The Magic of Turing

[From Rick Marken (960131.1230)]

Martin Taylor (960131 11:30) --

I think you quite misunderstand the nature of the Turing test. It is not
based on any concept of behaviour at all.

I think this is wrong. The Turing Test is clearly based on the idea that
behavior is visible output, not controlled (spontefacted) input. Turing never
suggested that his test was about determining whether or not you were dealing
with a system that is spontefacting its input.

Me:

There is no inkling [in the Turing Test] that some behavior might have
purposes that are not readily visible -- and some behavior may not.

Martin:

And how do you propose to perceive them if, by YOUR definition, you can't
perceive them?

I said that purposes are not _readily_ visible; I didn't sat that you can't
perceive them. Purposes can be made visible by performing The Test for the
Controlled Variable.

He [Turing] based his test on the foundational notion of PCT

Turing based his test on the notion that behavior is the spontefaction of
perception? How did I miss that?

that all you can tell of the world is what you can perceive of it.

Oy, vey, Martin. That's not the foundational notion of PCT (now PST, I
suppose). Check out the title of Bill's book; _that's_ the foundational
notion of PST.

Martin Taylor (960131 1331) --

Sarcasm has its place, I guess. But it reflects on both parties.

Who was being sarcastic?

If you are conducting a Turing test and want to use the PCT "Test", then
do so.

But then you would be doing The Test for the Controlled Variable, _not_ the
Turing Test. It's not really very classy to give one person (even a clever
and competant one like Turing) credit for the the dicoveries of another
(the cleverer and more competant W. T. Powers).

Now, since you insist on it, could you please expand on your method of
perceiving what is inside a "test object" without observing what it does?

When you do The Test for the Controlled Variable you observe the relationship
between what the organism does (actions and results of those actions) and
what happens in the organism's environment (disturbances). I don't think
I ever suggested that you could determine what is going on inside a "test
object" without observing what it does -- did I?

Since you seem to have a deep personal interest in _not_ learning that
magic doesn't work (it's been well over three years and there's been no
change in your understanding that I can detect;-)) it looks like you have,
indeed, committed yourself to looking at behavior from the conventional
(magical) perspective: so enjoy yourself in the "wonderfully intuitive"
world of internal appearances. You'll have lots of company;-)

Sorry. But I DO remember a series of interchanges about 3 years ago in
which you did commit yourself to a belief that retrofactive systems worked
by magic;-)

Don't be sorry. I sure remember those interchanges. I said that there is no
information in the perceptual signal regarding how to respond in order to
keep that signal under control. You said that this reflected my belief in
magic; what else could responses be based on but perceptual information about
the state of the controlled variable? Bill, Tom and I tried like mad to show
you that there is no information in the perceptual signal about how to
respond in order to keep that signal under control; it just doesn't work
that way in a closed negatve feedback loop. But, obviously, to no avail. In
a closed loop, the perceptual signal is always both a cause _and_ a result
of responses. The net result of the operation of this loop is that the
perceptual signal remains under spontefaction (uchh).

You seem to have a deep personal interest in maintaining your belief in
perception as _informative_ (rather than spontefacted) so it looks to me like
you have committed yourself to having only the most superficial understanding
of the "magic" of perceptual spontefaction.

Best

Rick

[Martin Taylor 960131 16:00]

Rick Marken (960131.1230)

You certainly have a most remarkenable way of twisting words around, don't
you?

How do you get from my (>>) statement to your (>) comment?

I think you quite misunderstand the nature of the Turing test. It is not
based on any concept of behaviour at all.

I think this is wrong. The Turing Test is clearly based on the idea that
behavior is visible output, not controlled (spontefacted) input. Turing never
suggested that his test was about determining whether or not you were dealing
with a system that is spontefacting its input.

Of course the Turing test is based on the idea that _all you can see_ is
visible output. To say that his test is "not based on any concept of
behaviour at all" is hardly to say "that his test was about determining
whether or not you were dealing with a system that is spontefacting its input."

Turing asks "Is there any way you can tell whether what you are communicating
with is human or not." If YOU decide that this is equivalent to the question
"Am I communicating with a spontefacting entity", YOU are quite entitled to
do so, and to use "the Test". But don't criticize Turing for not saying that
this was the question, or me for claiming that he did say it. He didn't, and
I didn't. If you believe that humans are spontefacting entities and machines
are not, then to use "the Test" is a guaranteed way for you to implement
the Turing Test successfully.

that all you can tell of the world is what you can perceive of it.

Oy, vey, Martin. That's not the foundational notion of PCT (now PST, I
suppose). Check out the title of Bill's book; _that's_ the foundational
notion of PST.

You wouldn't have "Behaviour: the Control of Perception" were it not for
the foundational notion that all you can tell of the world is what you can
perceive of it. If you could tell what was _really_ in the world, you
would have organisms that controlled what was _really_ in the world. The
book would have been "Behaviour: controlling the world."

Some foundations underly others.

If you are conducting a Turing test and want to use the PCT "Test", then
do so.

But then you would be doing The Test for the Controlled Variable, _not_ the
Turing Test. It's not really very classy to give one person (even a clever
and competant one like Turing) credit for the the dicoveries of another
(the cleverer and more competant W. T. Powers).

So who did that? I said that to conduct the Turing Test, you can do what you
want, limited only by the interaction being in language. You can't poke
the tested entity, and you can't look at it. You can do anything you want
with language, and you can interpret its language however you want. That's
what Turing said. He said nothing whatever about using Bill Powers' Test,
either to explicitly allow it or to prohibit it. And I never implied he did.

I don't think Bill P will want to be given credit for being either
cleverer or more competent than Alan Turing. Let Bill be satisfied with
being a genius. You don't have to give him any greater credit than that.
It might embarrass him.

ยทยทยท

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

More word twisting...

I sure remember those interchanges. I said that there is no
information in the perceptual signal regarding how to respond in order to
keep that signal under control.

Oh, no.... If that had been what you said, there would never have been
a discussion. What you said was that no information about the disturbance
waveform could be discovered in the perceptual signal, and that the output
of the control system was produced without the information being passed
through the perceptual function. That's an ENTIRELY different proposition.
You even used the term "magic" to say how the output changed in such a way
as to match the disturbance waveform without the information about the
disturbance waveform being processed through the perceptual function.

Bill, Tom and I tried like mad to show
you that there is no information in the perceptual signal about how to
respond in order to keep that signal under control; it just doesn't work
that way in a closed negatve feedback loop. But, obviously, to no avail. In
a closed loop, the perceptual signal is always both a cause _and_ a result
of responses.

Again, if that had been the gist of the argument, there would have been no
argument. The last sentence was _always_ a given in the discussion, and the
first was never at issue.

Discussions with you would be much easier if they didn't keep turning on
things never said or implied.

You seem to have a deep personal interest in maintaining your belief in
perception as _informative_ (rather than spontefacted)

And you seem to have a deep personal interest in maintaining your belief
that there is a contrast or even a contradiction here.

Maybe you could elaborate on your use of "rather than"? Why do you believe
that if perception is informative it cannot be "spontefacted", and
vice-versa? What I believe is that there is an essential unity between
perception being informative and perception being perfacted. At least,
perception can't be perfacted without being informative, although it
could be informative without being perfacted.

Oh, dear. I hope we aren't going to start into another fruitless "information
IN perception" round of postings. If there are to be postings on this topic,
could they be to the point, pretty please, and not about things not said?

Martin