Facts, Stress, Info, Emergence

[From Rick Marken (930618.0900)]

Mary Powers (930617.1530) --

I suspect that this [menstrual synchronization] is one of
the gadzillion physiological and psychological *facts* about
people in which some statistical probability has been elevated
to rock-hard 100% certainty.

I guess we men were all a bit too fast to accept the menstrual
phenomenon as "fact". But this does remind me of one topic
that should definitely be included in the BBS article (book?).
The topic would be something like:

How does PCT explain X?

Where X would be all the purported phenomena that behavioral
scientists think are phenomena (usually based on statistical
tests) but are not phenomena from a PCT perspective. It looks
like menstrual synchronization could be one of the prime examples
(maybe Jeff Hunter could post the relevent published reports on
this phenomenon). As Mary says, there are probabaly a gadzillion
"facts" of this kind to choose from: the social psych and sociology
literature would be particularly fruitful, I imagine: obediance,
bystander intervention, etc. All purported facts that are not facts.
Not all the non-facts would be statistical. The idea that reinforcement
increases the probability of behavior is another non-fact because it
is only true in special circumstances (when the animal is starved and
it's connection to food input is weak [a demanding schedule]). The "fact"
is that animals control food input; the apparent "effect" of that input
(reinforcement) on behavior depends on parameters of the feedback

Greg Williams (930618)--

So how would you assign blame, based on PCT ideas?

I would hope that people could cooperate to achieve their goals
in a business. I know that business people often violate this
cooperative relationship, setting up contingencies that can hurt
people. I know that business people can do bad things -- and I
blame them for it. But I also know that employees take advantage
of business people by claiming to have been caused stress when,
in fact, the stress, if they have it, is not caused. The assumption
that the environment causes stress is a problem, not merely because
it is false, but because it allows workers to perpetuate a fraud; all
they have to do is point to a "stress inducing" stimulus to which they
were exposed and the court is required to assume that the result
was stress. This is what I object to; using a fake psychological
model to make fraud easy.

I am not against workman's comp and there certainly are "stimuli"
out there that can cause bad results; if a heavy piece of equiment
drops on your foot then it will cause damage. I think a worker should
unquestionably be compensated for the effects of this kind of cause.
But exposure to "stress information" should not be a basis for
assuming that the behavioral result was stress -- because there
is no such thing as stressful information.

Your example of lying to me and causing a heart attack is cute; I would
be annoyed but I probably would not think of sueing you; I'm just not
a sueing kindda guy. But that is a situation where I might give the
"victim" of your "stressful stimulus" a sympathetic hearing -- at least
because you did apply that "stimulus" intentionally. I'm sure Linda
would have no problem accepting a person's stress claim if it were
the result of maliciousness on the part of someone else; I still think
the stress (if it exists) was not caused by the "stressful stimuli" applied
by the malicious person but that person did, intentionally, create the
circumstances where reorganization (and the attendant stress)
was possibly required.

The problem with workman's comp is that you can make a stress claim
just by showing that you were exposed to stressful information; it is
not necessary to show that this information was presented to you
maliciously. As I said, if someone left the door to your files locked by
accident and you can make a case that a locked file is a stressful
event for you, then you're in the money. That's what is no good about
the current Workman's Comp law -- it encourages fraud and makes it
VERY EASY to carry it out. I am, however, all for protecting workers
and think that they should be generously compensated for REAL on
the job misfortunes.

>Your wife couldn't be a Republican, could she?

We're both unclassifiable. She, like me, would probably call
herself a democrat, but with democrats like Sam Nunn
running around, and republicans like Goldwater (who wrote a
wonderful, very pro gays in the military editorial recently)
I don't know that either of us knows what to be. We both encourage
individual freedom tempered by communtarian cooperation. I suppose
I'm a libertarian socialist; and Linda's a tad to the right (left?) of that.

Martin Taylor (930617 20:30)--

It's the unreconstructible
part of the disturbance that leads to the need for more layers of ECSs
in the hierarchy.

This is news. I thought the layers were there to control perceptual aspects
of the environment that are not represented in lower level perceptual

If you can control in one layer, why go for more?

See above.

The question at issue (for me) is how much is there that needs to be
dealt with by higher layers.

They are dealing with completely different kinds of perceptions -- not
the residual variance of controlled variables at a lower level. At least,
in the HPCT model, that's how it works.

what is controlled (compensated for) can
be reconstructed using the perceptual signal, the reference, and the
output function.

And the feedback function.

Of course you can reconstruct the disturbance if you know all this; then
you know everything; All you are saying is that, in an control system,
d = 1/g(o) which we are all well aware of. Where does information
theory make its contribution? I thought that you were saying that
there is something about the information in p and r that would let
you compute the outputs that compensate for the disturbance. If
you are not saying that, then we have no argument. All we have is a
disagreement about the value of info theory for understanding control;
you think it has some, I think that it has none.

Since you know "exactly" how a control system works, I presume that tells
you "exactly" what can be reconstructed and what cannot.


Hans Blom (930617a) --

You are of course right: the individual control
laws together 'control' flockness, but in an emergent way

I think we are in violent AGREEMENT. The confusion (probably
mine) comes from my understanding of "emergent" as "irrelevant
side effect" or "uncontrolled result". Your "flockness" is a variable
controlled by the individual members and you agree that this is
the case. You go on to call this an emergent variable (meaning,
to me uncontrolled) -- I would prefer to call it a controlled result
of collective action.

But it is clear from you post that you do not imagine that there is
some godlike control system (the "social control system") looking
over the flock and keeping flockness under control. That's all that
Tom and Bill and I are objecting too -- the idea of an unseen social
control system that is "outside" the individuals in the collective.
There are variables (that you call emergent) that are controlled by
the collective and can only be controlled by the collective of
individuals. Perhaps people thought we were denying this. But this
is precisely what Tom Bourbon's experiments show. Our point has
only been that the only control systems involved in producing these
controlled variables are the individual ones in the collective;
the collective ITSELF does NOT become a control system; it seems
like it does because the collective controlled variable depends
on the actions of all the individuals -- but the collective of individuals
itself is not a control system because the reference signals for the
variables that must controlled in order to controlled the collective
controlled variable exist only in the heads of the individual members
of the collective. Any change in the individual goals of the members
ends control of the collective controlled variable.