[From Frank Lenk 2019.01.03.13:31 CST]
Ran into this line and had to send the link:
“To be alive, he says, is to act in ways that reduce the gulf between your expectations and your sensory inputs. Or, in Fristonian terms, it is to minimize free energy.�
The Genius Neuroscientist Who Might Hold the Key to True AI
WIRED
https://www.wired.com/story/karl-friston-free-energy-principle-artificial-intelligence/
Frank
I read it. It reeks of PCT. someone ought to clue Friston in.
Fred Nickols
Distance Consulting LLC
“Assistance at A Distance�
www.nickols.us
[Rick Marken 2019-01-03_15:29:58]
FN: I read it [the article on Friston’s Free Energy principal]. It reeks of PCT. someone ought to clue Friston in.
RM: I agree that it reeks, but not of PCT. Indeed, Friston’s Free Energy principle seems to be a direct denial of PCT. For example, the article suggests that a central insight of the Free Energy principle is that the apparent purposefulness of behavior is an illusion:Â
He turned over an old log and spotted several wood licesmall bugs with armadillo-shapedd exoskeletonsmoving about, he initially assumed, in a frantic searrch for shelter and darkness. After staring at them for half an hour, he deduced that they were not actually seeking the shade. âThat was an illusion,â? Friston says. âA fantasy that I brought to the table.â?
Â
RM: And what was Friston’s brilliant insight? It was that:Â
The creaturesâ movement was random; they simply moved faster in the warmth of the sun.
RM: So warmth (a stimulus) causes faster movement (the response). It seems to me that Friston is a bit confused about what is real (control) and what is illusion (S-R).Â
BestÂ
Rick
Â
On Thu, Jan 3, 2019 at 11:53 AM Fred Nickols csgnet@lists.illinois.edu wrote:
On Thu, Jan 3, 2019 at 2:35 PM Franklin Lenk csgnet@lists.illinois.edu wrote:
 [From Frank Lenk 2019.01.03.13:31 CST]
Â
Ran into this line and had to send the link:
Â
âTo be alive, he says, is to act in ways that reduce the gulf between your expectations and your sensory inputs. Or, in Fristonian terms, it is to minimize free energy.â?
The Genius Neuroscientist Who Might Hold the Key to True AI
WIRED
https://www.wired.com/story/karl-friston-free-energy-principle-artificial-intelligence/
Â
Frank
–
Fred Nickols
Distance Consulting LLC
âAssistance at A Distanceâ?
www.nickols.us
–
Richard S. MarkenÂ
"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.â?
                --Antoine de Saint-Exupery
[Richard Pfau (01.04.2019 13:15 EST)
I tend to agree with Fred. The thinking expressed in the article does seem to overlap with PCT.
Regarding the wood lice, (a) their random behavior matches PCT’s view of random behavior occurring when an organism’s behavior isn’t reducing error signals that are occurring, and (b) the boy’s conclusion that it was an illusion that the wood lice were not actually seeking the shade seems to simply indicate that the boy had focused on the wrong variable that the wood lice were controlling.
With Regards,
Richard Â
On Thu, Jan 3, 2019 at 6:31 PM Richard Marken csgnet@lists.illinois.edu wrote:
[Rick Marken 2019-01-03_15:29:58]
On Thu, Jan 3, 2019 at 11:53 AM Fred Nickols csgnet@lists.illinois.edu wrote:
FN: I read it [the article on Friston’s Free Energy principal]. It reeks of PCT. someone ought to clue Friston in.
RM: I agree that it reeks, but not of PCT. Indeed, Friston’s Free Energy principle seems to be a direct denial of PCT. For example, the article suggests that a central insight of the Free Energy principle is that the apparent purposefulness of behavior is an illusion:Â
He turned over an old log and spotted several wood licesmall bugs with armmadillo-shaped exoskeletonsmoving about, he initially assumed, in aa frantic search for shelter and darkness. After staring at them for half an hour, he deduced that they were not actually seeking the shade. âThat was an illusion,â? Friston says. âA fantasy that I brought to the table.â?
Â
RM: And what was Friston’s brilliant insight? It was that:Â
The creaturesâ movement was random; they simply moved faster in the warmth of the sun.
RM: So warmth (a stimulus) causes faster movement (the response). It seems to me that Friston is a bit confused about what is real (control) and what is illusion (S-R).Â
BestÂ
Rick
Â
On Thu, Jan 3, 2019 at 2:35 PM Franklin Lenk csgnet@lists.illinois.edu wrote:
 [From Frank Lenk 2019.01.03.13:31 CST]
Â
Ran into this line and had to send the link:
Â
âTo be alive, he says, is to act in ways that reduce the gulf between your expectations and your sensory inputs. Or, in Fristonian terms, it is to minimize free energy.â?
The Genius Neuroscientist Who Might Hold the Key to True AI
WIRED
https://www.wired.com/story/karl-friston-free-energy-principle-artificial-intelligence/
Â
Frank
–
Fred Nickols
Distance Consulting LLC
âAssistance at A Distanceâ?
www.nickols.us
–
Richard S. MarkenÂ
"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.â?
                --Antoine de Saint-Exupery
[Martin Taylor 2019,01,05.13.27]
On 2019/01/3 6:30 PM, Richard Marken
( via csgnet Mailing List) wrote:
[Rick Marken 2019-01-03_15:29:58]
On Thu, Jan 3, 2019 at 11:53 AM Fred Nickols<csgnet@lists.illinois.edu >
wrote:
FN: I read it [the article on Friston'sFree Energy principal]. It reeks of PCT. someone ought
to clue Friston in.
RM: I agree that it reeks, but not of PCT. Indeed,Friston’s Free Energy principle seems to be a direct denial
of PCT. For example, the article suggests that a central
insight of the Free Energy principle is that the apparent
purposefulness of behavior is an illusion:
The way I understand the "Free Energy Principle" is that is
describes what is happening during the process of reorganization,
though it does not describe a method. Friston clearly does not say
the the apparent purposefulness of behaviour is an illusion,
whatever a journalistic article about his work may seem to imply.
Rather, the purpose behind behaviour is interpreted differently.
As I described on CSGnet [Martin Taylor 2017.07.11.10.35], the
Friston circuitry actually implements the performance of the
Powers hierarchy (the circuits I showed there differ from
Friston’s in that mine incorporate an imagination mode of
operation). What Friston calls “predicted value”, Powers calls
“reference value”, which changes the purpose implied, from
improvement in prediction accuracy to improvement in the current
state of the external environment. That’s a significant difference
of purpose and one might expect it to lead to differences in the
experiments chosen to test the theories, but whether those
experiments could discriminate between the theories is
questionable.
He turned over an old log and spottedseveral wood lice—small bugs with armadillo-shaped
exoskeletons—moving about, he initially assumed, in a
frantic search for shelter and darkness. After staring at
them for half an hour, he deduced that they were not
actually seeking the shade. “That was an illusion,�
Friston says. “A fantasy that I brought to the table.�
Â
RM: And what was Friston's brilliant insight? It wasthat:Â
The creatures’ movement was random;they simply moved faster in the warmth of the sun.
And so do all cold-blooded creatures. It is a fair model of
non-e-coli reorganization, in which variables change randomly,
faster when error is large, slower when error is small, the result
of which is variable values (parameters) piling up where the
variables produce low error values. Powers’s earlier problem with
this model of reorganization, before he discovered e-coli, was
that when the number of variables exceeds three or four, the time
to converge gets very long. But the wood lice are moving in only
two dimensions, so scurrying around randomly when they are hot
gets them quickly to places where they are cool, if there are any
such places nearby (nearby, because the distance travelled from he
starting point of a random walk is proportional to the square root
of time, so if the nearest cool place is too far away, the bug
might die of heat stroke before finding it.
RM: So warmth (a stimulus) causes faster movement (theresponse).Â
Would you say that warmth is a stimulus that creates the response
of rising for a packet of air over a fire in a cool atmosphere?
Same deal. You have to say both or neither is stimulus-response.
Of course it may well be true that the wood-lice were controlling
a perception of body temperature, but we can’t know that just from
this observation. Maybe its a simple biochemical effect, making
the legs move erratically and fast when they get too hot rather
than in a coordinated way and more slowly when they are cooler and
controlling for finding food.
It seems to me that Friston is a bit confused about whatis real (control) and what is illusion (S-R).
It seems to me that to know whether this is true depends on
knowing what Friston is trying to achiever–what he is controlling
for. From the PCT angle, we want to know how a control hierarchy
reorganizes to arrange the outputs connections down through the
levels so that the real-world effects of actions reduce error.
Looking at it from the predictive coding viewpoint, one could say
that we want to know what is changing so that the actions produce
the predicted effects more accurately. The free-energy principle
says that what is changing is the match between on the one hand
the perceptual functions and the content of real reality, and on
the other hand the match between what should happen when we do X
and what real reality causes to happen when we do X. The free
energy principle doesn’t suggest mechanism, but then neither does
e-coli reorganization.
We (Warren, mostly) tried to interest Friston in PCT a couple of
years ago, but the interaction (at least what little I saw of it)
petered out. My impression was that he didn’t seem very
interested. As I see it now, there isn’t any obvious experimental
way to make a critical distinction between Friston and Powers,
despite the vast gulf between their interpretations of the
observables. My feeling is that its a bit like seeing the same
complex object from two very different angles, both of which
produce a different simplified concept of what the object is.
Neither is wrong, but both are incomplete. In this case, the two
might be close enough to permit a more accurate stereo view of he
object, which is the structure that allows organisms to exist and
do what they do.
Martin
BestÂ
Rick
Â
On Thu, Jan 3, 2019 at 2:35 PM FranklinLenk <csgnet@lists.illinois.edu >
wrote:
 [FromFrank Lenk 2019.01.03.13:31 CST]
Â
Raninto this line and had to send the link:
Â
“Tobe alive, he says, is to act in ways that
reduce the gulf between your expectations
and your sensory inputs. Or, in Fristonian
terms, it is to * minimize
free energy*.�
** The
Genius Neuroscientist Who Might Hold the
Key to True AI**WIRED
https://www.wired.com/story/karl-friston-free-energy-principle-artificial-intelligence/
Â
Frank
–
Fred
NickolsDistance Consulting LLC “Assistance at A Distance� [www.nickols.us](http://www.nickols.us)
–
Richard S. MarkenÂ
"Perfectionis achieved not when you have
nothing more to add, but when you
have
nothing left to take away.�
   Â
            --Antoine de
Saint-Exupery
[Rick Marken 2019-01-04_16:57:02]
[Martin Taylor 2019,01,05.13.27]
FN: I read it [the article on Friston'sFree Energy principal]. It reeks of PCT. someone ought
to clue Friston in.
RM: I agree that it reeks, but not of PCT. Indeed,Friston’s Free Energy principle seems to be a direct denial
of PCT. For example, the article suggests that a central
insight of the Free Energy principle is that the apparent
purposefulness of behavior is an illusion:Â
MT: The way I understand the "Free Energy Principle" is that isdescribes what is happening during the process of reorganization,
though it does not describe a method.
RM: So the Free Energy Principle is a description of a phenomenon (reorganization), not a model of how that phenomenon happens? Is there a paper where Friston shows how well his Free Energy Principle corresponds to an actual example of reorganization?
