Challenge 2

CSGnet Archive CSG1993
1

[From Bill Powers (930328.0930)]

Greg Williams (930327.1945 EST) ---

In your 930326-2) you said

...show me you understand my behavior and what it is being used
to control. But first, YOU tell ME what a passing grade at such
"understanding" would consist in.

I did respond to your request that I tell you what passing grade
we should use. Your request was a behavior of yours. When you
read my answer, you replied "Fine." Was that an indication that
my answer gave you the information you wanted? If so, I would
explain that behavior by saying that you issued the request for
information because you wanted to see my answer (explanation 1),
and that at another level, my answer met whatever criteria you
had set for an acceptable answer (explanation 2). Is either of
those explanations 99% accurate?

As to the main part of the challenge statement, I asked you what
effect you intended to produce by issuing the challenge. Your
reply was

I don't know. Perhaps you have a way to get me to say that I
"realize" some particular effect is the one I intended? The
question then is: how do we know that was REALLY the effect I
intended (yesterday), and not just some sort of after-the-fact
fiction than seems to make sense? How do we avoid "just-so"
opining?

Is the challenge still in effect, or did it end with your issuing
of it? If it's still in effect, then unless you consider that
I've successfully met it, there must still be some sort of error
signal in existence, and the reference signal must still be in
effect, right now. It shouldn't be necessary to remember
yesterday's motive. Even if the nature of the reference signal
changes in your mind, it's only the goal as it exists right now
that you or I can try to meet, because that is the one you will
compare the results against.

Your words indicate that if I suggested what was "really" your
intent, you would tend to reject it as guesswork. I agree. I'm
not a psychoanalyst who will interpret your own words to yourself
in some surprising way. If there is some hoped-for result in your
issuing the challenge, it is still here and in effect. Otherwise,
we're finished, aren't we? You would have withdrawn the
challenge.

Assuming you haven't withdrawn it, perhaps we can work out the
intended effects by considering it at different levels. I
presume, for instance, that by maintaining the challenge, you
expect some sign that I am accepting it (or not). Was (and is)
this one of the immediate goals in issuing the challenge? That
is, did you want to get back an indication of my acceptance or
rejection of it? It would be a reasonable interpretation on my
part to assume that you did (and do) want to know whether I (then
and now) accept or reject it, and I might even guess that your
preference was and is that I accept it. But you are the only one
who knows the truth of the matter. I'm out here. You're in there
where you can see what's going on right now.

So, do you still want to know whether I still accept the
challenge? And is your preference right now that I do or do not
accept it?

···

-------------------------------------------------------------
Best,

Bill P.

2

From Greg Williams (930329 - 2)

Bill Powers (930328.0930)

In your 930326-2) you said

...show me you understand my behavior and what it is being used
to control. But first, YOU tell ME what a passing grade at such
"understanding" would consist in.

I did respond to your request that I tell you what passing grade
we should use. Your request was a behavior of yours. When you
read my answer, you replied "Fine." Was that an indication that
my answer gave you the information you wanted? If so, I would
explain that behavior by saying that you issued the request for
information because you wanted to see my answer (explanation 1),
and that at another level, my answer met whatever criteria you
had set for an acceptable answer (explanation 2). Is either of
those explanations 99% accurate?

I don't know. I don't have direct access to my motivational control
structure, so I can't say what explanation is "99% accurate." I can
CONJECTURE that both of your explanations (and a lot of other possible
explanations) make some sense. But I think "from the outside" we'd
need to apply disturbances to see which error signal(s) I was REALLY
trying to keep minimized. As they say, hindsight tends to appear
20/20, even if it really (on the basis of more solid evidence) isn't.

As to the main part of the challenge statement, I asked you what
effect you intended to produce by issuing the challenge. Your
reply was

I don't know. Perhaps you have a way to get me to say that I
"realize" some particular effect is the one I intended? The
question then is: how do we know that was REALLY the effect I
intended (yesterday), and not just some sort of after-the-fact
fiction than seems to make sense? How do we avoid "just-so"
opining?

Is the challenge still in effect, or did it end with your issuing
of it?

Still in effect.

If it's still in effect, then unless you consider that I've
successfully met it, there must still be some sort of error signal in
existence, and the reference signal must still be in effect, right
now.

Yes, but not necessarily the same one(s) in effect when I originally
issued the challenge.

It shouldn't be necessary to remember yesterday's motive. Even if
the nature of the reference signal changes in your mind, it's only the
goal as it exists right now that you or I can try to meet, because
that is the one you will compare the results against.

