The Learn Drive

82

I said it is the reward deprivation that increases the addictive power

my SET game score drops from 1m30s to 2m20s if I drink coke for a couple of days straight and I get less sleep

This has all hallmarks of deprivation. It starts from variable reward of 1:30->2:20. Everyone will care about his score in anything. However, the trick is to care about important things more, and the importance is derived from a network of value, which is derived from learning, which requires the learn drive. In other words, healthy learn drive is one of the most powerful weapons against addictions.

Supermemo should be able to score userā€™s performance vs expected performance as a measure of cognitive ability

it is not sensitive enough (in healthy learning conditions) to provide reliable feedback (NB: in SuperMemo, you have a dozens of measures implemented, but each requires a great deal of data and is not easy to interpret). Where QS would work better is measuring cognitive violations. It is easy to get drunk or lose a night of sleep, but I bet we speak of more nuanced contexts.

natural instincts do not evolve nearly as fast as technology

luckily, protections are easy. I would start from preserving the learn drive, by removing all forms of coercion in learning. Micro-management of behaviors and emotions starts from the cradle. It needs to end.

If you would like to feel ā€œdigital dementiaā€ play League of Legends or Civilization for 9 hours straight

Please describe all side effects you know in as much detail as you care. This is very interesting from the point of view of the harms of combating homeostatic sleep drive.

For me, the feat is impossible. I would not survive a minute. I prefer typing this message. I am sure you can trace that choice back to the preserved learn drive. There are too many other rewards around too. In a moment I will play football. That might be the best reward of the day. I would not give it up for games.

You seem to speak of situations where addictions override homeostatic fatigue. A healthy brain rests or goes to sleep. It is not ā€œdigital dementiaā€. Digital technology is an enabler (in many areas of human activity). The true problem is the gambling addiction. In such cases, deprivation would cause problems one way or another. Perhaps it is better to find a digital outlet than to let one fall into a drug addiction, etc.

So the learning drive is not a conscious effort?

It is wired in the brain from embryonic times

83

I think there are some excellent points in here:

The fact that something drives us is no proof to its value. No doubt the ā€œcocaine driveā€ is a strong one! Therefore we need some way of evaluating drives as healthier or less healthy. As the ā€œno true scotsmanā€ statement implies, we canā€™t simply define a drive as good. Instead, we need to look at the effects of that drive. A criminal learning about technology in order to plan an attack may have a very healthy learn drive that powers their investigations. Interestingly, this brings us back to @Agaricusā€™s point about valuing knowledge. We wouldnā€™t want to be too agnostic with respect to terrorists plots.

In summary, while the learn drive itself may be healthy, other individual aspects may be unhealthy (e.g. bad relationships, addictions, etc.) which will affect the learning focus. Parents, teachers friends and society can strongly affect what a person seeks to learn.

84

In the case of ā€œhardā€ drugs the best drive is a severe fear of them.

In the case of lighter addictions like video games, none of my drives stood a chance when I was little. I guess I did not have a network of value at the time. But you said removing video games is a coercion. Would probably have been safer to limit as even today the video games pull at me. At least I would have learned something and the addiction would have been lighter. I must admit my whole situation is very rare and may have no applicability no anyone.

I use SET as a cognitive test. I just mean the lack of sleep reduced my cognitive abilities. Just a simple test like this was able to detect problems. That is something people ought to care about if only because their learning could become frustrating and hurt the learning drive. The effect size was so sever that little data was required. I guess that means drinking too much Coke is a cognitive violation. This too much Coke problem could be very common and so useful to test for everyone.

Speaking of more nuanced effects; Donā€™t users go through 100 flash cards a day? Is that not lots of data? More could be added by having common flashcards and mining other usersā€™ data. Finally, a second app could combine lots of other sources with supermemoā€™s but that would require timestamps.

I miss used ā€œDigital Dementiaā€. What I had was ā€œVideo Game Dementiaā€. The games also matter as some have negative health effects in addition to VGD (Lol). Maybe I still have DD but it is like water to a fish. Its actually been a while since I had VGD and my memory can be terrible. I think it best to go back to this subject only when I get a really good QS app so that I may test the effects. I had written down the details for my doctors but those papers are buried.

85

I meant there are no humans Woz defines as healthy. Everyone has unhealthy drives. Almost everyone would rather sit home and watch TV than bicycle, no? Some have it much worse than others.

I guess this is because of biology. Capitalism optimizes products to satisfy drives irregardless of health. Technology grows far faster than biology as evidenced by the number of moths on my outdoor lights. They were being eaten by frogs though. Even if a strong bad drive was caused by poor nurture it still exists and may easily override the learning drive then requiring mild coercion.

86

Drugs and alcohol can be considered technological artifacts (like a fridge, or night light). I did not put them on the list of control threats because it is relatively easy to keep kids away from cocaine (as opposed to nighttime tablets). Moreover, there are excellent defense mechanisms against drugs. Nearly all drugs are pretty untasty and it takes a valuation push to even try. With healthy valuations, side effects should be easily observable and addictions prevented.

