How does PCT explain where our basic characteristics come from?

_Boris_Hartman · December 18, 2011, 11:18am

BP : The mutations are what the organism does in response to the selection pressures, in an attempt to nullify their effects.

BH : This look like “full” Lamarkian theory if I understand it right. But as far as I know from rats, girafes and so on…individual changes during life, doesn’t count in creation of new organism, except if heredity informations are changed. There can probably be problems in ontogenesis of individuals as self-producting process of creating and maintaining the structrure and organization of organism is concerned. Maybe we have different understanding of what mutation means.

Mutations = new heredity characteristics (that’s how is translated in our vocabulary). I don’t know how it goes in american-american vocabulary. As I understand, it means any change in heredity characteristics not just as result of “nullifying efects”.

BP ealrier : Of course reorganization theory is strange in another way because it claims that mutations are driven by error signals inside the organism, not by cosmic rays or mutagenic chemicals or other noxious causes from outside (though it doesn’t rule out such effects).

BH : Maybe this kind of explanation could be exceptable (I admitt it is possibility) from some angle. But…

BP earlier : Selection pressures are surely caused by environmental events, but they do not directly result in mutations. The mutations are what the organism does in response to the selection pressures, in an attempt to nullify their effects.

BH : This is probably to much “exclusive”.

BP : It has been pretty well established that E. coli itself initiates changes in its own DNA when certain stresses occur. Here is an example from Science magazine:

http://www.bichat.inserm.fr/equipes/emi0339/publications_pdf/exhaustiveliste/2003science.pdf

BH : Your first statement talks about exclusive “selective pressure of environment which do not directly result in mutations”, while in your second statement you use terms as : “pretty well established”, “when certain stresses occur”. So it’s not so sure (not so much exclusive) that there are no “direct result in mutation” by external events. I could agree that in different occasions (different circumstances) also internal errors which are “caused” by “external” can maybe influence the structure of DNA. So it doesn’t prove to me, that “what’s going on” is THE ONLY MECHANISM" which is changing the structure of DNA. It can be changed in many ways. But I’m still sceptical about “exclusively” internal mechanism which is changing DNA as it’s “triggered” by CERTAIN external events (special conditions, strange concentrations).These are mostly laboratory experiments and we know at least from Ashby that laboratory experiments doesn’t show the same results in natural environments.

I think it’s your decision what to beleive. And mine to. My intention was only to find out whether we can establish what is determining “the reference levels of intrinsic variables” in any cell. And maybe how we can explain the “reduction” of intrisic errors ? If I remember right Rick’s statements were : “Intrinsic reference are reference specifications for the values of intrinsic variables”.
“The hierarchy of control … develops as a means of keeping intrinsic variables matching their references”.

How all this is concretly happening ? Not abstractly ?

BP : You have difficulty in seeing how a reference condition could be genetically specified. How about body temperature, or all the other homeostatically controlled variables you are interested in? Many of them have variable reference conditions, but then we can ask where the reference conditions at the next higher level come from. At some level we will find a constant reference signal, zero or nonzero, and that will be genetically specified. How else could it be specified?

BH : I have no difficulty in seeing how a reference condition could be genetically specified. I thought you have ? O.K., now as you mentioned it, how are they genetically specified by your opinion ?

BP : All that is necessary to establish a reference condition is some physiological characteristic that determines the level of perceptual input to a control system at which its output will be exactly zero. A simple threshold will suffice: no output until the input reaches some magnitude. Or it could be a real neural signal. There are many neurons that have a rate of spontaneous firing without any incoming signals. Those neurons are clearly inheritable parts of the nervous system. If the output of such a neuron enters a comparator neuron (as either the inhibitory or excitatory input), it will establish a reference condition that a control system will seek.

BH : Why did you limit yourself only to a behaviroal part of nervous system ? Although as you said is genetically determined. I thought we are talking about “reference specifications of intrinsic variables” ?

BP : I didn’t realize that what bothered you was the idea of a genetically specified reference level. I have taken that for granted, because there are so many ways it can come into existence.

BH : You have taken it for granted ? By all these talkings about “reorganization”, Lamarckism, changing DNA… ?Which are these “so many ways” ?

BP : I wonder if the problem here isn’t in your concept of a reference signal. Are you thinking that a reference signal has to have some specific meaning in addition to having a specific magnitude? That the reference signal for one controlled perception will somehow have a different nature from a reference signal for a different controlled perception?

BH : I suppose you are to narrow in thinking here. But maybe I could have problems in understanding you, as I always have. Which reference signal you are talking about ? Are we talking about “intrinsic reference signal” ?

BP : A reference signal is just a neural signal…

BH : Again which reference signal you have in mind ? Intrinsic reference signal or the one in behavioral hierarchy according to Figure 14.1.which is called only “reference signal” and is coming from nowhere to hierachy on 11.level ? Are there any other ?

BP : So basically a reference signal, whether actual or only a structural bias, can easily be inherited if it’s not the output of some higher-order system. It is simply whatever properties have to exist to determine how much perceptual signal is to produce zero error signal.

BH : Maybe I don’t understand right again, but are you saying that there could me more kind of reference signals in behavioral hierarchy ? Inherited or the output of some higher-order system ? Does inherited means “genetic” ?

BP : Is there some other
problem with the ideas I’m offering that you haven’t mentioned yet?

BH : I’m wondering if it couldn’t be better that all in PCT is explained with “connected” control units and “local stabilities” ? Why giving the “reorganization” so much importance ? Without that importance, I imagine that your PCT could work better or as Ashby said that the things could go for better…Reorganization looks like to me as some kind of “feed-back” to DNA self-production…O.K. if you keep term “reorganization” will it still have the place it has now according to Figure 14.1. p 191 (between output of genetic source and behavioral heirarchy) ?

I admitt I got “hint from behind” what could mean term “reorganization”. If I understand it now right, term “reorganization” is some “systematic” mechanism in ontogenesis and phylogenesis, which is “taking care for the things to go better” as Ashby noted by “trials and errors” or as my hint said (in short) : the bacterium (principle) goes in one direction for a while, then in another random direction and change direction when things stop “getting better” and start “getting worse”. We could call it something like “If it’s not getting better, try something else”.

If for example DNA is self-producting homeostasis in the cell, than, if I understand right, in the case that DNA is not self-producing “optimum” homeostasis (constancy of internal conditions) in certain external environment, “reorganization will seek” for better homeostasis, what could mean better “adapted” organism to certain external conditions (probably better ultra-stability). This is probably goal-seeking behavior as Ashby called it. In your terms it’s probably “goal-oriented” or “goal-directed” behavior. Do I understand right ? So reorganization has it’s goal ? And to attain that goal “reorganization” can even “change some genetic properties” of DNA ?

Best,

Boris

···

----- Original Message -----

From:
Bill Powers

To: CSGNET@LISTSERV.ILLINOIS.EDU

Sent: Thursday, December 15, 2011 5:24 PM

Subject: Re: How does PCT explain where our basic characteristics come from?

[From Bill Powers (2011.12.15.0919 MST)]

At 11:00 PM 12/14/2011 +0100, Boris Hartmann wrote:
BP earlier:I haven't spoken about intrinsic variables inside E. coli, or its homeostatic systems if it has any.

BH: I went through some titles in researches of E.coli. It says that this bacteria is one of the most well-explained as there were so  many experiments speccially long term. There are articles about how some substances are kept in steady-state by genetic codes.
BP: I agree that E. coli may have – does have – homeostatic systems, but those aspect of it were not what I was talking about.
BH: From some articles I got an impression that E.coli is normal cell maintaining it's homeostasis and showing Darwins characteristics of "natural selection".
BP: What is observed is that E. coli does evolve new characteristics, even inheritable characteristics. How that happens is not observed. Natural selection is a theory about unobserved causes of changes in characteristics.
Reorgnaization is a different theory about the same thing.
BP earlier : Of course reorganization theory is strange in another way because it claims that mutations are driven by error signals inside the organism, not by cosmic rays or mutagenic chemicals or other noxious causes from outside (though it doesn't rule out such effects). Selection pressures are surely caused by environmental events, but they do not directly result in mutations. The mutations are what the organism does in response to the selection pressures, in an attempt to nullify their effects.
BH : Are you saying that some "theoretical" process inside organism is changing DNA ?
It has been pretty well established that E. coli itself initiates changes in its own DNA when certain stresses occur. Here is an example from Science magazine:

http://www.bichat.inserm.fr/equipes/emi0339/publications_pdf/exhaustiveliste/2003science.pdfhttp://www.bichat.inserm.fr/equipes/emi0339/publications_pdf/exhaustiveliste/2003science.pdf

For more, just Google “stress-induced mutation”. I know that this sort of thing is still rejected by many people because it sounds like Lamarkian theory, but apparently Lamark was not as wrong as he seemed. You still can’t create mice with short tails by cutting off the parent’s tails, but at least some organisms are capable of reorganizing themselves, at the level of DNA, to counteract certain kinds of disturbances. And the changes are inheritable.

However, I have to repeat that reorganization inside E. coli is not the subject I was talking about, and I still don’t know if E. coli can do the kind of reorganizing under discussion here: changing organization within a single lifetime.
As you are persistingly avoiding asnwers on some crucial questions which usually involve intrinsic reference signal coming from "genetic source" it's maybe really useless to continue conversation.
You have difficulty in seeing how a reference condition could be genetically specified. How about body temperature, or all the other homeostatically controlled variables you are interested in? Many of them have variable reference conditions, but then we can ask where the reference conditions at the next higher level come from. At some level we will find a constant reference signal, zero or nonzero, and that will be genetically specified. How else could it be specified?

All that is necessary to establish a reference condition is some physiological characteristic that determines the level of perceptual input to a control system at which its output will be exactly zero. A simple threshold will suffice: no output until the input reaches some magnitude. Or it could be a real neural signal. There are many neurons that have a rate of spontaneous firing without any incoming signals. Those neurons are clearly inheritable parts of the nervous system. If the output of such a neuron enters a comparator neuron (as either the inhibitory or excitatory input), it will establish a reference condition that a control system will seek.

I didn’t realize that what bothered you was the idea of a genetically specified reference level. I have taken that for granted, because there are so many ways it can come into existence.
 When you will find time and explain how does control work from genetic source and how reference levels of intrinsic variables are set (as that was the problem in my conversation with Rick) than I suppose our converstaion will have any sense. It's obviously that some members are tyred of this kind of theorizing.
I guess we all are.
I wonder if the problem here isn’t in your concept of a reference signal. Are you thinking that a reference signal has to have some specific meaning in addition to having a specific magnitude? That the reference signal for one controlled perception will somehow have a different nature from a reference signal for a different controlled perception?
A reference signal is just a neural signal, if it’s anything more than a threshold effect or some other built-in structural bias. All it does is indicate a magnitude; it doesn’t even identify the variable that is to have that magnitude. The nature of the controlled variable is set by the organization of the perceptual input function, but the perceptual signal contains no information about that organization. It’s just a neural signal,. too. The comparator doesn’t know what the two signals entering it mean: they’re just two signal magnitudes, one excitatory and the other inhibitory. If the inhibition is less than the excitation, the net excitation produces an output signal indicating the magnitude of the difference, which we call the error signal.
So basically a reference signal, whether actual or only a structural bias, can easily be inherited if it’s not the output of some higher-order system. It is simply whatever properties have to exist to determine how much perceptual signal is to produce zero error signal.
Is there some other problem with the ideas I’m offering that you haven’t mentioned yet?

Best,

Bill P.

rsmarken · December 18, 2011, 5:29pm

[From Rick Marken (2011.12.18.0930)]

···

On Sun, Dec 18, 2011 at 3:18 AM, boris_upc <boris.hartman@masicom.net> wrote:

BP : The mutations are what the organism does in response to the selection
pressures, in an attempt to nullify their effects.

BH : Thisï¿½look like "full" Lamarkian theory if I understand it right.

You don't. And apparently you won't. So there's really no need to
waste time with the rest of your post.

Best

Rick
--
Richard S. Marken PhD
rsmarken@gmail.com
www.mindreadings.com

MartinT · December 18, 2011, 5:48pm

[Martin Taylor 2011.12 18. 12.23]

[From Bill Powers (2011.12.16.-220 MST)]

Martin Taylor 2011.12.16.00.36 --

[From Bill Powers (2011.12.15.1645 MST)]

[Martin Taylor 2011.12.15.16.36]

MT: I have two questions related to this thread, one long-standing, the other new because I think Bill has introduced something new into the conceptual space.

1. How does the e-coli process escape from a local optimum?

BP: The basically random variations will occasionally introduce large excursions that can help get out of local traps. I don't know what the limits are, or the kinds of local minima found in this application. I would expect that if being stuck in a local minimum causes increasing problems for the organism, reorganization will increase, the larger error producing faster and coarser changes. That might work.

So a bit a simulated annealing in conjunction with hill-climbing.

I don't know what "annealing" means in this context. Another far-fetched physicist's metaphor?

No more than "perception" in "PCT" is a metaphor. It has a well defined meaning in neural network simulation, which is what we are talking about. It means adjusting the probability distributions of large and small changes in parameters, increasing the probabilities of small excursions as the performance improves (i.e. as the optimum is approached).

There's another point, however, that I missed. We're talking about evolution here,

I wasn't. My question was about reorganization of the perceptual control hierarchy. When the environment contains considerable irregularity, including other control systems, the space of possible organizations must contain large numbers of local optima, local in two senses: (1) the best for controlling in this part of the environment, but not in other parts of the environment; (2) better than nearby possible organizations in this part of the environment but not the best that can be in this part of the environment ("nearby" means reachable by small changes in any or all parameters).

MT: 2. I understand what defines a direction in the reorganization of a control network, but what defines a "direction" in genetic space?

This point seems to be addressed by the paper Rick cited, which gave rise to a lot of other studies.

