···
Sent with ProtonMail Secure Email.
â?â?â?â?â?â?â? Original Message â?â?â?â?â?â?â?
On July 13, 2018 2:59 PM, Bruce Nevin bnhpct@gmail.com wrote:
[Bruce Nevin 2018-07-13_07:57:24 ET]
Joh Orengo 7.13.18.9:26 EEST –
So the rewarder is trying to control some desired variable by providing a reward to another so that that organism can to begin controlling, continue controlling, or better control the rewarder’s desired variable. This desired variable might relate in some way to the organism’s own goal. Even if it doesn’t relate to the organism’s own goal, it might eventually relate to the goal and, furthermore, may play a role in changing the goal.
The punisher is trying to control some desired variable by providing a punishment to another organism because the organism had a failure to control the rewarder’s desired variable and should attempt better control next time. Again, even if it doesn’t relate to the organism’s own goal, it might eventually relate to the goal and, furthermore, may play a role in changing the goal.
Is this an accurate summary?
As far as it goes, yes. The mechanism for this is of particular interest, within or adjunct to the PCT model. Part of a PCT explanation of learning.
These posts about rewards and punishment reminds me of an email exchange Kent and I had 8 months ago about the strategies of interpersonal control. (I hope you don’t mind if I post it here, Kent.) He wrote:
"2.0 The next main type of [interpersonal] interactions has not actually been given a PCT label yet, but for want of a better name Iâll call it Collateral Control: Occurs when two people in a shared environment control different perceptual variables, but the side-effects
of one personâs control of a perception help the other person to control a different perception. This âhelpâ? occurs because the first personâs control of a perception reduces the disturbances affecting the feedback path (the physical links between the personâs
output and sensory input) used by the second person in controlling his or her perception. With fewer disturbances to overcome, the second person can maintain tighter control of his perception and exert less physical effort.
2.1 Cooperative or Coordinated Collateral Control: Occurs when the interaction benefits both parties, so that the first personâs control of a perceptual variable helps the second person to control a different perceptual variable, and vice versa
in a mutually beneficial loop. Examples: Two people sing a duet (different melody lines but in the same key) or perform a partner dance,
where each personâs performance enables the other to perform his part. A
baby cries and the babyâs caretaker feeds it,
allowing the baby to control the perception of being fed and the caretaker to control the perception of not hearing the baby cry. Two people have a conversation, with each responding to messages conveyed by
the other. Martin Taylor describes collateral control
of this type as âprotocolsâ?, and he has written quite a lot on that topic in manuscripts that will eventually be included in the Living Control Systems IV volume. Commercial exchanges, where one person gives money to another in exchange for goods or services
also fit the definition of protocols and more broadly of coordinated collateral control.
Note: Many cooperative interactions involve cooperative collateral control (type 2.1) at one perceptual level and cooperative collective control (type 1.1) at a higher perceptual level. Examples: Two people sing different but coordinated melody
lines in order to control the higher-level perception of performing a duet together. Two people complete a project together but one person takes care of some of the tasks, while the other person takes care of the rest.
2.2 Oppositional or Exploitative Collateral Control: Occurs when an interaction between two people controlling different perceptions has side effects that benefit one party but not the other. This category includes coercion, deception, and manipulation.
2.2.1 Coercion: As I said in my previous email, coercion occurs when the perpetrator attempts to distract another person from controlling a perception in that personâs region A of active control by creating a large disturbance in the personâs
region B of background perceptions. If the coercion is successful, it benefits the perpetrator, because the victim had been causing some kind of disturbance to the perpetratorâs control of perceptions by what the victim was actively doing in his region A,
and the victimâs control of the perception formerly in region B does not disturb the perpetrator’s control of his own perception. The interaction is not beneficial to the victim, because it prevents him from controlling the perception he had been controlling,
and his control of the new perception may conflict with his control of other valued perceptions (like the safety of family members, for instance, if the perpetrator is threatening to harm to them).