Therefore we need some way of evaluating drives as healthier or less healthy

This sounds like a first step towards behavior micromanagement and limits on freedom on the grounds of immaturity. I posit a child has better defenses against consumption of illegal drugs than an adult. The young brain is simply a better balanced control device. We destroy that control overtime, and micromanagement is one of the key problems. A free six year old will absorb enough cultural memes on drugs and alcohol to be either cautious or even aversive toward the whole idea. In other words, an immature childā€™s brain may be stronger in terms of controls, defenses and even valuations.

A criminal learning about technology in order to plan an attack may have a very healthy learn drive that powers their investigations

I doubt it is possible to develop a criminal mindset in conditions of healthy learn drive combined with minimum natural empathy. Again I would see culprits in deprivation, environment, limited freedom, etc.

We have crime, terrorism, addictions, and suicides all over the place. However, in development, the remedy is more freedom, not more control.

87

Were you breastfed? Did you co-sleep? Were you sent to daycare or kindergarten? Did you go to school? This is a prime list of deprivation suspects in early childhood. In case of gaming, the school is the prime culprit. The gaming industry is powered by the slavery of schooling.

Donā€™t users go through 100 flash cards a day? Is that not lots of data?

SuperMemo strives at providing stable retention. In other words, it will attempt to compensate for all extrinsic influences. Moreover, few people do their learning when in poor cognitive shape. Even homeostatic fatigue is hard to measure because the user often gives up as soon as learning stops being fun. If you extend measurements over decades, you can collect enough violations of principle to see some effects (e.g. the circadian cycle). Hardly anyone learns in the middle of his circadian night, but once in a decade everyone does something which isnā€™t entirely reasonable.

my memory can be terrible

I hear it often and I always contest it. Interestingly, such a claim seems to also be a side effect of schooling. Teachers keep complaining and this has terrible effect on self-esteem for many. Without free explorations and free cognitive management, it is really hard to see oneā€™s true ability and to understand the limitations of the brain that affect everyone. Each time I ask for specifics, people list situations that happen to everyone, even memory experts. If you care, please provide evidence :slight_smile:

What you call ā€œVideo Game Dementiaā€ could simply be a result of ā€œSwiss cheese effectā€ when your gambling drive overrides your homeostatic drive that is supposed to protect your memory from abuse. It is a perfectly normal thing as long as you do not keep repeating the violation for decades. In the end, you can indeed end up with Alzheimerā€™s

I meant there are no humans Woz defines as healthy. Everyone has unhealthy drives. Almost everyone would rather sit home and watch TV than bicycle, no?

The love of sports is very effectively destroyed at school. Bad sleep, bad nutrition, bad brain hygiene, bad emotional management, bad learning, bad circadian timing, etc. I see kids decline drastically within the first 4-5 years of schooling. From an avid footballer, to a pizza gobbling videogamer who hates school. Even PE can ruin the love of sports. All it takes is to run it at 8 am on a sleepy head with a rude coach and a sport one does not fancy. As for me, I consider exercise a wonderful and precious addiction. A day without good sweat feels wasted. The brain loves that effort, but it needs to re-learn the habits and restore what was lost in schooled childhood.

Capitalism optimizes products to satisfy drives irregardless of health

It is not capitalism. It is human perfectionism and the wish to control the mind and the world with dismal support from evidence and dismal understanding of control theory. In communism, this perfectionism was driven to North Korean levels.

Technology grows far faster than biology as evidenced by the number of moths on my outdoor lights

Yes. We lose a million lives on roads every year and few people seem to do personal adjustments to account for the fact. I have not stepped into a car for 8 years now, and feel pretty liberated as a result :slight_smile: Bless the learn drive, all weird ideas are born from tinkering with models of the world :slight_smile:

88

Hi all. Ollie here, new contributor. Have been enjoying the chat so far. Especially some of the debate early on regarding school and coercion. Looking forward to reading the book sections about this for sure.

Seems like thereā€™s been about a 20hr break in the convo so I hope this is an ok time to ask a new question.

Iā€™ve been reading some of @Wozā€™s stuff about the learn drive recently and Iā€™m curious about the ideas therein in relation to the idea of learning and efficiency. In critiquing the ā€˜researchā€™ on discovery vs. explicit instruction, @Woz says,
ā€˜The learn drive mechanism is the best known optimizer of learning trajectoriesā€™ (from here)

When talking about optimising learning, I guess there are a few things we need to think about. We do need to think about learning trajectories, but we also need to think about search costs to find the info that populates that trajectory (search time/search costs).

For example: if itā€™s my goal to learn about quantum mechanics, perhaps a discovery approach will mean that the trajectory is optimised. That is, at each stage in the journey I learn some knowledge item that has a short semantic distance from something I already know. Challenge is, if Iā€™m self-directing that learning, it may take me a significant amount of time to identify what it is thatā€™s likely to be right on the edge of my current knowledge network.