Cairns, J., Overbaugh, J and Miller, S. (1988) The origin of mutants.
Nature, 335, 142-145

Here the genes themselves mutated so E. coli became able to live on galactose or glucose alone when that was all that was available. Of course I don't know if those specific mutations were done in the E. coli style, but they were definitely generated inside the bacterium as a stress response. There's little doubt about mutation rate being varied as a stress response in many organisms. Those with the shortest generation time are the best known, of course.

Yes, that this happens is well understood, I think. I'm not clear how it relates to my question, though.

[Elided discussion of continuing evolutionary changes in specific properties of species]...

BP earlier: The "direction," of course is a metaphor for varying relative rates of change in some set of parameters, each considered as a different dimension in some hyperspace.

MT: But what I find difficult is to see how you continue the direction if you have changed a C to a T in a DNA strand and "things get better". How do you do more of the same?

BP: That's a discrete change, of course. But what about insertions and deletions? These have the effect of changing the relationships between sequences that occur before and after the changed part. And there's also the effect on folding of the DNA and the proteins synthesized according to the sequences. We're talking about 20,000 genes, and each gene consists of a lot of base pairs, so the discrete changes quickly begin to look smooth as you back away from the close-up view.

I don't think the last clause follows at all from the preceding. If it does, you must have some chain of reasoning that leads from the different kinds of chromosomal changes to the idea that they all add up to smoothness as though the "law of large numbers" applied.

MT: Change more Cs to Ts? That would usually create more of the same effect, in the way that changing the weights in the hierarchy would usually do (until you got to a dynamic bifurcation point).

BP: Yes, that's what I thought, too.

Sorry, there was a crucial "not" missing in my post. I intended to imply that changing a C to a T in one gene would probably have a vastly different effect from the effect it would have if the change were in a different gene. But if you believe that all changes of C to T have the same effect, and I presume you think the same about changes of A to G, then the space available for evolution is only two-dimensional (I seem to remember that not all interchanges are allowed).

MT: Anyway, my question wasn't about all of that. It was just about how one would specify a direction in gene space so that subsequent mutations could expand on the benefits of a good mutation. I don't mean specify for our analytic purposes, but specify in a way that would make practical sense within the organism, in the way that it does for reorganizing the control hierarchy.

BP: This is probably a good place to switch from the reinforcement concept to the reorganization concept. Reorganization doesn't try to "expand on the benefits of a good mutation." It simply causes mutations when errors increase, and stops changing them when they decrease. Introducing direction as the mutated variable, we can see that as long as error is decreasing, the new direction of change will be followed. Sooner or later, the continuing change will stop reducing the error and will start making the error larger. Then there will be a change of direction, another mutation. Some dinosaurs will start getting smaller again, and growing more feathers.

Let me offer a background question that may help you interpret my question about how "direction" can be defined in a geneset, since your first three sentences simply restate the principle of natural selection augmented by the stress response.

The background to my question can be restated as: "Given that some genetic change increased the probability that this geneset will be included in a later population, and supposing that it is possible to define a "direction" in genespace, what is passed to the offspring that tells the offspring (a) that it is in some way "better" or "not better" than its parents, and (b) what direction of change made it better than its parents, and how does the offspring implement "more" of that direction of change?"

The offspring must have these two pieces of information if evolution is to implement the e-coli process.

Martin

···

On 2011/12/15 6:47 PM, Bill Powers wrote:

MartinT · December 18, 2011, 5:53pm

[Martin Taylor 2011.12.18.12.48]

[From Rick Marken (2011.12.18.0930)]

BP : The mutations are what the organism does in response to the selection
pressures, in an attempt to nullify their effects.
BH : This look like "full" Lamarkian theory if I understand it right.

You don't. And apparently you won't. So there's really no need to
waste time with the rest of your post.

What a delighfully Christmassy message of help for someone who, for very practical reasons, is seriously trying to understand.

You might try rewording what Bill is trying to get across, since you clearly have it very clearly in your mind. Bill's way of putting it either isn't getting across or isn't making sense to Boris. Maybe you could put it in some way that would allow Bill and Boris to determine whether there is a misunderstanding or a scientific disagreement.

Martin

···

On Sun, Dec 18, 2011 at 3:18 AM, boris_upc<boris.hartman@masicom.net> wrote:

_Boris_Hartman · December 18, 2011, 7:24pm

BH : This look like "full" Lamarkian theory if I understand it right.

RM : You don't. And apparently you won't. So there's really no need to
waste time with the rest of your post.

BH : Well what a hostile opinion. Do I feel anger in your "voice" ? I think the way you control your high valued perception surprised me a little. It's somehow not civilized or siutable. Disqualifying somebody without arguments. Well I saw your "hostile control behavior" quite some time, but I think you never controlled your perception in this way against me. I'll try to be more civilized and I'll try to put one more argument why I thought, what Bill wrote is Lamarckian theory. I'll help myself with your opinion.

RM much, much earlier (15.12.2009, 20:27) : "I think the E.coli "reorganization" model of evolution can be
considered somewhat Lamarckian... think this is reasonably good quick and dirty summary of the control model of evolution, as Bill described it some time ago."

BH : I just couldn't forget your critical remark Rick. I thought that you were right. After all you are some PCT authority. I think I've always appreciate your opinion to some extent. So we don't need to be hostile. We are "on the same boat".

Merry Christmas Rick, and many dancing success.

Boris

···

----- Original Message ----- From: "Richard Marken" <rsmarken@GMAIL.COM>
To: <CSGNET@LISTSERV.ILLINOIS.EDU>
Sent: Sunday, December 18, 2011 6:29 PM
Subject: Re: How does PCT explain where our basic characteristics come from?

[From Rick Marken (2011.12.18.0930)]

On Sun, Dec 18, 2011 at 3:18 AM, boris_upc <boris.hartman@masicom.net> wrote:

BP : The mutations are what the organism does in response to the selection
pressures, in an attempt to nullify their effects.

BH : This look like "full" Lamarkian theory if I understand it right.

You don't. And apparently you won't. So there's really no need to
waste time with the rest of your post.

Best

Rick
--
Richard S. Marken PhD
rsmarken@gmail.com
www.mindreadings.com

rsmarken · December 18, 2011, 7:30pm

[From Rick Marken (2011.12.18.1130)]

Martin Taylor (2011.12.18.12.48)--

A belated happy birthday to Beethoven and Jane Austen, two of my
favorite people, both born on Dec 16. Best Xmas presents of all time.
Oh, except for one, but no one really paid much attention to him
anyway;-)

BP : The mutations are what the organism does in response to the
selection pressures, in an attempt to nullify their effects.
BH : This look like "full" Lamarkian theory if I understand it right.

RM: You don't. And apparently you won't. So there's really no need to
waste time with the rest of your post.

MT: What a delighfully Christmassy message of help for someone who, for very
practical reasons, is seriously trying to understand.

Oh come on, Martin, have you be reading Boris' posts?

MT: You might try rewording what Bill is trying to get across, since you clearly
have it very clearly in your mind. Bill's way of putting it either isn't
getting across or isn't making sense to Boris. Maybe you could put it in
some way that would allow Bill and Boris to determine whether there is a
misunderstanding or a scientific disagreement.

Actually, Bill has been trying to get across what I was trying to get
across earlier. But whatever Bill says, Boris always comes back to my
statement that "Intrinsic references are reference specifications for
the values of intrinsic variables" and that "The hierarchy of control
.... develops as a means of keeping intrinsic variables matching their
references" and repeats them as though they were evidence that there
is something wrong with them or with the concept of reorganization.
It's kind of like a Fox News interview; do they have Fox News where
Boris is?).

Bill keeps explaining what my statements mean (they pretty much mean
what they say;-) and Boris keeps saying that the explanations are
insufficient. Bill has a version of several control models (the little
man, the inverted pendulum, I think) which learn to control using the
E. Coli reorganization algorithm so we know how the algorithm works
and that it does work. Anyone who is "seriously trying to understand"
reorganization would have understood it it by now. So when Boris says
"If I understand correctly..." in his posts it is clearly a con. He
doesn't understand because he doesn't want to understand; and he
doesn't want to understand because he thinks he already _does_
understand; he thinks Ashby has the answer. Boris, like Gavin Ritz, is
a good example of what happens when you come to PCT with a strong
prior agenda -- and it ain't pretty.

Boris isn't even that good at seeming like he's trying to understand.
I can't believe he's got you fooled but maybe I can;-)

Best

Rick

···

--
Richard S. Marken PhD
rsmarken@gmail.com
www.mindreadings.com

_Boris_Hartman · December 18, 2011, 7:53pm

MT : What a delighfully Christmassy message of help for someone who, for very
practical reasons, is seriously trying to understand.

You might try rewording what Bill is trying to get across, since you
clearly have it very clearly in your mind. Bill's way of putting it
either isn't getting across or isn't making sense to Boris. Maybe you
could put it in some way that would allow Bill and Boris to determine
whether there is a misunderstanding or a scientific disagreement.

BH : Martin beside that you are a great mind, I see you are a real gentlemen. Thanks for your comment. You are right. Me and Bill have to "determine whether there is a misunderstanding or a scientific disagreement". I'm sure that we all could come to some agreement and more "efficient PCT picture" if we'll be persistant at putting our arguments. I said many times that on most parts of PCT I feel highest respect toward Bill. I think that those "efficient" parts of PCT makes him immortal.
But I also think that Complex Perception of Bill's PCT has to be matched against other Complex Perceptions about how organism works.I think this is the way how knowledge grows.

Thanks again for your mediation Martin.

Merry Christmas,

Boris

···

----- Original Message ----- From: "Martin Taylor" <mmt-csg@MMTAYLOR.NET>
To: <CSGNET@LISTSERV.ILLINOIS.EDU>
Sent: Sunday, December 18, 2011 6:53 PM
Subject: Re: How does PCT explain where our basic characteristics come from?

[Martin Taylor 2011.12.18.12.48]

[From Rick Marken (2011.12.18.0930)]

On Sun, Dec 18, 2011 at 3:18 AM, boris_upc<boris.hartman@masicom.net> >> wrote:

BP : The mutations are what the organism does in response to the selection
pressures, in an attempt to nullify their effects.
BH : This look like "full" Lamarkian theory if I understand it right.

You don't. And apparently you won't. So there's really no need to
waste time with the rest of your post.

What a delighfully Christmassy message of help for someone who, for very practical reasons, is seriously trying to understand.

You might try rewording what Bill is trying to get across, since you clearly have it very clearly in your mind. Bill's way of putting it either isn't getting across or isn't making sense to Boris. Maybe you could put it in some way that would allow Bill and Boris to determine whether there is a misunderstanding or a scientific disagreement.

Martin

rsmarken · December 18, 2011, 8:02pm

[From Rick Marken (2011.12.18.1200)]

Martin Taylor (2011.12 18. 12.23)--

MT: But what I find difficult is to see how you continue the direction if
you have changed a C to a T in a DNA strand and "things get better". How do
you do more of the same?

I think you just don't change again, or at least reduce the
probability of a mutation.

MT: The background to my question can be restated as: "Given that some genetic
change increased the probability that this geneset will be included in a
later population, and supposing that it is possible to define a "direction"
in genespace, what is passed to the offspring that tells the offspring (a)
that it is in some way "better" or "not better" than its parents, and (b)
what direction of change made it better than its parents, and how does the
offspring implement "more" of that direction of change?"

I think the answer to (a) is that the genetic change results in some
phenotypic characteristic that allows the offspring to keep its level
of intrinsic error lower than it was in the parent generation so that
the probability of another mutation in the offspring generation is
reduced. The answer to (b) would be a phenotypic change that allows
better control and, thus, reduction of intrinsic error. An example
might be a genetic change that results in a slight change in the
orientation of the hip joint that allows support of the body while
standing, thus allowing the organism to control for walking while
using much less energy.

I really must write an evolution simulation. My little E. coli demo is
not a good analog of evolution. I've got some ideas about how to do it
and I might just have to do it to show (to myself, at least) the
difference between pure natural selection and the E. coli process.
Mainly I have to do this for myself because, on re-thinking it, I'm
pretty sure that my idea about pure natural selection not ending up in
a fairly steady state is wrong. I think the only difference between
pure natural selection and E. coli type natural selection will be
efficiency, which is what Bill has argued all along. But I think we
(I?) need a plausible simulation to see what really happens.

Best

Rick

···

--
Richard S. Marken PhD
rsmarken@gmail.com
www.mindreadings.com

_Boris_Hartman · December 18, 2011, 9:00pm

RM : "If I understand correctly..." in his posts it is clearly a con. He
doesn't understand because he doesn't want to understand; and he
doesn't want to understand because he thinks he already _does_
understand; he thinks Ashby has the answer. Boris, like Gavin Ritz, is
a good example of what happens when you come to PCT with a strong
prior agenda -- and it ain't pretty.

BH : Well Rick you look now like quite unrational in your "uncontrolled" actions
(emotional explosions). You are disqualifying the human freedom of thinking
and expresing their opinion.

RM : Bill keeps explaining what my statements mean (they pretty much mean
what they say;-) and Boris keeps saying that the explanations are
insufficient. Bill has a version of several control models (the little
man, the inverted pendulum, I think) which learn to control using the
E. Coli reorganization algorithm so we know how the algorithm works
and that it does work. Anyone who is "seriously trying to understand"
reorganization would have understood it it by now.

BH : Do you understand it Rick ? Did you understand it 2 and half years ago ?
Do you really think that you understand PCT ? Do you think that you know how organisms
function ?
It's not that I don't understand it and it's not that I understand it. It's something in the middle.
But beleive me that I'd like to understand how human mind (organism) works.
In the matter of fact I have to in some sense. I've got caught in study and in different
conversations about PCT and "control" in living beings.
Bill and my conversations with him has important role in what I'm doing.

We can not predict questions and answers when people clearly shows "control" characteristics
and "reorganize" to defend their positions. Does it look like familiar to you ?
So I have to be inovative. I have to "reorganize" in my way, if I want to defend Control position.
PCT doesn't have all answers. We've established that many times. You were one of them and Bruce
who doubted and Martin T, and Martin L. and Gregory and Gavin and so on....

