     This is a very interesting perspective, but how can you be sure
     whether you are just inventing a rationalization or whether you
     really raised your consciousness (going up a level) in order to see
     the higher order perception?

A fair question: which part of this perspective is observational, and
which is theoretical?

Perhaps you can help to answer the question. Let me ask you a question
in return.

Are you really interested in getting an answer to your question? That
is, do you really want to know how we can be sure, or is this just a way
of expressing a sense of skepticism? I'm not asking you to defend your
question, merely to report what you find to be the answer to my
question.
-----------------------------------------------------------------------
Bruce Gregory (960430.1020) --
Martin Taylor (960429 16:10) --

RE: situational awareness

Martin:
       For example, a fighter pilot's "situational awareness" is usually
     taken to mean the extent to which he is able to take control of
     different potentially important perceptions

Bruce:
     I was simply trying to say that an accident seems often to be the
     result of a loss of situational awareness triggered by the effort
     to control one anomalous variable.

It seems to me that situational awareness, as being defined here, refers
to whether or not a pilot is perceiving and controlling certain
variables (and not others), not whether the pilot is aware of perceiving
and controlling those variables. By disturbing situational variables and
observing whether a pilot's actions resist the disturbances, we can test
to see if there is control (and therefore, necessarily, perception) of
the variables. But how do you test whether this control involves
awareness (i.e., is conscious) without asking the pilots? You can infer
awareness by putting yourself in the same situation (in imagination),
but how do you test this inference?

I agree with Bruce in his observation that if there is no startup
transient when a pilot begins to control a variable, the error must have
been close to zero at the time of startup. Unless this is purely
fortuitous, control must have been going on all the time. If the error
was not zero, then there has to be a startup transient, an initial
action that brings the controlled variable to the vicinity of the
intended state.
-----------------------------------------------------------------------
Best to all,

Bill P.

2

[From Bruce Gregory (960501.1200 EDT)]

Bill Powers (960430.1700 MDT)
Rick "Bulldog" Marken (960430.1510)

Well, if you guys are going to demand a _real_ model, count me among
those who

  can say this with confidence: I don't know what awareness is or
  how it works.

(I think I favor the "fourth-dimensional pseudopod of a Cosmic
Panjandrum" mechanism, but I'm not willing to go out on a limb for
it...)

Bruce G.

3

[From Bruce Gregory (960501.1200 EDT)]

Bill Powers (960430.1700 MDT)
Rick "Bulldog" Marken (960430.1510)

Well, if you guys are going to demand a _real_ model, count me among
those who

can say this with confidence: I don't know what awareness is or
how it works.

(I think I favor the "fourth-dimensional pseudopod of a Cosmic
Panjandrum" mechanism, but I'm not willing to go out on a limb for
it...)

Bruce G.

Wow! I am really enjoying this discussion as a lurker. Let's see
if I can control my experience of my foot and mouth.

I also,
   >> don't know what awareness is or how it works.

but I hope you will consider the following hypothosis:

When I experience my sense of self, some part of me is aware
of my awareness of the world. But not in an infinite
regression of homunculii in space, but a single loop in
time. And that requires memory. Short term memory. But not
ordinary short term memory but an uncounscious memory that
only operates over the last 100 millisecs or less. The
memory of what I just thought, what I just felt.

This is functionally separate from the usual memory system
we use to remember a phone number for a few seconds. This is
a smaller loop that serves as a 50 millisec duration
reference point for our sense of self. We don't even know
it is there, since it is the mechanism that creates our self
awareness. (By the way, the 50 milliseconds is just a
guess).

It is a memory of being here a moment ago. I don't mean half
a second or one second or five seconds ago. I mean 10 to 100
millisecs ago - down below the lowest know granularity of
the human mind's ability to distinguish events in the
outside world. This memory has nothing to do with the
ability to recall and then talk about a piece of
information. We don't use this memory, we live by means of
it. We keep up with who we are by the sense of continuity it
gives us. (speaking from the unconscious) I am ME, because I
feel my self to be a being with duration. I am here now, I
can feel my duration in the moment (50 millisecsounds ) and
I know who I am because I know I have been this same person
in the immediate past. I came from the past - see my path
through time? I am the same person I was a moment ago with a
few slowly changing variables. (Remember, all this is
nonverbal knowledge).

