rotten apples

[Martin Taylor 950301 13:55]

Bill Powers (950301.0700 MST)

(I have a meetign at 14:00, so this must be quick)

When you are inferring a rotten apple from the perception of a brown
spot on the apple's surface, you are specifically imagining something
that you are NOT CURRENTLY PERCEIVING. You ARE perceiving the brown
spot; you ARE NOT perceiving the rottenness. To perceive the rottenness
you would have to cut the apple open and see that it is brown inside,
too, and perhaps taste it to find the perception you call "rotten."
There is a possibility that when you cut the apple open you will find
none of those signs of rottenness; the brown spot may simply be a
discoloration of the skin.

When you are inferring a rotten apple from the perception of a red round
thing with a brown spot, you are specifically imagining something
that you are NOT CURRENTLY PERCEIVING. You ARE perceiving the red round
thing; you ARE NOT perceiving the appleness. To perceive the appleness
you would have to cut the round thing open and see that it is pale inside,
or not, and perhaps taste it to find the perception you call "apple" or
"rotten." There is a possibility that when you cut the red round thing
open you will find none of those signs of appleness; the red round thing
may simply be a plastic shell with a brown spot.

In other words, I read you as saying that no perception above the intensity
level is a "perception."

···

===================
I infer from the rest of your posting that by "inference" you imply
perceptions at the program level only.

I'll go along with that if you want. Doesn't matter to me how we use the
words, provided that we can use them in a mutually understood manner.

Martin

[From Bill Powers (950203.1420 MST)]

Martin Taylor (950301.13:55)--

     When you are inferring a rotten apple from the perception of a red
     round thing with a brown spot, you are specifically imagining
     something that you are NOT CURRENTLY PERCEIVING. You ARE perceiving
     the red round thing; you ARE NOT perceiving the appleness. To
     perceive the appleness you would have to cut the round thing open
     and see that it is pale inside, or not, and perhaps taste it to
     find the perception you call "apple" or "rotten." There is a
     possibility that when you cut the red round thing open you will
     find none of those signs of appleness; the red round thing may
     simply be a plastic shell with a brown spot.

Right, that's what I'm getting at. Even when we make such an innocuous
identification as that of saying "It's an apple," we can be mistaken
because we're referring in part to imagined perceptions.

Of course this depends in part on what we mean. If by "That's an apple"
I mean only that the shape and color are those I classify as "apple,"
there is no problem. We need not be asserting anything unobservable;
that is something that each of us alone must determine.

     In other words, I read you as saying that no perception above the
     intensity level is a "perception."

Not at all. If I say "I see something spinning," the spinning IS the
appearance, and it is entirely perceivable. Even if I'm being fooled by
stroboscopic motion, I am simply reporting the presence of a perceptual
signal, which I don't have to imagine. This is, in fact, how we know
there is a level of perceptions pertaining to motion or change --
transitions. If I see two objects at an apparent separation from each
other, I can report the relationship of "near" or "beside" without
imagining anything -- even if one object is "actually" 20 light years
farther from me than the other. If, on the other hand, I see a double
star and report that the two stars are in orbit around each other, I am
not perceiving that orbital motion, but imagining it. The difference
between perception and inference is not the difference between "real"
and "unreal." It is strictly a question of whether there is a perception
there to notice, or whether we are supplying it ourselves in
imagination.

Perceptions are _appearances_. To be fully aware of the difference
between perceiving and inferring, one has to spend a lot of time looking
to see whether there is actually any apparent thing, property,
experience, what have you being perceived right now. When we have become
very familiar with a phenomenon, we often start to blur the distinction.
An old hand at operant conditioning speaks and thinks as if
"reinforcement" is something we can actually see occurring. An
electronics expert looks at a trace on an oscilloscope and imagines that
he can see a voltage changing over the space of a microsecond. When we
see an athlete trying for a world record with the shot-put we think we
can actually see and feel the enormous effort the person is putting
behind the heave. Yet all of these perceptions are imagined, not real-
time. If we were not imagining, we would perceive none of these things.