MT: Friston clearly does not saythe the apparent purposefulness of behaviour is an illusion,
whatever a journalistic article about his work may seem to imply.
Rather, the purpose behind behaviour is interpreted differently.
RM: It’s not that clear to me. Here’s what it says in the article:Â
Â
Wired: After staring at them for half an hour, he deduced that they were not actually seeking the shade. “That was an illusion,â€? Friston says. “A fantasy that I brought to the table.â€? Â
RM: The phrase "seeking shade"describes a purposeful behavior: the purpose of the “seeking” is to get to “shade”. Friston is quoted as saying that the appearance that the flies are seeking shade is an illusion. That’s how I got the idea that Friston believes that purposeful behavior is an illusion.Â
MT: As I described on CSGnet [Martin Taylor 2017.07.11.10.35], theFriston circuitry actually implements the performance of the
Powers hierarchyÂ
RM: Is the “Friston circuitry” something different than his Free Energy Principle? Because you just said that the Free Energy Principle describes what is happening during the process of reorganization. The Powers hierarchy is a model of skilled behavior (behavior that is already successfully reorganized) as the control of a hierarchy of different types of perceptual variables.Â
MT: What Friston calls "predicted value", Powers calls“reference value”,
RM: Really?So Friston explains that “predicted values” specify the intended states of perceptual variables? And that these predicted values are set by the outputs of higher level input control systems as the means of controlling higher level perceptual variables? If he doesn’t, then what Friston calls “predicted value” is not anything like what Powers calls “reference value”.Â
RM: And what was Friston's brilliant insight? It wasthat:Â
The creatures’ movement was random;they simply moved faster in the warmth of the sun.
MT: And so do all cold-blooded creatures. It is a fair model ofnon-e-coli reorganization, in which variables change randomly,
faster when error is large, slower when error is small, the result
of which is variable values (parameters) piling up where the
variables produce low error values.
RM: The “warmth of the sun” is not an error; it is simply a variable. “Error” implies a reference specification for a controlled variable and a closed loop, negative feedback control organization that acts to bring the controlled variable to the specified state, thus bringing the error variable to zero. If Friston meant what you say he meant then Friston is describing the PCT e. coli reorganization process, if the variable controlled by this process is the error variable in another control system and the parameters being varied are the parameters of that control system.Â
RM: So warmth (a stimulus) causes faster movement (theresponse).Â
MT: Would you say that warmth is a stimulus that creates the responseof rising for a packet of air over a fire in a cool atmosphere?
RM: Yes. Â
MT: Same deal. You have to say both or neither is stimulus-response.
RM: No, it’s not the same deal: The wood lice are closed loop (control) systems; the packet of air over a fire is an open loop (causal) system.Â
RM: It seems to me that Friston is a bit confused about whatis real (control) and what is illusion (S-R).Â
MT: It seems to me that to know whether this is true depends onknowing what Friston is trying to achiever–what he is controlling
for. From the PCT angle, we want to know how a control hierarchy
reorganizes to arrange the outputs connections down through the
levels so that the real-world effects of actions reduce error.
RM: Yes, and the way to figure that out is to do tests on real organisms to see what variables then control, how they control them and how they learn to control them.Â
MT: Looking at it from the predictive coding viewpoint, one could saythat we want to know what is changing so that the actions produce
the predicted effects more accurately. The free-energy principle
says that what is changing is the match between on the one hand
the perceptual functions and the content of real reality, and on
the other hand the match between what should happen when we do X
and what real reality causes to happen when we do X. The free
energy principle doesn’t suggest mechanism, but then neither does
e-coli reorganization.
RM: Actually, the e. coli reorganization mechanism is very clearly specified (see Marken,
R. S. and Powers, W. T. (1989) Random-Walk Chemotaxis: Trial-And-Error as a
Control Process. *Behavioral Neuroscience,*103, 1348-1355, which is reprinted in Mind Readings). If the Free Energy Principle doesn’t describe a mechanism of reorganization than it’s untestable and, thus, scientifically useless.Â
MT: We (Warren, mostly) tried to interest Friston in PCT a couple ofyears ago, but the interaction (at least what little I saw of it)
petered out. My impression was that he didn’t seem very
interested. As I see it now, there isn’t any obvious experimental
way to make a critical distinction between Friston and Powers,
despite the vast gulf between their interpretations of the
observables.
RM: I think that is only because Friston’s theory is untestable.Â
MT: My feeling is that its a bit like seeing the samecomplex object from two very different angles, both of which
produce a different simplified concept of what the object is.
Neither is wrong, but both are incomplete. In this case, the two
might be close enough to permit a more accurate stereo view of he
object, which is the structure that allows organisms to exist and
do what they do.
RM: I couldn’t agree with you less.Â
Best
Rick
–
Richard S. MarkenÂ
"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery
[Martin Taylor 2019.01.04.23.18]
[Rick Marken 2019-01-04_16:57:02]
…
RM: I couldn’t agree with you less.
It would be easier to have a scientific discussion if your
comments had some relation to at least one of (1) what I wrote in
the message on which you purport to comment, (2) the message from
2017 that I referenced, or (3) something Friston published.
Actually, you did comment properly on one thing I said “* The
free energy principle doesn’t suggest mechanism, but then
neither does e-coli reorganization*”. You said “* Actually,
the e. coli reorganization mechanism is very clearly specified
(see Marken, R. S. and Powers, W. T. (1989) Random-Walk
Chemotaxis: Trial-And-Error as a Control Process. Behavioral
Neuroscience, 103, 1348-1355, which is reprinted in Mind
Readings*).”
To see whether what you said it true, I went back and re-read
this paper. I found nothing at all that mentions the mechanism of
e-coli reorganization. I found in the introductory part some
description of a mechanism for the e-coli biological bacterium’s
behaviour, and some very nice simulation models of goal-seeking by
the e-coli process done by humans. Maybe I missed the mechanism in
my earlier reading and on my search for a mention of a mechanism
of e-coli reorganization in my re-reading, and I would be very
happy if you would point me to the page and relevant words in the
“Mind Readings” publication of that paper.
Since 2019 is a new year, should we try, perhaps, to keep
discussions on CSGnet a little more focused on science and less on
personality differences than in 2018? In your comments on my post
are we looking at one of the reasons CSGnet is so devoid of female
participants? Do females tend to be less combative and more
collaborative than males, as is the common notion?
Martin
[Frank Lenk 2019.01.05.10:29 CST]
Rick, I agree that the story about the wood lice rejects they have a purpose to seek shade. But the next paragraph, along with its footnote, modifies that:
[Wired] He realized that the movement of the wood lice had no larger purpose, at least not in the sense that a human has a purpose when getting in a car to run an errand. The creatures’ movement was random; they simply moved faster in the warmth4 of the sun.
4 Young Friston was probably right. Many species of wood lice will dry out in direct sunlight, and some respond to a rise in temperature with kinesis, an increased level of random movement.
I interpret this to mean that wood lice don’t consciously construct a purpose to be achieved like humans. Rather they have been organized by evolution to scatter randomly when light or heat rapidly increases. They aren’t seeking shade. They are seeking a reduction in error, which looks to us like they are seeking shade.
At least, this interpretation is consistent with the rest of the article’s discussion of “free energy�, from which I quote liberally below (I apologize for the length) and provide my translation into PCT terminology in square brackets where appropriate. As Martin suggests, the match isn’t one-to-one, but I still think there is a high degree of overlap:
Wired: Â Free energy is the difference between the states you expect to be in and the states your sensors tell you that you are in. Or, to put it another way, when you are minimizing free energy, you are minimizing surprise [FL: error].
According to Friston, any biological system that resists a tendency to disorder and dissolution will adhere to the free energy principle [FL: minimizing the error between perception and reference]—whetheer it’s a protozoan or a pro basketball team.
A single-celled organism has the same imperative to reduce surprise that a brain does. The only difference is that, as self-organizing biological systems go, the human brain is inordinately complex: It soaks in information from billions of sense receptors, and it needs to organize that information efficiently into an accurate model of the world [FL: world model]. “It’s literally a fantastic organ in the sense that it generates hypotheses or fantasies [FL: or references in imagination] that are appropriate for trying to explain these myriad patterns, this flux of sensory information that it is in receipt of,â€? Friston says. In seeking to predict what the next wave of sensations is going to tell it—and the next, and the next—the brain is constnstantly making inferences and updating its beliefs [FL: reorganizing] based on what the senses relay back, and trying to minimize prediction-error [FL: error] signals.
So far, as you might have noticed, this sounds a lot like the Bayesian idea of the brain as an “inference engine� that Hinton told Friston about in the 1990s. And indeed, Friston regards the Bayesian model as a foundation of the free energy principle (“free energy� is even a rough synonym for “prediction error� [FL: error]). But the limitation of the Bayesian model, for Friston, is that it only accounts for the interaction between beliefs and perceptions; it has nothing to say about the body or action. It can’t get you out of your chair.
This isn’t enough for Friston, who uses the term “active inferenceâ€? to describe the way organisms minimize surprise while moving about the world. When the brain makes a prediction [FL: brings its attention to a reference] that isn’t immediately borne out by what the senses relay back, Friston believes, it can minimize free energy in one of two ways: It can revise its prediction—absorb the surprise, concede the error, update its model of the world [FL: reorganize perceptual functions, reorganize the hierarchy, or establish new references]—or it can act to makke the prediction true [FL: engage output functions and reorganize if error persists]. If I infer that I am touching my nose with my left index finger, but my proprioceptors tell me my arm is hanging at my side, I can minimize my brain’s raging prediction-error signals by raising that arm up and pressing a digit to the middle of my face.
And in fact, this is how the free energy principle accounts for everything we do: perception, action, planning, problem solving. When I get into the car to run an errand, I am minimizing free energy by confirming my hypothesis—my fantasy—[FL: bringinging perceptions into agreement with references] through action.
For Friston, folding action and movement into the equation is immensely important. Even perception itself, he says, is “enslaved by action�: To gather information, the eye darts, the diaphragm draws air into the nose, the fingers generate friction against a surface. And all of this fine motor movement exists on a continuum with bigger plans, explorations,10 and actions [FL: acknowledgement of a hierarchy of perceptions and references].
10 Friston’s term for this kind of exploration is “epistemic foraging.� He is notorious among his colleagues for his coinages, known as Fristonese.
“We sample the world [FL: create a world model],� Friston writes, “to ensure our predictions become a self-fulfilling prophecy [FL: our references are always, or mostly, brought into alignment with our references].�
So what happens when our prophecies are not self-fulfilling? What does it look like for a system to be overwhelmed by surprise [FL: error]? The free energy principle, it turns out, isn’t just a unified theory of action, perception, and planning; it’s also a theory of mental illness [FL: Here, PCT seems much more fully developed, with its discussion of reorganization, conflict, and the ability to go up a level to resolve conflicts].
I was excited when I saw this article, because I thought maybe I had found a fellow traveler in this idea that living things are always trying to minimize the error between their perceptions and their references, and whose different starting point and differences in interpretation might lead us all to a more complete theory, particularly on the perception side of the control loop. I wasn’t aware, or had forgotten, that Martin and Warren attempted to start a conversation with Friston that unfortunately went nowhere. Disappointing.Â
Frank
From: “csgnet@lists.illinois.edu” csgnet@lists.illinois.edu
Reply-To: “rsmarken@gmail.com” rsmarken@gmail.com
Date: Friday, January 4, 2019 at 6:58 PM
To: “csgnet@lists.illinois.edu” csgnet@lists.illinois.edu
Subject: Re: WIRED: The Genius Neuroscientist Who Might Hold the Key to True AI
[Rick Marken 2019-01-04_16:57:02]
[Martin Taylor 2019,01,05.13.27]
FN: I read it [the article on Friston’s Free Energy principal]. It reeks of PCT. someone ought to clue Friston in.
