There’s an old English proverb about leading a horse to water — a proverb that has been in continuous use since the 12th century. Given what we know of how humans learn, we could presumably adapt this saying to more modern concerns: something like “you can lead a student to a mental model, but you can’t get it to take”.
@cedric Interesting post! I think I’ve shared other posts where I’ve used slightly different language to describe exactly what you’re saying here. Instead of Tacit Knowledge, I prefer Theo Dawson’s language “Skill Development.” But the exact same conclusions that you’re drawing remain: knowledge can be transferred fairly easily (i.e., through reading a book, watching a YT video, or listening to a podcast). Skill development however requires effort, practice, iteration, etc.
Theo’s belief is the reason why this is true is because Skill Development actually has neural correlates associated with those skills. It’s literally the embodiment of that knowledge in your nervous system and brain. So, the speed of skill development is bounded by your own biology (as well as other factors).
If I take this idea further, the emergent behavior of any Complex Adaptive System is a function of the nodes in the system and the connection between those nodes.
At some level of simple, you can predict how a system will work (I like Mini Motorway - a little iOS/Web game as an example). But beyond a certain level of nodes and connections, the complexity tips over and odd emergent behavior starts to show up.
My current belief is the same thing is happening in the brain. As we learn things, we create new neural wiring. That neural wiring create dense, interlocking connections that give us the emergent behavior of ‘getting’ concepts.
You use the example of understanding Dalio’s Principles and how it may not be available to some people at the bottom of those particular skillsets. I think you’re right. And, I would add it’s also possible the neural structures aren’t there to support this next lesson.
This begins to approximate Vygotsky’s Zone of Proximate Development. You can only really learn what’s at the edge of your current mapped world. I think this is why all of your writing about learning in ill-structured domains is so interesting - learning in those contexts is completely non-linear, so it requires a fairly excellent teacher to help you navigate that terrain efficiently. They need to know what your edge is, and increase the likelihood you’re exposed to experiences that are at your edge of proximate development AND help you interpret and make sense of those lessons. Hard stuff!
So how do you do this efficiently? I go back to Theo’s approach of VCoLing. I had an insight recently that what she calls VCoLing is basically “Labeling and Training” in AI systems. I’ve begun thinking of Labeling system objects as akin to increasing your Token Vocabulary. Once a system has enough tokens, it can then start to create connections/correlations/predictions between those tokens - but not before.
I’ll throw in just for fun: all of this, I believe, is directly related to Kegan’s work on Adult Development.
@richardhughesjones In case you’re curious, as this also feels related to other things we’ve discussed here.
This resonates with me - I’ve been teaching Product Management in Vietnam for the past two years, working with over a hundred students with great outcome (in my opinion). I agree that you can’t teach someone something they’re not ready to learn. But one way to get them ready is by immersing them in situations that expose the edges of their current knowledge.
In my course, we do this through a 2-week final project where students tackle a broad problem space, conduct product discovery, and ultimately propose a solution concept with supporting delivery artifacts. These are then presented to mentors (including myself) for review. There are many difficult parts: how to frame a good job story, how to conduct a good user interview, how to craft good hypothesis, how to take into account conceptual complexity when evaluating a solution, etc. Students usually struggle in either one or all of these tasks.
Overtime, we’ve built some scaffolds to help ease cognitive load without taking away the most challenging parts: The timeline includes clear milestones for when certain artifacts should be delivered, along with multiple checkpoints for feedback. It’s just enough structure to prevent paralysis, and just enough ambiguity to simulate real-world uncertainty.
By the end, most students have been stretched to their cognitive limits. They walk away with a much deeper appreciation for the complexity of product management - and with real experience that shapes how they think and work beyond the classroom.
However, I’ve also encountered cases where they seem to have a different neural architecture that is not suitable for the content of this course. In one particular, one student clearly used ChatGPT to frame research questions (he didn’t even put much efforts into prompting, and the result shows), so the questions didn’t really reflect what he wanted to know, but what the model “wanted” to know (heh). When confronted, he pushed for the narrative that the questions are his own, without even attempting to explain how his mind worked. I couldn’t get him to see that, this would only work if you truly grapple and struggle with this task, not over-indexing on the output itself.
That example you gave of the student is so interesting!
I sometimes quote (pre-crazy) Elon Musk, when talking about building the Tesla 3.
He said something like (paraphrasing poorly): it’s easy to build a fast, elegant, electric vehicle. What’s hard is building many of them, quickly, at a relatively low cost. In that respect, the factory is the most important product Tesla is building. The factory is the product.
If we apply this to human development, the underlying cognition is the thing we should be optimizing for, not the output that cognition can create. Because of course, we can now use these really powerful tools to produce seemingly useful outputs at incredibly low cost/time/resource expenditure. But that doesn’t do anything to improve our own cognition (the factory).
Kegan has a book on what he calls DDO’s - Deliberately Developmental Organizations. Companies that structure and organize themselves around accelerating learning of the individuals. I think that’s what is necessary for people to remain competitive - accelerate learning.