Transdisciplinary transitions and apprenticeships

[guitarplayer]

This thread brings to my mind the idea of co-figurative culture as spelled out by Margaret Mead in Culture and Commitment. The first time I was digesting that concept, I would imagine a group of people on a voyage to discover new land, then settling there and setting all up. Co-figurative is stacked against pre-figurative (roughly traditional) and post-figurative (rapid environmental change, offspring more well adapted than elders). It's a short book and a pleasant read if you've not read it.

[Walwen]

Right now I have three positions at my workplace: I'm a housekeeper, a maintenance man, and a cook. Everyone assumes I must be miserably overworked but I actually immensely enjoy it. I am learning so many different things, I know so many different people, and I'm kinda 3xing the amount of job experience I get. I mean, I could be 22 with two years experience as a cook, two years experience cleaning, and two years experience in maintenance. Also, talk about job security.

I just want to be a well-rounded person with a lot of skills, and I feel like my current job(s) is great for that. I don't think I could rise up the ranks like this, I don't think I could be both a kitchen manager and still a low-level housekeeper, but right now it's just like I'm a do-it-all kind of guy, and as a 20yo I'm happy to have a whole bunch of mentors from the different departments.

[karff]

Causal Breakdown: A method of transdisciplinarianism.

We are quite bad at causal thinking, but believe ourselves to be good at it.
See-

https://en.wikipedia.org/wiki/Moravec%27s_paradox

Similarly, Minsky emphasized that the most difficult human skills to reverse engineer are those that are below the level of conscious awareness. "In general, we're least aware of what our minds do best", he wrote, and added "we're more aware of simple processes that don't work well than of complex ones that work flawlessly"

Causal reasoning is done consciously, indicating we do it infrequently and aren’t very good at it, yet providing the illusion the opposite is true.

Most of our causal mental models are not the result of our own causal reasoning, but have been taught to us by others. When we use them, we are under the illusion we are doing original causal thinking, but we aren’t.
Causal chains tend to be stored in our minds as simple correlated temporal sequences, but we’re under the illusion we’re thinking causally when we use them.
Definitions of things are usually determined by some causal relationship, but we tend to store the definition without that relationship, and many times the definition came from another mind without the causal information.
We have a tendency to subconsciously conserve cognitive steps, consolidating the middle of causal chains into fewer links, and cutting the ends off. As this is done subconsciously, we do not realize all the causal information we are missing.
Some of us believe we have seamless 4 dimensional causal models of the universe in our heads. I was under that illusion myself until I began realizing how bad we are at causality.
Many of the components of our causal models are stored as spatial correlations, giving the illusion of an interconnected model, but in reality, it is fragmented and fractured, full of gaps and incongruencies we are unaware of.

For transdisciplinarianism

Break down a model into causal relationships, simple models forming a few chains, more elaborate models forming more complex webs. Look for any gaps, areas of low causal detail, inconsistencies. Especially look for loose ends, prior causes and further effects (the limits of your causal understanding). Also, links that can be broken down into further detail.
For many models, it actually takes very little breaking down to reveal insights. Because, even though we believe we already think this way, we don’t.
Areas of the model with gaps and low detail reveal, quite apparently, the kind of model that will fill in the gap.

Suppose you are looking to increase the efficiency of a household. Just a little causal breakdown of a household reveals it to be an economic entity with inputs and outputs. The discipline with the most detailed models for this is apparent, business. Business has quite detailed models of managing capital for efficiency.
Now having a model of household capital management, you can look at the loose ends and see how to understand where the input comes from, and the output goes. Economics, ecology, sociology.
Suppose you want to market this household capital management model. Look at various personality types, lifestyles and ideologies, and break those down into purely causal models. You can then see where they overlap with the system, or how a system of capital management could increase the particular output they are interested in.
Again, you don’t have to break the models down very much before starting to gain insight, as few are thinking this way, even though everyone thinks they are.

It works because causality is the common language of reality. Any model describing reality can be broken down this way, and compared with any other. Insights are immediate.
Again, we think we are already thinking this way, but we are not.

Categories of causal thinkers

Poor Learners learn causal relationships poorly, even when taught. They learn on the level of executing specific tasks with little causal understanding as to why. “Tips and tricks”, if you will.