RM: I agree that it reeks, but not of PCT. Indeed, Friston’s Free Energy principle seems to be a direct denial of PCT. For example, the article suggests that a central insight of the Free Energy principle is that the apparent purposefulness of behavior is an illusion:
MT: The way I understand the “Free Energy Principle” is that is describes what is happening during the process of reorganization, though it does not describe a method.
RM: So the Free Energy Principle is a description of a phenomenon (reorganization), not a model of how that phenomenon happens? Is there a paper where Friston shows how well his Free Energy Principle corresponds to an actual example of reorganization?
MT: Friston clearly does not say the the apparent purposefulness of behaviour is an illusion, whatever a journalistic article about his work may seem to imply. Rather, the purpose behind behaviour is interpreted differently.
RM: It’s not that clear to me. Here’s what it says in the article:
Wired: After staring at them for half an hour, he deduced that they were not actually seeking the shade. “That was an illusion,� Friston says. “A fantasy that I brought to the table.�
RM: The phrase "seeking shade"describes a purposeful behavior: the purpose of the “seeking” is to get to “shade”. Friston is quoted as saying that the appearance that the flies are seeking shade is an illusion. That’s how I got the idea that Friston believes that purposeful behavior is an illusion.
MT: As I described on CSGnet [Martin Taylor 2017.07.11.10.35], the Friston circuitry actually implements the performance of the Powers hierarchy
RM: Is the “Friston circuitry” something different than his Free Energy Principle? Because you just said that the Free Energy Principle describes what is happening during the process of reorganization. The Powers hierarchy is a model of skilled behavior (behavior that is already successfully reorganized) as the control of a hierarchy of different types of perceptual variables.
MT: What Friston calls “predicted value”, Powers calls “reference value”,
RM: Really?So Friston explains that “predicted values” specify the intended states of perceptual variables? And that these predicted values are set by the outputs of higher level input control systems as the means of controlling higher level perceptual variables? If he doesn’t, then what Friston calls “predicted value” is not anything like what Powers calls “reference value”.
RM: And what was Friston’s brilliant insight? It was that:
The creatures’ movement was random; they simply moved faster in the warmth of the sun.
MT: And so do all cold-blooded creatures. It is a fair model of non-e-coli reorganization, in which variables change randomly, faster when error is large, slower when error is small, the result of which is variable values (parameters) piling up where the variables produce low error values.
RM: The “warmth of the sun” is not an error; it is simply a variable. “Error” implies a reference specification for a controlled variable and a closed loop, negative feedback control organization that acts to bring the controlled variable to the specified state, thus bringing the error variable to zero. If Friston meant what you say he meant then Friston is describing the PCT e. coli reorganization process, if the variable controlled by this process is the error variable in another control system and the parameters being varied are the parameters of that control system.
RM: So warmth (a stimulus) causes faster movement (the response).
MT: Would you say that warmth is a stimulus that creates the response of rising for a packet of air over a fire in a cool atmosphere?
RM: Yes.
MT: Same deal. You have to say both or neither is stimulus-response.
RM: No, it’s not the same deal: The wood lice are closed loop (control) systems; the packet of air over a fire is an open loop (causal) system.
RM: It seems to me that Friston is a bit confused about what is real (control) and what is illusion (S-R).
MT: It seems to me that to know whether this is true depends on knowing what Friston is trying to achiever–what he is controlling for. From the PCT angle, we want to know how a control hierarchy reorganizes to arrange the outputs connections down through the levels so that the real-world effects of actions reduce error.
RM: Yes, and the way to figure that out is to do tests on real organisms to see what variables then control, how they control them and how they learn to control them.
MT: Looking at it from the predictive coding viewpoint, one could say that we want to know what is changing so that the actions produce the predicted effects more accurately. The free-energy principle says that what is changing is the match between on the one hand the perceptual functions and the content of real reality, and on the other hand the match between what should happen when we do X and what real reality causes to happen when we do X. The free energy principle doesn’t suggest mechanism, but then neither does e-coli reorganization.
RM: Actually, the e. coli reorganization mechanism is very clearly specified (see Marken, R. S. and Powers, W. T. (1989) Random-Walk Chemotaxis: Trial-And-Error as a Control Process. Behavioral Neuroscience, 103, 1348-1355, which is reprinted in Mind Readings). If the Free Energy Principle doesn’t describe a mechanism of reorganization than it’s untestable and, thus, scientifically useless.
MT: We (Warren, mostly) tried to interest Friston in PCT a couple of years ago, but the interaction (at least what little I saw of it) petered out. My impression was that he didn’t seem very interested. As I see it now, there isn’t any obvious experimental way to make a critical distinction between Friston and Powers, despite the vast gulf between their interpretations of the observables.
RM: I think that is only because Friston’s theory is untestable.
MT: My feeling is that its a bit like seeing the same complex object from two very different angles, both of which produce a different simplified concept of what the object is. Neither is wrong, but both are incomplete. In this case, the two might be close enough to permit a more accurate stereo view of he object, which is the structure that allows organisms to exist and do what they do.
RM: I couldn’t agree with you less.
Best
Rick
–
Richard S. Marken
"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
–Antoine de Saint-Exupery
[Rick Marken 2019-01-06_10:25:55]
[Martin Taylor 2019.01.04.23.18]
MT: Actually, you did comment properly on one thing I said “* The
free energy principle doesn’t suggest mechanism, but then
neither does e-coli reorganization*”. You said "* Actually,
the e. coli reorganization mechanism is very clearly specified
(see Marken, R. S. and Powers, W. T. (1989) Random-Walk
Chemotaxis: Trial-And-Error as a Control Process. Behavioral
Neuroscience, 103, 1348-1355, which is reprinted in Mind
Readings*)."Â
MT: To see whether what you said it true, I went back and re-readthis paper. I found nothing at all that mentions the mechanism of
e-coli reorganization. I found in the introductory part some
description of a mechanism for the e-coli biological bacterium’s
behaviour, and some very nice simulation models of goal-seeking by
the e-coli process done by humans. Maybe I missed the mechanism in
my earlier reading and on my search for a mention of a mechanism
of e-coli reorganization in my re-reading, and I would be very
happy if you would point me to the page and relevant words in the
“Mind Readings” publication of that paper.
RM: The model is the mechanism. A pseudo-code representation of the mechanism is on p. 92. This mechanism produces behavior that matches that of the subjects when the controlled variable is the component of velocity toward the target (statement 4. in the program on p. 92). The mechanism doesn’t work when the controlled variable is radial distance form the target. So an important aspect of how the e. coli navigation mechanism works is getting the controlled variable right.
rsm
Richard S. MarkenÂ
"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.â?
                --Antoine de Saint-Exupery
[Rick Marken 2019-01-06_14:17:16]
[Frank Lenk 2019.01.05.10:29 CST]
Â
FL: Rick, I agree that the story about the wood lice rejects they have a purpose to seek shade. But the next paragraph, along with its footnote, modifies that:
RM: I don’t think it’s particularly productive to try find what see like PCT compatible ideas in the work of others. Indeed, I think it’s counter-productive. My experience is that in order to see the ideas of others as being compatible with PCT requires that the essential elements of Powers work be ignored. In Friston’s case (and virtually every other case that I’ve encountered) you have to ignore the fact that a central insight of PCT is that behavior is control, that it is organized around the control of a hierarchy of different types of perceptual variables and that the reference states of these variables are set autonomously by the system itself. I guarantee you that there is no other theory of behavior contains any of these insights.
RM: My recommendation is to ignore all other theories of behavior besides PCT and concentrate on the facts that these other theories purport to account for. Then see if you can figure out how PCT would handle these facts. That, rather than looking for superficial similarities of other theories to PCT, would be a major contribution to PCT science.
BestÂ
Rick
–
Richard S. MarkenÂ
"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery
[Martin Taylor 2018.01.07.13.09]
[Rick Marken 2019-01-06_10:25:55]
[Martin Taylor 2019.01.04.23.18]
MT:Actually, you did comment properly on one thing I said "* The
free energy principle doesn’t suggest mechanism, but
then neither does e-coli reorganization* ". You said
“* Actually, the e. coli reorganization mechanism is
very clearly specified (see Marken, R. S. and Powers,
W. T. (1989) Random-Walk Chemotaxis: Trial-And-Error
as a Control Process. Behavioral Neuroscience, 103,
1348-1355, which is reprinted in Mind Readings*).”
MT: To see whether what you said it true, I went backand re-read this paper. I found nothing at all that
mentions the mechanism of e-coli reorganization. I found
in the introductory part some description of a mechanism
for the e-coli biological bacterium’s behaviour, and
some very nice simulation models of goal-seeking by the
e-coli process done by humans. Maybe I missed the
mechanism in my earlier reading and on my search for a
mention of a mechanism of e-coli reorganization in my
re-reading, and I would be very happy if you would point
me to the page and relevant words in the “Mind Readings”
publication of that paper.
RM: The model is the mechanism. A pseudo-coderepresentation of the mechanism is on p. 92. This mechanism
produces behavior that matches that of the subjects when the
controlled variable is the component of velocity toward the
target (statement 4. in the program on p. 92). The mechanism
doesn’t work when the controlled variable is radial distance
form the target. So an important aspect of how the e. coli
navigation mechanism works is getting the controlled
variable right.
The e-coli behaviour is seldom invoked as a way people or
organisms more complex than bacteria actually move toward a
visible target. Controlling perceptions of the distance to the
target in two or three dimensions is much more effective. Only
when the behaviour is restricted to tumbling or continuing in the
same direction do people who are actually controlling for
approaching a target use the e-coli technique.
I any case, how people or other living beings make something
approach a visible target is completely irrelevant to the question
of describing the mechanism of e-coli reorganization of the
control hierarchy.
Let me put the general question more precisely, by splitting it
into its component parts.
Q1. How does the set of parameter changes move in a straight
line? This question has three parts.
Q1a. What defines "continuing to move in the same direction"
when lots of parameters change together and keep changing either
continuously or at discrete moments?
Q1b. What causes a parameter to change?
Q1c. How are changes in many parameters coordinated to keep
the parameter state moving in whatever is defined as “the same
direction”?
Q2. What is the mechanism that determines that the set of many
parameters should stop “moving in the same direction” and instead
start moving in a new direction (tumble)? This question also has
three parts.
Q2a. What mechanism implements the random choice of new
direction?
Q2b. What determines the new value of any one of the myriad
parameters involved in the reorganization “tumble”.
Q2c. Is the "tumble" executed as a single change of direction
or as a curve caused by different parameters changing at different
moments?
If these questions are all answered in terms of known or proposed
properties of the neural or biochemical structure of any organism
more complex than a bacterium, I would agree that a mechanism for
reorganization by e-coli has been described. If they are answered
for an organism as complex as a hunting mammal or an octopus, I
would agree that a mechanism for e-coli reorganization has indeed
been suggested, whereas one based on the free energy principle has
not, so far as I know.
