[Eetu Pikkarainen 2019-04-11_06:51:45 UTC]
···
From: “Boris Hartman” csgnet@lists.illinois.edu
Sent: Wednesday, April 10, 2019 6:04 PM
[Eetu Pikkarainen 2019-04-10_13:17:31 UTC]
EP : I think that Rick did not claim that CV is in external environment.
HB : He does not claim now. But go and look in CSGnet archives. I know exactly what Rick is aiming at.
EP: Then he must have changed his mind?
[…snipped…]
EP : What I suggested was that there is (or can be) something in the real reality (not necessarily outside the controller, it can be also in her like in the
case of hunger or itch), which is perceived when the controller has a perception which she tries to control. And I also suggested that that something could be called RREV or just “object of perception”.
HB :And in Ricks case is "something in the real reality, which is perceived as CV or distance between “cursor and target” ? Is this what you wanted to say
? I also don’t understand what could be “object of perception” ?
EP : This RREV, or whatever you want to call it, is what is affected by the controller’s output if the control is successful.
HB : I’m sorry I don’t understand this either. “RREV” exist in outer environment only if effects of controllers output are successful ?
HB : Well it’s everything to abstract. Could you give some real life example how this works, beside Ricks “chewing” laboratory experiment" which has no use
in explaining other behaviors.
EP: I know this is very risky but still I want to try to invent a simple example of how I think the things go (unfortunately simple examples are seldom simple). I hope I can learn something
myself and also that I can at the same time reply Rick, too:
Say I have found a dog and I suspect that it can be hungry. So I offer some food to it. It starts to bite the food but I am still not sure so I try take the food away. Then the doc protests
by growling and threatening to bite my hand.
Now I infer: 1) The dog controls for a perception of having the food, because it resisted the disturbance I caused by trying to take the food away. 2) The dog is hungry which means that
it controls for the perception of not hungriness i.e. fullness.
In the step one, I can perceive that the dog has the food and I assume that also the dog perceives that it has the food. In this sense it can be said like Rick that I and the dog have
the same perception, the perception of this dog having this food. from my crude TCV I inferred that the dog was controlling
this perception. So it is possible to say that the dog is controlling the same perception which also I have. (Even though it sounds crazy to say that the dog controls my perception.) Of course this is actually very metaphorical way to talk because the
dog and I have our own perceptions and we have them probably in quite a different ways. But however it is understandable to say so. More accurate is to say that the dog seems (empirically) to control a perception which in some way corresponds to my perception
of “the dog having the food”.
The dog’s perception and my perception are similar in a same way as the number three is similar to the alphabet c: They have a similar position in their respective systems. I select
from my perceptions that one which I think that best fits with that perception which I assume the dog has. And I can check that fit with some kind of TCV operations.
The step two is a little different. I infer that the dog perceives itself (feels) hungry and I think that it is something similar than the perception which I have when I feel hungry,
but I do not feel hungry at the moment. So I cannot say that I have a same perception with the dog, but I have in my memory a perception which I assume that corresponds to the perception which the dog has at the moment. But also here I can (metaphorically)
say that these two perception are the same perception because I assume and infer and check that the hunger perception which I have is the best fit with the perception which I have inferred that the dog has.
There are two background premises which are required for those previous inferences to be rational. One is that I and the dog have sufficiently similar sensory functions. If they are
too different it is harder and harder to trust that my perceptions have any correspondence with the perceptions of that other creature. The second premise is still much more important because it is in a way also a premise for the first premise. This second
premise is about RREVs (Real Reality Environmental Variables) or the objects of the perceptions. According to it there must be something (we do not know what) in the real reality (either outside or inside the perceiving and controlling organisms) which can
be perceived as food or hunger or any perception at all and which can be affected by our actions (output) in such a consistent and predictable manner that the successful control is possible.