Good Learners can learn and understand causal relationships if they are taught. They can then execute causal functions while understanding the relationship.

Autodidacts can learn new causal relationships if it has been discovered before, but no one is teaching them. If they start at point A, and know someone else has got to point C, they will try to figure out B.

Original Thinkers will search for causal relationships even if no one has discovered them before, and there isn’t necessarily anything productive in the direction of inquiry.

Poor Learners are like technicians, executing specific tasks without necessarily knowing the causal relationship of them.

Good Learners are like engineers, executing causal relationships someone else has discovered and taught them.

Autodidacts are like good physics teachers, understanding how previously discovered causal relationships can be taught and learned.

Original Thinkers are like productive theoretical physicists, searching for causal relationships where there may not be any, where there is no path.


Poor Learners are like cooks who only follow recipes, not understanding the causal relationships of the techniques used.

Good Learners are like trained chefs, understanding the causal relationships of techniques as they were taught in culinary school.

Autodidacts are like self-taught chefs/cooks, figuring out and refining known techniques without being taught.

Original Thinkers are like the inventors of entirely new techniques and cuisines, like molecular gastronomy.

Causal Breakdown moves each group up one level.

Poor Learners become Good Learners, as the causal relationships can be broken into specific tasks, the Poor Learner can absorb it through repetition, becoming a Good Learner.

Good Learners become Autodidacts. The causal relationships of the causal breakdown method can be taught to the Good Learner, allowing them to execute the causal methods of the Autodidact.

Autodidacts become Original Thinkers, as breaking down models into the same causal language and linking them together provides the Autodidact with a gap of causality to cross. They can see points A and B, and then cross the gap, even if no one has done it before, becoming an Original Thinker.

Original Thinkers become Unified Thinkers.All their mental models are connected with the same causal language, allowing all the gaps and limits of their knowledge to be self-perceived. Unknown unknowns become known unknowns, with all the many jumping off points for discovery apparent. Communities of Unified Thinkers much more easily compensate for each other's gaps in understanding. For them, transdisciplinarianism would be the norm, not the exception.

[karff]

A demonstration.

Let’s look at the model of the k3-k4 transition.
K3s apparently want to conform.
What kind of model is that? A motivational model.
Motivational models can be expressed as vectors, having more causal detail than just “to conform”.
Motivational vectors generally have both a carrot and a stick.What would be such carrots and sticks for a conformity motivational model?
Here we can create a causal gap and try to bridge it. On one side are observations about people who conform, and on the other is a blank motivational model with a carrot and a stick.
Conformist people try to avoid ostracism.
Ostracism can be a stick in our motivational model.
Note: we could find a more detailed causal model of ostracism and fill in more causal detail, but let’s find the carrot motivation next.
Conformist people try to gain high esteem, or status in their group.
So, now we have a more detailed causal model of k3s attempting to avoid ostracism, and moving toward higher status, of wanting to move upwards within their social group.
What kind of behavior would that cause?
Here we can create another causal gap with observations of conformists.
Conformists are interested in superficial surface details. Why?
Connecting this back to our ostracism - status vector, they are likely paying attention to the details that would get them ostracized, and also those that afford high status.
A model of status is one of hierarchy, with multiple levels between ostracism and highest status.
To move upward, the k3 would need to pay close attention to identifying details of those groups, to avoid being those below, and to be those above.
To the k3, those details are not superficial, but significant, as they determine social position.
From observation, k3s believe themselves to be different, using the surface details as differentiation.
So, k3s are interested in the differences in people. They do not seem to look for the reasons why people are different. They are not thinking causally about differences. Is this just with people, or do they not think as causally altogether? From observations, they just don’t think causally about most things.

Let’s look at k4. They are known to perspective take. What’s a causal model of perspective taking?
I have an observation from my youth.
I clearly remember wondering why people believe different things. I noted that one person would have a belief, justify it with their own logic (post hoc, I started to realize), and another person would have a different belief, justifying it with their own logic. And really, they believed that just because they were raised that way.
So, as a young k4, I causally reasoned that people believe what they believe because of how they were raised, and then they make up a post hoc logical argument justifying those beliefs.
Why was I interested in this? I was curious about the commonality of people’s surface differences. I was building a single model of internal human differences. In other words, you perspective take with a single large “model of perspective”, and adjust smaller variables within the model to predict outcomes. Having completely separate perspectives for each person is not how mental models generally work. This jives with the observation that k4s are not nearly as conformist - there’s no need to try to be the same, everyone’s already similar anyway. In this model, k4s try neither to conform, nor to individualize, as the effort is pointless, all humans being somewhat similar, anyway.