At that point, I would ask Q3 of both e-coli and free-energy:
whether the proposed mechanism implies or relies on modularity
and/or scale-free structures in the control hierarchy.
Martin
[Rick Marken 2019-01-07_15:51:16]
[Martin Taylor 2018.01.07.13.09]
MT: The e-coli behaviour is seldom invoked as a way people ororganisms more complex than bacteria actually move toward a
visible target.
RM: So what? When it has been “invoked” it did a wonderful job of explaining behavior that was thought to be inexplicable by PCT (see its invocation on pp. 21-23 of Mind Readings in response to the Fowler and Turvey hatched job on Bill’s work). Â
MT: Let me put the general question more precisely, by splitting itinto its component parts…
Q1. How does the set of parameter changes move in a straightline? This question has three parts.
   Q1a. What defines "continuing to move in the same direction"when lots of parameters change together and keep changing either
continuously or at discrete moments?
   Q1b. What causes a parameter to change?
   Q1c. How are changes in many parameters coordinated to keepthe parameter state moving in whatever is defined as “the same
direction”? …
Q2. What is the mechanism that determines that the set of manyparameters should stop “moving in the same direction” and instead
start moving in a new direction (tumble)? This question also has
three parts.
  Q2a. What mechanism implements the random choice of newdirection?
  Q2b. What determines the new value of any one of the myriadparameters involved in the reorganization “tumble”.
  Q2c. Is the "tumble" executed as a single change of directionor as a curve caused by different parameters changing at different
moments?Â
RM: All of these questions can only be answered by observing reorganization taking place and then seeing which model fits the data best. That is, the reorganization model has to be tested against data to see how these aspects of the model should be implemented.Â
MT: If these questions are all answered in terms of known or proposedproperties of the neural or biochemical structure of any organism
more complex than a bacterium, I would agree that a mechanism for
reorganization by e-coli has been described.
RM: Fine with me. Disagree all you want. But the e.coli algorithm does describe the basic mechanism of reorganization in the PCT model.Â
rsm
Â
If they are answeredfor an organism as complex as a hunting mammal or an octopus, I
would agree that a mechanism for e-coli reorganization has indeed
been suggested, whereas one based on the free energy principle has
not, so far as I know.
At that point, I would ask Q3 of both e-coli and free-energy:whether the proposed mechanism implies or relies on modularity
and/or scale-free structures in the control hierarchy.
Martin
–
Richard S. MarkenÂ
"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery
Rick,
I promised that I’ll get involved when I’ll see your RCT. This is the moment.
From: Richard Marken (rsmarken@gmail.com via csgnet Mailing List) csgnet@lists.illinois.edu
Sent: Sunday, January 6, 2019 11:17 PM
To: csgnet csgnet@lists.illinois.edu
Subject: Re: WIRED: The Genius Neuroscientist Who Might Hold the Key to True AI
[Rick Marken 2019-01-06_14:17:16]
[Frank Lenk 2019.01.05.10:29 CST]
FL: Rick, I agree that the story about the wood lice rejects they have a purpose to seek shade. But the next paragraph, along with its footnote, modifies that:
RM: I don’t think it’s particularly productive to try find what see like PCT compatible ideas in the work of others.
HB : Rick. It’s a little bit too heavy when you talk in “black and white”, speccially when you are talking about compatibility of other theories with PCT. We both know (and probably) others too (after so many years), what I think about your compatibility with PCT. It seems that you are much more distant from PCT than Friston. His fundaments as far as I could understand are :
Thermodinamic Law about which I assume is in accordance to Wiener’s Cybernetics.
“Control of perception”
Behavior as random actions
He gave an example (observation) of animals which are functioning as Bill predicted (Franks’ description) which says that difference between perception of heat and reference drives random movements.
Well if there is something more please make suplementation. I think your S-R thinking is limiting you in wide view of how orgsnisms function and possible Friston’s support to PCT. I think we could be much stronger now. Here I meant those who are using PCT as basic theory of Living.
You Rick are anyway using psychological theory about “Control of behavior” which is controlling some “controlled aspect” of environment or “controlled variable” and finaly half of the circle ends with CPV (Controlled Perceptual Variable). You never explained how two “references” meet in comparator. But It seems that Bruce Nevin can help you.
RM : Indeed, I think it’s counter-productive.
HB : Well Rick be reasonable. Who is counter productive with RCT (Ricks Control Theory) ?
RM : My experience is that in order to see the ideas of others as being compatible with PCT requires that the essential elements of Powers work be ignored.
HB : I think Friston is using some “essential elements” compatibile with Powers work. But could it be possible that Friston “copied” the idea of “Control of perception” from Powers work ? Because I don’t see how he got it from “free energy” principle. How “free energy principle” could lead to “Control of Perception” ??? The only scientific explanation how we get to “Control of Perception” is by my oppinion hidden in Powers work.
Yournal :
Friston’s free energy principle says that all life, at every scale of organization—from single cells to the human brain, with its billions of neurons—is driven by the same universal imperative, which can be reduced to a mathematical function. To be alive, he says, is to act in ways that reduce the gulf between your expectations and your sensory inputs. Or, in Fristonian terms, it is to minimize free energy.
HB : The simple statement above does not tell me anything about relation between “Perceptual control” and whatever “minimized free energy” could mean as the whole article doesn’t. The difference between perception and “expectations” are scientifically proved by Bill and it’s happening in comparator (neuron, nerv-net). It’s odd that ingeneer proved existance of “biological error” with neurophysiological means and neurophysiologist proved existance of “error” with what ??? “imagined free energy” ?
Yournal :
First the bad news: The free energy principle is maddeningly difficult to understand. So difficult, in fact, that entire rooms of very, very smart people have tried and failed to grasp it. A Twitter account2 with 3,000 followers exists simply to mock its opacity, and nearly every person I spoke with about it, including researchers whose work depends on it, told me they didn’t fully comprehend it.
Yournal :
But they all have one reason in common for being here, which is that the only person who truly understands Karl Friston’s free energy principle may be Karl Friston himself.
HB : I don’t see the connection between his “Control of perception” and “minimization of free energy”. If I’m honest I don’t understand what it means. It’s quite fusy. But it could be also that Friston don’t understand his theory as well, because at least to me it has no sense except that bases is Thermodinymical Law (universal imperative). But who doesn’t know that ? (Rick Marken ?).
Yournal :
But often those same people hastened to add that the free energy principle, at its heart, tells a simple story and solves a basic puzzle. The second law of thermodynamics tells us that the universe tends toward entropy, toward dissolution; but living things fiercely resist it. We wake up every morning nearly the same person we were the day before, with clear separations between our cells and organs, and between us and the world without…
HB : To say that Thermodynamical Law is in the bases of Living processes is the same what many scientists were telling long before Friston. And that “Universe subspace” and it’s “hiden states” are influencing the way how organisms function is also known for centuries. It can be seen also in Science fiction movies.
Many theories explain why and how homeostasis is kept in organism (almost constant conditions), what explains also why we look most of the time the same. Among these theories are “autopoiesis”, PCT, physiological theories, biological theories, psychological theories (Lewin) etc.
RM : In Friston’s case (and virtually every other case that I’ve encountered) you have to ignore the fact that a central insight of PCT is that behavior is control,
HB : Ufff Rick. How many times do I have to tell you that PCT is not assembled around the central insight that “behavior is control”. Alison please help ?
You are so blind that you don’t see that Friston is our best alliance in promoting Bills’ genious idea that Perception is what is controlled not behavior.
RM : … that it is organized around the control of a hierarchy of different types of perceptual variables and that the reference states of these variables are set autonomously by the system itself. I guarantee you that there is no other theory of behavior contains any of these insights.
HB : Which Theories ? About “Control of behavior” or “Control of perception” ?
RM: My recommendation is to ignore all other theories of behavior besides PCT…
HB : So O.K. We’ll start to ignore (those who understand PCT) all other theories beside PCT what of course includes RCT (Ricks’ Control Theory). So please members on CSGnet please ignore RCT. I hope Rick will be helpfull. You are on the move now, Rick. It’s your’s recommendation.
RM : ….and concentrate on the facts that these other theories purport to account for. Then see if you can figure out how PCT would handle these facts. That, rather than looking for superficial similarities of other theories to PCT, would be a major contribution to PCT science.
HB : Well if I’m honest I agree about this one with you. I iether don’t see Friston’s contribution to PCT with “free energy”, although I see strong support to “Control of Perception” from enforced scientist, maybe even famous who is promoting PCT :
“Control of perception”
Behavior is “random”.
Let us see how PCT function with these principles :
Bill P :
Our only view of the real world is our view of the neural signals that represent it inside our own brains. When we act to make a perception change to our more desireble state – when we make the perception of the glass change from »on the table« to »near the mouth« - we have no direct knowledge of what we are doing to the reality that is the origin of our neural signal; we know only the final result, how the result looks, feels, smells, sounds, tastes, and so forth…It means that we produce actions that alter the world of perception…
HB : These are some of the main principles in PCT and they are all connected to physiological processes in organism which is science that works in practice. Even Neurophysiology is traped in Living organisms’ physiological functioning. Brains or soul or whatever, was not put into organisms by a god or any other “creature”, but was developed inside organism supporting homeostatical processes in cells organization through bilions of years. Some of these cells were developed into “neurons”.
I don’t understand why Friston didn’t connect everything theoretical about his view of the brain “functioning” with physiology. Physiology is explaining how organisms function and can be used practically in any explanation of organisms functioning including possible experiments. So if anybody see anything from Fristons’ work being in relation to physiological principles of how organisms fuction please inform me. That is the “connection point with PCT”. Both theories has to have the same root explanation of how organism works in practice. PCT has that with no doubt. But does Friston has it ? I can imagine conversation between Warren and Friston both talking one by another with their different terminologies. The common denominator is physiological language the one that works in practice and can be any time proved in practice. Whether patient lives or dies ? Phylosophy will not help much.
I think we should use Friston’s statements about “Control of perception” (if it is his, he is much younger from Bill) and promote that at least in this point they agree that Perception is controlled not behavior. I think this is the most important compatibility between them. I also think this could be a huge step in PCT promotion although Friston by my oppinion don’t understand much of organisms functioning and social functioning of organisms.
[FL: Here, PCT seems much more fully developed, with its discussion of reorganization, conflict, and the ability to go up a level to resolve conflicts].
HB : I think that PCT is much more developed in many sense in respect to Fristons’ “free energy”. But he can be welcome to PCT because of his reputation and basic understanding that “Perception is controlled” not behavior.
Yournal :
Friston isn’t just one of the most influential scholars in his field; he’s also among the most prolific in any discipline. He is 59 years old, works every night and weekend, and has published more than 1,000 academic papers since the turn of the millennium. In 2017 alone, he was a lead or coauthor of 85 publications3—which amounts to approximately one every four days.
HB : If this is not enough good evidence what kind of support could PCT has with kind relationship with Friston I don’t know what is the goal of CSGnet forum. Everyday talkings about whether “Behavior is controlled” or “Perception is controlled” ???
I agree with Martin. Show us some science Rick. You Rick are not scientific. I’m providing for years Bills’ evidences about PCT, you are providing almost nothing about PCT. Instead you are promoting your RCT which could be at the best shot equlized with Carvers’ book “On the Self-regulation of behavior” where behavior is with no doubt “controlled”.