If we could erase your 50 millisec memory loop just once,
you would be reset. You would look around like a tabla rosa
with no sense of who you are. Sure, you could remember
events from the past - if you had a reason to, but you
probably would not have any reason to remember until larger
events in the environment restored your connection to those
older events. Has anybody ever been hit over the head so
that when you came to, you didn't know who you were, where
you were or anything else for a few moments? You were reset
- newly conscious and possessing a sense of self but lacking
the normal data that would fill out that sense of self.

Take all this with a bit of salt and don't let the details
get in the way. I am trying to build a picture of something
I know no more about than any one else. I just see nothing
going on inside me, in terms of self awareness, that requires
any magic or any mechanism more complex than this proposed
memory loop.

This sense of self awareness - that we find so hard to
attribute to something we can build - like a robot, is
a self-reference memory loop that has a duration less
than we can grasp with our... primary mental fingers(?).
It's a hardware thing... microcode.

Any reactions are welcomed. John

John Burch, ProAccess Systems Inc. , (512) 458 1855
---- Lets try to leave this world a better place than we found it.----
- Considering the world situation, that should not be too hard to do.-

4

[from Peter Cariani (960501)]

[From Bill Powers (960430.1700 MDT)]

Nice distinction, but you got it backward. Rick and I are the Hard
school, Peter is describing the Soft school.

It matters not to me who is Hard and who is Soft. I think Greg means
it in terms of Hard AI vs. Soft AI, rather than Hard Science vs.
Soft Science.

Rick, Peter, and I, I presume, begin with the same observations: there
seems to be an observer and an observed, with the observer being some
sort of receiver of information from the observed. Rick and I do not
postulate any kind of receiver; we just say there is one. It might
consist of neural circuits (a higher-level system), or ectoplasm, or a
fourth-dimensional pseudopod of a Cosmic Panjandrum.
Who cares? We don't know what it is, so the best
we can do is try to characterize what it does.

This is fine as far as it goes, and this is the perspective that
I take in my work on the neural basis of pitch perception. But it
doesn't offer any explanation or account whatsoever of the relationship
between awareness and neural events.

The view Peter puts forth, on the other hand, attempts to explain the
nature of this receiver. It is said to consist of "reverberating
reentrant (or recursive) signals." It is "the whole system." It is
"coherence." It is "patterns." This, I submit, is the Soft, or Arm-
Waving, terminology. These terms don't actually represent descriptions
of mechanisms; they're just suggestive words unconnected to any specific
functions. They create the impression that one is talking about a
mechanism, but somehow the mechanism itself is not described, nor is it
explained how such a mechanism could result in the observed phenomena.
That is the essence of the Soft approach.

I offer the perspective as a heuristic for thinking about the problem,
and a general philosophical position that has not, to my knowledge,
been present in current discussions of mind-brain relations.
I don't claim that it specifies the answer,
only that the answer could involve an explanation of this sort.

The Hard approach doesn't fingo any hypotheticorum.

You assume consciousness, awareness, and perception without
specifying what any of these things are. What exactly
are these neural signals you are hypothesizing?
Where is the "observer"? You aren't even trying to
solve the same problem, so why compare the two explanations?
Heaping disdain on a rudimentary explanation of a phenomenon
is not helpful if you have no intention of even trying to
explain it and have no better one to offer.

If we don't know how
it works, we say so. We try to think of experiments that would at least
shed some light on the function of the apparent phenomenon. We don't
say, "Well, like, it's just the Whole Thing, man, it's all this
Coherence and Reverberation, you know, all that Reentrant Self-Reflexive
snake-up-its-own-ass stuff that you have to just sort of feel because
definitions are like restrictive, man. But it's all, you know, neurons."