In contrast, if Iā€™m working with a teacher (or other relevant domain expert), that teacher is likely to have pretty reasonable idea of the kind of learning progression thatā€™s likely to help a student to traverse short semantic distances in order to build a coherent understanding and (if they teach responsively), they can iteratively adjust their explanation/the learning trajectory in response to feedback from me (overt feedback ā€˜I donā€™t knowā€™ or more covert ā€˜confused facial expressionā€™). Further, theyā€™re likely to know which ā€˜nodesā€™ in a network are most relevant to building expertise. Granted, this approach wonā€™t establish as diverse a knowledge network as a discovery approach will, but itā€™ll likely help to build some of the major highways first between the most salient nodes first (time efficiency), from which further exploration can take place. To extend the metaphor which this post has seems to be heading towards, a ā€˜teacher as tour guideā€™ can help point out the most important sites within a knowledge domain, and support the student to work out how to get between those key sites, as a basis for further exploration.

In short, my assertion is:
-It seems unlikely to me that the benefits of a (potentially?) optimised learning trajectory through discovery would wholly and reliably counteract the increased search costs associated with a discovery approach (that is, I can see how this could happen in some scenarios, but feel itā€™s an overclaim if we suggest it happens in all cases).

In additon, my question is:
-What evidence is there that a discovery approach leads to optimised learning trajectories over that which is designed by a domain expert (who isnā€™t suffering from the curse of knowledge)?

Very much looking forward to peopleā€™s thoughts.

*(For context, Iā€™m a Yr 11/12 maths physics teacher. Have traditionally been very much in favour of explicit instruction. Getting more interested in independent learning recently as a result of chats with @Zon, some of @Wozā€™s writings, Peter Grayā€™s ā€˜Free to Learnā€™ (+ subsequent discussion with Peter on a podcast) and other explorations. But I think explicit instruction is extremely efficient when preparing students under limited time conditions for performance on exams with finite boundaries of required knowledge and performance. Whether those types of exams are a good idea is another point entirelyā€¦ *

89

Dear Ollie,

It is a great pleasure to chat with you directly. Your message carries so much value that I will need to spend some time offline analyzing and picking the best titbits. My prediction is that we will disagree a lot and then converge to a pretty solid consensus. I see our differences as a reflection of living in different worlds :wink: ā€¦ so there will be gaps to patch up on both sides :slight_smile:

Before I write in more detail, I would love to challenge you on that Quantum Mechanics scenario. Do you imagine that a free student without the pressures of the world would one day make a decision: ā€œI want to learn Quantum Mechanics?ā€. I rather see the QM quest as an encounter with some fascinating article in pop science press, and a gradual expansion from there in proportion to passions.

If I could singularly describe my professional interests, it would be the subject titled ā€œThe Brainā€. Despite years of my conventional schooling, I never took on a decision to study ā€œThe Brainā€. If there was a major at my college titled ā€œThe Brainā€, I might pick it only because I wanted to avoid military service. If I wanted to study the subject, I would rather look for fun books. Today, I would just google.

In other words, I believe your starting model is already tainted with your schooling experience. I posit that a free brain in a free course of adaptation to the modern world is not likely to say ā€œI want to study Quantum Mechanicsā€ (as a block of knowledge) or ā€œI want someone to optimize my trajectory towards knowing QMā€. I rather see the brain thinking ā€œQM is fun, letā€™s get some moreā€.

90

We do need to think about learning trajectories, but we also need to think about search costs to find the info that populates that trajectory (search time/search costs).

When you said ā€œsearch costsā€, my first thought was ā€œsearch for knowledgeā€ as in ā€œwhich book to chooseā€. The big difference between 1990 and 2020 as I experienced in my own learning is that I used to spend a lot of time searching for gold knowledge. Today gold is everywhere and the search cost is minimal. My search method is a lot of reading and sifting. In tons of books and articles there is always that breakthrough nugget. Sifting is costly too, but each sentence processed provides some value.

After reading your message further, I realize you speak of the search of the next knowledge node to which another piece of knowledge can be attached. This process is also relatively cheap in learning (as opposed to problem solving). the brain just activates a set of concept maps when confronted with new information. When a matching piece of jigsaw is found, the happy lightbulb turns on and we are ready to attach.

When I go deeper into your text, I realize that you rather speak of long shortcuts in the net of knowledge. From point A to point B. I would not label that ā€œsearchā€ because the shortcuts you have in mind are rather not findable. The distance is too long. This is where you see the role of the teacher. the problem with this approach is that for the shortcut to work as a viable memory, the goal B must be of high valuation, e.g. as you say ā€œknowing Quantum Mechanicsā€. However, how can be build valuations without explorations. Your scenario sounds like ā€œMy dad told me that QM will give me a good job in engineering. QM is the future. I want QMā€. This is unhealthy from the very start. Dadā€™s word can be used as guidance to make small steps in the direction of QM, and each step will affect further trajectory. perhaps competing pathways of exploration will tell the kid: I like Geometry better than QM.