So, now we have an interesting model. K3s are interested in the differences of people, because they do not think causally, and k4s are interested in the sameness of people, because they do think causally. K4s are less interested in conformity because they see the sameness of people over their surface differences (everyone is already the same, you don’t have to try to be). This more causally detailed and explanatory model diverges from Kegan’s.

Let’s see if there are any other differences.
Many k3s develop at a young age and stay that way all their lives.
Many k4s also develop at that same age, and report never having experienced k3ness.
There is a large number of individuals “stuck” in the transition between the two.
What kind of model does that fit?
It’s the model of a personality trait continuum, like extroversion and introversion.
It appears NOT to be a model of one development stage on top of another.
So, let’s look at some personality trait models.

https://en.wikipedia.org/wiki/Need_for_cognition
https://en.wikipedia.org/wiki/Closure_(psychology)

These match up fairly well, with need for closure being close to k3, and need for cognition being close to k4.
The relationship with causal thinking appears to be complex, which is where you might want to call in an expert for transdisciplinarianism. Really, anywhere in this whole model where there is low causal detail would be a place to look for an expert with a more complex understanding of that area.

The model as it stands is:
Both K3s and k4s are interested in people.
K3s are interested in the surface differences, because they don’t think very causally.
K4s are interested in internal sameness, to explain the surface differences, because they do think causally.
The propensity to think causally is related to the personality trait of need for cognition.
The propensity to not think causally is related to the trait of need for closure.
But which causes which? Might need an expert to figure that out.

What to do about it?
With personality traits, you don’t try to turn one into the other, you just teach one the skills of the other. You teach an introvert extrovert skills, and teach an extrovert to shut up and listen.
So, possibly, just teach k3s causal thinking skills.

Would, of course, want to test this model, but it has high explanatory power and fine causal detail.


That’s the way the causal breakdown method works. You can generate lots of interesting models to test like the k3-k4 one above.

If you want to dispute my sameness model of perspective taking, do a causal breakdown on just perspective-taking, to build a more causally detailed model.

[karff]

Imagine you want to understand an automobile. You examine all parts of it, inside and out, differentiating all the smaller parts, integrating them into larger pieces and systems. You do many many integration/differentiation cycles, making a very complex picture of the car.

Then someone starts it up, and drives off. “How does that happen?” someone asks.
“I have no idea” you say “What’s going on is a level of complexity above my understanding. I’ll try to go through a few more integration and differentiation cycles to figure it out.”

And you do. And you still can’t figure out how the damn thing starts up and moves.

What you need to do, is connect the parts causally.

Objects are constructs of separation. Even with integration, you can’t perceive the real relationship between the integrated parts.

Causality is the construct of connection. You don’t need to do any more differentiation or integration. You probably didn’t even need to do as much as you did. You need to start connecting things causally. Then you’ll know how the car runs.

[jacob]

@karff - One example (I believe from Weinberg's Introduction to General Systems Thinking https://www.amazon.com/Introduction-Gen ... 004VS9AUS/ ) serve to illustrate how a theoretical scientist thinks about causation.

It's all about reductionism. A good scientific experimental setup eliminates all connections but the one the scientist is interested in; then checks to see if there's a causal (A => B) relationship.

Back to the example: Consider the solar system. There are 9 major bodies (the sun and 8 planets, sorry Pluto). This makes for 9*8=72 possible connections, which is a lot. Indeed, understanding 72 degrees of freedom is still beyond what humans can do, the most complicated systems ever devised maxing out around 40 or so.

However, it is possible to intuit some simplifications, that is, apply a simpler structure by analogy and then posit an understanding of the more complex system by understanding the simpler/ideal system.

