Amazing!

[Bill_Powers]

[From Bill Powers (2003.06.02.0720 MDT)]

Bruce Gregory (2003.0601.1440)--

Observation. When I concentrate on one thing I cease attempting to
achieve competing objectives. The simplest explanation is inhibition of
those objectives. The gain appears to increase wherever attention goes,
but it may be that attention emerges wherever gain is highest.

What do you mean by "concentrating on one thing?" It seems to me that
"concentrating" requires an explanation just as much as the things you're
using it to explain. If you're seriously proposing "inhibition of competing
objectives," you need to explain what inhibition is, what does the
inhibiting, and how objectives can "compete". The term "attention" is not
self-explanatory, either, not is it clear how gain might depend on it, or
how it might depend on gain. Is this what you end up with after applying
Occam's Razor? A lot of pieces?

You're balking at a rather simple concept, that of viewpoint, but offering
as an alternative a collection of undefined terms with no apparent
organized idea behind them. What's the reason for this resistance? For it
seems to me that you _are_ resisting, rather adamantly, without really
getting to the core objection.

The basic question here is whether phenomena as we consciously experience
them require an experiencer as well as the perceptual signals that are
experienced. The puzzle lies in the fact that we can't simply observe the
Observer, because there is only one Observer in each of us (as various
people have independently put it), so when observing each of us is always
observing something else.

Looked on simply as a technical problem rather than a philosophical
conundrum, the answer appears to be fairly simple. A signal means nothing
until a receiver receives it. If you have a radio, you can use it to detect
radio waves, but of course it does not detect radios. Without the radio,
the radio waves will never reveal themselves to carry music, news, or
advertisements. A teenager listening to music on a radio does so without
ever wondering about the radio itself; it's only the music that is of
interest. The policeman reporting a robbery is not concerned about how
microphones work or how his voice reaches the dispatcher's ears. It's
getting help that matters.

It would be quite possible for any of these radio-users to be skeptical
about the function of the radio. After all, all they ever actually observe
is the output of the radio: the music, news, the dispatcher's voice. Why
complicate matters, they might say, by speaking of radios? The only
important questions are what music we like best, whether the news is about
a disaster that affects us, and what the ads tell us about how to achieve
happiness. Is Lawrence Welk a better musician than Ice-T? That is the
_important_ question.

I see that I could have improved this example by saying that radio users
are skeptical about the function of ears and the auditory parts of the
brain. We don't experience the auditory parts of the brain; we just hear
the sounds. But is that enough reason to avoid talking about hearing
itself? Shall we say that the sounds reaching our ears are real, but there
is nothing receiving the sounds?

You wouldn't have any trouble with the concept of viewpoint if it were
directly demonstrated to you in an MOL session. You wouldn't have any
trouble with it if you just relaxed and considered the many times every day
when your viewpoint naturally shifts: when you were attending to one aspect
of experience, and suddenly became aware of a quite different aspect,
without any change in the scene. In fact, I don't think we would have any
arguments if we simply described the phenomenon in question.

Where we depart seems to be in how we explain these changes. You seem to
want to explain them entirely in terms of the array of perceptual signals,
so if something new appears in awareness, a new perceptual signal must have
arisen. I, on the other hand, am claiming that sometimes the change is due
to the observer ceasing to receive information from one place and starting
to receive it from another place, while the perceptual signals themselves
remain the same as before. If the picture on the screen changes, is it
because the camera at the studio has pointed in a new direction, or because
somebody tuned your TV to a different channel while the camera went on
broadcasting the same scene?

How can we know if the perceptual signals have changed as opposed to
awareness tuning in different, but already-existing, perceptual signals?
One way is to establish that there is a control task going on in which the
perceptual signal in question plays an essential part. If control continues
when that perception disappears from conscious experience, the perceptual
signal must continue to exist while not in awareness. The same is true when
a perception appears in awareness which we suspect for various reasons must
have been there previously, but unnoticed (the sensations from the seat of
your pants right now).

This sort of evidence is indirect, but in the case of the Observer that
seems to be all that is available for rational consideration. It's quite
analogous to trying to find out if another person is perceiving something
we perceive. We can never get inside the other person's head to check for
ourselves, but by performing certain tests, and assuming that the other
person is organized basically the same way we are, we can make strong
inferences.

I imagine you're aware of the Gibsonians. They appear to believe that all
the brain does is "pick up" information that is actually there in the
environment. They don't want to talk about perception, interpretation, OR
viewpoints. All the information needed to guide movements, including goals,
says one person of this persuasion I have conversed with, is in the "optic
array," which has nothing to do with eyes or optic nerves, but is somehow
there in the way the visual field flows when you move. How there can be a
visual field in the first place somehow never comes up.

My feeling about your reactions to "viewpoints" is very reminiscent of the
feeling I get from Gibsonians in talking about perception. It's not that
Gibsonians think the nervous system doesn't do anything. They just don't
want to talk about it. It's not that you claim to be able to get along
without awareness. But somehow it doesn't seem important to ask how it works.

Best.

Bill P.

[Bruce_Gregory4]

[From Bruce Gregory (2003.0602.1312)]

Rick Marken (2003.06.01.2220)]

  > This does not seem to me to be the phenomenon of "point of view" that we

been discussing.

I hope not. I am trying to discourage the notion of point of view. I find no
need for it in the HPCT model.

The simplest explanation is inhibition of
those objectives. The gain appears to increase wherever attention

goes,

but it may be that attention emerges wherever gain is highest.

This does not seem to me to be a PCT explanation of the phenomenon you
describe.

O.K. What is?

[Richard_Marken2]

[From Rick Marken (2003.06.02.1330)]

Bruce Gregory (2003.0602.1513)--

Bill Powers (2003.06.02.1147 MDT)

I love the way you nitpick my suggestions and 'explain' consciousness by
postulating it ("the observer").

I didn't see any of that in Bill's post. All I saw were some pretty clear
explanations of what was wrong with your explanations of attention.

Sometimes I have to wonder if you are as open
to new suggestions as you would like to think you are.

I think Bill is very open to new suggestions. I think your problem comes from the
fact that he carefully evaluates new suggestions and incorporates them into his
modeling efforts only if they have scientific merit.

Best

Rick

···

--
Richard S. Marken, Ph.D.
Senior Behavioral Scientist
The RAND Corporation
PO Box 2138
1700 Main Street
Santa Monica, CA 90407-2138
Tel: 310-393-0411 x7971
Fax: 310-451-7018
E-mail: rmarken@rand.org

[Bruce_Gregory4]

[From Bruce Gregory (2003.0602.1332)]

Bill Powers (2003.06.02.0720 MDT)

What do you mean by "concentrating on one thing?"

I am looking over my should for oncoming traffic. I am not hearing the radio,
or worrying about what I will have for lunch. If you try to speak to me about
something, I will ignore you.

It seems to me that

"concentrating" requires an explanation just as much as the things you're
using it to explain.

When the gain on a control loop is high, this system inhibits (lowers the gain
on) competing control loops. I'm most familiar with the phenomenon as it is
related to vision, but it seems perfectly reasonable that complex system use
this mechanism to prevent unnecessary conflict.