I thought you were going to explain why I issued the challenge in the
first place, not why I continue to say you haven't succeeded in
meeting it.

Your words indicate that if I suggested what was "really" your
intent, you would tend to reject it as guesswork. I agree. I'm
not a psychoanalyst who will interpret your own words to yourself
in some surprising way. If there is some hoped-for result in your
issuing the challenge, it is still here and in effect. Otherwise,
we're finished, aren't we? You would have withdrawn the
challenge.

You are overlooking the possibility that an output can remain (at
least look) the same while the intended outcome is changing. My
motivations could be changing continuously for not withdrawing the
challenge. So what about an explanation for my originally issuing the
challenge?

Assuming you haven't withdrawn it, perhaps we can work out the
intended effects by considering it at different levels. I
presume, for instance, that by maintaining the challenge, you
expect some sign that I am accepting it (or not). Was (and is)
this one of the immediate goals in issuing the challenge?

Again, I don't know. It sounds reasonable, but how would I gather any
additional evidence beyond (my own) hearsay?

That is, did you want to get back an indication of my acceptance or
rejection of it?

It is possible. I'm not sure.

It would be a reasonable interpretation on my part to assume that you
did (and do) want to know whether I (then and now) accept or reject
it, and I might even guess that your preference was and is that I
accept it. But you are the only one who knows the truth of the
matter.

You have quite a bit more confidence in my "truth"-seeing than I do.
How do I KNOW -- not just GUESS -- what motivated me two days ago?

So, do you still want to know whether I still accept the
challenge? And is your preference right now that I do or do not
accept it?

Right now, I CLAIM that I want to know and that I prefer that you
accept it. So, what about explaining my original challenging?

···

-----

P.S. Are you going to address "understanding" vs. prediction?
_____

As ever,

Greg

3

[From Bill Powers (930330.0700)]

Greg Williams (930330) --

YOU are the one who is trying your method. I just work here (as
the judge of whether you've understood my behavior).

One of the assumptions I have to make in trying to discern a
control process is that there is some control process going on:
outputs that affect the external world to produce an effect
intended by the outputter. A person who chooses to play the role
of passive spectator and commentator is not obviously controlling
anything, so there is little to push against. However, it's not
hard to see the resistance to being drawn into conjecture, so I
assume that you have a low reference level for indulging in
conjecture (or introspection). This, I will admit, does make the
challenge a little harder (although I submit that the foregoing
analysis captures some of the situation -- is that right?).

It remains to be seen what credence I'll assign to my
impressions related to the challenge.

This suggests that you're waiting for something to reach a point
at which you will experience a definite impression one way or the
other. As long as there is no definite impression, you will
continue to feel that the challenge has not been resolved. It
also suggests a higher level, such that when some impressions
actually do occur, they will not be taken at face value, but will
be judged against some other criterion and only then given
credence or not. Do I interpret your meanings correctly?

[You be the judge?]Oh no, I would have won already. That would
be too easy.

And too unconvincing to observers, especially if I protested
your judgment?

That's what I meant -- too easy to be convincing, even to me.

Test questions:

My surface hypothesis was evidently wrong, because it led me to
wrong predictions of each of the answers I thought you would NOT
give, and also wrong predictions as to each of the answers I
thought you WOULD give. This is essentially what I expected
according to a deeper hypothesis.

Previously, when I asked questions about what you wanted, you
said that you had no contact with your inner motivational
structure and therefore didn't know, or could only guess. This
time, when I asked what you want, you gave a clear YES answer to
two of the questions, and when asked about success or failure
preferred to predict some degree of success between the extremes.

With regard to the first two questions, do the YES answers still
stand today?

If so, we would seem to have a glimpse of two controlled
variables: (1) a perception that I am questioning you in a
particular way, and a reference signal specifying that this
perception should continue rather than cease, and (2) a reference
signal specifying that a perception of a valid outcome of this
process would be desireable, and (judging from the fact that you
have not yet accepted the success of the process), a perception
that the outcome is not yet perceived as valid. The implied error
signals mean that you are not yet ready to judge that the
challenge has been met and will continue participating.

As to the third question, you indicated a preference for an
answer somewhere between absolute success and absolute failure.
May I correctly interpret this as a reference level regarding the
concepts of success and failure, indicating that you prefer a
continuous scale over a binary one in this case?

This is, of course, a model-based interpretation as are all my
interpretations. You're the only one in a position to say whether
the interpretations fit what you observe. How about it?

···

------------------------------------------------------------
Best,

Bill P.