If I dig into my own brain and wonder when I would personally ask for a teacher, I see only one major area: shopping advice. When I want to buy a device for recording video casts, I may decide that I do not plan to learn details of the technology that might die in 2-3 years. I would rather ask an expert: which device is best for me. In a few iterative interactive steps we can narrow my interests, and 5 minutes later I might have the device. I lose knowledge, I gain time. I accept that form of teaching.

There is also interaction with Ollie. You are teaching me your point of view. However, I would rather call it communication because it is a two-way system. We reconcile differences. My learn drive detects high value in your words. I go for a hunt for more.

I might be at a point in life where there are no major blocks of learning (like QM) that are well-defined to be ready to buy a book and study. I explore the edges of what is known and what is to be found about the brain. An 18-year-old might buy some Brain Bible and study page by page. Conversations with some great scientist in the field would be helpful too. However, I doubt an uninjured 18-year-old would enjoy being taught, unless on a partnership basis one-on-one with a great tutor.

For example: if itā€™s my goal to learn about quantum mechanics, perhaps a discovery approach will mean that the trajectory is optimised. That is, at each stage in the journey I learn some knowledge item that has a short semantic distance from something I already know. Challenge is, if Iā€™m self-directing that learning, it may take me a significant amount of time to identify what it is thatā€™s likely to be right on the edge of my current knowledge network

QM is vast and requires a great deal of support knowledge. Oneā€™s own explorations would be dramatically more valuable that interactions with a teacher. Shortcuts are great for speed, but a great deal of supplementary knowledge may be overlooked, customization will be absent, original thinking will be limited, creative exploration will be limited, and we may end up with a student fluent in QM. Instead, we want an open-minded experts that will uncover shortcomings in theories he read about in books. Thatā€™s the first step to progress.

if Iā€™m working with a teacher (or other relevant domain expert), that teacher is likely to have pretty reasonable idea of the kind of learning progression thatā€™s likely to help a student to traverse short semantic distances in order to build a coherent understanding and (if they teach responsively), they can iteratively adjust their explanation/the learning trajectory in response to feedback from me (overt feedback ā€˜I donā€™t knowā€™ or more covert ā€˜confused facial expressionā€™)

  1. if the choice of the teacher and the subject is voluntary, if the student sits in the driver seat, the above concept stands in compliance with the optimality of the learn drive
  2. I like to use the mountain climb metaphor to show how hard it is for an expert/teacher to see into the mind of the student. The greater the knowledge gap, the greater the illusion of superior guidance, and the greater the risk of harm.

theyā€™re likely to know which ā€˜nodesā€™ in a network are most relevant to building expertise

If you say ā€œbuilding a picture of QMā€, I would find it more acceptable. ā€œExpertiseā€ will always be highly customized. Experts are valued for they are hard to replace due to their uniqueness. A teacher can provide foundations of a block that helps understand other areas of knowledge, however, a teacher will not produce an expert. For that own explorations are indispensable.

Granted, this approach wonā€™t establish as diverse a knowledge network as a discovery approach will, but itā€™ll likely help to build some of the major highways first between the most salient nodes first (time efficiency), from which further exploration can take place

You are right, but why the hurry? Why not build best quality knowledge in a systematic manner? Why take shortcuts that may undermine coherence, stability, resistance to interference, applicability, creative substance, and the optimality of structure? I accept the reasoning for cases where knowledge is secondary (e.g. shopping assistance). QM sounds like a good material for deep exploration with a great deal of fruit to discover. Doing this in a hurry feels like an opportunity missed.

To extend the metaphor which this post has seems to be heading towards, a ā€˜teacher as tour guideā€™ can help point out the most important sites within a knowledge domain, and support the student to work out how to get between those key sites, as a basis for further exploration

If I was to study QM tomorrow, I would pick some great book, and chat interactively and occasionally with my best friends with solid QM background (oceans away). I somehow cannot imagine sitting down with some local QM expert and sense it was a good use of time (unless we could combine it with a dozen other interests and socializing :slight_smile:

It seems unlikely to me that the benefits of a (potentially?) optimised learning trajectory through discovery would wholly and reliably counteract the increased search costs associated with a discovery approach (that is, I can see how this could happen in some scenarios, but feel itā€™s an overclaim if we suggest it happens in all cases)

I claim that the learn drive is the best optimization device in the process. It may pick a teacher on the way to the goals. However, a child who never experienced schooling is far less likely to ever ask for ā€œteaching-likeā€ assistance. Humans have advantages over Google, but the way the brain likes to interact with ā€œeasily accessibleā€ knowledge has very little in common with what we see at school.