If you're seriously proposing "inhibition of competing
objectives," you need to explain what inhibition is, what does the
inhibiting,

See above.

and how objectives can "compete".

Higher level systems compete when they try to set differing reference levels
and gains for the same lower level systems.

The term "attention" is not
self-explanatory, either, not is it clear how gain might depend on it, or
how it might depend on gain.

I conjecture that attention is linked to gain. In loose language, which I
prefer to avoid, the gain of a loop determines whether the perception 'shows
up' in consciousness.

Is this what you end up with after applying

Occam's Razor? A lot of pieces?

No comment.

You're balking at a rather simple concept, that of viewpoint, but offering
as an alternative a collection of undefined terms with no apparent
organized idea behind them.

Sorry, but the terms are both defined and organized as far as I can tell. Can
you point out any inconsistencies in what I have said?

  What's the reason for this resistance? For it

seems to me that you _are_ resisting, rather adamantly, without really
getting to the core objection.

Funny, I have exactly the same thought about you.

The basic question here is whether phenomena as we consciously experience
them require an experiencer as well as the perceptual signals that are
experienced.

I find that adding the notion of an "experiencer" does nothing to make the
model more predictive.

The puzzle lies in the fact that we can't simply observe the

Observer, because there is only one Observer in each of us (as various
people have independently put it), so when observing each of us is always
observing something else.

A just-so story, and not a very convincing one as far as I am concerned. Stick
to the model.

You wouldn't have any trouble with the concept of viewpoint if it were
directly demonstrated to you in an MOL session. You wouldn't have any
trouble with it if you just relaxed and considered the many times every day
when your viewpoint naturally shifts: when you were attending to one aspect
of experience, and suddenly became aware of a quite different aspect,
without any change in the scene. In fact, I don't think we would have any
arguments if we simply described the phenomenon in question.

I think you have been seduced by the story that an experience requires an
experiencer. I don't find that story to be helpful. You apparently do.

Where we depart seems to be in how we explain these changes. You seem to
want to explain them entirely in terms of the array of perceptual signals,
so if something new appears in awareness, a new perceptual signal must have
arisen.

More likely the gain on a control loop has changed.

I, on the other hand, am claiming that sometimes the change is due

to the observer ceasing to receive information from one place and starting
to receive it from another place, while the perceptual signals themselves
remain the same as before.

Fine we are not aware of all possible perceptions. I propose that the reason
is associated with the gain on control loops (assuming the perception is
capable of being 'attended to.' What determines whether an observer "receives
information from one place" or another?

My feeling about your reactions to "viewpoints" is very reminiscent of the
feeling I get from Gibsonians in talking about perception. It's not that
Gibsonians think the nervous system doesn't do anything. They just don't
want to talk about it. It's not that you claim to be able to get along
without awareness. But somehow it doesn't seem important to ask how it
works.

I would welcome a model of awareness. I haven't come across anything but hand
waving so far.

[Bill_Powers]

[From Bill Powers (2003.06.02.1147 MDT)]

Bruce Gregory (2003.0602.1332)--

What do you mean by "concentrating on one thing?"

I am looking over my should for oncoming traffic. I am not hearing the radio,
or worrying about what I will have for lunch. If you try to speak to me about
something, I will ignore you.

Those are outcomes or consequences of concentrating. But what do you do in
order to make those outcomes occur?

It seems to me that

"concentrating" requires an explanation just as much as the things you're
using it to explain.

When the gain on a control loop is high, this system inhibits (lowers the gain
on) competing control loops.

What system is this? You're describing results of the operation of some
system, but not identifying the system or saying how it acts. As you
describe it, this unknown system can detect when the gain on a control loop
is high, can detect other control loops that are competing with it, and can
then do something that lowers the gain on the competing control loops. How
does it decide which is the preferred control loop, so it can know which
others to inhibit? Is this a system in the next higher level, or is it part
of the lower control loops? If in a higher level, what variable does it
control, and with respect to what reference level? Does every level act in
this way with regard to every subsidiary level? It seems to me that your
proposal creates rapidly-proliferating difficulties for the model.

I'm most familiar with the phenomenon as it is
related to vision, but it seems perfectly reasonable that complex system use
this mechanism to prevent unnecessary conflict.

You're speaking of the mutual inhibition between retinal cells, I presume,
which has the effect of enhancing edges. This has an effect on the relative
magnitudes of perceptual signals, but as far as I know has no effect on
varying loop gain. The utility of edge-enhancement is not understood,
especially considering that in the world of normal vision, edges do not
appear to be enhanced. Mach bands appear only under special circumstances;
most edges seem to be sharp and abrupt. It's possible that this enhancement
simply compensates for computational or optical blurring that would
otherwise occur at edges.

The problem with mutual inhibition among control systems is that this
creates interactions among them, so the perceptual signals in one system
can't easily be controlled independently of perceptual signals in others.
One main thesis growing out of HPCT is that control systems at a given
level will tend to become orthogonal, so altering one perception has the
least possible disturbing effect on others. If mere magnitude of a
perceptual signal is enough to cause inhibition in neighboring systems,
this orthogonality is destroyed, for we can no longer vary the magnitudes
of individual control systems independently of the others (without contlict).

Note that in the retina, the connections that create mutual inhibition are
permanently in place; this effect occurs all of the time with all visual
inputs. So there is no option about when the interaction will take place.

Higher level systems compete when they try to set differing reference levels
and gains for the same lower level systems.

In HPCT, they do not compete. They simply send the reference signals to the
target comparators where the effects add up to a net reference signal.
There's no "trying" about it. They do it. Conflict arises not in the
setting of a net reference signal, but in the inability of the lower system
to produce two different perceptual signals at once.

The term "attention" is not
self-explanatory, either, not is it clear how gain might depend on it, or
how it might depend on gain.

I conjecture that attention is linked to gain. In loose language, which I
prefer to avoid, the gain of a loop determines whether the perception 'shows
up' in consciousness.

Loop gain includes all gain factors encountered in one complete trip around
a control loop. It has relatively little effect on the magnitude of a
controlled perceptual signal if it is large enough for good control. For
example, if the reference signal is 100 units and the loop gain is 10, the
perceptual signal will be 10/11 of 100 or 90.9 units. If the loop gain now
rises to 50, or five times the former value, the perceptual signal will
become 50/51*100 or 98 units. That is a 7.8 percent increase in the
perceptual signal. The most the perceptual signal could rise would be for
infinite loop gain, which would make the perceptual signal equal to the
reference signal, or 100 units. That would be a rise of 10% over the size
of the perceptual signal when the loop gain is 5.

The question is, are you saying that it is the closeness of the perceptual
signal to the reference signal that determines whether it "shows up in
consciousness?" If so, we're talking about a 10% change in magnitude, which
doesn't seem enough. On the other hand, maybe you're saying that there is
something about loop gain itself that can determine whether a perceptual
signal will show up in consciousness. If so, we need to postulate some
mechanism for that, unless you want to say that the effect occurs without
any mechanism.

And of course you have said nothing about what consciousness is, that
_anything_ can show up in it.

