Perceptions are illusions ( was Re: Qualia)

[From Rupert Young (2018.03.17 14.50)]

Craic agus ceol!

(Rick Marken 2018-03-12_12:54:38]

Well, this is a bit of a philosophical discussion, but the nice

thing about philosophy, and ideas therein, is that they are like
hats. You can put them on, wear them, walk around in them, try them
out and see how they feel without having to accept them.

Some perceptions seem to clearly  be illusions, such as RDS's, the

Ames room, the face on Mars. Perhaps we should not consider these as
special cases or aberrations, but that all perceptions are of this
nature.

Consider more everyday perceptual experiences like watching TV or

movies, viewing photographs, using a computer screen. These all work
because perceptions can be induced without the thing which they
supposedly represent actually existing. I.e. there is no analog “out
there.”

Or consider much of our general experiences, for control of

perceptions like fear, honesty, love, socialism, believing that more
guns make us safer, believing that there is an invisible sky fairy
looking over us or thinking that Brexit is a good idea. None of
these represent things that actually exist in independently of the
perceiver. For some of them, even, there is no possible
analog that could exist. If there were no humans (or other animals)
in existence then these concepts (perceptions) wouldn’t exist
either.

The thing about perceptions, I think, is that they introduce another

dimension to experience that provides perspectives and
concepts (which we control) that having nothing to do with the hard
reality of the physical world. As humans we live in this
experiential “realm” of abstractions rather than the actual physical
world. Also*,* there are more things in heaven and earth than are
dreamt of in our philosophy. We only experience a minute fragment of
physical reality; only “visible” light, for example. If we were
suddenly able to directly perceive all the signals that are out
there we would be bombarded by an universe completely alien to our
experience. Of course, some perceptions, at the lower levels anyway, form
due to our interactions with the actual physical world. However,
perhaps, rather than considering

perceptions as analogs of states of affair "out there" with some

being illusions we should consider all perceptions as illusions with
some being analogs of states of affair “out there”. So, I’d say (to csgnet), try out this “all perceptions are
illusions” hat for a while and see how it feels. You never know, you
might decide you don’t want to take it off!

Regards,

Rupert
···

[From Rupert Young (2018.03.12 18.05)]

(Rick Marken (2018.03.12.1045)]

            RY: That's ok as far as it goes, for some cases,

but I don’t think it reflects the whole picture. As
RDS’s show you can have a perception for something that
doesn’t exist; there is no external analog. There is no
shape in the environment that corresponds to the
perception, though there are the basic constituents
which form the perception, via the perceptual function
(which is internal). All perceptions are illusions!

          RM: They are not illusions; they are analogs of

aspects of the reality that exists “out there”; the
reality that described by the models of physics and
chemistry. Even the perceptions that we consider to be
illusions are not, themselves, illusions. They are
illusions to the extent that they differ from what we we
“know”, on different grounds, to be out there. For
example, in the Ames room illusion, we perceive a person
to be growing or shrinking as they walk from one corner to
another. This is what we perceive; we consider it to be an
illusion because we know, on other grounds, that the
person is not actually shrinking or growing.

          RM: But on second thought, given this definition of

illusion, perhaps it is correct to say that all
perceptions are illusions, at least to the extent that
perceptions differ from what we know from physics and
chemistry to be what is actually going on “out there”. So
the taste of lemonade could be called an illusion since,
per chemistry, there is no unitary entity out there that
corresponds to this perception; what is actually out there
is concentrations of different types of molecules. Same is
true for this table top here, which looks like a solid
entity but which I kn ow from physics and chemistry to
also by a tightly bound collection of atoms that are
mainly empty space.

          RM: But on third thought, I think I would prefer to

reserve the term “perceptual illusion” for perceptions
that conflict with what we know based on other perceptions
rather than physical models. The reason is that
perceptions defined as illusions in the latter way are
adaptive; perceptions defined as illusions in the former
way are not. The Ames room, size change perception is
defined as an illusion in the formerway; because it
conflicts with what we know based on other perceptions. We
can see that people only change size when they are in the
Ames room. If the outside world were organized as it is
in the Ames room – with linear perspective consistently
conflicting with actual distance – we would be having
this illusion all the time. This would be maladaptive
since we would never be sure whether an approaching object
was large or small. However, perceiving collections of
atoms as a solid entity, like the table top, seems
perfectly adaptive; you know what will happen if you set
something on it (it won’t drop through), for example.

          RM: So I'm in favor of sticking with the definition of

perception as an analog of a states of affair “out there”
and referring to perceptions as “illusions” when we can
perceive them only under special circumstance, where the
perception differs from what we see in all other
circumstances.

                        RM: Why not just define perception the

way it’s defined in B:CP: A signal inside a
system that is a continuous analog of a sate
of affairs outside the system.

[Rick Marken 2018-03-17_18:42:12]

[From Rupert Young (2018.03.17 14.50)]

Craic agus ceol!

RM: Back at ya;-)>

RY: Some perceptions seem to clearly be illusions, such as RDS's, the Ames room, the face on Mars. Perhaps we should not consider these as special cases or aberrations, but that all perceptions are of this nature.