Perhaps try to make this judgement for me using your own brain: How would it feel to set a 45 min. time limit on your explorations with Google in where you can only choose a limited set of keywords (corresponding with the area of interest of an expert). To me that mental experiment approximates the interaction one-on-one with a single tutor. High quality teachers are still unbeatable. But the reality of schooling is left behind by Google far and fast. One-on-one is rare, quality feedback is rare, expertise is not as high as in your idealized case, pedagogy is lacking, freedoms are limited (e.g. bathroom visits), creativity is suppressed, explorations are suppressed, etc. I see the best teachers driven by the idealistic image that you present. However, after centuries of trying, we seem to be making things worse for kids who are better sensors of the future.

What evidence is there that a discovery approach leads to optimised learning trajectories over that which is designed by a domain expert

To optimize the trajectory, you need two pieces of information A and B, and an assessor of the value of the connection (which can only be done in the context of an individual brain). We have that assessor in the brain. On a high level, we need a similar comparator for streams of information X and Z. We have such a comparator in the brain. The devices are not optimum in their ability to pick the best pieces to learn from the universe of knowledge. However, they are best for evaluating knowledge at hand. There is no alternative (beyond erroneous extrinsic evaluators). This implies optimality. I wrote a few more words here

I think explicit instruction is extremely efficient when preparing students under limited time conditions for performance on exams with finite boundaries of required knowledge and performance

I agree. However, ā€œlimited timeā€ and ā€œperformance on examsā€ entails the whole spectrum of harms, and should never be a yardstick in learning. For me the only true yardstick is human opus vitae, and the optimality of the learn drive is easiest to appreciate from the lifetime perspective! I simplify to the pleasure of learning. As long as learning is fun, the future is secure

91

Welcome Ollie!

As teachers, I believe that you and I have more obstacles to overcome than the layperson as we discuss optimal learning, because we have some deep habits and ideas to unlearn. Therefore, while I totally agree with your comment, I believe the point you skipped over was actually the most important.

If explicit instruction serves exams, then we must begin by questioning the value of exams instead of accepting them as a fact of life. While exams are unavoidable in our day jobs, the question of optimal learning must go beyond school traditions. Until we question those underlying frameworks, we will never see how they could be improved.

So, rather than challenging the accuracy of your point, I challenge its importance. If we end up agreeing that exams are not useful - or even counterproductive - the original statement will become irrelevant. Of course, we might disagree about the importance of exams, but I think those basic goals of education are the place to focus our discussion.

Here is another teacher paradigm I think we should challenge ruthlessly. The idea of a ā€œlearning progressionā€ is based on an underlying model of knowledge as composed of discrete and linear components: K-12 Math, K-12 English, etc.

You would never seriously ask about the edge of the Earth unless you believed it was flat. If you believe itā€™s spherical, youā€™ll ask about its circumference instead. Likewise, when thinking about knowledge as a network with rich connections and relationships, the very idea of a ā€œlearning progressionā€ becomes questionable because there are so many possible directions to progress in, and so many unique combinations of knowledge.

Some people have learned management skills playing DnD while others learn probability, or both. Others learn to work with simple ratios while cooking (e.g. to double or halve recipes). In these scenarios, who would you say has progressed furthest? The point is not just rhetorical. Professional mathematicians each have their own area of expertise. You can find 5 PhD mathematicians and they will not all have the same knowledge in their disciplines. Taking off the teacher hat, it really is very questionable that they all needed to take the same route (i.e. linear learning progression) to get to those different destinations.

92

This is a very interesting point Woz. I like it. Youā€™ve pointed to an assumption that I hadnā€™t picked up on myself, that is, what it means to ā€˜learnā€™ something. The whole premise of my claim is that to ā€˜learnā€™ something is a pre-defined idea that has important ā€˜sitesā€™ (as I referred to them). I see that here youā€™re challenging that assumption and suggesting that instead the sites are what you make them. In the same way as in my 8 months of vagabonding around Europe I purposefully avoided many major sites (I termed it ā€˜shite seeingā€™ actually) and preferred to instead define for myself what was worth seeing (a squat in Prague, a festival in Belgium, a protest site in Scotland) perhaps accepting someone elseā€™s assumption of what it means to ā€˜learn quantum mechanicsā€™ is as bad as taking photos in front of the Eiffel Tower, Big Ben, and the Colosseum, and assuming that youā€™ve now ā€˜doneā€™ Europe.

This is indeed the first search cost that I was referring to. Your reflections in respect to the first are valuable too. Interesting to hear how this search cost has changed over time. Iā€™ve found these kinds of search costs can be minimised in conversation with a knowledgable other. Iā€™d be interested to hear how you categorise a conversation with an informal ā€˜teacherā€™ in terms of a learning episode (when is the line crossed into coercion?)

^Aha, you answered my question already!^

^Have uploaded to Dendro.