Bill has many answers and I'd like to understand what he thinks. But I also feel
that we could help each other understand more perfectly how organism works.
It's not only Ashby, Maturana that helped clarifying how "Living beings work",
it's whole physiology, neurophysiology, biology, psychology, physics, mathematics
and there's more other knowledge that will be by my opinion needed to understand
how organism and evolution works. O.K. PCT came very close. But it's not there.

Do you really think that blind beleiving in one paradigm is what makes human take
"control" over environment, make civilization ?
Well Rick if you beleive that, I'm sure that you are on wrong way.

RM : I really must write an evolution simulation. My little E. coli demo is
not a good analog of evolution. I've got some ideas about how to do it
and I might just have to do it to show (to myself, at least) the
difference between pure natural selection and the E. coli process.
Mainly I have to do this for myself because, on re-thinking it, I'm
pretty sure that my idea about pure natural selection not ending up in
a fairly steady state is wrong. I think the only difference between
pure natural selection and E. coli type natural selection will be
efficiency, which is what Bill has argued all along. But I think we
(I?) need a plausible simulation to see what really happens.

BH : That's the way to go. After you make simulation (laboratory)
you'll probably have to try it with free-living organisms.
I hope you'll show me results :)).

Best,

Boris

···

----- Original Message ----- From: "Richard Marken" <rsmarken@GMAIL.COM>
To: <CSGNET@LISTSERV.ILLINOIS.EDU>
Sent: Sunday, December 18, 2011 8:30 PM
Subject: Re: How does PCT explain where our basic characteristics come from?

[From Rick Marken (2011.12.18.1130)]

Martin Taylor (2011.12.18.12.48)--

A belated happy birthday to Beethoven and Jane Austen, two of my
favorite people, both born on Dec 16. Best Xmas presents of all time.
Oh, except for one, but no one really paid much attention to him
anyway;-)

BP : The mutations are what the organism does in response to the
selection pressures, in an attempt to nullify their effects.
BH : This look like "full" Lamarkian theory if I understand it right.

RM: You don't. And apparently you won't. So there's really no need to
waste time with the rest of your post.

MT: What a delighfully Christmassy message of help for someone who, for very
practical reasons, is seriously trying to understand.

Oh come on, Martin, have you be reading Boris' posts?

MT: You might try rewording what Bill is trying to get across, since you clearly
have it very clearly in your mind. Bill's way of putting it either isn't
getting across or isn't making sense to Boris. Maybe you could put it in
some way that would allow Bill and Boris to determine whether there is a
misunderstanding or a scientific disagreement.

Actually, Bill has been trying to get across what I was trying to get
across earlier. But whatever Bill says, Boris always comes back to my
statement that "Intrinsic references are reference specifications for
the values of intrinsic variables" and that "The hierarchy of control
.... develops as a means of keeping intrinsic variables matching their
references" and repeats them as though they were evidence that there
is something wrong with them or with the concept of reorganization.
It's kind of like a Fox News interview; do they have Fox News where
Boris is?).

Bill keeps explaining what my statements mean (they pretty much mean
what they say;-) and Boris keeps saying that the explanations are
insufficient. Bill has a version of several control models (the little
man, the inverted pendulum, I think) which learn to control using the
E. Coli reorganization algorithm so we know how the algorithm works
and that it does work. Anyone who is "seriously trying to understand"
reorganization would have understood it it by now. So when Boris says
"If I understand correctly..." in his posts it is clearly a con. He
doesn't understand because he doesn't want to understand; and he
doesn't want to understand because he thinks he already _does_
understand; he thinks Ashby has the answer. Boris, like Gavin Ritz, is
a good example of what happens when you come to PCT with a strong
prior agenda -- and it ain't pretty.

Boris isn't even that good at seeming like he's trying to understand.
I can't believe he's got you fooled but maybe I can;-)

Best

Rick

--
Richard S. Marken PhD
rsmarken@gmail.com
www.mindreadings.com

rsmarken · December 18, 2011, 10:45pm

[From Rick Marken (2011.12.18.1445)]

RM : "If I understand correctly..." in his posts it is clearly a con...

BH : Well Rick you look now like quite unrational in your "uncontrolled"
actions (emotional explosions).

I beg you pardon. My irrational explosions are completely controlled;-)

RM : Bill keeps explaining what my statements mean (they pretty much mean
what they say;-) and Boris keeps saying that the explanations are
insufficient. ..

BH : Do you understand it Rick ? Did you understand it 2 and half years ago?

Yes.

BH: Do you really think that you understand PCT ?

Yes, well enough to test it.

BH: Do you think that you know how organisms function ?

At a general level, yes. They control their perceptions.

BH: We can not predict questions and answers when people clearly shows "control"
characteristics and "reorganize" to defend their positions. Does it look like familiar to
you ?

I have no idea what you're talking about.

BH: So I have to be inovative. I have to "reorganize" in my way, if I want to
defend Control position.

Why not just try building actual models and testing them against
actual behavior. There is no need to _defend_ positions. This is
science, not politics.

BH: PCT doesn't have all answers.

Of course not. PCT is a general model of the organization of behavior.
It is therefore a possible model of how behavior works. It's an
"answer" to questions about behavior only after it has been tested and
passed the test.

BH: Bill has many answers and I'd like to understand what he thinks. But I also
feel that we could help each other understand more perfectly how organism works.

Sure we could. Be we can't do it by just talking. You have to go out
and actually build and test the models. If you want to see if the PCT
reorganization model is better than, say, Ashby's, then you have to 1)
learn what the PCT model of reorganization is (I presume you already
know Ashby's) and then 2) devise tests that will show which one is
better. You are still pre-number (1); you are arguing about what the
PCT model of organization _is_. We're trying to explain what it is and
how it works and you keep saying that the model is wrong. That's why
my patience has run out.

It's not only Ashby, Maturana that helped clarifying how "Living beings
work"

They haven't clarified how living systems work (from my perspective)
until they have tested working versions of their models empirically.
What they have done is provided some speculations; these are of no
value until their is empirical evidence to support them. There is lots
of empirical evidence that supports PCT which is why I find it so much
more interesting and compelling than other ideas about how living
beings work.

BH: Do you really think that blind beleiving in one paradigm is what makes human
take "control" over environment, make civilization ?

No, I believe only in empirically tested (non-blind) believing.

RSM

···

On Sun, Dec 18, 2011 at 1:00 PM, boris_upc <boris.hartman@masicom.net> wrote:
--
Richard S. Marken PhD
rsmarken@gmail.com
www.mindreadings.com

Bill_Powers1 · December 19, 2011, 11:11pm

[From Bill Powers (2011.12.19.1400 MST)]

BP : The mutations are what the
organism does in response to the selection pressures, in an attempt to
nullify their effects.

BH : This look like “full” Lamarkian theory if I understand it
right. But as far as I know from rats, girafes and so on…individual
changes during life, doesn’t count in creation of new organism, except if
heredity informations are changed.

Yes, and this relates to what Martin and Ted were talking about. Only
mutations that affect the DNA passed to the next generation matter; this
is why cutting the tails off mice does not result in offspring with short
tails. Lamark was wrong about some of the important details.

What I have to assume (in order to propose a reorganization theory of
mutation) is that the organism’s capacity to reorganize the genome works
specifically on the DNA that is transmitted, not on general body cells
(or if it does work on the latter, the effect would not be passed to the
next generation).

BP earlier : Of course
reorganization theory is strange in another way because it claims that
mutations are driven by error signals inside the organism, not by cosmic
rays or mutagenic chemicals or other noxious causes from outside (though
it doesn’t rule out such effects).

BH : Maybe this kind of explanation could be exceptable (I admitt it is
possibility) from some angle. But…

BP earlier : Selection pressures are surely caused by environmental
events, but they do not directly result in mutations. The mutations are
what the organism does in response to the selection pressures, in an
attempt to nullify their effects.

BH : This is probably to much “exclusive”.

BP: In one way, yes – there are certainly some selection pressures that
work directly on DNA, as in the case of teratogens. But in that case, the
mutations would not have any relationship to error signals in the
organismm, and the result would work like natural selection – very
inefficiently, compared with reorganization.

BP earlier: It has been pretty
well established that E. coli itself initiates changes in its own DNA
when certain stresses occur. Here is an example from Science
magazine:

http://www.bichat.inserm.fr/equipes/emi0339/publications_pdf/exhaustiveliste/2003science.pdf

BH : Your first statement talks about exclusive “selective pressure
of environment which do not directly result in mutations”, while in
your second statement you use terms as : “pretty well
established”, “when certain stresses occur”. So it’s not
so sure (not so much exclusive) that there are no “direct result in
mutation” by external events.

BP: That’s not a direct effect – all we know is that the mutation
follows the appearance of selection pressure. That alone doesn’t
establish the nature of the causal path. It’s the same situation we have
with the “behavioral illusion.” The appearance in that illusion
is that a disturbance simply causes an action. But it is quite possible
that the disturbance tends to alter some controlled variable, and that
the action is organized to oppose the effects of the disturbance and
stabilize the controlled variable against the disturbance. This is the
same situation that holds between selection pressures and mutations. The
appearance is that the selection pressure causes the mutations. But what
I am proposing is also possible: that the mutation is an action by a
reorganizing control system that is part of a control process that
resists the effects of the selection pressure. The Cairnes et. al.
experiments with E. coli show that in the case of a change in available
nutrients, the mutation rates increase by a factor of several hundred and
provide more variants. Some of the descendants are capable of
metabolizing the new nutrient and are able to pass this characteristic on
to offspring. Natural selection comes into play in eliminating the
descendants which did not mutate in this way, but the organism itself
reorganized the genome when starvation threatened so as to produce a
larger range of new organizations. It probably also produced more
failures, just as E. coli’s tumbles are just as likely to turn it in the
wrong direction as the right one.

BH: I could agree that in
different occasions (different circumstances) also internal errors which
are “caused” by “external” can maybe influence the
structure of DNA. So it doesn’t prove to me, that “what’s going
on” is THE ONLY MECHANISM" which is changing the structure of
DNA. It can be changed in many ways. But I’m still sceptical about
“exclusively” internal mechanism which is changing DNA as it’s
“triggered” by CERTAIN external events (special conditions,
strange concentrations).These are mostly laboratory experiments and we
know at least from Ashby that laboratory experiments doesn’t show the
same results in natural environments.

You’re reading the idea of “exclusive” reorganization into my
words, but I keep pointing out that natural selection still works and
plays some part in the process. But that part is mostly in eliminating
the failures; natural selection can’t do anything to cause more mutations
when a change is needed. My main point is based on comparing mutations
(as random jumps from one organization to another) with the E. coli
principle in which the spacing of mutations depends on the amount of
intrinsic error, and in which there is a steady change in system
parameters between mutations at a rate proportional to intrinsic error.
What is mutated is the direction of change, the vector in the parameter
hyperspace, rather than the values of parameters. This is a new principle
in evolutionary theory, I think, and evolutionists are going to have to
give credit to E. coli in their references.

I don’t know if it’s technically possible to tell the difference between
these two modes of random change in a living organism.

I think it’s your decision what
to believe. And mine too. My intention was only to find out whether we
can establish what is determining “the reference levels of intrinsic
variables” in any cell. And maybe how we can explain the
“reduction” of intrisic errors ? If I remember right Rick’s
statements were : “Intrinsic reference are reference specifications
for the values of intrinsic variables”.

“The hierarchy of control … develops as a means of keeping
intrinsic variables matching their references”.

How all this is concretly happening ? Not abstractly ?

I can’t point to the physiological mechanisms. The closest I have come to
a concrete demonstration is to show a model that behaves in the
postulated way, to demonstrate its properties. The model has such an
overwhelming advantage over the traditional concept of mutation that we
really have to investigate it to see if there is evidence for or against
it. It’s not a matter of believing or not believing; normal experimental
procedures will tell us which to do.

BH : I have no difficulty in
seeing how a reference condition could be genetically specified. I
thought you have ? O.K., now as you mentioned it, how are they
genetically specified by your opinion ?

I don’t know; there are many possible ways, and I have no idea which way
is the right one. Reference signals could be in the form of
concentrations of proteins or enzymes or other molecules, or in the form
of circuits containing spontaneously-firing neurons which emit steady
signals.

BP : All that is necessary to
establish a reference condition is some physiological characteristic that
determines the level of perceptual input to a control system at which its
output will be exactly zero. A simple threshold will suffice: no output
until the input reaches some magnitude. Or it could be a real neural
signal. There are many neurons that have a rate of spontaneous firing
without any incoming signals. Those neurons are clearly inheritable parts
of the nervous system. If the output of such a neuron enters a comparator
neuron (as either the inhibitory or excitatory input), it will establish
a reference condition that a control system will seek.

BH : Why did you limit yourself only to a behaviroal part of nervous
system ? Although as you said is genetically determined. I thought we are
talking about “reference specifications of intrinsic variables”
?

Yes, we are, but we don’t know their physical form. They could exist as
chemical concentrations or as neural signals. They could be in the form
of enzymes or energy-bearing molecules.

I am not limiting myself to the systems that use behavior to control
perceptual inputs. In fact I have often mentioned biochemical control
systems. But I know much more about the neural systems, or more exactly I
should say that my knowledge of biochemical systems is more severely
limited than my limited knowledge of the neuromuscular systems.

BP : I didn’t realize that what
bothered you was the idea of a genetically specified reference level. I
have taken that for granted, because there are so many ways it can come
into existence.

BH : You have taken it for granted ? By all these talkings about
“reorganization”, Lamarckism, changing DNA… ?Which are these
“so many ways” ?

I’ll say it once more.

A reference condition is anything that can determine how much input is
needed to bring the output exactly to zero. This can be done by injecting
signals into a control system anywhere in the loop, or by changing the
threshold at which a neuron will begin to fire or a chemical reaction
will start to take place.

BP : I wonder if the problem
here isn’t in your concept of a reference signal. Are you thinking that a
reference signal has to have some specific meaning in addition to having
a specific magnitude? That the reference signal for one controlled
perception will somehow have a different nature from a reference signal
for a different controlled perception?

