[From Bill Powers (950829.0930 MDT)]
Hans Blom (950829a) --
I do not think of the "new viewpoint" as a "person". I can compare
the "new viewpoint" best with the new insight that a "connection"
exists that had not been discovered before.
The observer, I think you might agree, is always the same "person." But
it's not really a person in the usual sense, a self. You can see the
difference when you're observing yourself. Anything you can observe is
not the observer.
Hans Blom (950829b) --
Interesting excerpts from Bombrich.
Questions:
1. How can we -- if we can do so at all -- discriminate whether
we project our internal model upon the world rather than adapt
our internal model to the world? In which conditions are we
most likely to project rather than to process?
The first question is one we can't answer; we can only make assumptions.
Finding "the" correct answer would require that we observe the world,
observe the internal model, and compare the two. But we have only our
perceptions and the models we build on them -- only one half of the data
needed to do the comparison.
Hypothetically, if the world model has properties different from the
world, then making the world-model work should require more output
effort and more continuous effort than if the world-model were the same
as the world. Also, the world-model itself would have to be modified
more frequently to keep errors small. Both of these signs of an
incorrect world model are detectable by the system as a whole, without
requiring knowledge of the external world.
However, this gives us only a relative measure of goodness of fit,
unless you postulate that (1) with a _perfect_ world model, no effort at
all would be needed to control the world, and (2) once a world model's
properties are matched to the world, the world will never change its
properties. Since we have no _a priori_ reason to believe either (1) or
(2), we're talking about finding local minima, not perfection. However,
it is interesting that at least theoretically there is a way to compare
world models for _relative_ goodness of fit with the invisible external
world -- under a mninimum set of assumptions.
2. How can we -- if we can do so at all -- discover whether we
hallucinate? In which conditions are we most likely to halluci-
nate rather than to perceive?
The second question is really only a problem for the lower levels of
perception, intensity and sensations. People who can imagine at these
levels can, indeed, have such vivid experiences that they have trouble
distinguishing present-time perceptions from imagined ones, including, I
would suppose, difficulty telling perception of real actions of their
own from imagined actions. Back in my infamous dianetics days, one of my
favorite clients was a raving paranoid homosexual man. He was quite
aware of how screwed up he was, but somehow, when he got to know me
better, he turned out to have a great sense of humor (of a certain act
of love, he said "It's all good protein"). He also had an eidetic memory
and a capacity for, as they say now, photorealistic hallucinations. His
greatest problem turned out to be that he couldn't tell when he was
hallucinating. Just playing it by ear, I got him to practice
manipulating the hallucinations, turning them on on purpose (you can
imagine he was pretty reluctant to try that) -- and then turning them
off again (which proved to be the reassurance he needed). When he
couldn't make _me_ disappear, he was pretty much cured of that.
I was pretty set up about that, of course, but then it turned out that
he already knew how to do this in one area: organic chemistry. For some
years he had made a living formulating new organic chemicals for an oil
company, by imagining molecules and pushing them around until they would
fit right. Talk about world-models!
At the higher levels, where we perceive relationships, categories,
sequences, programs, principles, and system concepts, there's not so
much of a problem because generally we _know_ we're constructing a world
at these levels. At least it's not such a shock to find that we've been
imagining, say, a relationship. Since you can see almost any
relationship you want to see, the problem is much more clearly that of
constructing _useful_ relationships, rather than "right" relationships.
And when you get to even higher levels, I think there's hardly any
problem at all. When you derive a mathematical theorem, is the theorem
real, or imagined? I mean, what's the difference?
3. What could it mean, in our daily lives, to KNOW that we,
sometimes or frequently, project and/or hallucinate?
I think it's significant that the projections described by Gombrich
occur mainly under very low signal-to-noise ratio conditions. At normal
stimulus levels, where we have most of our experiences, the real-time
invention of perceptions is much more constrained by the parts of the
experience that have high signal-to-noise ratios. Since we can clearly
see the front side of most objects under good lighting, what we imagine
the other side to be like is of less importance. We would have a real
problem if all perceptions occurred with a signal-to-noise ratio of 0.1.
Your question really has two parts. I answered the first part assuming
that we take the reality of normal clear perceptions for granted. The
other part of the question really is, what difference does it make if we
admit that even our clearest and most reliable perceptions are
constructed by the brain?
One difference is that we will quit going around pontificating about
objectivity. We'll make a lot more use of phrases like "It seems to me
..." or "As far as I can tell ..." or "The evidence seems to favor ..."
or even "I'm guessing that ...". We become much more aware of being
Observers who see the world as constructed by a brain -- namely _one_
brain, our own. This tends to introduce a touch of humility into
otherwise bold pronouncements of fact.
This is a problem I've been working on understanding for some time, and
I can't claim to have had the final "aha!" There's thin ice here; I can
see how a person who gets immersed in these problems too far could fall
through into very deep dark water.
···
------------------------------------
RE: method of levels.
And I believe that anything can be a focus. I sometimes compare
this to looking up the meaning of a word in a dictionary. That
meaning is expressed in terms of other words. Thus the word
connects to a primary meaning, new words, in a circle around the
original word. And so on, in ever expanding circles of meaning in a
"semantic network".
The problem of circularity came up early in my explorations with Kirk
Sattley. Along with it came the "semantic network" problem, where one
word just leads to other words, around and around. The problem with
words also came up in my early work with Bob Clark at about the same
time.
The key for me was realizing that words are pointers to experiences,
only a few of which are words. From the word "apple" you can go two
ways: either to "An edible round object that is dark to light red on the
outside and white inside," or to "sweet, crisp, juicy, round, red, and
delicious." In other words, you can go to the verbal definition or to
labels for the nonverbal experiences you have. The dictionary's circles
begin when you get down to the interface between the words and the
experiences they denote. The experiences aren't in the dictionary; only
their names are.
If the method of levels turns into a word-association game, where one
thought reminds you of another and that reminds of you still another,
you can indeed go in circles. I've been around some of the best of them.
Although many thoughts do occur in words, most don't. Starting with a
verbal expression like "safety first," you can again go two ways. You
can say "a principle that expresses giving priority to self-
preservation," or you say "a sense of being cautious, on the lookout,
alert to danger." The first way is the dictionary way, the second the
experiential way.
In the method of levels, it's the experiential way we are after. For
people who operate mainly at the logical program level, catching on to
the difference can be difficult, because they're used to dealing with
symbols and more or less ignoring the experiences behind them. But I
think that when a higher-level system comes into view, that problem more
or less takes care of itself.
As I now use it, the method of levels doesn't run into any problems of
circularity. When a person starts by talking about all the people in his
world who cause problems for him, and ten minutes later is talking about
how he tends to blame his problems on others, and ten minutes after that
is wondering why he's so critical of himself and others, there's
obviously no problem with circularity. The circularity comes in only
when you get hung up at the level of words and symbols.
I would hesitate to use the term "levels" here.
I don't try to be more specific about it than that. Specifically, I
don't try to bring in PCT or the particular levels of organization I've
proposed. I notice that you make frequent reference to your "world-
model" model, but that's just a thought, too, isn't it?
Hilgard describes something like it as "The Hidden Observer". Yoga
describes it as "The Witness".
Yeah. It kind of makes you think we're all trying to talk about the same
thing.
-----------------------------------------------------------------------
P.S. I just turned 69 today. Can you tell?
-----------------------------------------------------------------------
Best,
Bill P.