For example, lets assume that the force between e.g. Earth and Mars is the same as the force between Mars and Earth. (Note, we're posing the existence of something called "force", which is actually a non-obvious idea.) This cuts the problem down to 36 possible connections. Now, if you have F=ma (a giant leap of insight), it stands to reason that if one of those 9 bodies is much heavier than the others, it is the only important one(!). This means that the number of connections are down to 8: Sun--Mercury (=Mercury-Sun), Sun--Venus, ... and so on. Also, posit that no planet is special. If so, then all these forces are the same kind. Presto, one only needs to understand 1 connection. That was the genius of Newton. He went from 72 and all the way down to 1.

The process of simplification is what the latticework is about. The "trick" here was the idea of the "two-body interaction". The latticework has about 100 such different ideas.

By moving in the direction of induction, reduction, and deduction, an otherwise complex world can be simplified enough to become understandable and somewhat predictable.

A brain that is unconsciously competent is this matter of thinking has achieved vision-logic. A brain that is consciously-competent has achieved rationality. Note that rationality only requires learning a few tricks-of-the-trade. Vision-logic is much much broader.

Somewhere within https://www.amazon.com/One-Taste-Reflec ... 1570625476 the author provides examples of the internal thought process of different kinds of thinking. The book is worth reading just for that.

[karff]

An example.

The movie “Don’t Look Up” is supposedly a metaphor for a Current Environmental Crisis (CER).

Let’s take a look at CER deniers. What is the mechanism for them accepting or denying things?
Not going into all the logic, but examining the various crises they react to builds this model:

If they are told of a crisis, and it isn’t their fault, they readily accept it and want to blame, hold others responsible. (The beginning of the health crisis a few years ago, the current migrant crisis).

If they are told of a crisis, and they are at blame, they react with denial and assume you are stupid. Note, no conspiracy theories yet.

If you tell them they have to change their behavior, they will assume the intent is to control them, and develop conspiracy theories as to why.

With the health crisis of a few years ago, the CER deniers (basically the same group of people) first accepted it as true, as others were to blame (the country of origin). But then, when they were told to modify their behavior, they denied it’s existence/severity, and developed conspiracy theories as to why they were being controlled.

Now, looking at the CER awareness advocates and scientists, they are using more or less logical arguments about how the CER exists, and how it might manifest in the future. Also, behaviors that might ameliorate it. When faced with denialism, they double down on their logical explanations, which the CER deniers then interpret as more stupidity (CER activists are accusing their lifestyle of causing a problem, an untenable idea, and therefore stupid), and are more convinced they are trying to be controlled.

And the feedback loop goes on like that.

The problem with “Don’t Look Up” - The deniers are not to blame for the asteroid. If told about an impending crisis that’s not their fault, the CER deniers are the first to get all excited and want to do something about it. Tell them their lifestyle is to blame for the asteroid, and they will deny its existence, and think you’re stupid. Tell them they need to change their behavior because of the asteroid, and they will concoct conspiracy theories about how you’re trying to control them.
Tell them that some others are to blame for the asteroid, and they’ll accept it’s existence, overblow its threat, and demonize those responsible.

This is a fairly simple phenomena. Probably most people reading this grokked it fairly easily. Yet, no one working on current crises seems to have figured it out. (The producers of Don’t Look Up didn’t seem to understand it)

Society is full of these low hanging fruits, and no one seems to be applying basic causal reasoning to them.

Jacob, hard scientists seem to use the thinking processes you described for technical problems, then abandon them when thinking about everything else. What we need is a way to get them (and everybody) to apply those processes to the wider social world.

[7Wannabe5]

@karff:

What if you tell them nobody is to blame, because Free Will is an illusion?

Seriously, I think your interesting take on causality intersects fairly neatly with Sapolsky's "Determined: A Science of Life Without Free Will." Wherever we do not comprehend our own underlying motivations or the genesis that forms our intentions, we (when in modern mode) insert Free Will as cause and tend towards assigning blame to individual humans or groups of humans.

How might your form of analysis be applied to post-modern (or post-post-modern) "de-nihilists" ? For instance, the tendency towards "inclusiveness" might result in "dark" reaction/response to CER along the lines of "Yes, my behavior is contributing to the coming apocalypse, because I am a human, and all humans, very much inclusive of me, suck, and do not deserve a nice planet."