BH : I suppose you are to narrow in thinking here. But maybe I could have
problems in understanding you, as I always have. Which reference signal
you are talking about ? Are we talking about “intrinsic reference
signal” ?

Any reference signal. Reference signals are all alike; they are simply
indications of magnitude.

BP : A reference signal is just
a neural signal…

BH : Again which reference signal you have in mind ? Intrinsic reference
signal or the one in behavioral hierarchy according to Figure 14.1.which
is called only “reference signal” and is coming from nowhere to
hierachy on 11.level ? Are there any other ?

I was thinking of a neural circuit, but the concept includes any
reference signal in anty physical form. What kind of difference are you
looking for? There is nothing about a reference signal that distinguishes
it from any other kind of signal – a perceptual signal or an error
signal or an output from the output function that is sent to affect the
reference inputs to lower systems. In the nervous system, its just a
series of neural impulses moving along a nerve fiber and action on the
input to a comparator function made of neurons.

BP earlier : So basically a
reference signal, whether actual or only a structural bias, can easily be
inherited if it’s not the output of some higher-order system. It is
simply whatever properties have to exist to determine how much perceptual
signal is to produce zero error signal.

BH : Maybe I don’t understand right again, but are you saying that there
could me more kind of reference signals in behavioral hierarchy ?
Inherited or the output of some higher-order system ? Does inherited
means “genetic” ?

BP: Yes. To say a reference signal can be inherited is to say that its
value is preserved across generations. What could be passed to another
generation might be a particular kind of output function that generates
the error signal. In general, the signal itself would not be inherited;
only the physical components that produce the signal could be inherited.
Reference signals can exist as concentrations of chemicals, as neural
signals, as thresholds of firing. To explain how they could be inherited,
you have to know what is generating them. In the muscle spindles, a
neural signal causes contractions of small muscles at each end of the
spindle, stretching the central portion and affecting the nerve-endings
wrapped around it. That stretch is the comparator. The reference input to
the comparator is the length of the small muscles at the ends; the
perceptual signal is the length of the main muscle to which the muscle
spindle is attached.

BP earlier: Is there some
other problem with the ideas I’m offering that you haven’t
mentioned yet?

BH : I’m wondering if it couldn’t be better that all in PCT is explained
with “connected” control units and “local
stabilities” ? Why giving the “reorganization” so much
importance ? Without that importance, I imagine that your PCT could work
better or as Ashby said that the things could go for
better…Reorganization looks like to me as some kind of
“feed-back” to DNA self-production…O.K. if you keep term
“reorganization” will it still have the place it has now
according to Figure 14.1. p 191 (between output of genetic source and
behavioral heirarchy) ?

It’s been customary for quite a few decades to include learning as one
aspect of behavior that a theory has to handle. PCT is basically an
explanation of how behavior is carried out (or how “local
stabilities” are maintained); the reorganization part is an answer
to the question of how organisms become able to behave as they
do.

BH: I admitt I got “hint from behind” what could mean term
“reorganization”. If I understand it now right, term
“reorganization” is some “systematic” mechanism in
ontogenesis and phylogenesis, which is “taking care for the things
to go better” as Ashby noted by “trials and errors” or as
my hint said (in short) : the bacterium (principle) goes in one direction
for a while, then in another random direction and change direction when
things stop “getting better” and start “getting
worse”. We could call it something like “If it’s not getting
better, try something else”.

BP: That’s a good description of what reorganization does. My model is an
attempt to explain how it does that. I’d change your statement a little:
"“If it’s not getting better, try something else” is
almost the whole story, except I would say “If going in this
direction is making things worse, try a different direction of
change.”

BH: If for example DNA is self-producting homeostasis in the cell,
than, if I understand right, in the case that DNA is not self-producing
“optimum” homeostasis (constancy of internal conditions) in
certain external environment, “reorganization will seek” for
better homeostasis, what could mean better “adapted” organism
to certain external conditions (probably better ultra-stability). This is
probably goal-seeking behavior as Ashby called it. In your terms it’s
probably “goal-oriented” or “goal-directed” behavior.
Do I understand right ? So reorganization has it’s goal ? And to attain
that goal “reorganization” can even “change some genetic
properties” of DNA ?

Its goal is not that general. What you’re describing is the overall
effect of what reorganization accomplishes. But the mechanism by which it
does that is much simpler than the result is. The E. coli principle is
accomplished, in my models, with only about a dozen lines of Delphi
(Pascal) code. Isn’t this one of the goals of modeling behavior? We want
to reduce the complexity of what we observe by finding simple underlying
principles that require only the least amount of skill and intelligence
to carry out. That makes it easier to understand how such a system might
have evolved.

Best,

Bill P.

···

At 12:18 PM 12/18/2011 +0100, Boris Hartman wrote:

Bill_Powers1 · December 20, 2011, 8:07am

[From Bill Powers (2011.12.19,1156 MST)]

Martin Taylor 2011.12 18. 12.23 --

MT: So a bit a simulated annealing in conjunction with hill-climbing.

BP earlier: I don't know what "annealing" means in this context. Another far-fetched physicist's metaphor?

MT: No more than "perception" in "PCT" is a metaphor. It has a well defined meaning in neural network simulation, which is what we are talking about. It means adjusting the probability distributions of large and small changes in parameters, increasing the probabilities of small excursions as the performance improves (i.e. as the optimum is approached).

BP: What does that have to do with the original meaning of annealing, having to do with the effects of heating and cooling metals? That's what I meant by "metaphor." In PCT, perception is not a metaphor; it can be directly experienced without saying it is like something else.

BP earlier: There's another point, however, that I missed. We're talking about evolution here,

MT: I wasn't. My question was about reorganization of the perceptual control hierarchy. When the environment contains considerable irregularity, including other control systems, the space of possible organizations must contain large numbers of local optima, local in two senses: (1) the best for controlling in this part of the environment, but not in other parts of the environment; (2) better than nearby possible organizations in this part of the environment but not the best that can be in this part of the environment ("nearby" means reachable by small changes in any or all parameters).

BP: I think the concept of "best" simply doesn't apply, either in evolution or in reorganization during a single lifetime. Evolution and reorganization cease when they are good enough to eliminate the errors that gave rise to them. When new circumstances arise that make the previous organization insufficient to eliminate error well enough, the processes start up again and continue until once again the errors are eliminated -- that is, reduced to a level where they don't matter any more. This is how I see the problem of local minima being handled -- not by a search for the best possible solution, which almost by definition is futile.

MT: 2. I understand what defines a direction in the reorganization of a control network, but what defines a "direction" in genetic space?

BP earlier: This point seems to be addressed by the paper Rick cited, which gave rise to a lot of other studies.

Cairns, J., Overbaugh, J and Miller, S. (1988) The origin of mutants.
Nature, 335, 142-145

Here the genes themselves mutated so E. coli became able to live on galactose or glucose alone when that was all that was available. Of course I don't know if those specific mutations were done in the E. coli style, but they were definitely generated inside the bacterium as a stress response. There's little doubt about mutation rate being varied as a stress response in many organisms. Those with the shortest generation time are the best known, of course.

MT: Yes, that this happens is well understood, I think. I'm not clear how it relates to my question, though.

BP: Then please clarify your question. What problem do you see in defining a direction in genetic space?

I am seeing genetic space, as you call it, as a space defined by the parameters of the physical systems governed by genetics. Gene expression is continuously variable, and I assume that rearranging the base-pairs that make up genes also alters the way the parameters of the systems being governed by the genes vary. I'm not thinking in terms of either-or characteristics such as blue eyes or curly hair, but degrees of characteristics such as spectral reflectance and curvature of filaments. We habitually classify such continuous variations by binning them, ignoring changes that occur within certain broad boundaries and treating a whole range of variations as if they are constants. This is an artifact of the capacity to categorize, which has its uses but also its limitations. Categorizing has no physical significance. To paraphrase Bateson, a category encompasses differences that, for specific purposes, make no difference. But that is not to say that the differences actually make no difference. They always make a difference in some manner, even if we don't notice it.

BP earlier: The "direction," of course is a metaphor for varying relative rates of change in some set of parameters, each considered as a different dimension in some hyperspace.

MT: But what I find difficult is to see how you continue the direction if you have changed a C to a T in a DNA strand and "things get better". How do you do more of the same?

BP: That's a discrete change, of course. But what about insertions and deletions? These have the effect of changing the relationships between sequences that occur before and after the changed part. And there's also the effect on folding of the DNA and the proteins synthesized according to the sequences. We're talking about 20,000 genes, and each gene consists of a lot of base pairs, so the discrete changes quickly begin to look smooth as you back away from the close-up view.

I don't think the last clause follows at all from the preceding. If it does, you must have some chain of reasoning that leads from the different kinds of chromosomal changes to the idea that they all add up to smoothness as though the "law of large numbers" applied.

Yes, I suppose that is part of my idea. But chromosomal changes, as I understand them, are not simply a matter of discrete changes. Genes also change their effects on a smooth scale, the process called "expressing" a gene. This might happen as a result of geometric changes -- "folding" -- as well as interactions with enzymes and who knows what else. On the finest scales, there are no discrete processes -- the boundaries that look sharp from a distance become fuzzy and gradual when you look at them closely. Just looking at chromosomes, we draw outlines around the fuzzy boundaries and treat them as discrete units simply because they look that way to the eye, with its ability to sharpen edges. We ignore little differences like their exact length and width, and the angles they form with each other, all of which will have physical consequences on the molecular scale and all of which vary in a continuum.

I suppose you're thinking in terms of the Mendelian laws of combination, which are expressed in logical, digital terms. They do apply for characteristics seen from afar, but they don't explain what is causing even the differences we notice.

MT: Change more Cs to Ts? That would usually create more of the same effect, in the way that changing the weights in the hierarchy would usually do (until you got to a dynamic bifurcation point).

BP: Yes, that's what I thought, too.

MT: Sorry, there was a crucial "not" missing in my post. I intended to imply that changing a C to a T in one gene would probably have a vastly different effect from the effect it would have if the change were in a different gene. But if you believe that all changes of C to T have the same effect, and I presume you think the same about changes of A to G, then the space available for evolution is only two-dimensional (I seem to remember that not all interchanges are allowed).

At some level of observation they have "vastly" different well-separated effects. But that is only because we ignore changes on a smaller scale. Not all "blue eyes" are the same shade of blue with the same patterns in the iris. Not all hairs that are "curly" are congruent with each other. The actual variations are far more continuous than our classifications of them.

MT: Anyway, my question wasn't about all of that. It was just about how one would specify a direction in gene space so that subsequent mutations could expand on the benefits of a good mutation. I don't mean specify for our analytic purposes, but specify in a way that would make practical sense within the organism, in the way that it does for reorganizing the control hierarchy.

BP: This is probably a good place to switch from the reinforcement concept to the reorganization concept. Reorganization doesn't try to "expand on the benefits of a good mutation." It simply causes mutations when errors increase, and stops changing them when they decrease. Introducing direction as the mutated variable, we can see that as long as error is decreasing, the new direction of change will be followed. Sooner or later, the continuing change will stop reducing the error and will start making the error larger. Then there will be a change of direction, another mutation. Some dinosaurs will start getting smaller again, and growing more feathers.

MT: Let me offer a background question that may help you interpret my question about how "direction" can be defined in a geneset, since your first three sentences simply restate the principle of natural selection augmented by the stress response.

No, they don't. They are completely different from the principle of natural selection, because natural selection has no influence on the rate of mutations, whereas reorganization does. Natural selection assumes a more or less constant rate of mutation that is not a function of the selection process as it is in reorganization theory. To produce systematic effects via reorganization, it is only necessary to be able to cause a mutation or delay it, not to determine what kind of change happens. Reorganization simply keeps the changes going until one occurs that removes the reason to keep changing. Natural selection doesn't work that way: Darwin's hammer just eliminates the failures altogether and they play no further part in evolution. The random mutations go right on in the successful survivors. Reorganization does not require extinction; the system gets many chances to try again after a failure. And the survivors reorganize at a much lower rate.

MT: The background to my question can be restated as: "Given that some genetic change increased the probability that this geneset will be included in a later population, and supposing that it is possible to define a "direction" in genespace, what is passed to the offspring that tells the offspring (a) that it is in some way "better" or "not better" than its parents, and (b) what direction of change made it better than its parents, and how does the offspring implement "more" of that direction of change?"

The offspring must have these two pieces of information if evolution is to implement the e-coli process.

No, they needn't know anything about the directions of change in the parameters. All they need to do is judge the consequences of whatever changes occurred, and if they are not acceptable, initiate another random change. Those are the tumbles. As to the continuous phase of changes, the genetic drift, I don't yet see what the mechanisms would be, but the organism doesn't need to know what the current direction is, either, since the mutations simply change the direction at random. Knowledge of the current direction of changes in parameters would be useless. Only the effect of the changes matters, along with the reference conditions, inherited or learned, defining what is acceptable.

Best,

Bill P.

_Boris_Hartman · December 21, 2011, 9:20pm

RM : I beg you pardon. My irrational explosions are completely controlled;-)

BH : You are right. But on which level ? E. coli ? :)

BH earlier : Do you understand it Rick ? Did you understand it 2 and half years ago?

RM earlier : Yes.

BH : Well I don't quite understand what this means. Would you mind explaining this...?

RM earlier : "I think the E.coli 'reorganization' model of evolution can be
considered somewhat Lamarckian... think this is reasonably good quick and
dirty summary of the control model of evolution, as Bill described it some
time ago."

BH : Whether you understood "reorganization" model Bill descibed some time ago, or you understand it now, because it's different. Aproximately 2 years ago you wrote that Bill's theory of reorganization "can be considered somewhat Lamarckian". And now you said that I don't understand theory when I wrote, that is "full" Lamarckian. So what's gone be. Theory of "reorganization" is Lamarckian, or half Lamarckian or not ? What is that I don't understand ? Or what do you understand ? Your answer "yes" can't have the same value both times. But it could show that you are occasionaly changing your opinion :)

BH earlier : Do you think that you know how organisms function ?