MT : It would be easier to have a scientific discussion…
HB : So Rick no more “empty talkings”. Start scientific upgrade of diagram on p. 191 (B:CP, 2005) and make scientific contribution to PCT model of how organisms function.
Boris
Best
Rick
–
Richard S. Marken
"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.”
–Antoine de Saint-Exupery
[Martin Taylor 2019.01.08.11.44]
[Rick Marken 2019-01-07_15:51:16]
[Martin Taylor 2018.01.07.13.09]
…
RM: All of these questions can only be answered byobserving reorganization taking place and then seeing which
model fits the data best. That is, the reorganization model
has to be tested against data to see how these aspects of
the model should be implemented.
MT: If these questions are all answered in terms ofknown or proposed properties of the neural or
biochemical structure of any organism more complex than
a bacterium, I would agree that a mechanism for
reorganization by e-coli has been described.
RM: Fine with me. Disagree all you want. But the e.colialgorithm does describe the basic mechanism of
reorganization in the PCT model.
I guess we use the word "mechanism" differently. If I see a crane
lifting something and I want to know its mechanism, I would think
in terms of levers and gears and motors. According to your
response to my questions, the mechanism is that “a crane lifts
things.”
In the same vein, the mechanism of the "free energy principle" is
that the hierarchy of perceiving functions improve their accuracy
of modelling the real-reality environment, while the output
functions improve their accuracy of acting so that the state of
the outputs of the perceiving functions more accurately matches
the predictions. In your use of the term “mechanism”, that’s a
perfectly satisfactory specification of the mechanism. We need
look no further to find out how all this happens.
As with the early versions of the Mac operating systems, "It Just
Works" applies equally to both. The users of those early Mac
Operating systems could use their computers without, or with very
little, access to manuals. They didn’t need to know the mechanism.
As you describe yourself, you are a User of PCT that “Just Works”.
Friston and his colleagues are, so far as I can see, Users of a
“Free Energy Principle” that “Just Works”.
Martin
[Rick Marken 2019-01-08_15:01:18]
[Martin Taylor 2019.01.08.11.44]
MT: If these questions are all answered in terms ofknown or proposed properties of the neural or
biochemical structure of any organism more complex than
a bacterium, I would agree that a mechanism for
reorganization by e-coli has been described.
RM: Fine with me. Disagree all you want. But the e.colialgorithm does describe the basic mechanism of
reorganization in the PCT model.Â
MT: I guess we use the word "mechanism" differently. If I see a cranelifting something and I want to know its mechanism, I would think
in terms of levers and gears and motors. According to your
response to my questions, the mechanism is that “a crane lifts
things.”
RM: No, according to my answer to your question the mechanism of the crane would be described in terms of the essential components (the variables of a model), how they are arranged (their structural organization blueprint) and how they work (the functional relationships between the variables). This mechanism can then be implemented in many different ways just as the mechanism of e. coli navigation can be implemented in many different ways: as a computer program, a bacterium, a cockroach, etc.Â
rsm
Richard S. MarkenÂ
"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.â?
                --Antoine de Saint-Exupery
[Martin Taylor 2019.01.08.22.52]
[Rick Marken 2019-01-08_15:01:18]
[Martin Taylor 2019.01.08.11.44]
MT: If these questions are all answered interms of known or proposed properties of the
neural or biochemical structure of any
organism more complex than a bacterium, I
would agree that a mechanism for
reorganization by e-coli has been described.
RM: Fine with me. Disagree all you want. Butthe e.coli algorithm does describe the basic
mechanism of reorganization in the PCT model.
MT: I guess we use the word "mechanism" differently. IfI see a crane lifting something and I want to know its
mechanism, I would think in terms of levers and gears
and motors. According to your response to my questions,
the mechanism is that “a crane lifts things.”
RM: No, according to my answer to your question themechanism of the crane would be described in terms of the
essential components (the variables of a model), how they
are arranged (their structural organization blueprint) and
how they work (the functional relationships between the
variables). This mechanism can then be implemented in many
different ways just as the mechanism of e. coli navigation
can be implemented in many different ways: as a computer
program, a bacterium, a cockroach, etc.
You are changing your answer again, but still your definition of
mechanism is different from mine.
Nobody, least of all me, denies that the e-coli procedure of
hill-climbing optimization can be implemented in many ways in many
different physical media. The actual point at issue comes from
your response to what I thought was a non-controversial statement
in [Martin Taylor 2019.01.04.23.18], that “* The free energy
principle doesn’t suggest mechanism, but then neither does
e-coli reorganization*”. In response, you claimed that the *** actual*** mechanism used to implement the e-coli process of
reorganizing the control hierarchy was well understood and had
been described in a paper you published. But then things
changed…
if I may remind you of your answer in [Rick Marken
2019-01-07_15:51:16] about whether the mechanism of the e-coli
method of reorganization was known when I asked that question more
simply, as six individual questions for which the answers would
constitute much of a mechanism: “* RM: All of these questions can
only be answered by observing reorganization taking place and
then seeing which model fits the data best. That is, the
reorganization model has to be tested against data to see how
these aspects of the model should be implemented.”* Unless I
very much misread this answer, you are agreeing with me that the
mechanism of e-coli reorganization is, as I said, as unspecified
as is the mechanism (so far as I know) of the free energy approach
to what we would call reorganization and control.
At least we seem now to agree that there are many ways in which
e-coli optimization can be implemented, and apparently also that
the implementation in reorganization of a living control hierarchy
has not been described satisfactorily in any place either of us
know of.
Martin
Good hell! I’ve not been following the
list now for quite some time, for various reasons and don’t intend
to restart.
However, this 'fight' between Martin
and Rick has reached the point of absurdity!
Yes, absolutely, Perception is what is
being controlled! AND yes behaviour is control! BCP: =
Behaviour, the control of perception! The point being that
behaviour is NOT what is controlled (unless there is a perception
related to the behaviour that has a reference value). Perception
is what is controlled and behaviour is the means by which that
control is achieved… so behaviour IS control. Saying that does
NOT mean that one considers the entire control loop as not being
significant.
When you observe an individual acting
on the environment (or internally when it is possible to make the
necessary measurements), they are demonstrating that their is some
aspect of their perception of the environment that they have a
reference for and the perception is NOT matching the reference.
It is indeed that belief that spawned
the TEST! The truth of PCT is that a proper TEST achieves
correlations of 95% or greater. And of course, as Bill himself
said, an “outlier” means that you have fertile ground for further
research (and that their is some deficiency in you TEST
methodology!
So can we be done with this RCT vs PCT
BS?
bill
On 1/8/19 3:53 AM, "Boris Hartman"
( via csgnet Mailing List) wrote:
Rick,
Ipromised that I’ll get involved when I’ll see your RCT. This
is the moment.
From: Richard Marken
( via csgnet Mailing List)
Sunday, January 6, 2019 11:17 PM
csgnet Re: WIRED: The Genius Neuroscientist Who Might
Hold the Key to True AI
[Rick Marken 2019-01-06_14:17:16]
[Frank Lenk 2019.01.05.10:29 CST]
FL: Rick, I agree that the storyabout the wood lice rejects they have a purpose to
seek shade. But the next paragraph, along with
its footnote, modifies that:
RM: I don't think it's particularlyproductive to try find what see like PCT compatible
ideas in the work of others.
HB : Rick.It’s a little bit too heavy when you talk in “black and
white”, speccially when you are talking about
compatibility of other theories with PCT. We both know
(and probably) others too (after so many years), what I
think about your compatibility with PCT. It seems that
you are much more distant from PCT than Friston. His
fundaments as far as I could understand are :
ThermodinamicLaw about which I assume is in accordance to Wiener’s
Cybernetics.
"Controlof perception"
Behavioras random actions
Hegave an example (observation) of animals which are
functioning as Bill predicted (Franks’ description)
which says that difference between perception of heat
and reference drives random movements.
Well ifthere is something more please make suplementation. I
think your S-R thinking is limiting you in wide view of
how orgsnisms function and possible Friston’s support to
PCT. I think we could be much stronger now. Here I meant
those who are using PCT as basic theory of Living.
You Rickare anyway using psychological theory about “Control of
behavior” which is controlling some “controlled aspect”
of environment or “controlled variable” and finaly half
of the circle ends with CPV (Controlled Perceptual
Variable). You never explained how two “references” meet
in comparator. But It seems that Bruce Nevin can help
you.
RM : Indeed, I think it'scounter-productive.
HB : WellRick be reasonable. Who is counter productive with RCT
(Ricks Control Theory) ?
RM : My experience is that in orderto see the ideas of others as being compatible with PCT
requires that the essential elements of Powers work be
ignored.
HB : Ithink Friston is using some “essential elements”
compatibile with Powers work. But could it be possible
that Friston “copied” the idea of “Control of
perception” from Powers work ? Because I don’t see how
he got it from “free energy” principle. How “free energy
principle” could lead to “Control of Perception” ??? The
only scientific explanation how we get to
“Control of Perception” is by my oppinion hidden in
Powers work.
Yournal:
Friston’sfree energy principle says that all life, at every
scale of organization—from single cells to the human
brain, with its billions of neurons—is driven by the
same universal imperative, which can be reduced to a
mathematical function. To be alive, he says, is to act
in ways that reduce the gulf between your expectations
and your sensory inputs. Or, in Fristonian terms, it
is to * minimize
free energy*.
HB : Thesimple statement above does not tell me anything about
relation between “Perceptual control” and whatever
“minimized free energy” could mean as the whole article
doesn’t. The difference between perception and
“expectations” are scientifically proved by Bill and
it’s happening in comparator (neuron, nerv-net). It’s
odd that ingeneer proved existance of “biological error”
with neurophysiological means and neurophysiologist
proved existance of “error” with what ??? “imagined free
energy” ?
Yournal:
Firstthe bad news: The free energy principle is maddeningly
difficult to understand. So difficult, in fact, that
entire rooms of very, very smart people have tried and
failed to grasp it. A Twitter
account2 with
3,000 followers exists simply to mock its opacity, and
nearly every person I spoke with about it, including
researchers whose work depends on it, told me they
didn’t fully comprehend it.
Yournal:
Butthey all have one reason in common for being here,
which is that the only person who truly understands
Karl Friston’s free energy principle may be Karl
Friston himself.
HB : Idon’t see the connection between his “Control of
perception” and “minimization of free energy”. If I’m
honest I don’t understand what it means. It’s quite
fusy. But it could be also that Friston don’t understand
his theory as well, because at least to me it has no
sense except that bases is Thermodinymical Law
(universal imperative). But who doesn’t know that ?
(Rick Marken ?).
Yournal:
Butoften those same people hastened to add that the free
energy principle, at its heart, tells a simple story
and solves a basic puzzle. The second law of
thermodynamics tells us that the universe tends toward
entropy, toward dissolution; but living things
fiercely resist it. We wake up every morning nearly
the same person we were the day before, with clear
separations between our cells and organs, and between
us and the world without…
HB : Tosay that Thermodynamical Law is in the bases of Living
processes is the same what many scientists were telling
long before Friston. And that “Universe subspace” and
it’s “hiden states” are influencing the way how
organisms function is also known for centuries. It can
be seen also in Science fiction movies.