You're balking at a rather simple concept, that of viewpoint, but offering
as an alternative a collection of undefined terms with no apparent
organized idea behind them.

Sorry, but the terms are both defined and organized as far as I can tell. Can
you point out any inconsistencies in what I have said?

The terms may be organized and defined as you think of them, but if so you
haven't communicated what you're talking about clearly enough for me to
understand what you mean. Telling me what happens as a result of
"concentrating" doesn't tell me anything about what to do to reproduce
those results. Saying that consciousness of a perception depends on loop
gain in the system producing it tells me nothing either about the
connection between loop gain and consciousness, or what you mean by the
word "consciousness." Is consciousness simply a strong perceptual signal?
That doesn't hold water -- too many counterexamples leap to mind -- even
cases where the absence of a perceptual signal can be of tremendous
conscious importance ("Where's the baby?".

>I find that adding the notion of an "experiencer" does nothing to make the
>model more predictive.

Then you're not counting subjective experience as something that requires
predicting. The model as it stands is unable to predict even that there
will be such a thing as subjective experience.

The puzzle lies in the fact that we can't simply observe the

Observer, because there is only one Observer in each of us (as various
people have independently put it), so when observing each of us is always
observing something else.

A just-so story, and not a very convincing one as far as I am concerned. Stick
to the model.

A person who has only one telescope can't look at his telescope through a
telescope. A person with a radio can't use his radio to detect that radio
or any other radio. All our models are just-so stories, of course, and we
look for ways to test them all the time. But we don't get anywhere by
denying phenomena.

On the other hand, perhaps in this case I am speaking to a hierarchy of
control systems in which there is no Observer. That would be embarrassing,
sort of like starting to complain to a recorded telephone message before
realizing that there's no actual person there.

Best,

Bill P.

[Bruce_Gregory4]

[From Bruce Gregory (2003.0602.1513)]

Bill Powers (2003.06.02.1147 MDT)

I love the way you nitpick my suggestions and 'explain' consciousness by
postulating it ("the observer"). Sometimes I have to wonder if you are as open
to new suggestions as you would like to think you are.

  > On the other hand, perhaps in this case I am speaking to a hierarchy of

control systems in which there is no Observer. That would be embarrassing,
sort of like starting to complain to a recorded telephone message before
realizing that there's no actual person there.

Only the Test will tell.

[David_Goldstein1]

[From David Goldstein (2003.06.02.1507 EDT0}

[From Bill Powers (2003.06.02.1147 MDT)]

Bruce Gregory (2003.0602.1332)

The idea of "attention" is one that I have been interested in too.

Bruce Gregory wants to relate it the concept of loop gain. But Bill
Powers is pointing out problems with this approach.

I have always wondered what determined the control systems which were
"activated" at any given moment. It seems clear that not all the control
systems within a person are active all the time. The situation obviously
partly determines what is activated. I look at a scene and this
constrains to some degree what control systems are active. The higher
level goal of a person also obviously partly plays a role. I took my son
to a comic book convention in Philadephia. We agreed to meet at a
certain place at a certain time within the convention center. I was
looking for him at that time and place. I guess these two ideas are what
people mean when they say "top down" and "bottom up."

When one reads, the "top down" and "bottom up" processes are going on
simultaneously. If I am reading something that deviates from what I am
expecting, my eyes reread some of the material.

Rick's "mind reading" task may be a good place to think about attention
within a PCT framework. It seems clear that a person can see all the
different figures at one time. One can attend to all of the figures.
However, only one of the figures can have its movements controlled at
one time as the program is presently written, I believe. The program can
pick out the selected figure by looking at variability. It compares the
expected variability due to the disturbance to the observed variability.

I suppose that it would be possible to have two or more of the figures
controlled. Suppose that one defines some areas of the screen. If a
figure was within its designated area, then the figure is controlled.
The goal is to keep two or more of the figures within their own defined
areas. If this happened, then one could say that the figures were being
controlled. We could start off with two areas and build up from there.
We could find out the largest number of areas a person could monitor and
keep the figures controlled. This would correspond to what others call
"the breadth of attention."

I will stop here before I get too carried away.

···

-----Original Message-----
From: Control Systems Group Network (CSGnet)
[mailto:CSGNET@listserv.uiuc.edu] On Behalf Of Bill Powers
Sent: Monday, June 02, 2003 2:48 PM
To: CSGNET@listserv.uiuc.edu
Subject: Re: Amazing!

[From Bill Powers (2003.06.02.1147 MDT)]

Bruce Gregory (2003.0602.1332)--

What do you mean by "concentrating on one thing?"

I am looking over my should for oncoming traffic. I am not hearing the

radio,

or worrying about what I will have for lunch. If you try to speak to me

about

something, I will ignore you.

Those are outcomes or consequences of concentrating. But what do you do
in
order to make those outcomes occur?

It seems to me that

"concentrating" requires an explanation just as much as the things

you're

using it to explain.

When the gain on a control loop is high, this system inhibits (lowers

the gain

on) competing control loops.

What system is this? You're describing results of the operation of some
system, but not identifying the system or saying how it acts. As you
describe it, this unknown system can detect when the gain on a control
loop
is high, can detect other control loops that are competing with it, and
can
then do something that lowers the gain on the competing control loops.
How
does it decide which is the preferred control loop, so it can know which
others to inhibit? Is this a system in the next higher level, or is it
part
of the lower control loops? If in a higher level, what variable does it
control, and with respect to what reference level? Does every level act
in
this way with regard to every subsidiary level? It seems to me that your
proposal creates rapidly-proliferating difficulties for the model.

I'm most familiar with the phenomenon as it is
related to vision, but it seems perfectly reasonable that complex

system use

this mechanism to prevent unnecessary conflict.

You're speaking of the mutual inhibition between retinal cells, I
presume,
which has the effect of enhancing edges. This has an effect on the
relative
magnitudes of perceptual signals, but as far as I know has no effect on
varying loop gain. The utility of edge-enhancement is not understood,
especially considering that in the world of normal vision, edges do not
appear to be enhanced. Mach bands appear only under special
circumstances;
most edges seem to be sharp and abrupt. It's possible that this
enhancement
simply compensates for computational or optical blurring that would
otherwise occur at edges.

The problem with mutual inhibition among control systems is that this
creates interactions among them, so the perceptual signals in one system
can't easily be controlled independently of perceptual signals in
others.
One main thesis growing out of HPCT is that control systems at a given
level will tend to become orthogonal, so altering one perception has the
least possible disturbing effect on others. If mere magnitude of a
perceptual signal is enough to cause inhibition in neighboring systems,
this orthogonality is destroyed, for we can no longer vary the
magnitudes
of individual control systems independently of the others (without
contlict).

Note that in the retina, the connections that create mutual inhibition
are
permanently in place; this effect occurs all of the time with all visual
inputs. So there is no option about when the interaction will take
place.

Higher level systems compete when they try to set differing reference

levels

and gains for the same lower level systems.

In HPCT, they do not compete. They simply send the reference signals to
the
target comparators where the effects add up to a net reference signal.
There's no "trying" about it. They do it. Conflict arises not in the
setting of a net reference signal, but in the inability of the lower
system
to produce two different perceptual signals at once.

