Some examples of skills that I expect the vast majority[1] of adults at 2-3 SDs above mean intelligence plus some subskill specialization (in your ontology) to not become an expert in 2 years:
Learning a language from a different language family than your own, for monolingual people.
Adults learning to play chess to a professional or semi-professional level
Chess should be really easy for your model: perfect feedback, unlimited reps, pure reasoning, no (or very limited) institutional gatekeeping, no physical component
Yet approximately 0 people have learned chess post age 20 and gotten to GM or even IM (International Master), even people getting to IM/GM while learning it as teens are rare (and often played similar games like Go and Chinese Chess very seriously)
the only partial exception Claude can find is Rani Hamid, who peaked at WIM (Woman International Master).
This isn’t just a matter of interest; many people want to be good at chess!
Nor is it a matter of g, we don’t have great data on the intelligence of top chessplayers but what little information we do have suggest chess GMs aren’t very high in IQ-as-traditionally-measured.
radiology
much less extreme than chess but my impression is that radiology similarly has a ton of domain-specific pattern recognition that you need to load up on, requiring both specific cognitive subskills and learning time.
Piano
I think the manual dexterity alone is pretty hard, but also a bunch of accompanying skills will be difficult.
My guess is that adult prodigies are possible (unlike chess), but fairly rare and certainly randomly selected smart people won’t be it.
Starcraft and some other esports
I’m less sure about this one but my impression is that the manual dexterity and low-level cognitive requirements are too intense for most smart people (for context 250+ effective “actions per minute”, or 4+ distinct actions/second, is on the low end for professional play in Starcraft 2).
Research in pure math
already covered by other commenters. Basically I think math is a very deep field, such that to make nontrivial advancements in it requires a bunch of prior context, especially pre-AI.
I also suspect (though am less confident) >2 years is necessary for expertise in non-software forms of engineering (my impression is that software engineering is unusually g-loaded and most other subfields of engineering requires a lot more knowledge/practice/experience, and less raw smarts), as well as many non-management areas of interpersonal skills/social skills.
Note: I’m not saying this is impossible, just very difficult. My guess is that for the people where this is possible, it’s due to a knack/specific low-level cognitive skills that aren’t picked up by coarse measures of g; for example chess grandmasters might have +4 SDs of a certain type of pattern recognition and memorization.
fwiw I think piano pedagogy is exactly the kind of thing where an entrenched regime has propogated a suboptimal approach relative to most people’s goals on the instrument (and that there’s maybe only some single digit number of people in the country teaching outside of the small handful of dominant, not-especially-useful-to-most-people paradigms).
E.g., if what you want to do is play pop songs, a combination of ear training and a ‘simon says’ style app that reads midi off your keyboard and instructs you to play a ~random triad will basically get you there in ~100 hours of practice (assuming daily practice not to exceed ~3 hours/day). There are similarly straightforward training setups that I expect to be effective for other goals one may have on the instrument. I built ~all of my physical facility on the instrument in about three months of focused practice, and have similarly ‘cheated’ my way into my other capacities (I’m definitely missing things that other people 5 years into the instrument would have, but I also have a lot of things those folks don’t, and I prefer and deliberately pursued my skill profile over theirs).
I agree that few people who start playing piano as adults will ever play Rachmaninoff at competition level (but think very few people who enter into the pedagogical system designed to meet that end actually have that goal in practice).
I also (I think, although you don’t say so outright) agree that some tasks require developing wholly other senses — new channels for phenomenal sensation or new ways of comparing phenomenal sensation in an existing channel (e.g. audiation and relative pitch) — and that those aren’t well-captured in Oli’s ontology.
much less extreme than chess but my impression is that radiology similarly has a ton of domain-specific pattern recognition that you need to load up on, requiring both specific cognitive subskills and learning time.
I would expect that you could train those pattern recognition skills a lot faster with an app that provides good and fast feedback compared to the current way those skills are trained via apprenticeship plus tons of supervised case reading.
Some examples of skills that I expect the vast majority[1] of adults at 2-3 SDs above mean intelligence plus some subskill specialization (in your ontology) to not become an expert in 2 years:
Learning a language from a different language family than your own, for monolingual people.
Adults learning to play chess to a professional or semi-professional level
Chess should be really easy for your model: perfect feedback, unlimited reps, pure reasoning, no (or very limited) institutional gatekeeping, no physical component
Yet approximately 0 people have learned chess post age 20 and gotten to GM or even IM (International Master), even people getting to IM/GM while learning it as teens are rare (and often played similar games like Go and Chinese Chess very seriously)
the only partial exception Claude can find is Rani Hamid, who peaked at WIM (Woman International Master).
This isn’t just a matter of interest; many people want to be good at chess!
Nor is it a matter of g, we don’t have great data on the intelligence of top chessplayers but what little information we do have suggest chess GMs aren’t very high in IQ-as-traditionally-measured.
radiology
much less extreme than chess but my impression is that radiology similarly has a ton of domain-specific pattern recognition that you need to load up on, requiring both specific cognitive subskills and learning time.
Piano
I think the manual dexterity alone is pretty hard, but also a bunch of accompanying skills will be difficult.
My guess is that adult prodigies are possible (unlike chess), but fairly rare and certainly randomly selected smart people won’t be it.
Starcraft and some other esports
I’m less sure about this one but my impression is that the manual dexterity and low-level cognitive requirements are too intense for most smart people (for context 250+ effective “actions per minute”, or 4+ distinct actions/second, is on the low end for professional play in Starcraft 2).
Research in pure math
already covered by other commenters. Basically I think math is a very deep field, such that to make nontrivial advancements in it requires a bunch of prior context, especially pre-AI.
I also suspect (though am less confident) >2 years is necessary for expertise in non-software forms of engineering (my impression is that software engineering is unusually g-loaded and most other subfields of engineering requires a lot more knowledge/practice/experience, and less raw smarts), as well as many non-management areas of interpersonal skills/social skills.
Note: I’m not saying this is impossible, just very difficult. My guess is that for the people where this is possible, it’s due to a knack/specific low-level cognitive skills that aren’t picked up by coarse measures of g; for example chess grandmasters might have +4 SDs of a certain type of pattern recognition and memorization.
fwiw I think piano pedagogy is exactly the kind of thing where an entrenched regime has propogated a suboptimal approach relative to most people’s goals on the instrument (and that there’s maybe only some single digit number of people in the country teaching outside of the small handful of dominant, not-especially-useful-to-most-people paradigms).
E.g., if what you want to do is play pop songs, a combination of ear training and a ‘simon says’ style app that reads midi off your keyboard and instructs you to play a ~random triad will basically get you there in ~100 hours of practice (assuming daily practice not to exceed ~3 hours/day). There are similarly straightforward training setups that I expect to be effective for other goals one may have on the instrument. I built ~all of my physical facility on the instrument in about three months of focused practice, and have similarly ‘cheated’ my way into my other capacities (I’m definitely missing things that other people 5 years into the instrument would have, but I also have a lot of things those folks don’t, and I prefer and deliberately pursued my skill profile over theirs).
I agree that few people who start playing piano as adults will ever play Rachmaninoff at competition level (but think very few people who enter into the pedagogical system designed to meet that end actually have that goal in practice).
I also (I think, although you don’t say so outright) agree that some tasks require developing wholly other senses — new channels for phenomenal sensation or new ways of comparing phenomenal sensation in an existing channel (e.g. audiation and relative pitch) — and that those aren’t well-captured in Oli’s ontology.
I would expect that you could train those pattern recognition skills a lot faster with an app that provides good and fast feedback compared to the current way those skills are trained via apprenticeship plus tons of supervised case reading.