Why the hurry? A few reasons I guess. 1) If learning is occurring in a classic school/other time-bound setting (like trying to write a book for example), 2) Each minute entails an opportunity cost, so why not hurry? (To which (pt 2) the response is, ā€˜Whatā€™s the opportunity cost of forgetting? Given that the main time commitment within learning is retaining the infoā€™ā€¦ Ok, I see your point).

Thatā€™s enough for my brain for now! Much assumption challenging. Iā€™ll return shortly : )

Thanks for the ideas!

O.

93

Ok, this is a fair point. When we compare the average teacher to the average google search, a skillful learner is far more likely to get better bang for their (time) buck with a google search. I can agree with you on this.

It seems you also agree with the idea that when we compare optimum with optimum, an expert human guide is likely to win out (within a limited domain).

Which seems like a reasonable middle ground.

94

As far as I know, most people are introduced to drugs by their family and friends. This is where I see the ā€œno true scotsmanā€ effect in your reasoning. You say healthy valuations will not lead to drugs. In most contexts, socialisation is a healthy behaviour. But when it leads to drugs you will say that suddenly itā€™s not truly healthy.

To clarify, my own view is that the drive to socialise is healthy (like the learn drive), but may sometimes lead to bad outcomes. However, on the whole they are both very positive.

95

First: Thank you Ollie for joining this discussion. I notice in the high quality of your intervention a quality I associate with good teaching. Letā€™s call it ā€œtact.ā€ Iā€™ll return to this word later.

Second: Following Ollieā€™s questions, which lead to much clarification, Iā€™d like to try to articulate some things we agree on so as to clear them out of the way. If we attempt to make an argument that is comprehensive and convincing to people who may begin with beliefs that diverge from our common convictions, weā€™ll have to figure out our arguments for these convictions. But for now I think itā€™s useful to see if I can get them stated and stored so we can focus on where we donā€™t understand each other well and/or where we disagree.

  1. There is no delimitable set of ā€œthings worth learning.ā€
  2. Our human rights include freedom to learn.
  3. Our responsibilities to children include a responsibility to support them in learning.
  4. Learning should be pleasurable.

If these statements are considered obvious and a bit boring, then good! (If you donā€™t like them, then thatā€™s also good. Tell me why, and I will have learned something about a disagreement I didnā€™t know we had.).

Stipulating these common convictions, we have some obvious differences. @Woz is a radical. He seems to believe that school is always bad, that no explicit curricula or instructional program should ever be offered. @Woz have I made a mistake in characterizing your views? If I have, then tell me how. Of course I understand your examples of bad schooling, but I want to know if you truly believe that all schooling is bad schooling.

I still believe that schools can be good. What I hope for from schools is a situation in which children (and perhaps adults, as I sometimes wish I could still be in school!) participate in a learning situation with great teachers. What is learned in this situation is not primarily explicit content, of the kind that could be memorized with flashcards, but rather the values, frameworks, and actions associated with deepening our understanding of the topics that concern us most, both individually and socially. This tacit learning can only be done with our companions. To use Ollieā€™s example, I do not believe that it is remotely possible to deeply engage with the topic of quantum mechanics by looking up knowledge on the Internet. Progress in learning about quantum mechanics will require, sooner rather than later, doing research in quantum mechanics, research that will require comment (at least) from peers. Woz may prefer to call this ā€œcommunicationā€ rather than ā€œteachingā€ but Iā€™m not sure thatā€™s a great substitute, as the forms and rituals of this communication, the expectations and reciprocal social obligations that make it possible, are embodied in the specifics of ā€œteaching.ā€ Maybe I will offer my own radical position: There can be no quantum mechanics without teaching.

In using the word ā€œtactā€ to characterize Ollieā€™s intervention, I intentionally call on its relation to the sense of touch. The root domain of the word tact lies in physical experiences that are not formulated as propositions. A school is not a box for content. @woz if you want to abolish all schools, then I ask you to propose another way for students and teachers to be in close touch.

1 Like
96

Learn drive will optimally evaluate information in reference to prior knowledge. Socialization is largely based on the learn drive in that knowledge valuations are specifically affected by social forces, social instincts, pleasure of communication, love, etc.

The unhealthy factor in socialization is transferred by the peer pressure. There is always a link in the social network that goes bad and that link may propagate. The stronger the learn drive, the greater the resistance to contrary forces of social pressure.

Secondly, optimality of the learn drive does not imply it cannot lead you to get hit by a bus. Optimally generated decisions do not need to be optimal, and are not risk free.

When I speak of reward deprivation, I want to dispel that popular notion that if you give a kid bottles with sugar and with morphine, it will be the drug that conquers the brain. Or that videogames would consume lives (we see around that they do not). The brain always goes for maximum reward and the learn drive leads to extensive knowledge that provides an extra layer of protections in terms of valuations.