The term "attention" is not
self-explanatory, either, not is it clear how gain might depend on it,

or

how it might depend on gain.

I conjecture that attention is linked to gain. In loose language, which

I

prefer to avoid, the gain of a loop determines whether the perception

'shows

up' in consciousness.

Loop gain includes all gain factors encountered in one complete trip
around
a control loop. It has relatively little effect on the magnitude of a
controlled perceptual signal if it is large enough for good control. For
example, if the reference signal is 100 units and the loop gain is 10,
the
perceptual signal will be 10/11 of 100 or 90.9 units. If the loop gain
now
rises to 50, or five times the former value, the perceptual signal will
become 50/51*100 or 98 units. That is a 7.8 percent increase in the
perceptual signal. The most the perceptual signal could rise would be
for
infinite loop gain, which would make the perceptual signal equal to the
reference signal, or 100 units. That would be a rise of 10% over the
size
of the perceptual signal when the loop gain is 5.

The question is, are you saying that it is the closeness of the
perceptual
signal to the reference signal that determines whether it "shows up in
consciousness?" If so, we're talking about a 10% change in magnitude,
which
doesn't seem enough. On the other hand, maybe you're saying that there
is
something about loop gain itself that can determine whether a perceptual
signal will show up in consciousness. If so, we need to postulate some
mechanism for that, unless you want to say that the effect occurs
without
any mechanism.

And of course you have said nothing about what consciousness is, that
_anything_ can show up in it.

You're balking at a rather simple concept, that of viewpoint, but

offering

as an alternative a collection of undefined terms with no apparent
organized idea behind them.

Sorry, but the terms are both defined and organized as far as I can

tell. Can

you point out any inconsistencies in what I have said?

The terms may be organized and defined as you think of them, but if so
you
haven't communicated what you're talking about clearly enough for me to
understand what you mean. Telling me what happens as a result of
"concentrating" doesn't tell me anything about what to do to reproduce
those results. Saying that consciousness of a perception depends on loop
gain in the system producing it tells me nothing either about the
connection between loop gain and consciousness, or what you mean by the
word "consciousness." Is consciousness simply a strong perceptual
signal?
That doesn't hold water -- too many counterexamples leap to mind -- even
cases where the absence of a perceptual signal can be of tremendous
conscious importance ("Where's the baby?".

>I find that adding the notion of an "experiencer" does nothing to make
the
>model more predictive.

Then you're not counting subjective experience as something that
requires
predicting. The model as it stands is unable to predict even that there
will be such a thing as subjective experience.

The puzzle lies in the fact that we can't simply observe the

Observer, because there is only one Observer in each of us (as various
people have independently put it), so when observing each of us is

always

observing something else.

A just-so story, and not a very convincing one as far as I am

concerned. Stick

to the model.

A person who has only one telescope can't look at his telescope through
a
telescope. A person with a radio can't use his radio to detect that
radio
or any other radio. All our models are just-so stories, of course, and
we
look for ways to test them all the time. But we don't get anywhere by
denying phenomena.

On the other hand, perhaps in this case I am speaking to a hierarchy of
control systems in which there is no Observer. That would be
embarrassing,
sort of like starting to complain to a recorded telephone message before
realizing that there's no actual person there.

Best,

Bill P.

[Bruce_Gregory4]

[From Bruce Gregory (2003.0602.1651)]

Rick Marken (2003.06.02.1330)

I think Bill is very open to new suggestions. I think your problem comes from the
fact that he carefully evaluates new suggestions and incorporates them into his
modeling efforts only if they have scientific merit.

That's very reassuring. Exactly where does "the observer" fit into those
modeling efforts?

[bnhpct]

[From Bruce Nevin (2003.06.02 17:28 EDT)]

[From Bruce Gregory (2003.0602.1332)]

Bill Powers (2003.06.02.0720 MDT)

What do you mean by "concentrating on one thing?"

I am looking over my should[er] for oncoming traffic. I am not hearing the radio, or worrying about what I will have for lunch. If you try to speak to me about something, I will ignore you.

It seems to me that "concentrating" requires an explanation just as much as the things you're using it to explain.

When the gain on a control loop is high, this system inhibits (lowers the gain on) competing control loops. I'm most familiar with the phenomenon as it is related to vision, but it seems perfectly reasonable that complex system use this mechanism to prevent unnecessary conflict.

It may be helpful to flip this over and ask how to explain the phenomenon of ignoring perceptions. Concentrating, ignoring, distraction, and overlooking (failing to notice) are alternative perspectives on the same phenomenon.

Set up a situation where someone with you is talking while NPR is on the radio. Listen to what the person is saying. Listen to what the radio is saying. Listen to both at once. Shift from one to the other. Is this a controlled variable?

There is one system for hearing and comprehending speech. Is this system capable of hearing and comprehending the speech of more than one person at a time?

When the head and eyes are directed back toward oncoming traffic they cannot be simultaneously turned to watch Liv Tylor walking along the other side of the road. This perceptual input is a limited resource. When it is in use by a system controlling moving the car into the flow of traffic it cannot be used by a system controlling the pleasure of Ms. Tylor's company, and vice versa. If in peripheral vision you notice someone walking and briefly glance, perhaps the gain on merging into traffic plummets and the gain on enjoying Ms. Tylor's companionship rises, with the effect of staring across the flow of traffic at her. Perhaps you don't notice the gap in traffic that has just opened, or the car that has just arrived behind yours.

Perhaps "attention" is largely a function of means for control (perceptual inputs, effectors) that connect with many higher level systems, but can be 'used' by only one at a time. (Or perhaps by two or a few in some cases.) Your hands can't be in two places at once, holding something and tossing something else at the same time. Your eyes can't be looking in two directions at once.

That's what sequence perceptions are good for. But that suggests an ability to put arbitrary and unpredicted control tasks under sequence control. In the experiment of shifting from person to NPR and back, are you controlling a sequence? How could there be a generic sequence controller with ad hoc inputs? This 'scheduling' capability has puzzled me for some time. It seems a bit much to ask of our reorganization process.

But that's only a piece of what Bruce G. is driving at. Why do we turn the radio down or off when we're following directions to find someone's house? Why should auditory input compete with remembering the sequence of steps to find the house and looking for street names and house numbers?

Is this actually evidence that the cognitive capacities for language, for music, and for tasks like following driving directions, are general-purpose capacities, not specialized? Why else would these be in competition?

Or, for someone who is convinced of specialized modules, is this evidence that something other than competition for limited resources is at the root of concentrating, ignoring, being distracted, and overlooking?

No real answers, I'm afraid, just more questions.

The basic question here is whether phenomena as we consciously experience
them require an experiencer as well as the perceptual signals that are
experienced.

I find that adding the notion of an "experiencer" does nothing to make the
model more predictive.

For reasons stated previously I don't expect to find a place for the Observer or Witness within the hierarchy. The only evidence is inferred, a confection of memory and imagination.

         /Bruce Nevin

···

At 01:34 PM 6/2/2003, Bruce Gregory wrote:

[Bill_Powers]

[From Bill Powers (2003.,06.03.0614 MDT)]

Bruce Gregory (2003.0602.1513)--

>I love the way you nitpick my suggestions and 'explain' consciousness by
>postulating it ("the observer"). Sometimes I have to wonder if you are as
open
>to new suggestions as you would like to think you are.