RM earlier : At a general level, yes. They control their perceptions.

BH : ... and probably reorganize. Do you have a model to test where reference signal at a general level for a perception to be controlled is coming from ? Top-down, left-right, right-left, down-up, diagonal, from nowhere ... ? :)

RM earlier : Why not just try building actual models and testing them against actual behavior.

BH : How actual models are build in the first place that you can later test it ? With no testing ? What does it mean testing ? Can you describe what you are doing while you are testing ?

Do you think that Bill got all his knowledge only through scientific testing ?

RM earlier : There is no need to _defend_ positions. This is science, not politics.

BH : I think that in science and politics are pepole - Living Control System. So by my opinion it's hard to separate science and politics. All people are usually "defending their theories, models, opinion...." They try to controll perception as near as possible to their reference, using all kind of action to "achieve" that. It's after all goal-directed behavior. And every human has usualy her/his goalswhat probably depends from the "way how homeostasis" is capt in organism or as you say how "

I think there is a lot of politics in science. But I think also there is science in politics.

I present a case in history of science when Lamarckian theory "sunk". For me this case show how "scientific testing" depends from those who are testing. So findings of any testing are relative, specially if you have wrong model for testing. So here is "faulty" attempt (testing) to prove that Lamarckian theory was right :

"In the 1920s, experiments by Paul Kammerer on amphibians, particularly the midwife toad, appeared to find evidence supporting Lamarckism, but his specimens with supposedly acquired black foot-pads were found to have been tampered with. In The Case of the Midwife Toad Arthur Koestler surmised that the tampering had been done by a Nazi sympathiser to discredit Kammerer for his political views, and that his research might actually have been valid. However most biologists believe that Kammerer was a fraud and even among those who believe he was not dishonest most believe that he misinterpreted the results of his experiments." (Wikipedia)

See how politics and science work together. I hope you will not act like a typical "control unit" as Kammerer and do everything to prove that PCT is right theory. He comitted suicide 6 weeks after the critics on his work was published, probably because he was exposed as a fraud. Sorry Rick, but sometimes it seems to me, that you are acting like a PCT "fanatic".

What people wouldn't do (beside testing) to prove that they are right or realize their goals or realize what they really, really want. In politics or in science.

In the matter of fact I think it's difficult to separate human activities, as all people are operating as Living Control System, controlling for as you said "to keep its level of intrinsic error low" as possible. Ad for that they usually do whatever it takes.

As I see it, you are "defending" your "position" of PCT model against disturbances as I am, Gavin, Ford's RTC, Martin T. and L., etc., all who have some different controlled perceptions about PCT. Your "defence" or compensation of disturbances as I see it, is usualy "hostile" for the source of disturbances. There are many ways you could "defend your position" - reference. You could compensate disturbances with kind arguing, but you usualy choose "disqualifying" the person (as source of "bad" disturbances) instead of putting arguments. Or maybe I could say, that you are trying to "eliminate oponent" from a discussion, who don't agree with you (making more or less discrepancy to your "image"). It looks to me, that you try to "eliminate" any "bad" disturbances which are enlarging discrepancy from your reference.

I think that experimenting or testing can be tricky, relative, depending on who is testing and what is testing. Kammerer has a "bad" or "wrong" model, he has been testing. So he helped his "testing" with Indian ink. Who garantee that your model which you test aginst is right ? How your test model was formed ?

BH earlier : It's not only Ashby, Maturana that helped clarifying how "Living beings
work"...

RM earlier : They haven't clarified how living systems work (from my perspective)
until they have tested working versions of their models empirically.
What they have done is provided some speculations; these are of no
value until their is empirical evidence to support them.

BH : It's useless, as I see you didin't read their work or you read it superficially or long time ago and you forget crucial evidence in their theories, models. I see it also as COMPETITIVE TRIAL to prove how PCT is much better than those theories. Again I see it as "you are defending your PCT position" against theory of ultrastability and autopoiesis, which were recognized practically by whole world as relevant.

Can you once step out of your PCT perceptual world and try to accustome to other theories to see that maybe all theories have something what you could use in your model and what could help in testing your model.

Maturana wrote his books on findings in experiments with colors and frogs, and he has even mentioned experiments with E.coli. And Ashby made his "homeostat". Bill wrote many times about "homeostat" and ideas he got there. Ashby built his theory of ultrastability on natural case (cat and fire). And I think however you test his model of ultrastbility : with common sence or scientifically, you'll come to the same results. That's why he gave example of "cat and fire" as it is so widely known. There is no speculations. Everybody can test it. I think that specualtions can be present when model is not clear understandable or testing is to much subjective, like Kamerrer.

People can experiment (test) with what ever they want (arbitrary choosen variables) from the "real machine" as Ashby would say. They can construct systems from any set of variables they want and it's their controlled "picture". Maybe you could re-read what Bill wrote about that. He was affraid whether he "went" in solipsism.

Hierarchies of perceptual control enable people to test. That's natural ability. I think that people can test PCT model also in everyday life in every moment, and every individual can experiance it with his/her own controlled perceptions. I think that some laboratory testing is just addition to experiences of many people. It's not enough just experiances of one man. But if he is trying to convince others that he is right, "defending" only his "tested" position, he will probably rush into a conflict. Nobody can be perfectly right about anything. there is no such a thing as "objective world", everybody are "narrowed" in their worlds of perception. Every perception of Living Control System are limited to some aspects of the "reality" to some system that is integrated with choosen variables. So I think that cooperation of experiances are better. It can be one against other (trying to eliminate others) or people can put their expeirences together in some exceptable common "perceptual picture" for all.

I think once Bill described experiancing "baseball catch" in natural environment and report his findings. They were quite interesting and his experiance seemed to me something new in his way of arguing. He didn't talk about testing model. He simply spoke about his common sence experiances with PCT model.

RM : No, I believe only in empirically tested (non-blind) believing.

BH : So beleiving in one paradigm (model) and empirically testing it, is the only thing you beleive. Is this right ? Don't you think that exchange of experiances (tested or scientifically non tested) made and is making human progress - going for better ? Do you really think that only science made the civilization going on better ? As I see it, best ideas for testing are usualy not born in "pure science".

Best,

Boris

···

----- Original Message ----- From: "Richard Marken" <rsmarken@GMAIL.COM>
To: <CSGNET@LISTSERV.ILLINOIS.EDU>
Sent: Sunday, December 18, 2011 11:45 PM
Subject: Re: How does PCT explain where our basic characteristics come from?

[From Rick Marken (2011.12.18.1445)]

On Sun, Dec 18, 2011 at 1:00 PM, boris_upc <boris.hartman@masicom.net> > wrote:

RM : "If I understand correctly..." in his posts it is clearly a con...

BH : Well Rick you look now like quite unrational in your "uncontrolled"
actions (emotional explosions).

I beg you pardon. My irrational explosions are completely controlled;-)

RM : Bill keeps explaining what my statements mean (they pretty much mean
what they say;-) and Boris keeps saying that the explanations are
insufficient. ..

BH : Do you understand it Rick ? Did you understand it 2 and half years ago?

Yes.

BH: Do you really think that you understand PCT ?

Yes, well enough to test it.

BH: Do you think that you know how organisms function ?

At a general level, yes. They control their perceptions.

BH: We can not predict questions and answers when people clearly shows "control"
characteristics and "reorganize" to defend their positions. Does it look like familiar to
you ?

I have no idea what you're talking about.

BH: So I have to be inovative. I have to "reorganize" in my way, if I want to
defend Control position.

Why not just try building actual models and testing them against
actual behavior. There is no need to _defend_ positions. This is
science, not politics.

BH: PCT doesn't have all answers.

Of course not. PCT is a general model of the organization of behavior.
It is therefore a possible model of how behavior works. It's an
"answer" to questions about behavior only after it has been tested and
passed the test.

BH: Bill has many answers and I'd like to understand what he thinks. But I also
feel that we could help each other understand more perfectly how organism works.

Sure we could. Be we can't do it by just talking. You have to go out
and actually build and test the models. If you want to see if the PCT
reorganization model is better than, say, Ashby's, then you have to 1)
learn what the PCT model of reorganization is (I presume you already
know Ashby's) and then 2) devise tests that will show which one is
better. You are still pre-number (1); you are arguing about what the
PCT model of organization _is_. We're trying to explain what it is and
how it works and you keep saying that the model is wrong. That's why
my patience has run out.

It's not only Ashby, Maturana that helped clarifying how "Living beings
work"

They haven't clarified how living systems work (from my perspective)
until they have tested working versions of their models empirically.
What they have done is provided some speculations; these are of no
value until their is empirical evidence to support them. There is lots
of empirical evidence that supports PCT which is why I find it so much
more interesting and compelling than other ideas about how living
beings work.

BH: Do you really think that blind beleiving in one paradigm is what makes human
take "control" over environment, make civilization ?

No, I believe only in empirically tested (non-blind) believing.

RSM
--
Richard S. Marken PhD
rsmarken@gmail.com
www.mindreadings.com

_Boris_Hartman · December 21, 2011, 9:52pm

Hi Bill,

I must say that I’m a little surprised that you answered my previous post, as Rick already answered.

Earlier :

BH : This look like “full” Lamarkian theory if I understand it right.
RM: You don’t. And apparently you won’t. So there’s really no need to waste time with the rest of your post.

BH : As I usualy don’t know who is speeking in who’s name, I thought that he answered for both of you. His attack was quite offending (not only on the level of civilized behavior which I haven’t experienced with him in the past, but also on personal level as he asserted that I don’t understand and it’s waste of time continue talking with me). So I first didn’t know which opinion to consider for relevant.

But as Rick use to say that you are the master of PCT, and you are also the author, T thought that probably you are the one, I should talk to, not Rick who “jumped” into discussion.

I must admitt that your post is on the level worth of your name. I see it as relativization of our knowledge. I see no usual “facts” and “one way of thinking”. I see it as a perfect ground for continuing discussion. I’m glad about your answer which by my opinion means quite a progress relative to some our previous discussions. I’ve got again feeling that this post is in the style of “our old conversations”, full of ideas.

Continuing it would be a real challenge. If not with me, it can be continued with others.

I would like to express, that I have a strange feeling that most of CSGnet is somehow “against me” (I was also on the “court” once, and Rick’s attack, members tyred of my posts…). So I start to wonder whether it has any sense to continue discussion which could possibly lead to some corrections of Figure 14.1. p. 191. And people hardly change their opinion.

If we only imagine Rick, his “irrational” acting on any actual perception that doesn’t fit his wanted perception about how PCT model should look like. He is acting like a child to whom toys were taken away (far from his reference).

For now I answered to Rick, as answer was already almost written. It’s an “old easy type” of conversation with sarcasm, competing opinion, possible conflict, some arguments, peronality estimations, jokes,…chating like “old friends” in cofee time, you know Rick. No “heavy” new informations. Just “light”, superficial, unpretended discussion.

I really don’t know what’s happening to me, but I’m losing energy (motivation) for continuing our discussion. And that’s happening, as I see it now, in the moment, when I think that we are closing to the center of the problem and possible new cognition about how organism might work. An upgrade model, which Rick could maybe test.

Maybe is that so, because knowing you, we could slide into converstaion which will require from me a lot of time, extra reading, talking to “hints” and energy. And that is, what I’m a little short, as quite some money is “in the game”, if I don’t finish my PhD in time. And it’s taking me a lot of energy and time.

I would like to say, that maybe I will not be able to continue our conversation immediatelly. But I will look forward with great pleasure for continuing it, if you will agree.

Best,

Boris

···

----- Original Message -----

From:
Bill Powers

To: CSGNET@LISTSERV.ILLINOIS.EDU

Sent: Tuesday, December 20, 2011 12:11 AM

Subject: Re: How does PCT explain where our basic characteristics come from?

[From Bill Powers (2011.12.19.1400 MST)]

At 12:18 PM 12/18/2011 +0100, Boris Hartman wrote:
BP : The mutations are what the organism does in response to the selection pressures, in an attempt to nullify their effects.



BH : This look like "full" Lamarkian theory if I understand it right. But as far as I know from rats, girafes and so on...individual changes during life, doesn't count in creation of new organism, except if heredity informations are changed.
Yes, and this relates to what Martin and Ted were talking about. Only mutations that affect the DNA passed to the next generation matter; this is why cutting the tails off mice does not result in offspring with short tails. Lamark was wrong about some of the important details.

What I have to assume (in order to propose a reorganization theory of mutation) is that the organism’s capacity to reorganize the genome works specifically on the DNA that is transmitted, not on general body cells (or if it does work on the latter, the effect would not be passed to the next generation).
BP earlier : Of course reorganization theory is strange in another way because it claims that mutations are driven by error signals inside the organism, not by cosmic rays or mutagenic chemicals or other noxious causes from outside (though it doesn't rule out such effects).

BH : Maybe this kind of explanation could be exceptable (I admitt it is possibility) from some angle. But...

BP earlier : Selection pressures are surely caused by environmental events, but they do not directly result in mutations. The mutations are what the organism does in response to the selection pressures, in an attempt to nullify their effects.