Drugs and videogames are always brought up in the contexts when I speak about the freedom for the kids. If theoretical models are unconvincing, you can always look at the outcomes of unschooling.

97

I am working on an article called ā€œ50 bad habits learned at schoolā€. I think I know which habits shape your opinions (e.g. as hinted by the mere use of the term ā€œflashcardā€). If your wish is to go back to school for a while, I have no reasons to question that. You clearly hold a romantic and idealized version in your mind. I discuss the origins of this ā€œglorificationā€ in many contexts. One of the most powerful forces in that glorification is generalization that will always drive the brain to a model that will be colored with a label ā€œbadā€ or ā€œgoodā€. Once that label is attached, the memory will nicely clean up all contradictory evidence. Without good retrospective analysis, and great deal of evidence, it is hard to build a true model. Secondly, you need a good model of a childā€™s brain and the conceptualization process. Child brains are not adult brains in miniature. They call for very different strategies of learning. Adults cannot empathize with a childā€™s brain, not even with their own brain from early childhood. They can only understand it via theory (or just trust kid brains). Adults tend to glorify school, give school credit for their own brainā€™s effort, for their parents good influence, etc. Having studied the impact of school on my own brain and the impact of free learning on my own brain, I have a nice safety clause: I moved from glorification to more accurate appreciation of the bad and the good. Having a good comparison of that analysis with free learning, my verdict for the school system is devastating. You call it radical, I see it as well-modeled. In contrast, your romantic notion of going back to school feels like a Stockholm syndrome to me. It has nothing to do with my modeling, theory or analysis. I viscerally abhor the idea of someone telling me what to learn, and how to learn. I predict that if you study the bad habits inherited from school, and try to unlearn some of the mythology injected therein, you will likely agree with me: free learning trumps learning at school by a galactic margin. All you need is cool head, patience, and a great deal of analysis.

You will need to quote me on that. I cannot imagine a context in which I would make such a claim. I said the opposite many times above, including my first reply to your post (search for ā€œthe child is always rightā€).

no explicit curricula or instructional program should ever be offered

no. big problems arise from compulsory curriculum

There can be no quantum mechanics without teaching

Nobody has ever taught me anything about the brain theory. Is QM different than ā€œThe Brainā€ theory? Who taught QM to its fathers?

abolish all schools

My Grand Education Reform does not include such a postulate.

To sum it up, I see the school system as a highly harmful form of slavery of children. Let children make their own decisions about learning. If they opt to go to school (e.g. due to being nagged or convinced by their parents), they should be free to see the reality of schooling, and be free to change their mind.

1 Like
98

My original answer at 3 am characteristically lacked some creative insight. When speaking of the school subject called ā€œthe Brainā€ (in lieu of QM that I studied only superficially), I should instantly come up with a staggering example of how direct instruction harms innovation.

Not so long ago, I criticized a couple of the smartest scientists I know. The scientists who are fathers of my own knowledge about the brain. I criticized connectionists. I have no doubt that the brain conceptualizes in the direction of concept neurons. Concept neurons were jokingly called grandmother cells. The term now slowly becomes derogatory in the mouth of connectionists. If I studied a school subject ā€œthe Brainā€, I would probably espouse the connectionist model, which has very good theoretical grounding in artificial neural networks. The power of free learning comes from layering knowledge differently and arriving at different architectures of the concept network where connections are characterized by different valence, stability, retrievability, and more. This means that different brains will come up with different models, even if their entire abstract model of the universe was identical. The architecture and the properties of the connectivity are essential in determining the outcomes. I arrived at the theory of ā€œThe Brainā€ from the perspective of spaced repetition that necessitated the two component model of memory (that few neuroscientists even heard of). The two component model provides an entirely different light on the brain. I guess it is the most productive model in my mind. Our original difference also has its roots in the model. When I speak of retrievability or stability of connections that determine outcomes, I use the language of the model. This reasoning would not be possible if my brain model was delivered in a textbook manner at school.

perhaps accepting someone elseā€™s assumption of what it means to ā€˜learn quantum mechanicsā€™ is as bad as taking photos in front of the Eiffel Tower, Big Ben, and the Colosseum, and assuming that youā€™ve now ā€˜doneā€™ Europe

Excellent metaphor! You are right. If you swim with the school, you rarely get a thrill of TRUE discovery!

search costs can be minimised in conversation with a knowledgable other

You need the right other and in the right context. No doubt, when you have the interaction of two intelligences, things can get sparky. This is what makes this discussion so thrilling. If you voluntarily choose your tutor as superior to books&google, the same is true. It is the learnā€™s drive decision.

Have uploaded to Dendro

More proof of the power of the context in prior learning. I heard of Dendro many times, but your commitment is what truly jars my attention. The concept triples its value and its activations will come easier.

Why the hurry? If learning is occurring in a classic school/other time-bound setting; like trying to write a book ā€¦

I think the main value of any book is its creative content. In that light, learning for the sake of writing would be the last area where I would want to be guided. That would be exactly the field where I would want to slow down, branch out, and stay free!