I don't treat your suggestions any more critically than I treat ideas of my
own that I intend to offer for serious consideration. If you did more
nitpicking of your own ideas before going public with them you'd have fewer
bugs to apologize for.

I have never attempted to "explain" consciousness. I'm simply trying to pin
down the phenomenon so we can agree on what we're talking about. You
haven't offered any explanations, either.

This is my first experience of arguing, live, with someone who didn't
believe himself to be a perceiver or experiencer (if that's not a correct
understanding of your position, please correct me). Of course I've read
about such things, especially the popular criticisms of Descartes and his
"mind-body split" idea. But I've always thought that was basically an
argument with religious interpretations, not an actual denial that
individuals experience the world and themselves as conscious observers.
It's hard to understand how anyone could extend such a theoretical position
to his OWN experiences.

The only factual arguments I can bring to bear on this have to do with
perceptions or implied perceptual signals that can be present both with and
without awareness of them. That seems to me a sufficient reason for saying
that awareness and the existence of perceptual signals are not synonymous.
The method of levels has provided what appears to be corroboration of this
phenomenon at a number of levels of abstraction. Furthermore, most people
with whom I have discussed this phenomenon say they have a single
experience of observing that persists across all the different mental and
perceptual content that they can see going on inside them. That is, they
see the world as if from one consistent place or viewpoint, though nobody,
apparently, can turn around and examine that viewpoint. It's analogous to
looking at the outside world with your eyes; you can't look at your own
eyes directly, though you may occasionally see reflections of them. Yet all
the evidence supports the idea that you need your eyes to do any looking at
all.

Enough people have denied this idea to make me pause and wonder what I'm
missing. Is there some simple argument that I might grasp that shows what's
wrong with the idea of an Observer? Nothing I have heard sounds like
anything more convincing than an insistent denial. No development of an
argument with premises, reasoning, and conclusions. Just a lot of emotional
pressure against anyone so stupid as to think the mind and body are
different. If I'm being stupid about this I'd like to stop, but I'm sure
not going to stop just because someone tells me I'm being stupid. That
would indeed be stupid.

Best,

Bill P>

···

> On the other hand, perhaps in this case I am speaking to a hierarchy of

control systems in which there is no Observer. That would be embarrassing,
sort of like starting to complain to a recorded telephone message before
realizing that there's no actual person there.

Only the Test will tell.

[Bill_Powers]

[From Bill Powers (2003.06.03.0702 MDT)]

Bruce Nevin (2003.06.02 17:28 EDT)--

It may be helpful to flip this over and ask how to explain the phenomenon
of ignoring perceptions. Concentrating, ignoring, distraction, and
overlooking (failing to notice) are alternative perspectives on the same
phenomenon.

I agree, but add that any explanation, to merit the name, has to offer at
least some kind of mechanism, not merely a coincidence or sequence of
happenings in time or space. I don't think we're ready for explanations
yet. If the loop gain of a control system seems elevated at the same time a
person reports higher awareness of the associated perception(s), which is
cause and which is effect -- if either? It's possible, of course, that both
are effects of some unrecognized third process; you'd have to be able to
manipulate one and observe the effects on the other to decide. And that
would merely establish which phenomenon it is that requires explanation.

In speaking of "ignoring perceptions" you say, "Concentrating, ignoring,
distraction, and overlooking (failing to notice) are alternative
perspectives on the same
phenomenon."

Yes, they are, and in each case, the implication is that the perception
continues to be there while not being consciously observed. You don't say
you're "ignoring" a pink elephant in the corner if there is no pink
elephant in the corner. You can't be "distracted" from a perception if it's
not still present even after your -- something -- has been distracted. You
can't overlook something if it's not still there while you're overlooking
it. And from the other angle, you don't say you've "concentrated" on one
sound if it simply becomes louder relative to other sounds.

So the basic phenomenon is still that there is a difference between the
observer and the observed. For the observed to be ignored, it must be
present while the observer is not observing it. And for the observed to be
concentrated upon, some action by the observer is called for, not a change
in the observed. When you concentrate on reading the words in a book, the
words you see don't change. The manner of observing them changes.

It would be interesting to do some experiments with "attention," or rather
with distractions (since we can't observe or measure attention directly).
It's not hard to get people to shift attention from one perception to
another; the question we can answer experimentally is what effect this
shift might have on the parameters of control when the two (or more)
perceptions are continuously controlled? My image of the experiment was
nicely captured in a recent TV commercial: a waitress pouring coffee into a
cup is distracted by a good-looking laptop computer, and the cup overflows
until she notices it and stops the flow. Of course that situation involves
moving the eyes away from a controlled variable that really needs vision to
be controlled, but perhaps we could think of two different aspects of the
same field of vision that could be under simultaneous control (by X and Y
motions of a mouse?). We can measure control parameters pretty well, so
maybe we could see any effects of attention shifts.

But that's only a piece of what Bruce G. is driving at. Why do we turn the
radio down or off when we're following directions to find someone's house?
Why should auditory input compete with remembering the sequence of steps to
find the house and looking for street names and house numbers?

This is an important question and needs to be investigated -- not so much
the "why" as the "what." Clearly, perceptions can interfere with each
other. But that is not a matter of attention; it's a signal-to-noise ratio
problem, isn't it? You turn down what is considered noise to make a needed
signal more distinct from the background. That doesn't require any shifts
of attention, although they could also be going on.

Clearly, there appear to be limits on the number of control processes that
can be going on at once in conscious experience. What happens, however, as
those control processes occur again and again until they become so familiar
that they are carried out automatically, meaning without conscious
attention? Do the same limits then hold? Is the limit really a limit on
simultaneous control processes, or is it a limit on what can be
simultaneously in process of being learned? The "five chunks plus or minus
two" doesn't seem to apply to automatic control processes, of which
hundreds might be going on at once.

And what is the difference between a control process being carried out
consciously and the same process being carried out automatically?

Of course we have to separate limits imposed by the bandwidth of attention
from those caused by simple mechanical conflicts. The waitress could have
overflowed the coffee cups simply because her eyes couldn't be looking in
two directions at once. An alternative scenario might have her daydreaming
about something while she poured the coffee, and letting the cup overflow
because the perceptual channel was being supplied from imagination instead
of real-time perception. The only real "distraction" case is the one in
which her attention is on some other perception, such as what someone
behind her is saying about her, while the visual channel is operating
normally. Would we then see the same effects? I would guess not.

The same commercial shows a waiter using the refilling of water glasses as
an excuse to look over the shoulder of the person operating the laptop. In
the final scene, the waiter has poured about four glasses full already, and
is drinking from the fifth glass -- all of which, however, have been
competently filled without spilling anything. The distraction in his case
is apparently at a higher level. I think maybe we should hire that script
writer.

For reasons stated previously I don't expect to find a place for the
Observer or Witness within the hierarchy. The only evidence is inferred, a
confection of memory and imagination.