Note: Coercion sometimes occurs when two people are engaged in conflictive collective control (type 1.2) of some perceptual variable, because in conflictive control each partyâs actions create a disturbance for the other personâs control of that
variable. By using coercion, the perpetrator can change the nature of the interaction, distracting the victim from attempts control the disputed variable and involving him instead in collateral control of whatever variable the perpetrator has threatened, thus
gaining an advantage over the other person in the dispute over collective control of the contested variable.
(2.2.2 Is ‘deception’ but I don’t think it’s really relevant to the discussion at hand.)
2.2.3 Manipulation: Occurs when one person creates disturbances intended
to focus a second personâs attention on controlling a perceptual variable that will have side effects which facilitate the first person’s
control of a different variable,
while the first personâs control of his own variable does not have reciprocal benefits for the second person. By this definition, one can see that manipulation clearly fits into the category of oppositional collateral control interactions. Examples: A parent
offers a candy bar to a child to get the child to stop crying. One person tells lies to another to evade punishment for a misdeed. An advisor flatters a politician to get him to accept a policy initiative that the advisor advocates."
So both rewards and punishments can be a type of ‘Cooperative or Coordinated Collateral Control’ or ‘Oppositional or Exploitative Collateral Control’ to use Kent’s terminology depending no the situation.
For example, if a parent rewards a child a lollipop so that the child can stop screaming then while it is a reward it would fall under the category of ‘manipulation’, a type of ‘Oppositional or Exploitative Collateral Control’.
If a parent punishes a teenager for not cleaning up his/her room while it might not be one of the teenagers goals it could still benefit the teen by teaching him/her responsibility and/or keeping him/her from contracting a flesh-eating disease due to poor hygiene. Thus, it could be seen as a type of ‘Cooperative or Coordinated Collateral Control’.
Perhaps, it’s in these ways that rewards and punishments fit into the PCT model.
Joh
On Fri, Jul 13, 2018 at 3:30 AM Joh Orengo joh.orengo@protonmail.com wrote:
[Joh Orengo 7.13.18.9:26 EEST]
Sent with ProtonMail Secure Email.
â?â?â?â?â?â?â? Original Message â?â?â?â?â?â?â?
On July 13, 2018 4:06 AM, Bruce Nevin csgnet@lists.illinois.edu wrote:
[Bruce Nevin 2018-07-12_20:39:46 ET]
Joh Orengo 7.12.18.10:34 EEST –
How do you think this could relate to rewards?
Reward and punishment are also loaded notions in PCT discourse.
We recognize that rewards and punishments do not control behavior (stimuli do not cause responses), but it is a reasonable inference from results of S-R experiments that rewards and punishments are influential on reference values for control and that they can influence structural consequences of learning and of reorganization such as the inputs to perceptual input functions, the relative contributions of those inputs, and other interconnections of control loops, and, in a very much related way, they can influence associative memory. I have no demonstration of any of that, but I suspect that much of the necessary work has actually been done but has not been properly understood, lost in the miasmas of S-R and S-[cognitive process]-R rationalizations.
This makes sense to me.
I suggested that learning, understood as the refinement of input functions (‘recognizers’) and their reference inputs, etc., can also be talked about as the refinement of predictions. It appears that changes to the perceptual input functions, reference input functions, and the distribution of error output to diverse reference inputs can be influenced by success or failure of control of other variables that are linked by some environmental contingency. From the point of view of the subject organism, reward is control of a desired variable that is contingent upon perceiving and controlling in some way that the rewarder is controlling by means of the reward, and a punishment is conversely failure of control in a corresponding way.
So the rewarder is trying to control some desired variable by providing a reward to another so that that organism can to begin controlling, continue controlling, or better control the rewarder’s desired variable. This desired variable might relate in some way to the organism’s own goal. Even if it doesn’t relate to the organism’s own goal, it might eventually relate to the goal and, furthermore, may play a role in changing the goal.