Also, the Trad/Modern take of the Deniers may take on a more attractive hue when compared/contrasted with possible Level Red or Level Purple reaction/response such as "Let it burn! I will rise majestic from the ashes as prophesized by the rolling of the bones!"

[karff]

Seriously, I think your interesting take on causality intersects fairly neatly with Sapolsky's "Determined: A Science of Life Without Free Will." Wherever we do not comprehend our own underlying motivations or the genesis that forms our intentions, we (when in modern mode) insert Free Will as cause and tend towards assigning blame to individual humans or groups of humans.

You’ve hit the nail on the head. When I started thinking about this stuff, I thought “Wait a minute, the metacrisis is caused by a belief in free will?” If you break down all the problems of the metacrisis, they appear to be the result of humans not extending chains of causation back into/through/beyond other humans. But after looking into it a little more, it was more likely that humans are just bad at causation in general. Human agents are just a nice neat ending to causal chains for the subconscious mind.

As evidence it’s more of a general causal thinking problem, the same group of people seem to have high future discount, which would result from not being able to link present actions with far future effects. (This makes me wonder about the marshmallow test - Were the 1 marshmallow kids actually more impulsive, or were they just less able to link present behavior to future outcomes causally?)

My own hypothesis is that this could be remedied by bringing more awareness to it and teaching everybody to think more causally.

The “de-nihilists”. Their reasoning sounds like moral reasoning. What sort of moral logic (or emotions/intuition) are they using to determine “humans suck”? How do they arrive at a determination of “sucking” (The Deniers have the opposite logic - they are completely good, therefore they cannot be responsible for anything bad like an environmental crisis.)
“Do not deserve a nice planet” - What is their concept of “deserve”? By what mechanism do they determine who deserves what?
And then, when those mechanisms have been determined, see what the mechanisms behind those are. The end goal, to link those mechanisms to the rest of reality. Then, you get a sense of where they fit into everything.
Or, more intuitively, it sounds like they've just "given up", and are rationalizing it. Might look in that direction for deeper mechanisms too.

[7Wannabe5]

My own hypothesis is that this could be remedied by bringing more awareness to it and teaching everybody to think more causally.

My perspective is more like a probabilistic field. Might just be semantics, but "causation" seems too simple and mechanistic. For instance, I have observed that when humans move from Michigan to Colorado, they often lose weight, and I can list the factors I believe may be associated with this effect, but I will quickly note that my ability to summon up this thought which may lead to the intention of "moving to Colorado in order to become more healthy" already places me within the field of influence.

Also, I currently consider the marshmallow test to be almost pure self-referential propaganda.

Or, more intuitively, it sounds like they've just "given up", and are rationalizing it. Might look in that direction for deeper mechanisms too.

I think it's just in aligment with those who are the most idealist in youth (primary/secondary Fi) becoming the most cynical with age. INTJs with their tertiary Fi are kind of protected from that by their secondary Te. For better or worse, those of us with primary Ne will never give up on "finding" a solution. Is "ERE" a possible solution? Yes, if everybody (or even most everybody) did it would likely approximate graceful fall to Green Anarchy. Does/will most everybody do it? No. Back to the search loop.

[karff]

My perspective is more like a probabilistic field. Might just be semantics, but "causation" seems too simple and mechanistic. For instance, I have observed that when humans move from Michigan to Colorado, they often lose weight, and I can list the factors I believe may be associated with this effect, but I will quickly note that my ability to summon up this thought which may lead to the intention of "moving to Colorado in order to become more healthy" already places me within the field of influence.

That you even consider causal factors puts you apart from a lot of the population. Many people don’t even have a causal “slot” behind phenomena. And, part of causal thinking is realizing that every effect has multiple causes (might not have been apparent from the way I was reducing everything to simple interactions).
Some people I know with type 2 diabetes believe it’s some unfortunate thing that just happened to them. You couldn’t tell them it was lifestyle, because they have no causal slot to put the information. I can vividly recall an elderly gentleman with the affliction, standing with fast food sandwich in hand, asking rhetorically “How does one get diabetes?”, as if perplexed how he had somehow acquired a noncommunicable disease.