Right. But we might hope to find some effects on the hierarchy when people
do things we identify as changes in the relationship of Witnessing to the
hierarchy.

Best,

Bill P.

[BJORN_SIMONSEN]

[From Bjorn Simonsen (2003.06.03,21:10 EST)]

From David Goldstein (2003.06.02.1507 MDT)

The idea of “attention” is one that I have been interested in too.

Bruce Gregory wants to relate it the concept of loop gain. But Bill
Powers is pointing out problems
with this approach. I have always wondered what determined the control systems which were >“activated” at any given moment. It seems clear that not all the control systems within a person are active all >the time. The situation obviously partly determines what is activated.

Is it so clear that not all the control systems within a person is active all the time?

All control systems include a muscle cell or a gland cell. I disregard the control systems which are active when we use memory in the Imagination Mode. The negative feedback lead to p equal to the reference signal. By so doing the e is zero and also the qo. You may have a practical example when a person sits calm in a chair reading a book. He is not moving his muscles because the qo is zero. The important point here is that the golgi cells still send feed back variables to the input function which create the p which is still equal to the reference signal. I see a picture for myself where the values in the loop are: p,e-p,zero,qi,p,e-p,zero,… . I think this is an active system.

If the person stand up and stand still without moving any muscles the same systems have different reference signals and different p, but the qo is still zero and the feedback variables are different from when the person was sitting calm in the chair.

I have the understanding that all our control systems are active always as long as we live, I have figurative picture of our nervous system quite like our circulatory system, it is always active.

Am I wrong?

[bnhpct]

[From Bruce Nevin (2003.06.03 15:57 EDT)]

Bill Powers (2003.06.02.1147 MDT)--
At 02:47 PM 6/2/2003, Bill Powers wrote in reply to Bruce Gregory (2003.0602.1332):

On the other hand, perhaps in this case I am speaking to a hierarchy of
control systems in which there is no Observer. That would be embarrassing,
sort of like starting to complain to a recorded telephone message before
realizing that there's no actual person there.

Since the Observer is not a control system in the hierarchy, and does not control or behave in any way that we know of, what difference does it make? Everything that you can know of a person, or that a person can express as a person, is in the hierarchy. What do you envision being missing if you converse with a complete, intact human with no Observer present? What are you assuming that the Observer does or contributes?

         /Bruce Nevin

[bnhpct]

[From Bruce Nevin (2003.06.03 16:28 EDT)]

Bill Powers (2003.,06.03.0614 MDT)–
At 09:00 AM 6/3/2003, Bill Powers wrote in reply to Bruce Gregory
(2003.0602.1513):

… an actual denial that

individuals experience the world and themselves as conscious
observers.

It’s hard to understand how anyone could extend such a theoretical
position

to his OWN experiences.

Enough people have denied this idea to make me
pause and wonder what I’m

missing. Is there some simple argument that I might grasp that shows
what’s

wrong with the idea of an Observer?

I have proposed how the perception of oneself as an observer is an
illusion based on the observer-like relationship between present
perception and memory.

Since there is no second observer (nor the infinite regress that this
entails), we know that this perception of oneself as a conscious observer
must be illusory.

Therefore, some account of how we so consistently do create this illusion
is called for. If not the above, then propose another.

The method of levels has provided what appears
to be corroboration of this

phenomenon at a number of levels of abstraction. Furthermore, most
people

with whom I have discussed this phenomenon say they have a single

experience of observing that persists across all the different mental
and

perceptual content that they can see going on inside them. That is,
they

see the world as if from one consistent place or viewpoint, though
nobody,

apparently, can turn around and examine that viewpoint.

It’s kind of like Moses only being able to see the ‘hinder parts’ of God.
We can perceive where awareness has been, but not where awareness is.
Then we construct from this a story about what must be going on
now as I attend to … whatever it is I’m presently controlling.

The only factual arguments I can bring to bear
on this have to do with

perceptions or implied perceptual signals that can be present both with
and

without awareness of them. That seems to me a sufficient reason for
saying

that awareness and the existence of perceptual signals are not
synonymous.

One of the consequences of meditation is to become more immediately aware
of perceptions, and to do less controlling unconsciously (scratching,
shifting weight, cocking the jaw, twiddling fingers, etc.).

This is part of why I think the question might not be how does awareness
work but rather how do we become unaware. This is what the Buddhists mean
when they talk about ignorance. They’re not talking about education or
knowledge of what somebody has said or taught. They’re talking about
ignoring. Which is being not awake.

I realize this is not an answer. I doubt there is anything to be said
about the hierarchy that can constitute an explanation of awareness.

    /Bruce

Nevin

[Bill_Powers]

[From Bill Powers (2003.06.03.1953 MDT)]

Bruce Nevin (2003.06.03 16:28 EDT)--

I have proposed how the perception of oneself as an observer is an
illusion based on the observer-like relationship between present
perception and memory.

If it's a perception of oneself as an observer, it has to be in and of the
hierarchy, and so is a construction based on whatever we consider as
evidence of an observer. It's clearly not the observer. However, there are
certain actions that we can take that look as if they are at least
associated with the observer, the main one being the shifting of attention.

Look at the 0 below, and while fixating on it, shift attention to the X.
Stay fixated on the O; do not fixate on the X.

0 X

Now do it again, but after having moved attention to the X, while still
fixated on the O for a few more seconds, let your eyes jump to the X. This
is an illustration of selecting a reference position without activating it,
and then putting it into effect. Eye movement measurements show that this
jump is a true saccade (perhaps followed by one or two small corrective
saccades). The "putting into effect" of the reference signal appears to be
instantaneous and it's accurate within a degree or two.

The act of moving attention within the visual field is detectable by me,
but just barely. Ditto for the act of putting the reference position into
effect. It feels like doing something, but what I do is beyond description.
It's certainly not like controlling a "real" perception. The _result_ of
this action is also not like a real perception; I'm aware of the region
where the X is, and of the X shape, but it's certainly not like looking
directly at it. Something moves from the O to the X. but what moves is
definitely not a thing; it's more like the place where I'm getting
information, a two-dimensional gate. If you position the mouse cursor next
to the X and then, while fixating on the O but attending to X, move the
cursor, the motion will stand out very sharply without any further changes
in the attention point.

Other experiments can show similar phenomena in other modalities. If you
wear red and green glasses like the ones used for 3-D pictures, you can
shift attention back and forth between the green field and the red field --
or give them equal attention, at which point the apparent color changes to
a strange shade of white or gray with red and green shimmers. You can cover
one eye, and attend either to the one that sees or the blacked-out one. You
can attend to sensations from your right big toe or your left big toe, or
your tongue. or your visual field, or the auditory field. Or you can see
yourself as probably looking as if you're in a trance to any person
watching you, or you can attend to thoughts about the significance of the
exercise.

The mobility of attention within a relatively fixed perceptual world
convinces me that there is some system that is not related to the hierarchy
of perceptions as a higher level is to a lower level.The changes I perceive
as it operates seem unrelated to any structure of goals that may be in
effect, and they seem to be mostly interior changes, changes in which part
of the hierarchy I am paying attention to (where "I" in this case
detinitely does not refer to the I who is trying to describe all this). The
phenomenon seems no different whether I am attending to a pressure on my
skin or an idea like the United States Government, or the grammar of the
sentence I'm writing at this instant.