BH : This is probably to much "exclusive".
BP: In one way, yes – there are certainly some selection pressures that work directly on DNA, as in the case of teratogens. But in that case, the mutations would not have any relationship to error signals in the organismm, and the result would work like natural selection – very inefficiently, compared with reorganization.
BP earlier: It has been pretty well established that E. coli itself initiates changes in its own DNA when certain stresses occur. Here is an example from Science magazine:
http://www.bichat.inserm.fr/equipes/emi0339/publications_pdf/exhaustiveliste/2003science.pdf
BH : Your first statement talks about exclusive "selective pressure of environment which do not directly result in mutations", while in your second statement you use terms as : "pretty well established", "when certain stresses occur". So it's not so sure (not so much exclusive) that there are no "direct result in mutation" by external events.
BP: That’s not a direct effect – all we know is that the mutation follows the appearance of selection pressure. That alone doesn’t establish the nature of the causal path. It’s the same situation we have with the “behavioral illusion.” The appearance in that illusion is that a disturbance simply causes an action. But it is quite possible that the disturbance tends to alter some controlled variable, and that the action is organized to oppose the effects of the disturbance and stabilize the controlled variable against the disturbance. This is the same situation that holds between selection pressures and mutations. The appearance is that the selection pressure causes the mutations. But what I am proposing is also possible: that the mutation is an action by a reorganizing control system that is part of a control process that resists the effects of the selection pressure. The Cairnes et. al. experiments with E. coli show that in the case of a change in available nutrients, the mutation rates increase by a factor of several hundred and provide more variants. Some of the descendants are capable of metabolizing the new nutrient and are able to pass this characteristic on to offspring. Natural selection comes into play in eliminating the descendants which did not mutate in this way, but the organism itself reorganized the genome when starvation threatened so as to produce a larger range of new organizations. It probably also produced more failures, just as E. coli’s tumbles are just as likely to turn it in the wrong direction as the right one.
BH: I could agree that in different occasions (different circumstances) also internal errors which are "caused" by "external" can maybe influence the structure of DNA. So it doesn't prove to me, that "what's going on" is THE ONLY MECHANISM" which is changing the structure of DNA. It can be changed in many ways. But I'm still sceptical about "exclusively" internal mechanism which is changing DNA as it's "triggered" by CERTAIN external events (special conditions, strange concentrations).These are mostly laboratory experiments and we know at least from Ashby that laboratory experiments doesn't show the same results in natural environments.
You’re reading the idea of “exclusive” reorganization into my words, but I keep pointing out that natural selection still works and plays some part in the process. But that part is mostly in eliminating the failures; natural selection can’t do anything to cause more mutations when a change is needed. My main point is based on comparing mutations (as random jumps from one organization to another) with the E. coli principle in which the spacing of mutations depends on the amount of intrinsic error, and in which there is a steady change in system parameters between mutations at a rate proportional to intrinsic error. What is mutated is the direction of change, the vector in the parameter hyperspace, rather than the values of parameters. This is a new principle in evolutionary theory, I think, and evolutionists are going to have to give credit to E. coli in their references.

I don’t know if it’s technically possible to tell the difference between these two modes of random change in a living organism.
I think it's your decision what to believe. And mine too. My intention was only to find out whether we can establish what is determining "the reference levels of intrinsic variables" in any cell. And maybe how we can explain the "reduction" of intrisic errors ? If I remember right Rick's statements were : "Intrinsic reference are reference specifications for the values of intrinsic variables".

"The hierarchy of control .... develops as a means of keeping intrinsic variables matching their references".

How all this is concretly happening ? Not abstractly ?
I can’t point to the physiological mechanisms. The closest I have come to a concrete demonstration is to show a model that behaves in the postulated way, to demonstrate its properties. The model has such an overwhelming advantage over the traditional concept of mutation that we really have to investigate it to see if there is evidence for or against it. It’s not a matter of believing or not believing; normal experimental procedures will tell us which to do.
BH : I have no difficulty in seeing how a reference condition could be genetically specified. I thought you have ? O.K., now as you mentioned it, how are they genetically specified by your opinion ?
I don’t know; there are many possible ways, and I have no idea which way is the right one. Reference signals could be in the form of concentrations of proteins or enzymes or other molecules, or in the form of circuits containing spontaneously-firing neurons which emit steady signals.
BP : All that is necessary to establish a reference condition is some physiological characteristic that determines the level of perceptual input to a control system at which its output will be exactly zero. A simple threshold will suffice: no output until the input reaches some magnitude. Or it could be a real neural signal. There are many neurons that have a rate of spontaneous firing without any incoming signals. Those neurons are clearly inheritable parts of the nervous system. If the output of such a neuron enters a comparator neuron (as either the inhibitory or excitatory input), it will establish a reference condition that a control system will seek.

BH : Why did you limit yourself only to a behaviroal part of nervous system ? Although as you said is genetically determined. I thought we are talking about "reference specifications of intrinsic variables" ?
Yes, we are, but we don’t know their physical form. They could exist as chemical concentrations or as neural signals. They could be in the form of enzymes or energy-bearing molecules.

I am not limiting myself to the systems that use behavior to control perceptual inputs. In fact I have often mentioned biochemical control systems. But I know much more about the neural systems, or more exactly I should say that my knowledge of biochemical systems is more severely limited than my limited knowledge of the neuromuscular systems.
BP : I didn't realize that what bothered you was the idea of a genetically specified reference level. I have taken that for granted, because there are so many ways it can come into existence.

BH : You have taken it for granted ? By all these talkings about "reorganization", Lamarckism, changing DNA... ?Which are these "so many ways" ?
I’ll say it once more.

A reference condition is anything that can determine how much input is needed to bring the output exactly to zero. This can be done by injecting signals into a control system anywhere in the loop, or by changing the threshold at which a neuron will begin to fire or a chemical reaction will start to take place.
BP : I wonder if the problem here isn't in your concept of a reference signal. Are you thinking that a reference signal has to have some specific meaning in addition to having a specific magnitude? That the reference signal for one controlled perception will somehow have a different nature from a reference signal for a different controlled perception?

BH : I suppose you are to narrow in thinking here. But maybe I could have problems in understanding you, as I always have. Which reference signal you are talking about ? Are we talking about "intrinsic reference signal" ?
Any reference signal. Reference signals are all alike; they are simply indications of magnitude.
BP : A reference signal is just a neural signal...

BH : Again which reference signal you have in mind ? Intrinsic reference signal or the one in behavioral hierarchy according to Figure 14.1.which is called only "reference signal" and is coming from nowhere to hierachy on 11.level ? Are there any other ?
I was thinking of a neural circuit, but the concept includes any reference signal in anty physical form. What kind of difference are you looking for? There is nothing about a reference signal that distinguishes it from any other kind of signal – a perceptual signal or an error signal or an output from the output function that is sent to affect the reference inputs to lower systems. In the nervous system, its just a series of neural impulses moving along a nerve fiber and action on the input to a comparator function made of neurons.
BP earlier : So basically a reference signal, whether actual or only a structural bias, can easily be inherited if it's not the output of some higher-order system. It is simply whatever properties have to exist to determine how much perceptual signal is to produce zero error signal.

BH : Maybe I don't understand right again, but are you saying that there could me more kind of reference signals in behavioral hierarchy ? Inherited or the output of some higher-order system ? Does inherited means "genetic" ?
BP: Yes. To say a reference signal can be inherited is to say that its value is preserved across generations. What could be passed to another generation might be a particular kind of output function that generates the error signal. In general, the signal itself would not be inherited; only the physical components that produce the signal could be inherited. Reference signals can exist as concentrations of chemicals, as neural signals, as thresholds of firing. To explain how they could be inherited, you have to know what is generating them. In the muscle spindles, a neural signal causes contractions of small muscles at each end of the spindle, stretching the central portion and affecting the nerve-endings wrapped around it. That stretch is the comparator. The reference input to the comparator is the length of the small muscles at the ends; the perceptual signal is the length of the main muscle to which the muscle spindle is attached.
BP earlier: Is there some *other* problem with the ideas I'm offering that you haven't mentioned yet?

BH : I'm wondering if it couldn't be better that all in PCT is explained with "connected" control units and "local stabilities" ? Why giving the "reorganization" so much importance ? Without that importance, I imagine that your PCT could work better or as Ashby said that the things could go for better...Reorganization looks like to me as some kind of "feed-back" to DNA self-production...O.K. if you keep term "reorganization" will it still have the place it has now according to Figure 14.1. p 191 (between output of genetic source and behavioral heirarchy) ?
It’s been customary for quite a few decades to include learning as one aspect of behavior that a theory has to handle. PCT is basically an explanation of how behavior is carried out (or how “local stabilities” are maintained); the reorganization part is an answer to the question of how organisms become able to behave as they do.
BH: I admitt I got "hint from behind" what could mean term "reorganization". If I understand it now right, term "reorganization" is some "systematic" mechanism in ontogenesis and phylogenesis, which is "taking care for the things to go better" as Ashby noted by "trials and errors" or as my hint said (in short) : the bacterium (principle) goes in one direction for a while, then in another random direction and change direction when things stop "getting better" and start "getting worse". We could call it something like "If it's not getting better, try something else".
BP: That’s a good description of what reorganization does. My model is an attempt to explain how it does that. I’d change your statement a little: "“If it’s not getting better, try something else” is almost the whole story, except I would say “If going in this direction is making things worse, try a different direction of change.”
BH: If  for example DNA is self-producting homeostasis in the cell, than, if I understand right, in the case that DNA is not self-producing "optimum" homeostasis (constancy of internal conditions) in certain external environment, "reorganization will seek" for better homeostasis, what could mean better "adapted" organism to certain external conditions (probably better ultra-stability). This is probably goal-seeking behavior as Ashby called it. In your terms it's probably "goal-oriented" or "goal-directed" behavior. Do I understand right ? So reorganization has it's goal ? And to attain that goal "reorganization" can even "change some genetic properties" of DNA ?
Its goal is not that general. What you’re describing is the overall effect of what reorganization accomplishes. But the mechanism by which it does that is much simpler than the result is. The E. coli principle is accomplished, in my models, with only about a dozen lines of Delphi (Pascal) code. Isn’t this one of the goals of modeling behavior? We want to reduce the complexity of what we observe by finding simple underlying principles that require only the least amount of skill and intelligence to carry out. That makes it easier to understand how such a system might have evolved.

Best,

Bill P.

rsmarken · December 21, 2011, 11:47pm

[From Rick Marken (2011.12.21.1545)]

BH earlier : Do you understand E. coli reorganization] Rick ? Did you
understand it 2 and half years ago?

RM earlier : Yes.

BH : Whether you understood "reorganization" model Bill descibed
some time ago, or you understand it now, because it's different.
Aproximately 2 years ago you wrote that Bill's theory of reorganization
"can be considered somewhat Lamarckian". And now you said that I
don't understand theory when I wrote, that is "full" Lamarckian. So
what's gone be. Theory of "reorganization" is Lamarckian, or half
Lamarckian or not ? What is that I don't understand ? Or what do you
understand ? Your answer "yes" can't have the same value both times.
But it could show that you are occasionaly changing your opinion :)

Boris, I think you are indeed better off just discussing this with
Bill. I don't know how to answer your questions. When this discussion
started I was really trying to explain reorganization to you.
Apparently you didn't like my explanation because of the words -- like
"references" instead of "reference signals". Now you seem to have a
problem because I once said the reorganization model applied to
evolution is "somewhat" Lamarckian rather than "fully" Lamarckian (as
you seem to prefer, but which I think is wrong). So you are
"discussing" this with me in a way that I find incredibly irritating,
which led me to become sarcastic and rude. I should have just ducked
out earlier. I'm sure Bill can provide you with the answers your
looking for. If there is one thing Bill is great at it's making people
think he agrees with them. I have no talent for that at all.

RSM

···

<boris.hartman@masicom.net> wrote:
--
Richard S. Marken PhD
rsmarken@gmail.com
www.mindreadings.com

Bill_Powers1 · December 22, 2011, 6:56am

[From Bill Powers (2011.12.21.2245 MST)]

Ted Cloak (2011.12.16.1030 MST) 00

TC: Natural Selection, unlike
any other process proposed as a cause of

evolution, does not require an underlying mechanism. All NatSel
requires

is abiotic background conditions that support life, and unimaginably
deep

time. (And, of course, the fundamental laws of
physics.)

BP: 3 x 10^9 years is not unimaginably deep. My computer can count that
many years in a few seconds. Of course you can add that the same process
is being carried on by a few billion separate organisms in parallel,
which improves the chances. But I still wonder if that’s long enough to
accomplish what standard natural selection seems to have accomplished, if
we stick strictly to the principle of random mutation and selection by
survival to reproduce.
This doesn’t mean I’m ready to accept supernatural guidance; that would
be even harder to explain, and in fact is a way of giving up all attempts
to explain. Lazy thinking. But I claim that natural selection is not
sufficient to account for what has happened, except in a willing
imagination.
One of the difficulties I see with assuming that random mutations are
simply sifted to remove all failures is that a great deal of the
phenomenon of evolution doesn’t look at all random. Consider the neck of
the giraffe, often cited as an mistaken example of apparent systematic
guidance toward a goal, and then casually claimed to be a rather simple
thing for natural selection to have generated.
The problem is that the neck didn’t just randomly get more curved or
twisted or made of more or different bones, or shoot off sideways, or any
of a million other random possibilities. If that had happened, fossil
beds would be littered with the bones of the failures. They aren’t. What
we see is mainly a progression of elongating vertebrae, vertebrae which
are present in about the same numbers as in other vertebrates and which
have the same general forms of articulation, attached to a body at one
end and a head at the other. One main parameter of neck construction was
drifting slowly with time throughout this process, and random mutation is
simply not an adequate explanation of that. Of course natural selection
remained in play, but since the ability of the putative giraffes to feed
themselves and live to reproduce kept improving as a result of these
continuing changes in one particular direction, Darwin’s Hammer did not
fall on the giraffe. There was nothing to be naturally selected
out.
Those are the two main drawbacks that I can see to the theory of natural
selection. There should be remains of far more weird failures than I have
ever heard of being found, and there is definite evidence of prolonged
periods of systematic changes of particular characteristics. Neither of
those fits the theory that there are random mutations and only the more
successful ones win the struggle to survive.
Backing this up is my observation that models meant to demonstrate
natural selection do not in fact stick to the principles of natural
selection, but take advantage of intelligent designs provided by the
people who devised those models. In the models, organisms are allowed to
reproduce if they change in some way which, if continued long enough,
would allow survival. I mentioned giving credit for turning in the
general direction of food, as Beer did in his book about cockroaches.
That actually promotes survival only if the organism aims directly at the
food and the first generation after the mutation gets all the way to the
food and eats it. If the first generation only turns part of the way and
gets no closer, there will be no second generation. Isn’t that the next
thing to self-evident?
The theory and principle of reorganization does not suffer from any of
these defects. In the first place, the principle does not require
extrapolating from apparently beneficial effects and attemping to improve
on them or even create them again. Reorganization is based on the actual
present-time effects on critical variables of behaving (and being)
a certain way, those variables having been discovered in the usual
classical way of natural selection. Acquiring the ability to control
those variables must have had a dramatic effect on the ability to live to
reproduce, or at all. According to my World Futures essay, the most
critical of these variables must be the ones that must be protected
against disturbances in order for reproduction to remain accurate (the
temperature of the testicles, for example). Any change that can oppose
disturbances of those variables will automatically be preserved, since
the accuracy of replication (the opposite of mutation) is preserved. That
is a very neat and fundamental relationship and I have wondered why more
people haven’t done a double-take or an ohmygawd about it, as I did when
it occurred to me. Did I explain it poorly?
From that we get the basic principle of reorganization: error drives
changes, and correction of error stops the changes. Stopping the changes
leaves the organization in a new form. Nothing has to identify the new
form as “good” and do something to select it; it has already
been selected simply because the changes have stopped. I call the
condition that drives changes “error” because I’m thinking of
how control systems are acquired, but actually any condition that causes
changes would do the same job, if a result of the changes was to remove
that condition. Of course that’s negative feedback and negative feedback
is control, so I guess that amounts to the same thing.
I have noticed a tendency among scientists to bristle at any criticism of
Darwinian theory. I don’t think, however, that this is just because they
are generally defensive about their own beliefs. It’s more likely to be a
result of assuming that anyone who criticizes Darwin must be a proponent
of the Other Side in the ancient wars with religion, or is making life
easier for those opponents of old – aiding and abetting the enemy. This
means that the substance of the criticism gets swept aside without much
thought since it presumably, ipso facto (or do I mean prima
facie?) derives from false premises about the supernatural. The Arabs
have a saying for this: the enemy of my friend is my enemy.