My sights for the general population is just the awareness that phenomena have causes, and those have causes behind those. At least half the population doesn’t seem to think this way very much at all. They don’t need to adopt causal reasoning skills, just realize that causation is a thing that exists.

In the current conflict in the Middle East, many of the participants (and their supporters around the world), have little to no causal slots for why anyone is doing what they are doing. The other side just has the noncaused quality of evil, and therefore can be mistreated. Just adding a causal slot in some of their minds could help ameliorate such problems.

My own thinking is, this is the single problem. This is the metacrisis. I’m just trying to get some awareness amongst those trying to solve it to at least consider it as part of the problem.

Also, I think the awareness of one’s own level of causal thinking is similar to stages in human development models. Each level thinks they are at the level above. Everyone believes they are thinking more causally than they actually are.
Those at the bottom think they are thinking causally when they are just correlating.
Those in the middle think they are doing original causal reasoning when they are just accepting causal chains originated in another mind.
Those doing original causal reasoning think they are doing it all the time with everything, but are only doing it on specific problems they’re motivated to solve.


A proposed definition of “causal awareness”
Awareness of the nature of causality, and having the mindset of applying that when considering phenomena. But also, the awareness of how we are not very causal thinkers.

For instance, when the mind stores a causal sequence of A causes B causes C causes D, causal awareness allows us to realize that the brain has bundled it up into a neat object, where we subconsciously do not look for causes prior to A, or effects after D. Also, with causal awareness, we realize that each A, B, C, D have multiple causes and effects, approaching infinity. The brain has subconsciously pruned off the ones unimportant to us. Also, the individual sequences of A - B, B - C, can be broken down into smaller causal relationships nearly endlessly.

The method of causal breakdown was developed to ameliorate the last one, especially. The brain seems to store “A causes B” as a neat object, and we don’t even consider there’s something to uncover deeper. If you take a phenomenon you think you know, then look critically at the links stored in your mind, understanding that each link could be broken down further and other causes and effects are connected, it reveals the areas of your own ignorance. It turns unknown unknowns into known unknowns. Now you know where to look, where to focus your investigative efforts.

Causal awareness is on a continuum, with just moving some of the population a ways along it would help.

[7Wannabe5]

Although it may help in some cases, I don't think that causal awareness is necessarily enough to alter behavior. For instance, China has higher literacy rates than the U.S. and much higher savings rates, but also much higher rates of tobacco consumption. IOW, whether or not we will "pass" any given "marshmallow test" depends on the overall field. Another example would be that affluent humans who choose to be vegetarian generally take more international flights. Another example would be the top British epidemiologist who broke out of the Covid isolation he recommended in order to knock boots with his lover. I'm not pointing this out to communicate that it is hopeless, just that plenty of humans who have moved well beyond concrete thinking still engage in all sorts of seemingly stupid and counterproductive behaviors for a wide variety of reasons. Often humans simply choose to not think about what they know to be the likely consequences of their behavior. I think many or most humans who understand the CER are simply hoping for a Cat in the Hat clean-up crew in the form of a wondrous new technology to be implemented before things get too bad. They're okay with not being the one to figure it out, and optimistic that somebody else will. Most people are just trying to deal with the problems currently in front of them. Gotta get to work, rev up the SUV. Going to work sucks, pick up a donut on my way. etc. etc. etc.

[karff]

Although it may help in some cases,

Then it's worth looking into, no? That's all I'm asking.

Also, the most important part of causal awareness is being aware of how bad we are at causal thinking. I think hardly anyone is aware of that at this point.

[7Wannabe5]

Yes, worth looking into. Yes, most humans, including me for sure, unaware of how crappy at this task.

For some reason, I can't quite explain although I feel it relates, I would like to recommend the climate novel "Mobility" by Lydia Kiesling to you.

[jacob]

Also, I think the awareness of one’s own level of causal thinking is similar to stages in human development models. Each level thinks they are at the level above. Everyone believes they are thinking more causally than they actually are.
Those at the bottom think they are thinking causally when they are just correlating.
Those in the middle think they are doing original causal reasoning when they are just accepting causal chains originated in another mind.
Those doing original causal reasoning think they are doing it all the time with everything, but are only doing it on specific problems they’re motivated to solve.