The other side of this kind of experience is that as each new focus of
attention occurs, the previous one disappears; it simply ceases to exist.
It's hard to remember without doing it again.

There is nothing in PCT or in HPCT that can produce this sort of
phenomenon. That's why I have no explanation for it.

Finally, my experiences as subject in the MOL have shown me that there is a
state of mind in which all the ordinary objects of attention that I would
identify as perceptions of one level or another cease to occupy my
attention. In that state I am wide awake and alert, and know that there are
perceptions going on somewhere, but there is nothing in focus, nothing with
which I am consciously engaged. I am aware without an object of awareness.
At least that is how I describe what vague memories of that state remain
afterward. Almost by definition, one can't remember it except by doing it
again.

You describe something that I can agree with:

>This is part of why I think the question might not be how does awareness
work but >rather how do we become unaware. This is what the Buddhists mean
when they talk about >ignorance. They're not talking about education or
knowledge of what somebody has said >or taught. They're talking about
ignoring. Which is being not awake.

Nor is it being asleep. And "ignoring" is entirely too active a word,
because there is no temptation to notice that has to be resisted.

I realize this is not an answer. I doubt there is anything to be said
about the hierarchy that can constitute an explanation of awareness.

Yes. I mean no. I mean I agree. But it's exceedingly important to know
about this phenomenon, so that when it turns up we don't waste our time
trying to explain it with clever block diagrams and fancy postulates. I
wouldn't be excessively surprised if the things we're talking about go with
some brain structure that hasn't been identified yet. Nor would I be
uncomfortable with finding that the brain is a three-dimensional interface
between reality and an Observer who is truly disembodied. In a world of
entangled atoms and 9-dimensional string theory, anything is conceivable.

Best,

Bill P.

[David_Goldstein1]

[From David Goldstein (2003.06.04.1942)]

[From Bjorn Simonsen
(2003.06.03,21:10 EST)]

Very nice to hear from
you.

As I sit in my chair
and am not moving a muscle, I am still sitting and not falling over. That is a
state which requires activitity in control systems to maintain.

As I sit in my chair
and type at the keyboard, I am not aware of all the possible experiences I
could be experiencing in this environment. I can ask myself to tune into the
pressure of the seat against my body and now this comes into awareness.

When I say “activated”
I guess I am saying that a perceptual signal which was there changes from an
unaware state to an aware state. I believe that this is the result of the
Observer.

The Observer has a
limit on how many different experiences can enter awareness at once. I believe
this corresponds to what people call capacity of short term memory or working
memory. A quick way of assessing this capacity is through the backward digit
span. The size of short-term memory increases in a linear fashion with
chronological age as follows.

Age 3 2 items.

Age 5 3 items.

Age 7 4 items.

Age 9 5 items.

Age 11 6 items.

Age 13 7 items. This is the adult capacity.

Pacual-Leone has hypothesized
that the capacity change in short-term memory is the basis for the qualitative
changes in intelligence which Piaget has described.

So, you may not be
wrong. I guess that I was talking about the limit on the number of items one
can become aware of at one time. If I am at my limit, and I want to attend to something
else, then something has to go.

I think the Observer
can do this. This is “top-down” processing. However, some items of information
can force themselves on awareness. If someone pricks me with a pin, I am sure
that my awareness will go to that experience. This is “bottom-up” processing.

So, “activated” means a perceptual signal shifts from
an unaware to an aware state. The perceptual signal is always there, as you say.
I think this may be a primary role of the Observer.

···

-----Original Message-----
From: Control Systems Group
Network (CSGnet) [mailto:CSGNET@listserv.uiuc.edu] On Behalf Of Bjørn Simonsen
Sent: Tuesday, June 03, 2003 3:05
PM
To: CSGNET@listserv.uiuc.edu
Subject: Re: Amazing!

[From Bjorn Simonsen
(2003.06.03,21:10 EST)]

From David Goldstein
(2003.06.02.1507 MDT)

The idea of
“attention” is one that I have been interested in too.

Bruce Gregory
wants to relate it the concept of loop gain. But Bill Powers is pointing out
problems >with this approach. I have always wondered what
determined the control systems which were
“activated” at any given moment. It seems clear that
not all the control systems within a person are active all >the time.
The situation obviously partly determines what is activated.

Is it so clear that
not all the control systems within a person is active all the time?

All control systems
include a muscle cell or a gland cell. I disregard the control systems which
are active when we use memory in the Imagination Mode. The negative feedback
lead to p equal to the reference signal. By so doing the e is zero and
also the qo. You may have a practical example when a person sits calm in a chair
reading a book. He is not moving his muscles because the qo is zero. The
important point here is that the golgi cells still send feed
back variables to the input function which create the p which is still
equal to the reference signal. I see a picture for myself where the values in
the loop are: p,e-p,zero,qi,p,e-p,zero,… . I think this is an active
system.

If the person stand up
and stand still without moving any muscles the same systems have different
reference signals and different p, but the qo is still zero and the feedback
variables are different from when the person was sitting calm in the chair.

I have the
understanding that all our control systems are active always as long as we
live, I have figurative picture of our nervous system quite like our
circulatory system, it is always active.

Am I wrong?

[Bill_Powers]

[From Bill Powers (2003.06.04.0701 MDT)]

David Goldstein (2003.06.04.1942)--

>So, "activated" means a perceptual signal shifts from an unaware to an
aware state. The perceptual signal is always >there, as you say. I think
this may be a primary role of the Observer.

This way of saying it implies that there is something about the perceptual
signal that is changing. There must be a way of talking about this that
makes it clear that the receiver is tuning in to different signals, while
the signals themselves are not changing. Unless, of course, you can specify
just what sort of change in the perceptul signal you're talking about, if
that's what you really mean to say.

And why the heck has Eudora decided that I want 6-point type? I seem to
have no control over fonts. I've been trying Mozilla's mail program and
it's pretty good. Haven't decided to switch, though.

Best,

Bill P.

Best,

Bill P.

[David_Goldstein1]

[From David Goldstein(2003.06.04.2145)]

Hi Bill,
One way of talking about it is the way information processing people do.
They talk about short-term memory and working memory. A copy of the
perceptual signal is put in a special place in the brain. I believe
there are specialized brain structures for this. If a person has the
hippocampus on both sides of the brain destroyed, the person loses the
ability to learn anything new.
Making a copy of the perceptual signal does not change it.
However, I know that awareness itself can bring about a change in the
experience of a perceptual signal. If one is at the dentist, and the
dentist is drilling a tooth, and one spreads awareness to include as
many different bodily and environmental experiences as possible at the
same time as the tooth is being drilled, the intensity of the pain is
diminished greatly. On the other hand, if one focuses awareness 100% on
the tooth and the spot where the drill is working, the pain experience
intensifies.
Les Fehmi, a NJ Psychologist, has emphasized that how we pay attention,
Open versus Narrow Focus, makes a difference in the way we experience
something.
Always great to talk to you,
David

···

-----Original Message-----
From: Control Systems Group Network (CSGnet)
[mailto:CSGNET@listserv.uiuc.edu] On Behalf Of Bill Powers
Sent: Wednesday, June 04, 2003 9:06 AM
To: CSGNET@listserv.uiuc.edu
Subject: Re: Amazing!

