> Using simple simulations,we show that this pattern arises naturally from collider bias when selection into elitesamples depends on both early and adult performance. Consequently, associationsestimated within elite samples are descriptively accurate for the selected population,but causally misleading, and should not be used to infer developmental mechanisms
If the paper is true, then, yeesh! That's a pretty big miss on the part of Güllich et al.
Reading through the very short paper there, it seems to not have gone through review yet (typos, mispellings, etc). Also, it's not clear that the data in the tables or the figure are from Güllich's work or are simulations meant to illustrate their idea (" True and estimated covariate effects in the presence of simulated collider bias in the full and selected samples"). Being more clear where the data is coming from may help the argument, but I likely just missed some sentence or something.
I'll be interested to see where this goes. That Güllich managed to get the paper into Science in the first place lends credence to them having gone through something as simple as Berkson's Paradox and have accounted for that. It's not everyday you get something as 'soft' as that paper into Science, after all. If not, then wow! standards for review really have slipped!
That's not true. It didn't replicate and Norvig has said as much somewhere on HN, IIRC.
(I also agree with the other criticisms that this 'old vs young' setup in OP is obviously at least partially, and perhaps entirely, regression to the mean and Berkson.)
Motivation if you feel like you're young and failing
edit: went back a few more years, lots of NHLers in the top 5 in scoring in the tournament, but some years are more miss than hit.
Yeah, I'm really unconvinced by the paper.
1) Adult chess GMs all come from super advanced kids, now. Period. GM Ben Finegold talks about this at length.
2) We know that, for example, hockey success is correlated with birth month. This means that juniors who happen to be slightly larger and promising get more attention and coaching and so wind up being the world class adults, too.
I could be more convinced by academics and music that you need multi-disciplinary education to be world class, but I'd need to see a lot more evidence.
Nevertheless, the single thing I extract from my anecdata is that being top ten in a world class field requires a dedication bordering on psychotic mania. You have to be willing to not just go the extra mile but the mile beyond that and beyond that and ...
Most people are completely put out by the minimal amount of effort to get to the top 1% so they would be stunned by the amount of work you have to put in to get to top 10.
[0] https://xcancel.com/AlexGDimakis/status/2002848594953732521
In the canonical example, you have uncorrelated attributes, eg skill and attractiveness in actors, forming a round scatter plot with no correlation. Selecting a subpopulation of top actors who are either skilled or attractive, you get a negative correlation. You can visualize this as chopping the top-right of the round scatter plot off: the chopped off piece is oriented in roughly a line of negative correlation.
In this example, if you look in the linked paper inside the post by Dimakis, there is a positively correlated scatter plot: You can tell the shape is correlated positively between youth and adult performance. But in this case, if you condition on the extremes of performance, you end up selecting a cloud of points that has flat to slight negative correlation.
Uncorrelated attributes:
y
│ ∙
│ ∙∙ IIIIIII
│ E∙∙IIIIIIII
│ EEEE∙∙IIIIIII
│ EEEEEE∙∙IIIII
│ EEEEEEEE∙∙III
│ EEEEEEEEE∙∙
│ EEEEEEE ∙∙
│ ∙
└───────────────────x
Correlated attributes:
y
│ ∙
│ ∙∙ IIII
│ ∙∙IIIIII
│ E∙∙IIIII
│ EEEE∙III
│ EEEEEE∙∙
│ EEEE ∙∙
│ E ∙∙
│ ∙
└─────────────────x
To put it another way: If selection was restricted to people who performed highly in either their youth or in adulthood (or both), Berkson's Paradox explains the result. If selection was restricted to people who performed highly in their youth, or if selection was restricted to people who performed highly in adulthood, Berkson's doesn't explain it.
100% correct. For traits x and y, selecting for datapoints in the region x + y > z will always yield a spurious negative correlation for sufficiently uncorrelated data, since the boundary of the inequality x + y > z is a negatively sloping line.
>But in TFA, surely the "high performance" selection filter applies only to the adult performance level?