It is. See https://en.wikipedia.org/wiki/Model_of_ ... complexity

Causal reasoning ("if-then"-chains) thinking occurs at the formal level (11). Insofar one's "effective MHC" is the level at which one will actively and spontaneously process (e.g. "speaking German fluently") and not just passively comprehend in a contextual box (e.g. "understanding a technical paper in German"), I think humans should get far less credit than they usually get with most humans, most of the time and in most places, operating at the concrete (9) and abstract (10) levels.

Factoids and slogans can and does substitute for reasoning or comprehensive understanding at the concrete and abstract level. For example, people admitting to a writer or a speaker who saying things that they themselves are thinking but do not know how to formulate (formalize). Later on they can just repeat this. Perhaps this explains the growing popularity of OpEds, podcasts, and talk shows. Listening to other people talk substitutes for doing one's own thinking.

Causal awareness is on a continuum, with just moving some of the population a ways along it would help.

This is Kegan's "In over our Heads" problem writ-large. Kegan's posit is that our problem (the meta-crisis) is that humanity does not live long enough to get beyond a weighted average of Kegan3.2 or so. Whereas, many social problems are only solvable at Kegan4+; political change only at Kegan5. Analogously this also exist for MHC where the meta-crisis requires level 13-15 thinking.

There are people (directions) within the meta-crisis sphere working on evolving the complexity of human thinking (e.g. Schmactenberger, the Consilience Project) just like there are people working for social complexity (to Kegan5) trying to form functional communities ... and there's ERE1 trying to lift individuals lo MHC level 12 at a household level ... and ERE2 at the community level.

However, I'm not aware of any magical short cut to speed this up. Indeed, it often feels like an uphill struggle against a world that has been designed to extract value from maintaining humans at a somewhat simple-minded level.

[7Wannabe5]

Analogously this also exist for MHC where the meta-crisis requires level 13-15 thinking.

This makes sense to me, because I only maybe rise to Level 12 a couple mornings a week around 10 am after at least 2 large mugs of coffee, and I have little clue how to solve the meta-crisis.

Anyways, to clarify what I meant about the marshmallow test, I keep imagining numerous variations on the theme offered up in different contexts. For instance, what if you give each kid a marshmallow and tell them that they can either eat it right away or stake it in arm-wrestling match with another kid in attempt to win additional marshmallow? What if you initially give some kids 5 marshmallows and other kids zero marshmallows with median of around 2 and then proceed with original test? What if you give each kid a marshmallow and two kids an empty marshmallow bag and tell the kids that they will get their marshmallow back if they choose the kid/bag that ends up with the most marshmallow votes? What if you give each kid a marshmallow, but tell them that the marshmallows are cursed (or sugar is bad-will kill them)? How is the marshmallow test different from the "test" offered in my Girl's Friendly Society group, in which we were asked to silently tell God that we love him and then listen for his response?

However you design the "test" when you check back in on the kids who did well on it twenty years later, there is bound to be some coherence with how well the test represents the form of society in which the kid-to-adult is functioning, and post-modern level theory informs us that the adults who design the "test" are also humans without free will functioning within that which can be known in their field. So, by extension, if I am capable of considering a move from Michigan to Colorado in order to promote weight loss, it is also the case that I am already absent-free-will-finding-myself within a field in which my consideration of the matter represents a feedback loop in the system.

What is the overlap between humans who have read Hanzi Frienacht, practice polyamory, and hold the words "isomorphic" and "commensurable" in their semi-active vocabularies? IOW, how can things be unlikely, but also entirely predictable? Isn't it just one "cave' or another all the way "up"?

[jacob]

In terms of thought-modes, causal thinking is but one example. My definition of causality is "if there's ~100% correlation between two isolated(*) events, A and B, separated in time, then A causes B"), but how do I know if others understand this or whether they just pretend to understand or believe they understand. For example, does chatGPT understand causal thinking?

(*) Therein lies the problem. Especially when it comes to complex system, where A and B can not be isolated from other factors.