This distinction is particularly important in theorizing about
observable phenomena. If we fail to be alert to this difference, we will
not be able to report observations accurately. Some of what we report
will be supplied by our own imaginations, as when we report that an
organism received 10 reinforcements per minute. That is quite different
from reporting that the organism received 10 pellets of food per minute.

In a tracking experiment, we see a person moving a handle to make a
cursor follow a target. Hidden in this observation is a certain degree
of imagination. For example, when we refer to the "target" we tend to
imagine that the subject wishes the cursor to be at the target position.
The Test is designed, in part, to relieve us of such illusions created
by our own imaginations. If you apply the Test properly, you may well
find that the participant's behavior is maintaining the mark we call the
cursor 10 cm to the right of the position of the mark we call the
target, or the participant might be keeping the cursor at the same
distance as the target and in the opposite direction from an imaginary
place on the screen. We think we are observing a target, but all we are
actually seeing is a moving mark on the screen. This becomes immediately
apparent when something happens to reveal what we are imagining.

      I infer from the rest of your posting that by "inference" you
     imply perceptions at the program level only.

     I'll go along with that if you want. Doesn't matter to me how we
     use the words, provided that we can use them in a mutually
     understood manner.

I don't really need to be humored. I think there is a serious point here
that goes beyond word usage. There are many levels of inference, some
involving reason, but they all rest on inserting imagined perceptions
among the real-time ones, the ones that are not dependent on imagination
alone. Many of our problems with conventional psychology arise because
psychologists have become so used to certain terms that they think they
are observing phenomena when they are mostly imagining them. The demos
are supposed to create a conflict with the imagined perceptions, so they
can be separated from the real-time ones. But of course that doesn't
happen automatically; the person involved must be willing for it to
happen.

···

-----------------------------------------------------------------------
Best to all,

Bill P.

<[Bill Leach 950301.20:30 EST(EDT)]

[Martin Taylor 950301 13:55]

... You ARE perceiving the red round thing; you ARE NOT perceiving the
appleness.

Your assertion muddies the point but is in an absolute sense correct as
long as "appleness" includes perceptions beyond visual examination of the
outside... and I think that it does.

I have more than once 'hefted' a "wax apple" to discover that it was not
a "real apple"... this is, it was not then perception which to me is
"appleness".

You in essence made Bill P.'s point in disagreeing with him. You posting
further emphasizes the ease with which "add to" what we directly perceive
WITHOUT recognizing the addition(s).

In other words, I read you as saying that no perception above the
intensity level is a "perception."

Ouch!! No, absolutely incorrect (though I have to agree that this IS
what Bill P. said)! Inferences ARE perceptions. The point is that there
is a difference between perceptions that are 'just due to external
stimulus' and perceptions derived by 'consideration' of external stimulus.
I believe that the distinction is useful even if exceedingly difficult to
apply. That is, almost as soon a we 'perform' any catagorization we
drawing upon _some_ measure of inference (as your 'round red' points
out).

Bill's choice of 'appleness' may thus have been a mistake (though maybe
we will both find out otherwise later). I can reasonably posit that even
the perception of "redness" or "roundness" is not absolute in the sense
that both can be 'fooled' under the correct circumstances.

When you 'see something', that perception is always subject to being
'fooled' at sometime or in someway. I think that in Bill's example, he
implying possibly additional perception in support of 'appleness' such as
to preclude plastic or wax apples.

For example, when I pick up and examine an apple in a grocery store,
while as you assert there is _some_ measure of inference there, my degree
of certainty that the object that I am observing IS what I call an apple
is very high. The 'brown spot' OTOH that 'I perceive as rotteness' is
directly an inference at the point of just visual examination.