The punisher is trying to control some desired variable by providing a punishment to another organism because the organism had a failure to control the rewarder’s desired variable and should attempt better control next time. Again, even if it doesn’t relate to the organism’s own goal, it might eventually relate to the goal and, furthermore, may play a role in changing the goal.
Is this an accurate summary?
Joh
Stories like *The Good Soldier vejk *and Catch-22 illustrate some aspects of the unreliability of rewards and punishments, as likewise Sgt. Shaftoe in Crytonomicon and his ancestor ‘Half-cocked’ Jack Shaftoe in the preceding books of The Baroque Cycle. But there are countless other examples. Pre-literate peoples have recognized ironies of would-be ‘prediction and control of behavior’ in their oral literature. But though rewards and punishments are not causative, they are influential, no doubt. Ask anyone who has been involved with a cult, an authoritative and punitive family, the military, an abusive relationship, etc.
/Bruce
On Thu, Jul 12, 2018 at 3:40 PM Joh Orengo joh.orengo@protonmail.com wrote:
[Joh Orengo 7.12.18.10:34 EEST]
Sent with ProtonMail Secure Email.
â?â?â?â?â?â?â? Original Message â?â?â?â?â?â?â?
On July 12, 2018 8:46 PM, Bruce Nevin csgnet@lists.illinois.edu wrote:
[Bruce Nevin 2018-07-12_09:59:50 ET]
Bart, I hope it is OK that I reply to the list. And I am glad that you’re aiming to be at the conference in Evanston. I’m looking forward to seeing you again!
Bart Madden Jul 12, 2018, 9:19 AM –
Bart: Here is an excellent summary of a trend in neuroscience to treat the brain as a prediction machine that minimizes error.
https://www.quantamagazine.org/to-make-sense-of-the-present-brains-may-predict-the-future-20180710/
Screams out PCT.
Is anyone in PCT working with this stuff?
‘Prediction’ has become a loaded term in PCT discourse, apt to be a trigger for dismissal without further examination, but I agree with you, this looks worthy of closer consideration. A PCT theory of learning is still in a rudimentary state–a baby, if you will, and throwing out the bathwater of behaviorism has slowed its development while simultaneously raising hurdles for being perceived as properly understanding PCT if you come to it from that background.
“Every want is a prophesy.” Instead of construing a desire (a reference perception with poorly controlled input) as a specification of lack, it can be understood as a prediction of what will be experienced when control is good. Learning as the refinement of input functions (‘recognizers’) and their reference inputs can also be talked about as the refinement of predictions. Non-PCT jargon, but on first glance it does sound quite well aligned with PCT.
“Every want is a prophesy” is a useful reversal of what may be habitual ways of thinking about wants and lacks. Construing a desire (a reference signal in a comparator with poorly controlled perceptual input) as a specification of lack is self-defeating if that perception of lack is a stable part of the context for controlling other perceptions. This is an important practical aspect of ‘the power of positive thinking’. The converse is a basis of ‘self-fulfilling prophesy’.
Excellent post, Bruce! How do you think this could relate to rewards?
Joh
In the same way, an expectation that no one understands PCT correctly, taken as a stable part of the perceived environment, can become a part of the environmental feedback function for controlling other perceptions, and its destabilization may actually be resisted. Looking back over the past almost 30 years I am wondering if this might sometimes have been an instance of collective control. I have no doubt that it has been sometimes an instance of individual control, because I have seen it in myself. Perceptions controlled by means of this perceived environmental stability (“others do not understand PCT correctly”) include perceiving myself (and perceiving myself being perceived) as part of the in-group rather than those “others”, and probably yes some of the ‘invigorating’ body states associated with conflict, so you are certainly not alone in that, Rick.
Warren has some remarkable skills in this area, which I admire and which I want to observe more closely in the hope of developing more refined and accurate perceptual input functions (‘recognizers’) for them, so that maybe I might improve my control of like perceptions. PCT learning theory applied–what a concept!