Manytheories explain why and how homeostasis
is kept in organism (almost constant conditions) ,
what explains also why we look most of the time the
same. Among these theories are “autopoiesis”, PCT,
physiological theories, biological theories,
psychological theories (Lewin) etc.
RM : In Friston's case (and virtuallyevery other case that I’ve encountered) you have to
ignore the
fact that a central insight of PCT is that behavior is
control,
HB : UfffRick. How many times do I have to tell you that PCT is
not assembled around the central insight that
“behavior is control”. Alison please help ?
You areso blind that you don’t see that Friston is our best
alliance in promoting Bills’ genious idea that
Perception is what is controlled not behavior.
RM : … that it is organized aroundthe control of a hierarchy of different types of
perceptual variables and that the reference states of
these variables are set autonomously by the system
itself. I guarantee you that there is no other theory of
behavior contains any of these insights.
HB : WhichTheories ? About “Control of behavior” or “Control of
perception” ?
RM: My
recommendation is to ignore all other theories
of behavior besides PCT…
HB : SoO.K. We’ll start to ignore (those who understand PCT)
all other theories beside PCT what of course includes
RCT (Ricks’ Control Theory). So please members on CSGnet
please ignore RCT. I hope Rick will be helpfull. You are
on the move now, Rick. It’s
your’s recommendation.
RM : ….and concentrate on the factsthat these other theories purport to account for. Then
see if you can figure out how PCT would handle these
facts. That, rather than looking for superficial
similarities of other theories to PCT, would be a major
contribution to PCT science.
HB : Wellif I’m honest I agree about this one with you. I iether
don’t see Friston’s contribution to PCT with “free
energy”, although I see strong support to “Control of
Perception” from enforced scientist, maybe even famous
who is promoting PCT :
"Control
Behavior
Let us see how PCT function withthese principles :
Bill P:
Ouronly view of the real world is our view of the neural
signals that represent it inside our own brains. When
we act to make a perception change to our more
desireble state – when we make the perception of the
glass change from »on the table« to »near the mouth« -
we have no direct knowledge of what we are doing to
the reality that is the origin of our neural signal;
we know only the final result, how the result looks,
feels, smells, sounds, tastes, and so forth…It means
that we produce actions that alter the world of
perception…
HB : Theseare some of the main principles in PCT and they are all
connected to physiological processes in organism which
is science that works in practice. Even Neurophysiology
is traped in Living organisms’ physiological
functioning. Brains or soul or whatever, was not put
into organisms by a god or any other “creature”, but was
developed inside organism supporting homeostatical
processes in cells organization through bilions of
years. Some of these cells were developed into
“neurons”.
I don't understand why Friston didn'tconnect everything theoretical about his view of the
brain “functioning” with physiology. Physiology is
explaining how organisms function and can be used
practically in any explanation of organisms functioning
including possible experiments. So if anybody see
anything from Fristons’ work being in relation to
physiological principles of how organisms fuction please
inform me. That is the “connection point with PCT”. Both
theories has to have the same root explanation of how
organism works in practice. PCT has that with no doubt.
But does Friston has it ? I can imagine conversation
between Warren and Friston both talking one by another
with their different terminologies. The common
denominator is physiological language the one that works
in practice and can be any time proved in practice.
Whether patient lives or dies ? Phylosophy will not help
much.
I think we should use Friston'sstatements about “Control of perception” (if it is his,
he is much younger from Bill) and promote that at least
in this point they agree that Perception is controlled
not behavior. I think this is the most important
compatibility between them. I also think this could be a
huge step in PCT promotion although Friston by my
oppinion don’t understand much of organisms functioning
and social functioning of organisms.
[FL: Here, PCT seems much more fullydeveloped, with its discussion of reorganization,
conflict, and the ability to go up a level to resolve
conflicts].
HB : I think that PCT is much moredeveloped in many sense in respect to Fristons’ “free
energy”. But he can be welcome to PCT because of his
reputation and basic understanding that “Perception is
controlled” not behavior.
Yournal:
Fristonisn’t just one of
the most influential scholars in his field; he’s also
among the most prolific in any discipline. He is 59
years old, works every night and weekend, and has
published more than 1,000 academic papers since the
turn of the millennium. In 2017 alone, he was a lead
or coauthor of 85 publications3 —which
amounts to approximately one every four days.
HB : If this is not enough goodevidence what kind of support could PCT has with kind
relationship with Friston I don’t know what is the goal
of CSGnet forum. Everyday talkings about whether
“Behavior is controlled” or “Perception is controlled”
???
I agree with Martin. Show us somescience Rick. You Rick are not scientific. I’m providing
for years Bills’ evidences about PCT, you are providing
almost nothing about PCT. Instead you are promoting your
RCT which could be at the best shot equlized with
Carvers’ book “On the Self-regulation of behavior” where
behavior is with no doubt “controlled”.
MT : It would be easier to have ascientific discussion…
HB : So Rick no more "emptytalkings". Start scientific upgrade of diagram on p. 191
(B:CP, 2005) and make scientific contribution to PCT
model of how organisms function.
Boris
Best
Rick
–
Richard S.Marken
"Perfectionis achieved not when you have
nothing more to add, but when youhave nothing left to take away.”–Antoine de Saint-Exupery
boris.hartman@masicom.net
rsmarken@gmail.comcsgnet@lists.illinois.edu
Sent:
To:csgnet@lists.illinois.edu
Subject:
Good hell! I’ve not been following the list now for quite some time, for various reasons and don’t intend to restart.
However, this ‘fight’ between Martin and Rick has reached the point of absurdity!
Yes, absolutely, Perception is what is being controlled! AND yes behaviour is control! BCP: = Behaviour, the control of perception! The point being that behaviour is NOT what is controlled (unless there is a perception related to the behaviour that has a reference value). Perception is what is controlled and behaviour is the means by which that control is achieved… so behaviour IS control. Saying that does NOT mean that one considers the entire control loop as not being significant.
When you observe an individual acting on the environment (or internally when it is possible to make the necessary measurements), they are demonstrating that their is some aspect of their perception of the environment that they have a reference for and the perception is NOT matching the reference.
It is indeed that belief that spawned the TEST! The truth of PCT is that a proper TEST achieves correlations of 95% or greater. And of course, as Bill himself said, an “outlier” means that you have fertile ground for further research (and that their is some deficiency in you TEST methodology!
So can we be done with this RCT vs PCT BS?
bill
From: Bill Leach (wrleach@cableone.net via csgnet Mailing List) [mailto:csgnet@lists.illinois.edu]
Sent: Thursday, January 10, 2019 5:03 AM
To: csgnet@lists.illinois.edu
Subject: Re: WIRED: The Genius Neuroscientist Who Might Hold the Key to True AI
Hi Bill
Where did you come from ? Middle Ages ? J
You exploaded again in wrong direction. Behavior is generally not controlled. Show us how, not just with wording and phylosophing, but with evidences ? With cheep sentences you’ll hardly prove that »Behavior is control«. I would say that this is an insult for PCT. It seems that you think that PCT is cheep psychological »Behavior is control« theory.
Well I advice you to start reading and studying PCT. Whatever you are doing is just making more confussion.
You are denying beasic principles of PCT. For you »Behavior can be controlled« but that doesn’t mean that it is. You can’t control muscle tension directly. Let us see some PCT »facts« :
Bill P. (B:CP ) :
Rather, the central problem has been to find out a plausible model which can behave at all…. For example it will be shown later that the brain does not command the muscles to act. That concept implies properties that the neuromuscular system simply does not have… There is just no way the brain can select a muscle tension that will produce one and only one behavioral effect, even if that tension is accurately produced. The result of this approcah is a model nearly devoid of specific behavioral content.
HB : There is not only basic PCT text denying your idea of »Control of behavior«, but there are also physiologocal evidences which you obviously »jumped over«. And there is a problem of generality. PCT contol loop works for every behavior the same. Not just for some of them in »control mode« and selectivelly. See diagram LCS III. It’s general PCT diagram about »Control of perception«.
I’ll just expose some of your wrong statements.
BL : Yes, absolutely, Perception is what is being controlled! AND yes behaviour is control! BCP: = Behaviour, the control of perception!
The answer to your question is simple. In PCT control loop it is obviously that behavior is coming after »Control of perception« in »comparator«. So »Behavior : the control of perception« does not mean that »with behavior we control perception«, because we don’t. But we can also affect perception among other effects. With control of perception we produce behavior, random movements.
Bill P :
Our only view of the real world is our view of the neural signals that represent it inside our own brains. When we act to make a perception change to our more desireble state – when we make the perception of the glass change from »on the table« to »near the mouth« - we have no direct knowledge of what we are doing to the reality that is the origin of our neural signal; we know only the final result, how the result looks, feels, smells, sounds, tastes, and so forth…It means that we produce actions that alter the world of perception…
HB : When perception is changed there is no knowledge of what you did to reality. There are just perceptions of effects of movements in environment. So conclussion when you see that »actions« are beiing controlled is exclusivelly yours. It’s an ilussion. I just don’t understand how you got it from Bills’ literature ???
Behavior is consequence of »Control of perception and it is »blind«, »random« like reorganization in nervous system is. You don’t know what you are doing to reality until you perceive it. So you don’t control your movements and thus you can’t control your perception with behavior. Movements are random, like Friston saw it.
Show us in Bills’ diagram how »Control of behavior« works ??w
Diagram (LCS III) :
HB : Diagram shows that Behavior is produced by »Control of perception« so Behavior is consequence of »Control of perception«. I hope you see it in diagram. First comparator and »Control of perception« than behavior. Behavior does not control anything. Show me where do you see it ?
HB : There is generally no »controlled variable« in environment so that »Control of behavior« could control something outside. We of course talk about PCT. What are you talking about ?
BL : It is indeed that belief that spawned the TEST! The truth of PCT is that a proper TEST achieves correlations of 95% or greater.
TCV (Test for the controlled variable) in PCT is not used for researching what is controlled outside so it’s NOT in use to determine whether »behavior is control« or not, but what was controlled inside (see definition of control). How TCV works :
Bill P (B:CP) :
The TCV is method for identifying control organization of nervous system….
There will be ambiguous cases : the disturbance may be only weakly opposed. That effect could be due not to poor control system but to a definition of actions that are only remotely linked to the actual controlled quantity.
For example : if when you open the window I sometimes get up and close it, you might conclude that I am controlling the position of the window when in fact I only shut it if the room gets too chilly to suit me. I could be controlling sensed temperature very precisely, when necesarry, but by a variety of means : shutting the window, turning up the termostat, putting on a sweater, or exercising. You are on the track of the right controlled quantity, but haven’t got the right definition yet. It is safest to assume that an ambiguous result from TCV is the fault of the hypotehsis and to continue looking for a better definition of the controlled quantity
HB : I could show you how this works in practice with some series of experiments, but it’s more probable that we will never meet. So you’ll have to beleive Bill’s literature.
Bill P (B:CP):
CONTROL : Achievement and maintenance of a preselected state in the controlling system, through actions on the environment that also cancel the effects of disturbances.
HB : So whatever you say about control in area of PCT it has to be in accordance to basic definition of control. And Basic definition talks only about control in organism not outside. Definitions of control loop in B:CP clearly show that actions are not controlled. Show us from Bills definitions of control loop that outisde the controlling system controls. Where do you see control outside ???