There is, I think, a very real difference in the nature of the inference.
That there is an inference associated with 'appleness' is indeed
appearently correct if for no other reason that many of us admit to
having 'been fooled' by decorative 'apples' thus admitting to the idea
that 'declaration of appleness' based exclusively upon visual examination
alone does involve inference at _some_ level.

-bill

[Martin Taylor 950302 13:00]

Bill Powers (950203.1420 MST)

+Bill Leach 950301.20:30 EST

Me, quoted by Bill P.:

     I infer from the rest of your posting that by "inference" you
    imply perceptions at the program level only.

    I'll go along with that if you want. Doesn't matter to me how we
    use the words, provided that we can use them in a mutually
    understood manner.

I don't really need to be humored.

I had no intention of doing so. Sorry if you took it that way. I only
proposed to accept your terminology because it should ease discussion if
we had such an agreement.

I think there is a serious point here that goes beyond word usage.

Your posting makes it clear that there is. Let's try to generate a discussion
on that point.

I take "perception" to represent the output of a perceptual function. But
"perception of" is something else again. "Perception of" is an assertion by
an an outside agent (i.e. another perceptual system, whether in the same
hierarchy or another) whose input is not only the perceptual signal, but also
some aspect of the "real" world. "Perception of" is a statement that some
perceptual signal correlates with some other perception of the real world.

So, I say that if I have a perceptual input function that most of the time
produces a high output when there is an apple in view, then I have a"perception
of" appleness. But that's irrelevant to the perceptual function that
generates the "appleness" signal. Those labels are applied from outside.
The same applies to the perceptual signal (if any) for "rottenness." And
if I happen to have in my hierarchy a PIF that gives a strong output in
the presence of rotten apples, but not of rotten grapes, I claim that I
truly perceive a rotten apple--whether or not it is a scented plastic
imitation. The facts of the real world are never known to me, but my
perceptions may change when new sensory data come available. What I
perceive NOW is truly what I perceive.

And this is where I strongly disagree with Bill L., who seems to argue
that I cannot "perceive" anything that is not true in the real world:

+I have more than once 'hefted' a "wax apple" to discover that it was not
+a "real apple"... this is, it was not then perception which to me is
+"appleness".

···

+
+You in essence made Bill P.'s point in disagreeing with him.

In fact, I thought I was agreeing with him in all respects except for the
scope of the word "inference." I think both Bill P. and I would at least
agree that whether we perceive something or not cannot be determined by
the facts of what exists in "reality." Perception is an unknowable
transformation on reality (though it can be constrained).

Now, what should we understand by "inference?" That's where the serious
discussion lies, because obviously the result of the inferencing process,
as well as much of what enters into that process, is perception. But
a process that says "I perceive red. I perceive round. I perceive dark
small line protruding from round... Therefore I have an apple" is quite
different from perceiving an apple.

I can't pursue this now. I have another meeting that started 3 minutes ago.

But it should perhaps be pursued.

Martin

<[Bill Leach 950303.01:26 EST(EDT)]

[Martin Taylor 950302 13:00]

Except that we were talking 'past' each other again, I find your thoughts
on this set of subjects to be excellent.

I would like to see how Bill feels about your rather precise distinction
between "perception" and "perception of", personally I consider that to
be a very clear distinction.

Trying to 'say this back to you':
A 'perception of' is a perception containing an inference produced by
comparing a (and achieving a satisfactory match between a) perception and
a symbol or concept.

And this is where I strongly disagree with Bill L., who seems to argue
that I cannot "perceive" anything that is not true in the real world:

Actually I blew the grammer so badly in writing that one, that even I am
not longer sure exactly what I was trying to say however, I should think
that one would have to 'stretch' to infer that I assert what you accused
me of asserting. By now, I fully believe that we know of no way to know
what is 'real' in any absolute sense.

I believe that some sort of consistancy in the real world causes us to
experience a measure of consistancy of perception in that real world and
that it is quite likely that others not only experience the consistancy
but the perceptions might be quite close to the same (though probably not
identical ever).

-bill