/Bruce
On Thu, Jul 12, 2018 at 9:19 AM Bartley Madden bartjm43@gmail.com wrote:
I’ll try to make the Evanston conference.
Here is an excellent summary of a trend in neuroscience to treat the brain as a prediction machine that minimizes error.
https://www.quantamagazine.org/to-make-sense-of-the-present-brains-may-predict-the-future-20180710/
Screams out PCT.
Is anyone in PCT working with this stuff??
Bart
On Fri, Apr 27, 2018 at 8:53 PM, Bruce Nevin bnhpct@gmail.com wrote:
Ah, yes, I used the address that gmail pulled out of its memory, and I should have copied the one from Dag’s fwd.
The conference is earlier in October, the 12th and 13th. Hope you can make it. Good luck with your worthy legislative agenda.
/Bruce
On Fri, Apr 27, 2018 at 8:33 PM, Dag Forssell dag@livingcontrolsystems.com wrote:
Bart,
I forwarded your email but not your gmail address. Bruce Nevin must have
had an old yahoo address. Interestingly enough, you replied using the
yahoo address.
With this mail, Bruce gets your gmail address.
I am glad I referred you to Bruce. I wish you could benefit from my
presentation. It is carefully structured to explain to a lay audience –
anyone – and get the significance of PCT across.
At 04:43 PM 4/27/2018, Bart Madden wrote:
Bruce
This is helpful. Thank you.
I am not sure if I can make the PCT conference.
I am committed to a major campaign now to pass my Free To Choose Medicine
proposal in the U.S.
I know that I’ll be going to make presentations to large groups in late
Oct and Nov … don’t know exact
dates yet.
FTCM has some “control” aspects to it …
An early version of these ideas was translated into Japanese and
apparently played an important role
in Japan’s 2014 passage of FTCM-type legislation for regenerative
medicine drugs. This and other new
developments are reported in my new 3rd edition FTCM book
https://www.amazon.com/Free-Choose-Medicine-Better-Sooner/dp/1934791679/ref=sr_1_1?ie=UTF8&qid=1524872193&sr=8-1&keywords=free+to+choose+medicine
Finally, please use my gmail address noted above.
Bart
On Fri, 4/27/18, Bruce Nevin bnhpct@gmail.com wrote:
Subject: short non-technical summary
To: “Bart Madden” bartmadden@yahoo.com
Cc: “Dag Forssell”
dag@livingcontrolsystems.com
Date: Friday, April 27, 2018, 6:05 PM
Hi, Bart,
As Dag has told you, he passed your email on to
me. You say you need a one-sentence non-technical statement
about PCT. Your first cut:
Human behavior
is best understood, not as responses to stimuli, but as
taking actions to control the perceptions of variables that
are important in keeping us on track to achieve our
goals.Ã
The best way to say it is the best way for your
reader to understand it, and that depends on who you’re
talking to. This talks to someone who presupposes that
behavior is responses to stimuli. But for your intended
“nontechnical” audience that might not be so. The
typical response to me is “well, isn’t that kind of
obvious?”
Your sentence above has the technical words
responses, stimuli, perceptions, variables, and (though the
reader doesn’t know it yet) control.
Where does the limit to one sentence come from?
Whether one sentence or six, “nontechnical” calls
for simple, direct, familiar language, and trying to
shrink-wrap too much in one bundle makes that harder to do.
Here are five sentences, for
example:Ã To state
the blindingly obvious, when something isn’t the way we
want it to be, we act so as to fix that. Can you think of
any human purposes that don’t follow that general rule?
But all too often we get at cross purposes. We can even get
at cross purposes with ourselves. With PCT, learn how to see
the purposes behind behavior and resolve
conflicts.
I’m not proposing that you use this. You need
your own words that say what your book is about in a direct,
non-technical way.
I hope this is helpful, Bart. Are you planning on
coming to the conference at Northwestern in
October?
/Bruce