Bill P (B:CP) :
OUTPUT FUNCTION : The portion of a system that converts the magnitude or state of a signal inside the system into a corresponding set of effects on the immediate environment of the system
Bill P (LCS III):…the output function shown in it’s own box represents the means this system has for causing changes in it’s environment.
Bill P (LCS III):
FEED-BACK FUNCTION : The box represents the set of physical laws, properties, arrangements, linkages, by which the action of this system feeds-back to affect its own input, the controlled variable. That’s what feed-back means : it’s an effect of a system’s output on it’s own input.
Bill P (B:CP) :
INPUT FUNCTION : The portion of a system that receives signals or stimuli from outside the system, and generates a perceptual signal that is some function of the received signals or stimuli.
HB : These are definitions of outside part of »control loop«. So where do you see that »Behavior is controlled« ???
But of course you can talk and spread rumors arround that »behavior is selectively controlled« :
BL : The point being that behaviour is NOT what is controlled (unless there is a perception related to the behaviour that has a reference value).
But you have to add also that this is yours BLCT (Bill Leach Control Theory).
HB : So you think that muscle tension is working in »double mode« ???
If I try to translate what you wrote it seems that you are saying : When there is »no reference value« for muscle tension, behavior is NOT controlled. But when muscle tension has reference valuue, then »Behavior IS controlled«. Is that what you wanted to say ?
Bill P (LCS I) :
There is one explanation for existance of reference states that has been proposed over and over for centuries : they are determined by the intentions of the behaving system. The driver has inside him, the intention that the door be open. He acts to achieve this purpose, doing whatever is required (if possible) to achieve it.
O.K. Show us your evidences for »Behavior is control« and references outside and so on ???
BL : So can we be done with this RCT vs PCT BS?
HB : RCT (Ricks’ Control Theory) stays. He and we can’t erase what he wrote. Partial home made thinking like yours is, just cause troubles in understanding how PCT really works. And the only one who can confrim what PCT is, is the author. If author is not present himself, we have to rely on his literature. So start reading…. And I hope we could make some more constructive discussion.
Boris
From: Bill Leach (wrleach@cableone.net via csgnet Mailing List) [mailto:csgnet@lists.illinois.edu]
Sent: Thursday, January 10, 2019 5:03 AM
To: csgnet@lists.illinois.edu
Subject: Re: WIRED: The Genius Neuroscientist Who Might Hold the Key to True AI
On 1/8/19 3:53 AM, “Boris Hartman” (boris.hartman@masicom.net via csgnet Mailing List) wrote:
Rick,
I promised that I’ll get involved when I’ll see your RCT. This is the moment.
From: Richard Marken (rsmarken@gmail.com via csgnet Mailing List) csgnet@lists.illinois.edu
Sent: Sunday, January 6, 2019 11:17 PM
To: csgnet csgnet@lists.illinois.edu
Subject: Re: WIRED: The Genius Neuroscientist Who Might Hold the Key to True AI
[Rick Marken 2019-01-06_14:17:16]
[Frank Lenk 2019.01.05.10:29 CST]
FL: Rick, I agree that the story about the wood lice rejects they have a purpose to seek shade. But the next paragraph, along with its footnote, modifies that:
RM: I don’t think it’s particularly productive to try find what see like PCT compatible ideas in the work of others.
HB : Rick. It’s a little bit too heavy when you talk in “black and white”, speccially when you are talking about compatibility of other theories with PCT. We both know (and probably) others too (after so many years), what I think about your compatibility with PCT. It seems that you are much more distant from PCT than Friston. His fundaments as far as I could understand are :
Thermodinamic Law about which I assume is in accordance to Wiener's Cybernetics.
"Control of perception"
Behavior as random actions
He gave an example (observation) of animals which are functioning as Bill predicted (Franks' description) which says that difference between perception of heat and reference drives random movements.
Well if there is something more please make suplementation. I think your S-R thinking is limiting you in wide view of how orgsnisms function and possible Friston’s support to PCT. I think we could be much stronger now. Here I meant those who are using PCT as basic theory of Living.
You Rick are anyway using psychological theory about “Control of behavior” which is controlling some “controlled aspect” of environment or “controlled variable” and finaly half of the circle ends with CPV (Controlled Perceptual Variable). You never explained how two “references” meet in comparator. But It seems that Bruce Nevin can help you.
RM : Indeed, I think it’s counter-productive.
HB : Well Rick be reasonable. Who is counter productive with RCT (Ricks Control Theory) ?
RM : My experience is that in order to see the ideas of others as being compatible with PCT requires that the essential elements of Powers work be ignored.
HB : I think Friston is using some “essential elements” compatibile with Powers work. But could it be possible that Friston “copied” the idea of “Control of perception” from Powers work ? Because I don’t see how he got it from “free energy” principle. How “free energy principle” could lead to “Control of Perception” ??? The only scientific explanation how we get to “Control of Perception” is by my oppinion hidden in Powers work.
Yournal :
Friston’s free energy principle says that all life, at every scale of organization—from single cells to the human brain, with its billions of neurons—is driven by the same universal imperative, which can be reduced to a mathematical function. To be alive, he says, is to act in ways that reduce the gulf between your expectations and your sensory inputs. Or, in Fristonian terms, it is to minimize free energy.
HB : The simple statement above does not tell me anything about relation between “Perceptual control” and whatever “minimized free energy” could mean as the whole article doesn’t. The difference between perception and “expectations” are scientifically proved by Bill and it’s happening in comparator (neuron, nerv-net). It’s odd that ingeneer proved existance of “biological error” with neurophysiological means and neurophysiologist proved existance of “error” with what ??? “imagined free energy” ?
Yournal :
First the bad news: The free energy principle is maddeningly difficult to understand. So difficult, in fact, that entire rooms of very, very smart people have tried and failed to grasp it. A Twitter account2 with 3,000 followers exists simply to mock its opacity, and nearly every person I spoke with about it, including researchers whose work depends on it, told me they didn’t fully comprehend it.
Yournal :
But they all have one reason in common for being here, which is that the only person who truly understands Karl Friston’s free energy principle may be Karl Friston himself.
HB : I don’t see the connection between his “Control of perception” and “minimization of free energy”. If I’m honest I don’t understand what it means. It’s quite fusy. But it could be also that Friston don’t understand his theory as well, because at least to me it has no sense except that bases is Thermodinymical Law (universal imperative). But who doesn’t know that ? (Rick Marken ?).
Yournal :
But often those same people hastened to add that the free energy principle, at its heart, tells a simple story and solves a basic puzzle. The second law of thermodynamics tells us that the universe tends toward entropy, toward dissolution; but living things fiercely resist it. We wake up every morning nearly the same person we were the day before, with clear separations between our cells and organs, and between us and the world without…
HB : To say that Thermodynamical Law is in the bases of Living processes is the same what many scientists were telling long before Friston. And that “Universe subspace” and it’s “hiden states” are influencing the way how organisms function is also known for centuries. It can be seen also in Science fiction movies.
Many theories explain why and how homeostasis is kept in organism (almost constant conditions), what explains also why we look most of the time the same. Among these theories are “autopoiesis”, PCT, physiological theories, biological theories, psychological theories (Lewin) etc.
RM : In Friston’s case (and virtually every other case that I’ve encountered) you have to ignore the fact that a central insight of PCT is that behavior is control,
HB : Ufff Rick. How many times do I have to tell you that PCT is not assembled around the central insight that “behavior is control”. Alison please help ?
You are so blind that you don’t see that Friston is our best alliance in promoting Bills’ genious idea that Perception is what is controlled not behavior.
RM : … that it is organized around the control of a hierarchy of different types of perceptual variables and that the reference states of these variables are set autonomously by the system itself. I guarantee you that there is no other theory of behavior contains any of these insights.
HB : Which Theories ? About “Control of behavior” or “Control of perception” ?
RM: My recommendation is to ignore all other theories of behavior besides PCT…
HB : So O.K. We’ll start to ignore (those who understand PCT) all other theories beside PCT what of course includes RCT (Ricks’ Control Theory). So please members on CSGnet please ignore RCT. I hope Rick will be helpfull. You are on the move now, Rick. It’s your’s recommendation.
RM : ….and concentrate on the facts that these other theories purport to account for. Then see if you can figure out how PCT would handle these facts. That, rather than looking for superficial similarities of other theories to PCT, would be a major contribution to PCT science.
HB : Well if I’m honest I agree about this one with you. I iether don’t see Friston’s contribution to PCT with “free energy”, although I see strong support to “Control of Perception” from enforced scientist, maybe even famous who is promoting PCT :
- “Control of perception”
- Behavior is “random”.
Let us see how PCT function with these principles :
Bill P :
Our only view of the real world is our view of the neural signals that represent it inside our own brains. When we act to make a perception change to our more desireble state – when we make the perception of the glass change from »on the table« to »near the mouth« - we have no direct knowledge of what we are doing to the reality that is the origin of our neural signal; we know only the final result, how the result looks, feels, smells, sounds, tastes, and so forth…It means that we produce actions that alter the world of perception…
HB : These are some of the main principles in PCT and they are all connected to physiological processes in organism which is science that works in practice. Even Neurophysiology is traped in Living organisms’ physiological functioning. Brains or soul or whatever, was not put into organisms by a god or any other “creature”, but was developed inside organism supporting homeostatical processes in cells organization through bilions of years. Some of these cells were developed into “neurons”.
I don’t understand why Friston didn’t connect everything theoretical about his view of the brain “functioning” with physiology. Physiology is explaining how organisms function and can be used practically in any explanation of organisms functioning including possible experiments. So if anybody see anything from Fristons’ work being in relation to physiological principles of how organisms fuction please inform me. That is the “connection point with PCT”. Both theories has to have the same root explanation of how organism works in practice. PCT has that with no doubt. But does Friston has it ? I can imagine conversation between Warren and Friston both talking one by another with their different terminologies. The common denominator is physiological language the one that works in practice and can be any time proved in practice. Whether patient lives or dies ? Phylosophy will not help much.
I think we should use Friston’s statements about “Control of perception” (if it is his, he is much younger from Bill) and promote that at least in this point they agree that Perception is controlled not behavior. I think this is the most important compatibility between them. I also think this could be a huge step in PCT promotion although Friston by my oppinion don’t understand much of organisms functioning and social functioning of organisms.
[FL: Here, PCT seems much more fully developed, with its discussion of reorganization, conflict, and the ability to go up a level to resolve conflicts].
HB : I think that PCT is much more developed in many sense in respect to Fristons’ “free energy”. But he can be welcome to PCT because of his reputation and basic understanding that “Perception is controlled” not behavior.
Yournal :
Friston isn’t just one of the most influential scholars in his field; he’s also among the most prolific in any discipline. He is 59 years old, works every night and weekend, and has published more than 1,000 academic papers since the turn of the millennium. In 2017 alone, he was a lead or coauthor of 85 publications3—which amounts to approximately one every four days.
HB : If this is not enough good evidence what kind of support could PCT has with kind relationship with Friston I don’t know what is the goal of CSGnet forum. Everyday talkings about whether “Behavior is controlled” or “Perception is controlled” ???
I agree with Martin. Show us some science Rick. You Rick are not scientific. I’m providing for years Bills’ evidences about PCT, you are providing almost nothing about PCT. Instead you are promoting your RCT which could be at the best shot equlized with Carvers’ book “On the Self-regulation of behavior” where behavior is with no doubt “controlled”.