I guess what I’m trying to say is that if you don’t like my arguments,
you’re going to have to come up with some reasons why not.

Best,

Bill P.

···

*Darwin’s abbreviation in his notebooks.

================================

"Please tell me, if we used Republican Eisenhower’s top tax rate of
90% for

billionaires, how many billionaires do you think would decide to stop
being

billionaires?" – David Pease

tedcloak · December 23, 2011, 9:13pm

[From Ted Cloak (2011.12.16.1030 MST) = TC2]

[From Bill Powers (2011.12.21.2245 MST)]

Ted Cloak (2011.12.16.1030 MST) 00

TC: Natural Selection, unlike any other process proposed as a cause of
evolution, does not require an underlying mechanism. All NatSel requires
is abiotic background conditions that support life, and unimaginably deep
time. (And, of course, the fundamental laws of physics.)

BP: 3 x 10^9 years is not unimaginably deep. My computer can count that many years in a few seconds. Of course you can add that the same process is being carried on by a few billion separate organisms in parallel, which improves the chances.

TC2: That’s an important addition; you’re referring to a population or a gene pool. One common definition of “evolution” is “variation in gene frequencies in a population over time”. That allows sheer luck, known as genetic drift, to be a driving force in evolution, along with NatSel. I think that’s true only in a tiny population or, more realistically, when a novel variant of a gene is in very short supply. Even a potentially very adaptive variant, when it first appears, is much more likely to be wiped out by chance than to eventually succeed. It needs to drift to 30 or 40 copies before NatSel becomes safely effective; then, of course, it may become the normal or even exclusive variant in that population with remarkable speed.

BP: But I still wonder if that’s long enough to accomplish what standard natural selection seems to have accomplished, if we stick strictly to the principle of random mutation and selection by survival to reproduce.

This doesn’t mean I’m ready to accept supernatural guidance; that would be even harder to explain, and in fact is a way of giving up all attempts to explain. Lazy thinking. But I claim that natural selection is not sufficient to account for what has happened, except in a willing imagination.

One of the difficulties I see with assuming that random mutations are simply sifted to remove all failures is that a great deal of the phenomenon of evolution doesn’t look at all random. Consider the neck of the giraffe, often cited as an mistaken example of apparent systematic guidance toward a goal, and then casually claimed to be a rather simple thing for natural selection to have generated.

The problem is that the neck didn’t just randomly get more curved or twisted or made of more or different bones, or shoot off sideways, or any of a million other random possibilities. If that had happened, fossil beds would be littered with the bones of the failures. They aren’t. What we see is mainly a progression of elongating vertebrae, vertebrae which are present in about the same numbers as in other vertebrates and which have the same general forms of articulation, attached to a body at one end and a head at the other. One main parameter of neck construction was drifting slowly with time throughout this process, and random mutation is simply not an adequate explanation of that. Of course natural selection remained in play, but since the ability of the putative giraffes to feed themselves and live to reproduce kept improving as a result of these continuing changes in one particular direction, Darwin’s Hammer did not fall on the giraffe. There was nothing to be naturally selected out.

TC2: OK, here’s the scenario: Let’s assume for this example that we’re starting with a population of ancestors which look somewhat like a okapi; i.e. like a giraffe but with shorter neck and legs. Along comes a random variation in a gene in the reproductive apparatus of one of these guys that, say, makes the neck of any offspring it may have a little shorter or weaker or crookeder. In the competition for food (leaves) such offspring are at a disadvantage. If drift doesn’t get that variant, NatSel certainly will. And, of course, there will always be competition (cite Malthus here). Given the unlikelihood of fossilization, it’s extremely unlikely that a carrier of that variant will be found in the fossil record.

OTOH, suppose a random variant that makes the neck of offspring a little longer or stronger or straighter. If that variant survives drift (see above), it will propagate through the population fairly rapidly, by NatSel, outcompeting the regular version and becoming the norm or even becoming fixed in the gene pool. Its carriers are (more) likely to appear in the fossil record.

And as long as there is competition for leaves, the above two processes will repeat, and the normal neck will continue to become longer, stronger, and straighter over evolutionary time.

And that’s what causes (the appearance of) direction in evolution.

BP: Those are the two main drawbacks that I can see to the theory of natural selection. There should be remains of far more weird failures than I have ever heard of being found, and there is definite evidence of prolonged periods of systematic changes of particular characteristics. Neither of those fits the theory that there are random mutations and only the more successful ones win the struggle to survive.

Backing this up is my observation that models meant to demonstrate natural selection do not in fact stick to the principles of natural selection, but take advantage of intelligent designs provided by the people who devised those models. In the models, organisms are allowed to reproduce if they change in some way which, if continued long enough, would allow survival. I mentioned giving credit for turning in the general direction of food, as Beer did in his book about cockroaches. That actually promotes survival only if the organism aims directly at the food and the first generation after the mutation gets all the way to the food and eats it. If the first generation only turns part of the way and gets no closer, there will be no second generation. Isn’t that the next thing to self-evident?

TC2: Yes and No. If I accept your premises, yes. The way evolution actually works, No: Large mutations are very rarely successful; successful mutations almost always yield very small improvements in the adaptation of their carrying population.

The big problem with evolutionary models, and this is recognized by the modelers, is that they specify an end result that the model is to achieve. In the wild, however, it’s never clear what change will be successful, because the environment is beyond complex.

What the models demonstrate is that sometimes random variation and selective retention in the lab, outside of the scientist’s head, can be more effective than random variation and selective retention inside the scientist’s head (here cite D. T. Campbell).

BP: The theory and principle of reorganization does not suffer from any of these defects. In the first place, the principle does not require extrapolating from apparently beneficial effects and attemping to improve on them or even create them again. Reorganization is based on the actual present-time effects on critical variables of behaving (and being) a certain way, those variables having been discovered in the usual classical way of natural selection. Acquiring the ability to control those variables must have had a dramatic effect on the ability to live to reproduce, or at all. According to my World Futures essay, the most critical of these variables must be the ones that must be protected against disturbances in order for reproduction to remain accurate (the temperature of the testicles, for example). Any change that can oppose disturbances of those variables will automatically be preserved, since the accuracy of replication (the opposite of mutation) is preserved. That is a very neat and fundamental relationship and I have wondered why more people haven’t done a double-take or an ohmygawd about it, as I did when it occurred to me. Did I explain it poorly?

From that we get the basic principle of reorganization: error drives changes, and correction of error stops the changes. Stopping the changes leaves the organization in a new form. Nothing has to identify the new form as “good” and do something to select it; it has already been selected simply because the changes have stopped. I call the condition that drives changes “error” because I’m thinking of how control systems are acquired, but actually any condition that causes changes would do the same job, if a result of the changes was to remove that condition. Of course that’s negative feedback and negative feedback is control, so I guess that amounts to the same thing.

TC2: Bill, I think what you’ve got here is a metaphor for evolution through natural selection, unless you’re willing to specify that there is some existing machinery underlying the process, analogous to the neurological machinery that underlies PCT. And then you’d have to explain the existence of that machinery. To repeat: “All NatSel requires
is abiotic background conditions that support life, and three billion years (was: unimaginably deep time). (And, of course, the fundamental laws of physics.)”

BP: I have noticed a tendency among scientists to bristle at any criticism of Darwinian theory. I don’t think, however, that this is just because they are generally defensive about their own beliefs. It’s more likely to be a result of assuming that anyone who criticizes Darwin must be a proponent of the Other Side in the ancient wars with religion, or is making life easier for those opponents of old – aiding and abetting the enemy. This means that the substance of the criticism gets swept aside without much thought since it presumably, ipso facto (or do I mean prima facie?) derives from false premises about the supernatural. The Arabs have a saying for this: the enemy of my friend is my enemy.

I guess what I’m trying to say is that if you don’t like my arguments, you’re going to have to come up with some reasons why not.

TC: Parsimony. NatSel can do the job based on straight physics, with no additional premises. I’m attaching an article in which I went to great lengths to explain this.

Best regards,

Ted

Best,

Bill P.

CLONS3.pdf (746 KB)

···

*Darwin’s abbreviation in his notebooks.

================================
“Please tell me, if we used Republican Eisenhower’s top tax rate of 90% for
billionaires, how many billionaires do you think would decide to stop being
billionaires?” – David Pease

Bill_Powers1 · December 23, 2011, 11:18pm

[From Bill Powers (2011.12.23.1448 MST)]

Ted Cloak (2011.12.16.1030 MST)

BP earlier: 3 x 10^9 years is not unimaginably deep. My computer can count that many years in a few seconds. Of course you can add that the same process is being carried on by a few billion separate organisms in parallel, which improves the chances.

TC2: That's an important addition; you're referring to a population or a gene pool. One common definition of "evolution" is "variation in gene frequencies in a population over time". That allows sheer luck, known as genetic drift, to be a driving force in evolution, along with NatSel.

BP: Genetic drift is not based on sheer luck: it is a systematic change in a particular direction. The direction itself may have been generated at random, but if it is beneficial and no further random changes occur, it will continue to improve something for the organism for some time, while if the results are adverse, they will most likely keep getting worse. This greatly amplifies both benefits and penalties compared with the effects of single point-mutations, and I think largely accounts for the great efficiency-advantage of the reorganization principle over the random mutations of standard natural selection theory.

TC2: I think that's true only in a tiny population or, more realistically, when a novel variant of a gene is in very short supply. Even a potentially very adaptive variant, when it first appears, is much more likely to be wiped out by chance than to eventually succeed. It needs to drift to 30 or 40 copies before NatSel becomes safely effective; then, of course, it may become the normal or even exclusive variant in that population with remarkable speed.

BP: You're probably right, and this is exactly the drawback of the standard theory. Continued improvements ion the standard theory require multiple point-mutations one after the other, all favorable. The chances of that are very slim. On the other hand, the drift, if allowed to continue as long as some error continues to decrease, has a very good chance of continuing the improvement that the last "tumble" started generating at least for a short time and sometimes for a long time.

TC2: OK, here's the scenario: Let's assume for this example that we're starting with a population of ancestors which look somewhat like a okapi; i.e. like a giraffe but with shorter neck and legs. Along comes a random variation in a gene in the reproductive apparatus of one of these guys that, say, makes the neck of any offspring it may have a little shorter or weaker or crookeder. In the competition for food (leaves) such offspring are at a disadvantage. If drift doesn't get that variant, NatSel certainly will. And, of course, there will always be competition (cite Malthus here). Given the unlikelihood of fossilization, it's extremely unlikely that a carrier of that variant will be found in the fossil record.

BP: No less likely than that a fossil of a favorable variant will be found, unless you're proposing that fossilization is less likely for unfavorable variants despite their (according to your argument) vastly greater numbers.

Your description of successive changes actually supports reorganization theory, not the theory of natural selection.

TC: OTOH, suppose a random variant that makes the neck of offspring a little longer or stronger or straighter. If that variant survives drift (see above), it will propagate through the population fairly rapidly, by NatSel, outcompeting the regular version and becoming the norm or even becoming fixed in the gene pool. Its carriers are (more) likely to appear in the fossil record.

BP: If the variant makes the neck a little longer or stronger or straighter, isn't the next mutation most likely to destroy that result? You're relegating drift to the role of something that can spoil the result of a favorable mutation, whereas I am focusing on it as the very mechanism that preserves and repeats the beneficial effects. In my model it is the direction of drift that mutates until it is favorable. There is a clear difference in models here. And speaking of models, where is the model behavior behind your statements of (expected) fact? You say "it will propagate through the population fairly rapidly, by NatSel, outcompeting the regular version and becoming the norm or even becoming fixed in the gene pool. Its carriers are (more) likely to appear in the fossil record." Are all those assertions demonstrably true? If so, how do you know they're true? Do you have a model you can run to demonstrate that this result will occur? Or are you just describing what you imagine will be the result, without having verified that this can be shown to be the expected result? I'm putting you on the spot, pal, now come on and admit it. And I'm not using "pal" sarcastically.

TC: And as long as there is competition for leaves, the above two processes will repeat, and the normal neck will continue to become longer, stronger, and straighter over evolutionary time.