Think that there were two somewhat similar RREVs which both we perceive as food: they look like a food, they taste like food, feel and smell like a food. But still they are different
so that if you eat the first one your sugar level of the blood rises and you feel full and nice, but if you eat the other one you instead get sick or you can even die. In the history of evolution this kind of situations may have happened often: Others have
perceived wrongly and they have suffered and died. Others have had slightly different perceptual functions and have been able to make the difference beforehand and they stayed alive and procreated. In this way the RREVs have constrained our perceptual functions
depending on the ways we as species have had to make our living. That is why we humans can have so similar perceptual functions that it is reasonable to say in sufficiently simple cases that we have “same” perceptions. But if there are different species or
complicated situations with hierarchically higher and learned perceptions then it is much better to talk about similar perceptions than
same perceptions – and even that similarity may be often susceptible. But even in those cases I can assume that possibly the other subject perceives the same or similar RREV that I perceive in my perception. I can test this assumption by trying to control
my perception to somewhat different value (by affecting something real in the common environment of me and the other). If I perceive that the other one then affects the reality so that it cancels my preceding effect and returns my perception back to its initial
value then I can infer that whatever is the perception of the other one it is still about the same RREV than my perception.
This same applies also to internal perceptions like hunger. We have a perceptual function which make it possible to us to perceive whether we are hungry or not. But sometimes we can
perceive wrongly. For example I often feel hungry if I am tired. If I am really hungry (there is in me a RREV which is the right object of hunger) then I can control that perception by eating some food (if that happens to be a “right food”), but if I am tired
then eating doesn’t help but instead I have to take some rest or sleep.
As I said earlier, we don’t know what these RREVs are but we have models of them in our perceptions, just like sciences have models of them in their theories. The duty of a model is
to correspond its object, that which it is a model of. The better model correspond better and worse correspond worse. This goodness of models can be often called truth. There are (at least) two ways how the goodness of a model can be evaluated. The first is
control: the better model enables better control. The second is comparing the models: a model which cannot be controlled may be good if it suits well with models which can be controlled. I believe that the physics (so called fundamental physics) has a special
duty to try to find out the basic particles and building blocks of our reality, but we may be quite far from finding them at the moment. But also the models of other sciences, like biology, psychology, economy, geology etc. can as well have (more or less)
true models, which are hardly replaceable by the models of physics in practice. Especially I think that PCT should have a similarly basic position among sciences of alive sphere that the theory of relativity or quantum mechanics have in the inorganic spheres.
Eetu
- Please, regard all my statements as questions,
no matter how they are formulated.
Eetu
Boris
Eetu
From: “Boris Hartman” csgnet@lists.illinois.edu
Sent: keskiviikko 10. huhtikuuta 2019 16.10
Sorry Eetu that I jumped in…
There is no “CV” in external environment as you can’t explain too many behaviors with such an approcah. But you can cause deeper misunderstandng of PCT and you
can cause that PCT will has less and less value showing where Bill was contradicting himself. Anyway Rick, I think that you give a dame about PCT. But you are taking care of your ass and nonsesne RCT.
From: Richard Marken (rsmarken@gmail.com via csgnet Mailing List) csgnet@lists.illinois.edu
Sent: Tuesday, April 9, 2019 7:33 PM
To: csgnet csgnet@lists.illinois.edu
Subject: RE: goal of our researchgate project
[Rick Marken 2019-04-09_10:25:07]
[Eetu Pikkarainen 2019-04-09_05:34:30 UTC]
EP: Hmm, just a thought: The control (any control of perception, except temporarily in imagination) is constrained by “Real Reality” (RR). Powers says in the quotation: “When we apply a disturbance, we apply it to CV, not to p.” This applies also to
observer: she has a perception p of the CV. When she applies a disturbance (because of TCV), she applies it to CV not to her p.
RM: I think the problem is in thinking of a CV as something
to be perceived. In fact, the CV is a perception for both the observer and the controller. It is an
actual perception for the observer; the distance between cursor and target, for example, is a perception for me as an observer; it is also a perception
in theory for the controller. The theoretical nature of the controller’s perception is indicated by the fact that the CV is represented as the theoretical variable p in the model of the controller.
HB : Observer will perceive whatever is outside in his way and controller in his way. What observer will perceive and control is question for TCV.
But I’m sure that most people will not perceive “CV”. And I’m sure you wouldn’t be one of them.
EP: Thus that which is called here CV is outside of both the controller and the observer.
RM: The CV actually exists only as a perception inside the observer (in fact) and in the controller (in theory). The CV certainly can “look like” it is “outside” of the controller and observer,
just as the distance between cursor and target appears to be “outside” both. But according to PCT the CV is inside the brains of both the observer and controller.
HB : How would you know that “CV” is inside observer ? In theory ? Or in practice.
EP: Where is it then? I think it is in RR.
RM: According to the PCT model, the CV is a perception (in both the observer and controller) inasmuch as it is a FUNCTION of variables in real reality (RR).