For a very interesting book on the early emergence of ego and possibly analytical thought, read the book described here https://en.wikipedia.org/wiki/The_Origi ... meral_Mind

I'm not sure what to do in practice in terms of developing more widespread usage of causal thinking. Insofar Jaynes's thesis is correct, it is possible for humanity to radically change the predominant thought-mode. It should be noted that the existence bicameral thought mode still exists today. There is a surprisingly large (estimated at 10% --- hard to estimate, because not everybody will admit to it because it sounds schizophrenic which carries a stigma) number of people (especially religious) who hear voices as if they were as real as some else else (or a radio) talking to them.

[7Wannabe5]

In terms of thought-modes, causal thinking is but one example. My definition of causality is "if there's ~100% correlation between two isolated(*) events, A and B, separated in time, then A causes B"), but how do I know if others understand this or whether they just pretend to understand or believe they understand. For example, does chatGPT understand causal thinking?

Based on my very brief experimentation with handiest AI, it "understands" that causality is akin to your definition and it knows how to apply this definition to a variety of problems or questions. However, it is also the case that, like most humans, it is lazy at research unless prompted and prodded. It informed me that it is highly reasonable to assert that human activity has caused an increase in average global temperature, but it is rather debatable whether regular exercise will reduce likelihood of contracting the common cold. When I asked it about recent research on complex relationship of BMI, comorbidities, and longevity, it initially only cited one study which was a follow-up to bariatric surgery, and it misinterpreted the concept of hazard ratio until I prompted it to research it further. More specific questions generated somewhat better results, but I believe this was mostly due to the fact that this forced it to do more research. It's fairly obvious that any bias held by the human prompter is likely to be reflected in the responses, but currently held consensus will be its center of gravity. Therefore, as I asked questions in alignment with the bias found in my recent reading on the topic of Intuitive Eating/ Gentle Nutrition (dieting to lose body weight is bad/counterproductive, but focus on nutrition,meditation, and movement to counter co-morbidities is good), the answers the AI provided became more nuanced. For instance, it "independently" brought the issue/factor of "frailty" into the discussion.

OTOH, answering a purely mechanistic question such as "How does a microwave oven cause food to be cooked?" would be trivial for an AI, but very difficult for majority of humans. So, if an AI had robotic access to a microwave oven and some food, it could eventually figure out how to peform that task without having to resort to an exact algorithm. OTOH, it could also do this by simply "watching" videos of humans cooking food in microwave ovens.

[zbigi]

OTOH, answering a purely mechanistic question such as "How does a microwave oven cause food to be cooked?" would be trivial for an AI, but very difficult for majority of humans. So, if an AI had robotic access to a microwave oven and some food, it could eventually figure out how to peform that task without having to resort to an exact algorithm. OTOH, it could also do this by simply "watching" videos of humans cooking food in microwave ovens.

I don't think it would be trivial. The LLM-based AIs, after all, don't "understand" anything, but are merely doing operations on texts they were trained on. So, even if the sentence "microwaves can be used to cook food" is something the LLM can generate, it still does not know what a microwave or food actually are. For LLM, they're just abstract word tokens. We would need another technological leap to map those tokens to actual physical objects (e.g. microwave), so that the AI could potentially manipulate them. But, even if we had that, the LLM does not understand the sentence structure, so all it would know is that, in the sentence "microwaves can be used to cook food", the token "microwave" refers to the physical microwave object in front of it. It would still not understand that microwaves can actually be used to cook food. Even if we wanted to circumvent that, by giving the AI an explicit task like "cook me some food" (so that the word "cook" in the instruction can be directly mapped to the word "cook" in the models, thus going around the problem of understanding what cooking is), it wouldn't still be able to do much without understanding sentences and essentially grammar (which has been a subject of study for the past 60 years or so, with little to show for it yet) - how would it otherwise discern between two trivial examples "microwaves can be used to cook food" vs "microwaves cannot be used to cook food".

In other words, LLM are not generating knowledge that is actionable for the computer. It is only actionable for humans because we, unlike AI, can map the words in generated sentences onto actual real-world objects and concepts, and also can parse the propositional logic encoded in sentences.

[ducknald_don]

how would it otherwise discern between two trivial examples "microwaves can be used to cook food" vs "microwaves cannot be used to cook food".

I'm always amazed by how many people can't distinguish between should and shouldn't or could and couldn't.