It is possible to distinguish systems that are self-producing from
those that are not, and it is possible to develop operational
definitions for deciding when a signalling system has coherence,
reverberations, and when it regenerates its own signals, i.e.
when the system exhibits a given mode of organization. The
systems-theoretic classification problems are tractable. The
problem is identifying and getting access to the neural signals
involved, rather than specifying how to operationally define
a functional organization. As I said before, I am actively
investigating the neural signals subserving pitch perception,
and have a very good idea of what these are like at several
stations in the auditory system, and I am trying to think through
how these signals might be processed in higher centers,
what kinds of neural organizations/architectures/processing
might be involved. It's not as if my statements are
flippant or unconsidered. Once one has operational
definitions of putative neural organizations that might support
conscious awareness, then one has empirically-testable hypotheses.
One looks to see what effects disruptions in the organization have
in terms of the awareness of the subject. This is anything
but gratuitous hand-waving, despite the present difficulties in
subjecting this proposal to actual empirical test.

I can say this with confidence:
I don't know what awareness is or how it
works, and neither does anyone else.

Well, I agree with you, I don't know either, but I'm actively trying
to understand using every bit of evidence I can find.

We are all of us in the gutter,
but some of us are looking at the stars.

Peter Cariani

5

[From Richard Thurman (960501.1315)]

Bill Powers (960430.1700 MDT)

Richard Thurman (960430.0815)

    So I ask, once again, is awareness necessarily the same as
    attention and/or consciousness?

Here are the theoretical distinctions I make.

1. Perception. The existence of a signal in an afferent pathway at any
level in the hierarchy. ....

2. Awareness. Receiving information about the activitiies of some system
in the hierarchy (by what, I refuse to try to say). ....

3. Consciousness. The state that consists of awareness coupled to
specific segments of the hierarchy. ....

Thanks Bill. At least now I know how you are using the terms. This will
help me to interpret your posts better. I do not see, however, a
difference between awareness and consciousness. I must assume you really
mean it when you use awareness and consciousness interchangeably.

I think that Rick and I and others agree that the center from which one
observes the world seems to stay with the center of awareness -- the
receiver -- rather than with any particuilar objects of awareness (such
as thoughts).

Yes, I believe I can agree with that. "I" am not my thoughts.

If you want to say that this awareness thing is a neural network, go
ahead, but you'll have to explain how it makes the world appear to us as
it does.

No, I would say awareness is the reception of input from the environment.
It can be inferred that one is aware of certain perceptions via the test.

Now if you had written 'this consciousness thing' above, instead of "this
awareness thing" I would have agreed completely with you. Consciousness
seems to be external to PCT. It is not within the model.

If you can't explain that, you might as well propose that it's
a Soul, or a Universal Oneness of Being, or a Cosmic Synchronicity, or
any other words that make you feel good.

That would be beyond the realm of PCT and irrelevant to this list.

But it's easiest to answer such
questions as Mark Twain answered the lady who asked him a very important
and difficult question. "Madame," he said (more or less), "I am
delighted to say that I can answer your question immediately. I don't
know."

Yes. It is easier to answer such questions that way.

···

-----------------------------------------------------------------------

RE: situational awareness

It seems to me that situational awareness, as being defined here, refers
to whether or not a pilot is perceiving and controlling certain
variables (and not others), not whether the pilot is aware of perceiving
and controlling those variables. By disturbing situational variables and
observing whether a pilot's actions resist the disturbances, we can test
to see if there is control (and therefore, necessarily, perception) of
the variables.

I like that reasoning. It seems to me that one can test for situational
awareness by doing the test for the controlled vairable. If the subject
(say in a flight simulator) can keep several situational variables under
control despite disturbances to those variables (or at least is resisting
those disturbances) then one can infer that the pilot is situationally
aware. Even if the subject can not articulate what he was doing, or can
not indicate conscious effort, one could show that he was attempting to
keep the situation under control.

Maybe this is too big a leap but it seems to me that studies of SA should
be redirected into studies of perceptual control.

But how do you test whether this control involves
awareness (i.e., is conscious) without asking the pilots? You can infer
awareness by putting yourself in the same situation (in imagination),
but how do you test this inference?

I would be content to infer that if the perception is under control then
at least one control system can perceive and is aware of the variable.

I agree with the assertion that one should not be too hopeful of verifying
if situational control is conscious (or unconscious) at this point.

Rich

--------------------------------------------------
Richard Thurman
Armstrong Lab
6001 S. Power Rd. BLDG. 558
Mesa AZ. 85206-0904
(602)988-6561
Thurman@hrlban1.aircrew.asu.edu
---------------------------------------------------