BP: That is a slow and extremely inefficient method, considering that, as you say, the chances of a favorable mutation are minuscule -- and the chance of continuing changes in the same direction by strictly random mutation are minuscule-squared, -cubed, and so on.

TC: And that's what causes (the appearance of) direction in evolution.

BP: Using the same form of rhetoric, I retort, "No, it's not." So that makes me right, right? We could take turns doing this for quite a while.

BP earlier: If the first generation only turns part of the way and gets no closer, there will be no second generation. Isn't that the next thing to self-evident?

TC2: Yes and No. If I accept your premises, yes. The way evolution actually works, No: Large mutations are very rarely successful; successful mutations almost always yield very small improvements in the adaptation of their carrying population.

BP: But small mutations by themselves are insufficient to produce the new forms that are actually needed for survival. The same might go for reorganization, except that the small improvements last far, far longer since they occur many more times before the result becomes unfavorable again. In some of my reorganization models, the coefficient determining how much a given parameter changes in one iteration is on the order of 1/100,000. If the changes are made too large, convergence is very slow if it happens at all. I can account for your observation that small mutations are more likely to succeed than large ones, but my reasons are far different.

TC: The big problem with evolutionary models, and this is recognized by the modelers, is that they specify an end result that the model is to achieve.

BP: Yes, precisely my objection to them. They are successful only because the end result is known and is used as a criterion for survival. A model without external guidance based on strictly random mutations, I claim, would fail utterly. However, to refute that claim, all you have to do is produce a model that doesn't fail under those conditions. So have at it (note gauntlet where I dropped it next to your left foot).

In the reorganization model, the final organization of behavior or form resulting from the mutations is not known and does not need to be predicted. All that has to be known is the state of some critical variable that will be affected by the consequences of whatever random changes occur. The mutations of the direction of drift will occur whenever the result makes the variable change unfavorably, and will be postponed as long as the result is favorable. That is all that is required -- no understanding of why the mutations produced a favorable result is needed. No knowledge of the behavior or form needed to achieve that result is needed.

TC2: Bill, I think what you've got here is a metaphor for evolution through natural selection, unless you're willing to specify that there is some existing machinery underlying the process, analogous to the neurological machinery that underlies PCT.

BP: See my working models. They show the functions that need to be carried out. They are not metaphors, mere claims that one phenomenon is "like" another one in some unspecified way.

TC: And then you'd have to explain the existence of that machinery.

BP: The machinery I envision was probably produced by natural selection. But once it existed, natural selection was no longer the dominant process, by several orders of magnitude.

TC: To repeat: "All NatSel requires is abiotic background conditions that support life, and three billion years (was: unimaginably deep time). (And, of course, the fundamental laws of physics.)"

BP: No, NatSel requires far more than that to work as it must. That is what the models purporting to use natural selection show: the models can produce acceptable results only if they are guided toward a final form that a real organism has no way of knowing about. You just said so yourself. I am claiming that a model without guidance can't work right. And I'm offering working models in which there is non-supernatural guidance that seems to work very well indeed. Are you going to call my bluff? If you do, you'll have to show your cards. I've already shown mine.

BP earlier. I guess what I'm trying to say is that if you don't like my arguments, you're going to have to come up with some reasons why not.

TC: Parsimony. NatSel can do the job based on straight physics, with no additional premises. I'm attaching an article in which I went to great lengths to explain this.

BP: A parsimonious explanation is good only if it actually explains. The theory of natural selection flunks that test. You keep saying that NatSel can "do the job," and I keep saying that it can't. We can't both be right. I've shown the model on which I base my statements. Your turn.

Please don't make me read that whole article. Just repeat the basic arguments. If they were right when you wrote the article, they will still be right now. And the contrary.

Best,

Bill

Kennaway_Richard_Dr · December 24, 2011, 11:46am

[From Richard Kennaway (2011.12.24.1134 GMT)]

BP: Genetic drift is not based on sheer luck: it is a systematic
change in a particular direction. The direction itself may have been
generated at random, but if it is beneficial and no further random
changes occur, it will continue to improve something for the organism
for some time, while if the results are adverse, they will most
likely keep getting worse.

That is not what the theory of natural selection says. It says that there is no direction in the mutation process. The resulting drift of phenotypes is the result of differential reproduction of the mutated organisms, not of any direction in the mutation process. Direction is supplied by whatever factors in the environment favour some phenotypical features over others.

I've just written a simulation of that process. Each member of the population, in each generation, makes a random mutation, resulting in a small random step in the fitness landscape, with all directions equally likely. Its probability of reproducing into the next generation depends on that fitness compared with the rest of the population.

What I observe is the population occupying a localised blob on the fitness landscape that slowly moves up the gradient.

A complete fossil record of this history would show no substantial departures from that trajectory. Anything that departs too far fails to reproduce any further, before it can get substantially different from the rest of the population.

Of course this model is grossly simplified, and could be extended in many ways, but it shows the essential idea: random diffusion over a fitness landscape producing apparently directed motion.

You can't tell what natural selection is doing just by watching what it is doing.

The program is written in Matlab, because it's easy to bang out quick simulations like this, but it's expensive commercial software. I'll generate some movies from it later today, so I can exhibit the behaviour more vividly, but right now I've got to go out and lay in supplies until the shops open again.

BP: No, NatSel requires far more than that to work as it must. That
is what the models purporting to use natural selection show: the
models can produce acceptable results only if they are guided toward
a final form that a real organism has no way of knowing about.

The guidance is provided by whatever physical factors make an organism more or less likely to reproduce.

-- Richard

···

--
Richard Kennaway, jrk@cmp.uea.ac.uk, Richard Kennaway
School of Computing Sciences,
University of East Anglia, Norwich NR4 7TJ, U.K.

CARTER_CLOYD · December 24, 2011, 1:06pm

Hi Bill

Merry Christmas and happy holiday

Carter

···

Sent from my Verizon Wireless BlackBerry

-----Original Message-----
From: Bill Powers <powers_w@FRONTIER.NET>
Sender: "Control Systems Group Network (CSGnet)" <CSGNET@LISTSERV.ILLINOIS.EDU>
Date: Fri, 23 Dec 2011 16:18:42
To: <CSGNET@LISTSERV.ILLINOIS.EDU>
Reply-to: "Control Systems Group Network (CSGnet)"
<CSGNET@LISTSERV.ILLINOIS.EDU>
Subject: Re: How does PCT explain where our basic characteristics come from?

[From Bill Powers (2011.12.23.1448 MST)]

Ted Cloak (2011.12.16.1030 MST)

BP earlier: 3 x 10^9 years is not unimaginably deep. My computer can
count that many years in a few seconds. Of course you can add that
the same process is being carried on by a few billion separate
organisms in parallel, which improves the chances.

TC2: That's an important addition; you're referring to a population
or a gene pool. One common definition of "evolution" is "variation
in gene frequencies in a population over time". That allows sheer
luck, known as genetic drift, to be a driving force in evolution,
along with NatSel.

BP: Genetic drift is not based on sheer luck: it is a systematic
change in a particular direction. The direction itself may have been
generated at random, but if it is beneficial and no further random
changes occur, it will continue to improve something for the organism
for some time, while if the results are adverse, they will most
likely keep getting worse. This greatly amplifies both benefits and
penalties compared with the effects of single point-mutations, and I
think largely accounts for the great efficiency-advantage of the
reorganization principle over the random mutations of standard
natural selection theory.

TC2: I think that's true only in a tiny population or, more
realistically, when a novel variant of a gene is in very short
supply. Even a potentially very adaptive variant, when it first
appears, is much more likely to be wiped out by chance than to
eventually succeed. It needs to drift to 30 or 40 copies before
NatSel becomes safely effective; then, of course, it may become the
normal or even exclusive variant in that population with remarkable speed.

BP: You're probably right, and this is exactly the drawback of the
standard theory. Continued improvements ion the standard theory
require multiple point-mutations one after the other, all favorable.
The chances of that are very slim. On the other hand, the drift, if
allowed to continue as long as some error continues to decrease,
has a very good chance of continuing the improvement that the last
"tumble" started generating at least for a short time and sometimes
for a long time.

TC2: OK, here's the scenario: Let's assume for this example that
we're starting with a population of ancestors which look somewhat
like a okapi; i.e. like a giraffe but with shorter neck and legs.
Along comes a random variation in a gene in the reproductive
apparatus of one of these guys that, say, makes the neck of any
offspring it may have a little shorter or weaker or crookeder. In
the competition for food (leaves) such offspring are at a
disadvantage. If drift doesn't get that variant, NatSel certainly
will. And, of course, there will always be competition (cite Malthus
here). Given the unlikelihood of fossilization, it's extremely
unlikely that a carrier of that variant will be found in the fossil record.

BP: No less likely than that a fossil of a favorable variant will be
found, unless you're proposing that fossilization is less likely for
unfavorable variants despite their (according to your argument)
vastly greater numbers.

Your description of successive changes actually supports
reorganization theory, not the theory of natural selection.

TC: OTOH, suppose a random variant that makes the neck of offspring
a little longer or stronger or straighter. If that variant survives
drift (see above), it will propagate through the population fairly
rapidly, by NatSel, outcompeting the regular version and becoming
the norm or even becoming fixed in the gene pool. Its carriers are
(more) likely to appear in the fossil record.

BP: If the variant makes the neck a little longer or stronger or
straighter, isn't the next mutation most likely to destroy that
result? You're relegating drift to the role of something that can
spoil the result of a favorable mutation, whereas I am focusing on it
as the very mechanism that preserves and repeats the beneficial
effects. In my model it is the direction of drift that mutates until
it is favorable. There is a clear difference in models here. And
speaking of models, where is the model behavior behind your
statements of (expected) fact? You say "it will propagate through the
population fairly rapidly, by NatSel, outcompeting the regular
version and becoming the norm or even becoming fixed in the gene
pool. Its carriers are (more) likely to appear in the fossil record."
Are all those assertions demonstrably true? If so, how do you know
they're true? Do you have a model you can run to demonstrate that
this result will occur? Or are you just describing what you imagine
will be the result, without having verified that this can be shown to
be the expected result? I'm putting you on the spot, pal, now come on
and admit it. And I'm not using "pal" sarcastically.

TC: And as long as there is competition for leaves, the above two
processes will repeat, and the normal neck will continue to become
longer, stronger, and straighter over evolutionary time.

BP: That is a slow and extremely inefficient method, considering
that, as you say, the chances of a favorable mutation are minuscule
-- and the chance of continuing changes in the same direction by
strictly random mutation are minuscule-squared, -cubed, and so on.

TC: And that's what causes (the appearance of) direction in evolution.

BP: Using the same form of rhetoric, I retort, "No, it's not." So
that makes me right, right? We could take turns doing this for quite a while.

BP earlier: If the first generation only turns part of the way and
gets no closer, there will be no second generation. Isn't that the
next thing to self-evident?

TC2: Yes and No. If I accept your premises, yes. The way evolution
actually works, No: Large mutations are very rarely successful;
successful mutations almost always yield very small improvements in
the adaptation of their carrying population.

BP: But small mutations by themselves are insufficient to produce the
new forms that are actually needed for survival. The same might go
for reorganization, except that the small improvements last far, far
longer since they occur many more times before the result becomes
unfavorable again. In some of my reorganization models, the
coefficient determining how much a given parameter changes in one
iteration is on the order of 1/100,000. If the changes are made too
large, convergence is very slow if it happens at all. I can account
for your observation that small mutations are more likely to succeed
than large ones, but my reasons are far different.

TC: The big problem with evolutionary models, and this is recognized
by the modelers, is that they specify an end result that the model
is to achieve.

BP: Yes, precisely my objection to them. They are successful only
because the end result is known and is used as a criterion for
survival. A model without external guidance based on strictly random
mutations, I claim, would fail utterly. However, to refute that
claim, all you have to do is produce a model that doesn't fail under
those conditions. So have at it (note gauntlet where I dropped it
next to your left foot).

In the reorganization model, the final organization of behavior or
form resulting from the mutations is not known and does not need to
be predicted. All that has to be known is the state of some critical
variable that will be affected by the consequences of whatever random
changes occur. The mutations of the direction of drift will occur
whenever the result makes the variable change unfavorably, and will
be postponed as long as the result is favorable. That is all that is
required -- no understanding of why the mutations produced a
favorable result is needed. No knowledge of the behavior or form
needed to achieve that result is needed.

TC2: Bill, I think what you've got here is a metaphor for evolution
through natural selection, unless you're willing to specify that
there is some existing machinery underlying the process, analogous
to the neurological machinery that underlies PCT.

BP: See my working models. They show the functions that need to be
carried out. They are not metaphors, mere claims that one phenomenon
is "like" another one in some unspecified way.

TC: And then you'd have to explain the existence of that machinery.

BP: The machinery I envision was probably produced by natural
selection. But once it existed, natural selection was no longer the
dominant process, by several orders of magnitude.

TC: To repeat: "All NatSel requires is abiotic background
conditions that support life, and three billion years (was:
unimaginably deep time). (And, of course, the fundamental laws of physics.)"

BP: No, NatSel requires far more than that to work as it must. That
is what the models purporting to use natural selection show: the
models can produce acceptable results only if they are guided toward
a final form that a real organism has no way of knowing about. You
just said so yourself. I am claiming that a model without guidance
can't work right. And I'm offering working models in which there is
non-supernatural guidance that seems to work very well indeed. Are
you going to call my bluff? If you do, you'll have to show your
cards. I've already shown mine.

BP earlier. I guess what I'm trying to say is that if you don't like
my arguments, you're going to have to come up with some reasons why not.

TC: Parsimony. NatSel can do the job based on straight physics, with
no additional premises. I'm attaching an article in which I went to
great lengths to explain this.

BP: A parsimonious explanation is good only if it actually explains.
The theory of natural selection flunks that test. You keep saying
that NatSel can "do the job," and I keep saying that it can't. We
can't both be right. I've shown the model on which I base my
statements. Your turn.

Please don't make me read that whole article. Just repeat the basic
arguments. If they were right when you wrote the article, they will
still be right now. And the contrary.

Best,

Bill