MT : It would be easier to have a scientific discussion…
HB : So Rick no more “empty talkings”. Start scientific upgrade of diagram on p. 191 (B:CP, 2005) and make scientific contribution to PCT model of how organisms function.
Boris
Best
Rick
–
Richard S. Marken
"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.”
–Antoine de Saint-Exupery
Damn Boris! I DID NOT SAY THAT
BEHAVIOR IS CONTROLLED! I said that behavior IS control (implied
in that is that it is control of perception)!
I to recognize that behavior CAN be
controlled… if you have a reference value for a perception of
some specific behavior.
The reason Bill classic work on this
subject was titled “Behavior: the Control of Perception” is that
what we see if behavior. We don’t see the reference value, we
don’t see the control loop, etc. we see the result of what that
control is doing. You ABSOLUTELY have to do the TEST to even
achieve an estimation of what perception is being controlled and
things like what the reference for that perception might be. All
of the results from the TEST are still not conclusive but at least
with well constructed TESTs we obtain results that are so far
beyond the results of any other theory to render comparison
ludicrous.
While meeting with Bill Powers
privately a number of times and having dinner with Bill and Mary
does not ‘make me an expert’ nor does it remove my human tendency
to error, I am quite certain that I do understand the most basic
and fundamental principles of PCT.
On 1/11/19 1:59 AM, "bBoris Hartman"
( via csgnet Mailing List) wrote:
From: Bill Leach (
via csgnet Mailing List)
Thursday, January 10, 2019 5:03 AM
Re: WIRED: The Genius Neuroscientist Who
Might Hold the Key to True AI
Hi Bill
Where didyou come from ? Middle Ages ? J
Youexploaded again in wrong direction. Behavior is generally
not controlled. Show us how, not just with wording and
phylosophing, but with evidences ? With cheep sentences
you’ll hardly prove that »Behavior is control«. I would say
that this is an insult for PCT. It seems that you think that
PCT is cheep psychological »Behavior is control« theory.
Well Iadvice you to start reading and studying PCT. Whatever you
are doing is just making more confussion.
You aredenying beasic principles of PCT. For you »Behavior can be
controlled« but that doesn’t mean that it is. You can’t
control muscle tension directly. Let us see some PCT »facts«
:
boris.hartman@masicom.net
wrleach@cableone.netmailto:csgnet@lists.illinois.edu
Sent:
**To:**csgnet@lists.illinois.edu
Subject:
Bill,
I already answered you in my previous post about your statement
BL : I to recognize that behavior CAN be controlled… if you have a reference value for a perception of some specific behavior.
For a discussion I’ll adapt it a little so to see what is the problem :
Behavior can be controlled if Perception of behavior is controlled (there is a reference for perception of some specific behavior)
If I put it right, statements are wrong from PCT perspective. And I already presented you with PCT means (diagram, definitions of control) in my previous post why is that so. But it seems that you don’t read what I wrote. You are not the only one. There are quite some members which are repeating themselves despite the evidences of PCT literature. So I also have to repeat all over again the same “PCT story”, like to little choldren. Absolute winner in this category is Rick to whom PCT evidences were presented arround 50 x. And he is still “whisling” his RCT story.
Bill I appreaciate your concern for PCT. I beleive that you were close to Bill and Mary. And I beleive that you understand PCT in fundamentals about perception being matched to references what is producing “error” signal and drive muscles to act. But it’s not controlled acts (behavior, output) that “Output function” is producing. Just effects to environment.
Bill P (B:CP) :
OUTPUT FUNCTION : The portion of a system that converts the magnitude or state of a signal inside the system into a corresponding set of effects on the immediate environment of the system
Bill P (LCS III):…the output function shown in it’s own box represents the means this system has for causing changes in it’s environment.
HB : in PCT the mantra is that “OUTPUT IS NOT CONTROLLED”. So generally speaking behavior in PCT is not controlled and can’t be controlled. And there of course can’t be some specific behavior that can be controlled if "Perceptin of behavior is controlled. PCT diagram is valid for all behaviors not for some of them. It’s always that “Perception of the behavior is controlled” not behavior itself.
From PCT diagram (LCS III) …
… we can see that Bill gave a name to “output” : actions or observed behavior. It’s general. That is how any behavior can be observed and analyzed. So we can not talk (I’m talking strictly about PCT diagram) selectivelly about behaviour (output) so that once “behavior is not controlled” and once it can be controlled, something like :
BL : I DID NOT SAY THAT BEHAVIOR IS CONTROLLED!
BL : I to recognize that behavior CAN be controlled… if you have a reference value for a perception of some specific behavior.
HB : In one breath you caused contradiction by my oppinon. These statements are not compatible and from the perspective of diagram LCS III only one is right : I DID NOT SAY THAT BEHAVIOR IS CONTROLLED! If you are saying that behavior can’t be controlled (or is not controlled), both statements in PCT describe “Output function” which is not producing controlled behavior or can’t produce contrrolled behavior. But then you can’t say in the same breath that some specific “Behaviors can be controlled” because “Output function” simply does not have that ability. Whether “Behavior is controlled or not”. It’s general.
Generally speaking from perspective of diagram LCS III, behavior can’t be controlled. It’s general mantra in PCT that “OUTPUT IS NOT CONTROLLED and can not be control”. So it can control anything in environment.
But as you said : Perception of behavior can be controlled. But this doesn’t mean that “Behavior is controlled” and that it causes controlled effects to immediate environment as Rick thinks.
Rick wrote in last years many nonsense, but I’ll expose just some of them :
All events in control loop happen at the same time. So when “Control of perception happens” also “controlled effects” to environment take effect in the same time.This is aproximatelly possible only with Telekinesis. Do you beleive that ?
To the extend that “Perception is controlled” also “controll of external environment happens” (cannonical principle)…
Behavior is control
There is some “aspect of environment” or “Controlled variable” that is controlled by controlled behavior"
There is some Controlled Perceptual variable" or CPV formed in receptors and afferent nerv
All statements are wrong from the perspective of PCT. They show wrongly outside part of the control loop om the basis of “Behavior is control”. Non of these things can we see in diagram LCS III and definitions of control loop (B:CP).
Because you don’t attend to CSGnet discussions frequently if I understood you right, I’ll repeat discussion between Rupert Young and Rick Marken some years ago :
RY earlier : Sure, a perceptual signal (q.i*g) may correspond to, or be a function of, variable aspects of the environment (q.i) but it is the perceptual signal that is controlled not the variable aspects of the environment
RM earlier : In my rush to show that this is not the case I came up with what has to be the dumbest rebuttal of all time – outdoing even myself in stupidity;-) I claimed that it was q.i*g rather than q.i that is the Input Quantity in the LiveBlock demo. This was simply, utterly wrong.
Rupert later even supported his briliant insight :
RY : I still have in mind the example of RDSs (random dot sterograms), where, it seems to me, no aspects of the environment vary while the perception is being controlled, so how can it be that “aspects of the environment” are being controlled?
So Bill I can’t agree with your statement :
BL : I to recognize that behavior CAN be controlled… if you have a reference value for a perception of some specific behavior.
As I said before. If I could show you some series of true life experiments (not Ricks imagined demos and tests), I’m quite sure you would understand immediatelly where is the problem. I’m also sure that there wouldn’t be any problems if we would sit behind the table and talk as you did with Bill and Mary. The only difference would probably be in the way I would argument PCT. Bill did sometimes change his mind what is by my oppinion logical when you construct new theory. And he “protected” his friends, so he did “look through fingers” to his friends. He speccially protected Rick. Whatever Rick wrote or said Bill supported him even he probably knew that Rick was wrong. I proved that several times. If you have time please make a tour through archives.
Martin and I noticed protective “mode” in “live” discussions with Bill and Rick.
Because Bill was also very gentle with Fred (I’m sure he deserved it, and I think Fred is with no doubt another person with whom I’d gladly sit and talk) Fred’s diagram does not fit into diagram LCS III. It’s totaly different story. Anyway I think Bill was wonderfull person, gentle to his familiy and friends, and genious scientist. But friendship and gentility could affect his objectivity. I hope you understand what I meant. And if we would sit and talk, we could have fine time (I’m sure) but I would not deviate from Bills’ diagram LCS III and his B:CP definitions of control loop, no matter of the kind of relation we would create.
And that’s my point. I’d like that we (all members including Powers ladies) would agree in main points of PCT which are represented with LCS III diagram and definitions of control loop whic I’m presenting for years, so to equalize our understanding of PCT. In this way myriad of phylosophical approaches (it seems that every member has own theory about how behavior function) would not affect originality of Bills theory. Rick is the first one who is counter productive in this matter despite the fact that it’s now more than five years since I’ve been warning him that his control loop (RCT) is just opposite to what Bill created as PCT.
Judge for yourself from LCS III diagram what is right and what is wrong about PCT. LCS III diagram is one of the latest in Bills literature so I assume it’s showing his determined knowledge of PCT.
Friends ?
Best regards,
Boris
From: Bill Leach (wrleach@cableone.net via csgnet Mailing List) csgnet@lists.illinois.edu
Sent: Saturday, January 12, 2019 10:00 AM
To: csgnet@lists.illinois.edu
Subject: Re: WIRED: The Genius Neuroscientist Who Might Hold the Key to True AI
Damn Boris! I DID NOT SAY THAT BEHAVIOR IS CONTROLLED! I said that behavior IS control (implied in that is that it is control of perception)!
I to recognize that behavior CAN be controlled… if you have a reference value for a perception of some specific behavior.
The reason Bill classic work on this subject was titled “Behavior: the Control of Perception” is that what we see if behavior. We don’t see the reference value, we don’t see the control loop, etc. we see the result of what that control is doing. You ABSOLUTELY have to do the TEST to even achieve an estimation of what perception is being controlled and things like what the reference for that perception might be. All of the results from the TEST are still not conclusive but at least with well constructed TESTs we obtain results that are so far beyond the results of any other theory to render comparison ludicrous.
While meeting with Bill Powers privately a number of times and having dinner with Bill and Mary does not ‘make me an expert’ nor does it remove my human tendency to error, I am quite certain that I do understand the most basic and fundamental principles of PCT.
On 1/11/19 1:59 AM, “bBoris Hartman” (boris.hartman@masicom.net via csgnet Mailing List) wrote:
From: Bill Leach (wrleach@cableone.net via csgnet Mailing List) [mailto:csgnet@lists.illinois.edu]
Sent: Thursday, January 10, 2019 5:03 AM
To: csgnet@lists.illinois.edu
Subject: Re: WIRED: The Genius Neuroscientist Who Might Hold the Key to True AI
Hi Bill
Where did you come from ? Middle Ages ? J
You exploaded again in wrong direction. Behavior is generally not controlled. Show us how, not just with wording and phylosophing, but with evidences ? With cheep sentences you’ll hardly prove that »Behavior is control«. I would say that this is an insult for PCT. It seems that you think that PCT is cheep psychological »Behavior is control« theory.
Well I advice you to start reading and studying PCT. Whatever you are doing is just making more confussion.
You are denying beasic principles of PCT. For you »Behavior can be controlled« but that doesn’t mean that it is. You can’t control muscle tension directly. Let us see some PCT »facts« :