RM: What's an RDS? Â
Â

RY: Consider more everyday perceptual experiences like watching TV or movies, viewing photographs, using a computer screen. These all work because perceptions can be induced without the thing which they supposedly represent actually existing. I.e. there is no analog "out there."

RY: Or consider much of our general experiences, for control of perceptions like fear, honesty, love, socialism, believing that more guns make us safer, believing that there is an invisible sky fairy looking over us or thinking that Brexit is a good idea. None of these represent things that actually exist in independently of the perceiver. For some of them, even, there is no possible analog that could exist. If there were no humans (or other animals) in existence then these concepts (perceptions) wouldn't exist either.

RM: I think of perception as the analog, and what it is an analog of is some variable aspect of the environment. The perception of honesty, for example, is ultimately a function of aspects of the environment such as what people say when they move their lips (or, as in the case of Trump, . >

RY: The thing about perceptions, I think, is that they introduce another dimension to experience that provides perspectives and concepts (which we control) that having nothing to do with the hard reality of the physical world. As humans we live in this experiential "realm" of abstractions rather than the actual physical world.

RM: I think all perceptions, those that are most abstract, like principles and system concepts, as well as those that are most concrete, like shape and color, are based on the hard reality of the physical world. That is the PCT view, anyway. It was hard for me to get my head around the idea that what you call abstract perceptions, like the principle "do unto others" are are perceptions that are as much a function of the hard reality of the physical world as are concrete perceptions, like the position of a cursor. But in order to control adhering to a principle you have to be able to perceive the degree to which that principle is being followed. And that perception, like all perceptions, must be a function of variables in the hard reality of the physical world. Or, at least, so says the PCT model of behavior.

RY: So, I'd say (to csgnet), try out this "all perceptions are illusions" hat for a while and see how it feels. You never know, you might decide you don't want to take it off!Â

RM: I don't care for that idea because I think it's anti-scientific. But I don't mind if you want to play around with it because you want to design robots based on PCT and when you start building robots that can vary the degree to which they follow rules in order to control for principles you'll see that you'll have to build perceptual functions that can construct perceptions of rules and principles; and you will probably find that those functions can't construct those perceptions from nothing.They will have to construct them from all the lower level perceptions that your Lego sensors construct. But maybe by that time Lego will include rule and principle sensors in their kits;-)
Best
Rick

···

--
Richard S. MarkenÂ
"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery

[From Rupert Young (2018.03.18 15.50)]

(Rick Marken 2018-03-17_18:42:12]

Random dot stereogram.

I think we are talking about different things, to some extent. I am

saying that the perceptions themselves do not exist out there, even
though the constituents might. Perceptions require
perceptual functions to exist, and they can only be found in brains.
Where does the principle “do unto others” exist if there is no
perceptual function? The external world doesn’t contain perceptual
functions. The perceptions for the above illusions are induced via
the constituent elements by perceptual functions, even though the
thing to which the perception represents doesn’t exist. However,
this is the case for all perceptions, in that they could all (in
principle) be induced by their constituents without the thing to
which they represent actually existing. So, we could think about all
perceptions as illusory. I don’t see any contravention of PCT.

Well, It worked pretty well for Einstein when he imagined how light

travelled from different frames of reference. And, I guess, it
worked pretty well for Bill when he re-imagined behaviour as the
control of perceptions rather than stimulus-response. Thought
experiments are often the starting point for scientific revolutions.
But maybe you meant something different.

What I am saying doesn't contradict any of that paragraph and I

agree with it. Except for the last sentence, but I assume you were
being humorous, unless you think that rules and principles exist
“out there”, and can be sensed directly?

Regards,

Rupert
···

[From Rupert Young (2018.03.17 14.50)]

Craic agus ceol!

RM: Back at ya;-)

            RY: Some perceptions seem to clearly  be illusions, such

as RDS’s, the Ames room, the face on Mars. Perhaps we
should not consider these as special cases or
aberrations, but that all perceptions are of this
nature.

RM: What’s an RDS?

            RY: Consider more everyday

perceptual experiences like watching TV or movies,
viewing photographs, using a computer screen. These all
work because perceptions can be induced without the
thing which they supposedly represent actually existing.
I.e. there is no analog “out there.”

            RY: Or consider much of our general experiences, for

control of perceptions like fear, honesty, love,
socialism, believing that more guns make us safer,
believing that there is an invisible sky fairy looking
over us or thinking that Brexit is a good idea. None of
these represent things that actually exist in
independently of the perceiver. For some of them, even,
there is no possible analog that could exist. If
there were no humans (or other animals) in existence
then these concepts (perceptions) wouldn’t exist either.

          RM: I think of perception as the analog, and what it is

an analog of is some variable aspect of the environment.
The perception of honesty, for example, is ultimately a
function of aspects of the environment such as what people
say when they move their lips (or, as in the case of
Trump, .

            RY: The thing about perceptions, I think, is that they

introduce another dimension to experience that
provides perspectives and concepts (which we control)
that having nothing to do with the hard reality of the
physical world. As humans we live in this experiential
“realm” of abstractions rather than the actual physical
world.