HB : According to PCT “CV” is not a perception. You are lying Rick.
There is no “Controlled Variable” in external environment and no “Controlled Perceptual Variable” in observer according to PCT.
EP: What the observer infers and claims to be controlled is thus the object of her own perception, something in RR which she assumes (because of the empirical
findings) to be also the object of the perception of the controller.
RM: The observer doing the TCV does not have to infer anything about the “object of her own perception” that the controller is controlling.
HB : So observer doing TCV is some expert for TCV ?
RM : The observer doing the TCV simply observes that something she perceives – such as the distance between cursor and target – is being controlled in the sense that it is being protected from
disturbance by the actions of the controller.
HB : You are breaking World Record in talking nonsense. And how controller is “protecting distance between cursor and target”.
With Telekinesis ? Or with yome “magnetic field” ? Or maybe he is protecting socket that somebody wouldn’t pull out electrical cable and cause the end of “experiment”. I must say that I agree with
you.
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;-)
RM : The observer may think of that perception – the CV-- as an objective variable in the outside world; the distance between cursor and target, for example, certainly seems like it is “out there”.
But the variable that appears to be “out there” is actually the result of a perceptual computation – a FUNCTION of sensory input – that produces that apparent reality. For example, the distance between cursor and target is the result of a perceptual computation
that involves taking the difference between two sensory inputs – one from the cursor and the other from the target.
HB : Sorry Rick PCT is not working in this way. How do you imagine that “two sensory inputs” look like. X and Y axis like in your Toy Helicopter
experiment with Schaffer ? Target through left and cursor through right eye ???
RM : So the FUNCTION that results in the appearance of the CV as the distance between cursor and target is a subtraction.
HB : So you perceived subtraction ?
EP: Thus, I think that RREV (or simply just “object of perception”) is a useful and necessary concept because without it one must say something like this:
“The controller is controlling the same perception which the observer has”, which literally means that there is one perception which is common to and shared between controller and observer.
RM: I consider that a feature, not a bug. When an observer has successfully tested to determine the variable the controller is controlling, the observer can be said, for all intents and purposes,
to be perceiving what the controller is perceiving – the CV.
HB : Did you try this with anybody or you are just imagining and dreaming ?
EP: Yet we know that those two subjects certainly have both their own perceptions – they have their own perceptual signals and those signals are not necessary
similar.
RM: If the perceptual signal that corresponds to the variable controlled by the controller is not the same as the perceptual signal that corresponds to the variable that the observer thinks is
being controlled, the observer will realize this immediately (because the controller will not be systematically resisting disturbances applied to this variable). In this case, the observer will change her hypothesis about the variable that is actually being
controlled. The observer will continue to change hypotheses about the CV until she hits on one that is protected from all disturbances.
HB : Are there also bullit disturbances ? Describe to us hypothesis about CV being “protected” from all disturbances ??? Which are all these disturbances
that you should protect “distance” from ? He,he. What an imagination ?
RM : In that case, she has discovered a variable the controller is controlling – the CV – and, for all intents and purposes, is, perceiving what the controller is perceiving.
HB : So controlled and observer are perceiving the same “CV” ?
EP: If the bat perceives a fly using its ultra sounds and sensible ears it must be very different perception compared the visual perception of the bat researcher
who sees those ultrasounds from a measuring device (or the fly visually). What is common to the perceptions of the bat and the researcher is not the perceptions as such but the object of these perceptions, some RREV, even though this RREV is described in the
research report by using the perceptions of the researcher (and her assumptions of the possible perceptions of the readers).
RM: The observer doesn’t have to experience the CV in the same way as the controller does in order to successfully determine what perception the controller is controlling.
HB : So now you are saying that observer and controller do not perceive CV in the same way ?
RM : This can be illustrated with the compensatory tracking task. The observer of the behavior in this task doesn’t need to actually see the difference between cursor and target on the screen
as the controller does that task.
HB : So observer has covered eyes with yomething so that he doesn’t perceive the same “CV” as controller ??? He,he. What an imagination.
RM : The observer can tell from a plot of the difference between the position of the cursor are target varying over time (along with a plot of the disturbance and the controller’s output) that
the perception (cursor - target) is the CV.
HB : Controller must be a good “teacher” to explain to observer what she/he is perceiving ?
Boris
Be
Rick
Eetu
-
Please, regard all my statements as questions,
no matter how they are formulated.
–
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