Each minute entails an opportunity cost, so why not hurry?

Each path untraversed entails a missed opportunity cost, so why hurry? Cleary we need a good balance, and the learn drive is the best guidance for charting that trajectory. A teacher or a textbook on a specific subject are unlikely to be of much help in my own endeavors. Naturally, a teacher like you is a goldmine of expertise on the subject of teaching. I love to chat here about teaching, I would love a chat on anything over beer, but I am not sure I would want to sit in a classroom listening to Ollieā€™s lecture on the art of teaching. I would rather have it registered, uploaded to YouTube, and let it compete with other threads of learning in incremental video. That my fastidious approach influenced by years of free learning.

It seems you also agree with the idea that when we compare optimum with optimum, an expert human guide is likely to win out (within a limited domain)

There is also a cost of finding an expert. This is why the search for an expert would most likely be initiated (1) at times of getting stuck where the expertise is likely to exist (2) at times when a new large area of study needs to be consumed promptly, (3) when learning was secondary to the ultimate decision/solution, and some not so vital circumstances. When I scan my brain for learning decisions in recent days, I see many interactions with experts. They either provided curious knowledge that I would never want to google for, or provided local knowledge that is not available elsewhere, or provided knowledge that is fun to get from a human as opposed to the net (for social reasons).

Woz: ā€œHigh quality teachers are still unbeatableā€. Ollie: ā€œWhich seems like a reasonable middle groundā€

you cannot leave my phrase unqualified without context. We spoke about tutoring one-on-one, in which there must be a coincidence of prior knowledge, goal valuations, and expertise with a stamp of approval from the learn drive. We can then get to unprecedented progress in learning, creativity, or problem solving. Thatā€™s when two brains are unbeatable.

p.s. I do not comment on Georgeā€™s prior post because we nearly always agree :slight_smile:

99

I again find myself curious about what we disagree about. Thatā€™s where the gold mine is. There are many rough spots that seem like disagreement, but Iā€™m skeptical whether there is treasure there. For instance:

This certainly seems like a big disagreement. But the ease with which I agree with your statements tells me that we donā€™t have anything too valuable here. If I conveyed an overly romantic vision of school or seem to have been brainwashed by my kidnappers, then I will definitely correct this impression. I think any school that does not give pleasure in learning is harmful, and I have unfortunately long experience with this kind of school. However, I have also experienced another kind of school, and I donā€™t want to forget this, either. Since I agree with you that pleasure is the key to learning Iā€™d be foolish to forget how much I loved it. We may have some slight misimpression of our argument, or slip into rhetorical opposition, but if Iā€™ve said anything that can be mistaken for a general argument that the reality of most childrenā€™s school experience is good and not harmful, then please allow me to withdraw it.

Where I think there may be some useful disagreement is in your use of concepts like ā€œcompulsory curriculumā€ and in the paired terms coercion/freedom. If by compulsion and coercion you refer to a prescribed course of study that children must follow regardless of interest, then of course I agree with you. And while Iā€™m certain that you donā€™t include all social factors that make mastery of certain topics rewarding under the category of compulsion and coercion, Iā€™m struggling to get a clear picture of where the boundary lies. Or maybe I do have a picture: the boundary is at the door of the school.

Before I write another word, I should check with you. Is this characterization more fair than my statement that ā€œ@woz wants to abolish schools?ā€

100

I am not arguing that free children will become addicted to videogames and drugs. Rather, I am arguing that a dogmatic valuation of the learn drive as 100% ā€œgoodā€ and ā€œhealthyā€ in all cases is debatable. It may be the best of imperfect options - which I believe correctly represents your claim of ā€œoptimalityā€ - but it could still lead you astray now and again. Therefore, I still claim that a healthy learn drive could lead to unhealthy interests, just as @Agaricus observed when he described the unschooled child he knows:

However, none of this is an argument against free learning. It is just a caveat to accompany a strong recommendation for it. You seem to agree when you say:

101

I would like to phrase this one a bit more strongly as: ā€œOur human rights include freedom to choose what, and how, to learnā€. I think this is what you meant, but it could also be interpreted similarly to the United Nationā€™s idea that children have the ā€œrightā€ to a ā€œcompulsory educationā€, where education means much of what we are arguing against: cookie-cutter curriculums with compulsory standardised tests, and so on. As it currently stands, I fear this statement could be misused by calling a child lazy because they are not making good use of their ā€œfreedomā€ to learn in maths class, science class, etc.

I would also suggest changing this one to ā€œEffective learning is pleasurableā€. The difference is between a subjective, moral statement (i.e. ā€œshould beā€) and an objective summary of what we know about learning (i.e. ā€œisā€). The reasons why we learn (e.g. exam marks vs curiosity) can absolutely impact both our learning in the moment and the likelihood that we will return to build on that learning in future.