          RM: I think all perceptions, those that are most

abstract, like principles and system concepts, as well as
those that are most concrete, like shape and color, are
based on the hard reality of the physical world. That is
the PCT view, anyway. It was hard for me to get my head
around the idea that what you call abstract perceptions,
like the principle “do unto others” are are perceptions
that are as much a function of the hard reality of the
physical world as are concrete perceptions, like the
position of a cursor. But in order to control adhering to
a principle you have to be able to perceive the degree to
which that principle is being followed. And that
perception, like all perceptions, must be a function of
variables in the hard reality of the physical world. Or,
at least, so says the PCT model of behavior.

            RY: So, I'd say (to csgnet), try

out this “all perceptions are illusions” hat for a while
and see how it feels. You never know, you might decide
you don’t want to take it off!

      RM: I don't care for that idea because I think it's

anti-scientific.

      But I don't mind if you want to play

around with it because you want to design robots based on PCT
and when you start building robots that can vary the degree to
which they follow rules in order to control for principles
you’ll see that you’ll have to build perceptual functions that
can construct perceptions of rules and principles; and you
will probably find that those functions can’t construct those
perceptions from nothing.They will have to construct them from
all the lower level perceptions that your Lego sensors
construct. But maybe by that time Lego will include rule and
principle sensors in their kits;-)

[Rick Marken 2018-03-19_12:34:46]

[From Rupert Young (2018.03.18 15.50)Â

RY: I think we are talking about different things, to some extent. I am saying that the perceptions themselves do not exist out there, even though the constituents might. Perceptions require perceptual functions to exist, and they can only be found in brains.

RM: Of course. Or in artificial systems (like thermostats and robots and imagine and speech recognition systems) that can produce analogs of functions of physical variables.
Â

RY: Where does the principle "do unto others" exist if there is no perceptual function? The external world doesn't contain perceptual functions. The perceptions for the above illusions are induced via the constituent elements by perceptual functions, even though the thing to which the perception represents doesn't exist.

RM: Yes, of course.Â
Â

RY: However, this is the case for all perceptions, in that they could all (in principle) be induced by their constituents without the thing to which they represent actually existing. So, we could think about all perceptions as illusory. I don't see any contravention of PCT.

RM: It just doesn't seem useful to me to call perceptions illusions. The term "illusion" connotes error. And I don't think of perceptions themselves as being in error; they just are what they are. The perception of a stick bending in water, for example, is not an error; that's what I see. The error turns out to be thinking that it is actually bent. That error is in how I interpret the perception, not in the perception itself. And I still have the perception of a bent stick even when I know it's not bent. So the perception stays what it is; the interpretation of it changes from being in error (an illusion) to being correct. This is what happens with behavioral illusions too. I still see the patellar reflex as a response to the tap stimulus even though I know that it is an (unsuccessful) control action. Perceptions don't change when you are able to correctly interpret them.Â

RY: So, I'd say (to csgnet), try out this "all perceptions are illusions" hat for a while and see how it feels. You never know, you might decide you don't want to take it off!Â

RM: I don't care for that idea because I think it's anti-scientific.

RY: Well, It worked pretty well for Einstein when he imagined how light travelled from different frames of reference. And, I guess, it worked pretty well for Bill when he re-imagined behaviour as the control of perceptions rather than stimulus-response. Thought experiments are often the starting point for scientific revolutions. But maybe you meant something different.

RM:: Yes. What you are describing is a reinterpretation of what the perceptions mean, taking the perceptions themselves to be illusions. Einstein re-imagined what our perceptions of measures of the speed of light might mean; he didn't perceive those measures differently. Bill re-imagined what our perceptions of measures of behavior might mean; he didn't perceive those measures in a new way. For me, perceptions themselves are not illusory; it's our interpretation of those perceptions that can be illusory; in error.Â
Â

RM: But I don't mind if you want to play around with it because you want to design robots based on PCT and when you start building robots that can vary the degree to which they follow rules in order to control for principles you'll see that you'll have to build perceptual functions that can construct perceptions of rules and principles; and you will probably find that those functions can't construct those perceptions from nothing.They will have to construct them from all the lower level perceptions that your Lego sensors construct. But maybe by that time Lego will include rule and principle sensors in their kits;-)

RY: What I am saying doesn't contradict any of that paragraph and I agree with it. Except for the last sentence, but I assume you were being humorous, unless you think that rules and principles exist "out there", and can be sensed directly?

RM: I think the bases of our perception of rules and principles are "out there". In order to control them we have to be able to perceive the world in terms of rules and principles; that is, we have to be able to perceive when a rule or principle is being followed or not. A robot that can do what we do -- control for rules and principles -- will have to be able to perceive rules and principles. Those perceptions will be based on simpler perceptions that are ultimately provided by the more elementary sensory receptors that are provided by Lego. I was kind of joking about rule and principle sensor. What it actually could be is some software -- maybe 2 million LOC -- that creates perceptions of rules and principles from the lower level perceptions derived, ultimately, from the Lego sensors.
BestÂ
Rick

···

--
Richard S. MarkenÂ
"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery