Is it modeled in the new, improved version of the collective control model? If not, then that statement is irrelevant to the discussion since you have been claiming that the modeling of collective control has gone beyond what was done by Tom Bourbon and Bill Powers. I think the collective control modeling gone backwards since Tom and Bill’s work, into data-free blathering.
No, my model demonstrated that when a population of people who pronounce things the same way (on average) are divided into non-interacting subgroups, each subgroup stabilizes at a different average pronunciation.
Does your new, improved model of collective control explain the topic of Labov’s paper, accounting for the data the Labov collected?
Looks like Labov already explained whatever it was he wanted explained. What he didn’t explain is what my model explained.
Do they have a place in yours? And, if so, could you should me how your model works and the data it accounts for.
No way. You’re the linguist. And besides, as I noted above, you don’t seem to know what my model was about. So it’s all yours. I’d really like to see how your new, modern collective control model accounts for some actual DATA!
It appears to me that for you this conversation is about defending true PCT from ideas that don’t qualify as PCT.
For me, this conversation is a continuation of a decades-long request for help understanding how to model control of higher-level ‘abstract’ perceptions for which qi, qo, and d are not easy to isolate as quantities.
I’m sorry that you are unable to be helpful. Others will have to work on these methodological problems by trial and error without your help.
If you are grumpy about it that will be trying in a different way, but if things you are grumpy about are not things we are doing it might cease to be a disturbance.
For me, this conversation was about getting examples of real would social situation that are accounted for by Kent’s collective control model, where N agents act as a giant virtual controller keeping some variable in a virtual reference state. I never got a clear answer on this – no examples of the application of that model to any real world situation other than tug of wars – and I’m sure I never will.
Well that’s a complete surprise to me.
I’m sorry you’ve found my work on control of higher order variables unhelpful. I’ve tried my best, with papers such as the “Hierarchical Behavior of Perception” paper in MORE MIND READINGS, suggestions for research on higher level perceptions in Chapter 8 of The Study of Living Control Systems and my on-line demonstrations of control of higher level perception, here and here. You didn’t seem to find any of this particularly useful so I guess you’ll just have to go to other sources for you information about control of higher level perceptions.
Unless the law and the customary obedience to that law have changed, in California when you step off the curb of a street an oncoming car stops while you cross.
What are the controlled perceptual variables and how do you quantify and model the control of them by the pedestrian, the driver, and onlookers?
Good luck trying that in New York or Boston. What is different and how would you model the difference?
I’m sorry, I don’t see how this relates to Kent’s model. Who are the agents simultaneously controlling the same variable relative to different references so that that variable is maintained in a virtual reference state? What is the virtual reference state of that variable?
Oh, I thought you were giving me an example of a real world social situation that is accounted for by Kent’s collective control model. But in answer to your question I bet you could come up with a reasonable simulation of stopping for people crossing the street using a variant of Bill’s CROWD program, where the agents would be controlling for things like stopping a fixed distance from agents crossing in front (if they are drivers) and getting to the other side of the street (if they are pedestrians).
I’m starting with phenomena. Phenomena first, then data. I thought the first-order problem is how to quantify such phenomena so that they are characterized by numerical data. But you do not perceive the phenomena.
I therefore thought that I first had to draw your attention to everyday examples of such phenomena. I described how certain physical aspects of our environment are in the feedback paths through which a large number of people control more abstract, higher-level perceptions. Often the higher-level abstract perceptions that one individual controls differ from those that any other individual controls, but for each of them successful control depends in some way upon also perceiving that aspect of the environment in the environmental feedback path.
However, you do not perceive the higher levels of perceptual control in these examples, only the lower perceptual levels of the control loops. Since you don’t perceive the phenomena, there is no point asking you for help figuring out how to quantify and model them.
I am now abandoning any wish for help from you understanding how to model control of higher-level ‘abstract’ perceptions for which qi, qo, and d are not easy to isolate as quantities. I believe you when you say that you are convinced that you have provided such help in the places that you have named. I believe you are convinced that you are talking about the same phenomena. But all of that is irrelevant. Your inability to perceive what we are talking about is no longer a disturbance to any perception that I am aware of controlling.
I’m sorry that you are unable to be helpful, but I accept that this is the way it is. Others will work out these methodological problems by trial and error without your help. This topic is not about a test of collective control at any but the lowest levels of the perceptual hierarchy. My participation in it is at an end.
An example of a collective control phenomenon is the price of a stock publicly traded. That it is collectively controlled (depends on the control actions of many people) seems hard to deny, or rather, I cannot find a way to interpret it in any other way.
Yes. The market value of a stock is in the environmental feedback function of many control loops, at the intersection of a huge number of ‘gossamer threads’.
What questions should be asked about this? What perceptual variables are controlled, and by what controllers?
For example, what are the different types of traders (professional on behalf of institutions, professional on behalf of retail clientele, day-traders, retail, influencers controlling to affect opinion that is relevant to the given stock, …). How much is each able to invest?
What are conditions affecting ROI, who perceives these conditions and who doesn’t, etc.
For each trader, the usual questions: What is my purpose or goal in investing (retirement, buy a house, financial autonomy, …)? What is my risk tolerance? Can I afford to lose it? What if it does go to zero? What is my investment time frame? Will I have to pull it out and spend it, and how soon? What are my criteria for selling this investment? Am I diversified?
Yes, I agree. The stock market is an example of a kind of collective control, and it is the kind of collective control performed by Kent’s model. Indeed, I think any market is an example of Kent’s kind of collective control, where there are many agents controlling the same variable – the current price of a good or service – relative to different references – agents acting as buyers would typically have a lower reference for the price of a good or service than sellers. The current price at which the good or service is selling is the current virtual reference state of that variable. Of course, in stock markets this virtual reference state is not stable but varies continuously, though the secular trend is generally upward.
I think Kent’s model applies to behavior in markets because a market is much like a tug of war, with buyers and sellers being equivalent to the two “teams” in a tug of war and the price of goods or services being the equivalent of the position of the flag. Of course, in order to properly model a market one has to take into account the resources and needs of the agents – buyers and sellers – as well as the rules and regulations of the market (such as “counterfeiting not allowed”), which are aspects of the feedback connection between the agents and the controlled variable.