[From Bill Powers (2003.06.04.0701 MDT)]

David Goldstein (2003.06.04.1942)--

>So, "activated" means a perceptual signal shifts from an unaware to an
aware state. The perceptual signal is always >there, as you say. I think
this may be a primary role of the Observer.

This way of saying it implies that there is something about the
perceptual
signal that is changing. There must be a way of talking about this that
makes it clear that the receiver is tuning in to different signals,
while
the signals themselves are not changing. Unless, of course, you can
specify
just what sort of change in the perceptul signal you're talking about,
if
that's what you really mean to say.

And why the heck has Eudora decided that I want 6-point type? I seem to
have no control over fonts. I've been trying Mozilla's mail program and
it's pretty good. Haven't decided to switch, though.

Best,

Bill P.

Best,

Bill P.

[bnhpct]

[From Bruce Nevin (2003.06.03 10:10 EDT)]

Bill Powers (2003.06.03.1953 MDT)--

I agree, that is, all your descriptions ring true for me.

One additional step:

certain actions that we can take ... look as if they are at least
associated with the observer, the main one being the shifting of attention.

Look at the 0 below, and while fixating on it, shift attention to the X.
Stay fixated on the O; do not fixate on the X.

0 X

In general, one can shift focus from foveal to peripheral vision and back. But it is also possible to attend to both the foveated O and the peripheral X (or vice versa) at the same time. The peripheral image could be the cabinet in the corner, or the ceiling light, or the frame of my glasses.

There is a tendency for the foveated image to become blurry which can be resisted, and which appears to me to be relaxation of the focus of the lenses of the eyes. So there may actually be control processes involved, and gain being reduced on control of the foveated image. I suppose a congenial optometrist or opthalmologist could help verify this. However, the fact that this de-focusing can be resisted (keeping the foveal image clear while shifting attention to the peripheral image) indicates that the change of gain is not inherent to attention; so it cannot explain it.

(I don't know of corresponding control of focus for peripheral vision; does such control exist? Or are there skills that can be cultivated? E.g. by those people jogging through the canyons of Arizona on cloudy, moonless nights. Recall that they put a speck of phosphor on the tip of a wire projecting from the bill of their cap, providing a fixed point to foveate, which they ignored.)

An interesting thing about peripheral vision is that attention can move around within the peripheral field without moving the eyes, and can be made more inclusive (both the O and the cabinet in the corner, now add the glasses frame -- oops! don't lose clarity of the foveated X -- and so on). By the way, while trying this, don't forget to breathe.

In general, attention can be narrow or broad. An interesting result of meditation practice with a very narrow focus, such as anapana (close attention to sensations at the nostrils due to breathing) is that the acuity of perception that develops is retained when attention is broadened and moved through different regions of the body in succession. The practice is to return to anapana as an anchor to refocus when one finds that memories and imaginings have supplanted simple awareness of sensations in the body. After much practice, one is able to move awareness smoothly through the body, clearly attending to sensations in each region in turn, sweeping from head to foot and back, without interruption by memories or imaginings. Awareness can further expand, I am told, to clear awareness of sensations throughout the body at the same time, without shifting awareness from one region to another, and all with that same acuity. So the field of attention can be expanded or contracted.

Now do it again, but after having moved attention to the X, while still
fixated on the O for a few more seconds, let your eyes jump to the X. This
is an illustration of selecting a reference position without activating it,
and then putting it into effect. Eye movement measurements show that this
jump is a true saccade (perhaps followed by one or two small corrective
saccades). The "putting into effect" of the reference signal appears to be
instantaneous and it's accurate within a degree or two.

Try putting the reference for foveal vision into effect and activating it (a saccade) while maintaining attention on the peripheral image of the frame of your glasses.

0 X

You have one reference for foveal vision and another for peripheral vision. Foveation is not the same as attention.

The act of moving attention within the visual field is detectable by me,
but just barely.

It's easier to detect moving attention around within the peripheral vision field because the mechanics of saccade and focus don't confuse the issue.

Similarly, shifting attention from one voice to another in a room where several conversations are going on at once. The particular complex of voice qualities comes into focus, and voices with other characteristics are disturbances rather than controlled perceptions.

The other side of this kind of experience is that as each new focus of
attention occurs, the previous one disappears; it simply ceases to exist.
It's hard to remember without doing it again.

The difference in what I am saying is that it is possible to attend within a broadened field that includes both, within rubbery limits that are extensible by practice.

>the question might not be how does awareness work but rather
>how do we become unaware. This is what the Buddhists mean
>when they talk about ignorance. They're not talking about
>education or knowledge of what somebody has said or taught.
>They're talking about ignoring. Which is being not awake.

Nor is it being asleep. And "ignoring" is entirely too active a word,
because there is no temptation to notice that has to be resisted.

Yes. Either 'ignoring' with this caveat or 'ignorance' with the caveat that it doesn't mean uneducated.

it's exceedingly important to know
about this phenomenon, so that when it turns up we don't waste our time
trying to explain it with clever block diagrams and fancy postulates. I
wouldn't be excessively surprised if the things we're talking about go with
some brain structure that hasn't been identified yet.

Isn't there a lot of brain-image stuff showing activation of different parts of the brain correlated with shifts of attention?

Nor would I be
uncomfortable with finding that the brain is a three-dimensional interface
between reality and an Observer who is truly disembodied. In a world of
entangled atoms and 9-dimensional string theory, anything is conceivable.

Indeed. We could all be centers of expression for a single Observer, constructing the illusion of being separate observers in our separate memories and imaginations.

         /Bruce Nevin

···

At 10:39 PM 6/3/2003, Bill Powers wrote:

[Bill_Powers]

[From Bill Powers (2003.06.04.0906 MDT)]

Bruce Nevin (2003.06.03 10:10 EDT)--

The act of moving attention within the visual field is detectable by me,
but just barely.

It's easier to detect moving attention around within the peripheral vision
field because the mechanics of saccade and focus don't confuse the issue.

I didn't mean the _result_ of the act, but the act itself. HOW do you cause
the region of attention (or its size) to change? You just make some
indescribable effort, and it happens. It was the effort I was referring to.

>Indeed. We could all be centers of expression for a single Observer,
>constructing the illusion of being separate observers in our separate
>memories and imaginations.

That introduces something contrary to experience (not just outside of
experience) which has to be accepted as an act of faith. You and I both
have arms and legs, but this doesn't mean we share the _same_ Cosmic Arms
and Cosmic Legs. You can call the apparent separateness of our arms and
legs illusory, but that's a mere assertion, The "three-dimensional
interface" idea requires only suspension of disbelief, as in all good
science fiction. But those rules don't require suspension of reason. But
the "Same Observer" hypothesis requires that we ignore all the evidence of
differentness. The best conjectures are those that can stand up longest
against nit-picking.

As is common with this subject, one does run out of things to say, and
wants to get back to business.

Best,

Bill P.