Doesn't seem that way. Reading the full paper [0], they say:
In sports, several predictor effects on early junior performance and on later senior world-class performance are not only different but are opposite. [...] The different pattern of predictor effects observed among adult world-class athletes is also evident in other domains. For example, Nobel laureates in the sciences had slower progress in terms of publication impact during their early years than Nobel nominees. Similarly, senior world top-3 chess players had slower performance progress during their early years than 4th-to 10th-ranked senior players, and fewer world top-3 than 4th- to 10th-ranked senior chess players earned the grandmaster title of the International Chess Federation (FIDE) by age 14.
[0] https://www.kechuang.org/reader/pdf/web/viewer?file=%2Fr%2F3...
Savantism can be pretty damned weird. I've known a few, including a couple who will never have an adult career beyond local gigs because of their mental disabilities in other, non-music areas. The Oliver Sacks book "Musicophelia" has fascinating case stories about it.
Another thought - Einstein had reviewed thousands of patents when he worked on the train - that's a hell of data set for an LM to start with.
Magnus Carlsen, Lang Lang, Terence Tao all were precocious and achieved elite performance in their youth.
Unintuitively, strength is a general adaptation that applies to all specific movements. A muscle is either strengthened across a range of motion, or it isn't; a muscle cannot be strong swinging a club while not strong lifting a weight, nor can it be strong holding a rock while weak holding a bar. It is optimal for most sports to train for general strength via barbells, and then to practice that strength via the sport. The rock climber should do heavy deadlifts and chinups to train his grip (and everything else), not special rock grip exercises, for the latter are difficult to progress in small increments and are inefficient in a time sense. A man who can do chinups with 150 pounds hanging from his waist, and who can hold a 550 pound barbell, will not have a problem hanging onto the bouldering wall; he need only practice his technique.
To the article's point, you should get "strong" in everything until you decide to practice that strength in one thing.
"slack around as kid, it will make you great later!"
but
"prodigy youth doesn't guarantee greatness later, as well as non-prodigy youth doesn't prevent you from becoming grat later".
Early exceptional performers and later exceptional performers within a domain are rarely the same individuals but are largely discrete populations over time... and Most top achievers (Nobel laureates and world-class musicians, athletes, and chess players) demonstrated lower performance than many peers during their early years. Together.
A simple explanation: high performance requires quite a bit of specific preparation. But "exceptional" performance is mostly random relative to the larger population of high performers in terms of the underlying training-to-skills-to-achievement "equation". Especially, being at the top tends to get someone more resources than those nearly at the top who don't have visible/certified achievements.
I'd that billing your work "the study of the very best" really gives you strong marketing spin and that makes people tempted to find simplistic markers rather than looking at the often random processes involved in visible success. IE, I haven't touched on reversion to mean (https://en.wikipedia.org/wiki/Regression_toward_the_mean).
That was quite a shock to realize that I had to do the exercises and the homeworks if I wanted to pass. And since I was not use to efforts, I was no longer the top performer in classss where you have to do the exercises to really understand the matter.
I was recognized as extremely clever by teachers and other students but let me assure you that over long enough, discipline (witch I don't really have) and consistent efforts beats cleverness.
A-level is like the stepping stone to university. It's optional and you only do subjects you like. Suddenly you had to actually understand the stuff. It wasn't just a memory test. The first exams I got B all around. I wasn't happy with this so I started doing homework etc and actually understanding things.
Coincidentally, my breakthrough also came with Newtownian mechanics. I was quite good at integration, but I didn't understand why I was doing it. Suddenly the whole thing started to make sense when I realised it was about rates of change etc.
https://press.princeton.edu/books/paperback/9780691133805/ol...
That book used artwork valuation as a performance measure and analyzed it over top artist's lifetimes finding two patterns. The "Young Genius" where an artist has a vision and realizes some innovation and their most valuable works center around that with value tapering off over their life. Picasso. (Who had two peaks but still fit the pattern.) Contrast to the "Old Master." This is someone who keeps refining their craft and their most valuable works and innovations are their late life works. Cézanne.
And yet their access restriction not only removes the public from consuming the fruits of their labor, but it also systematically harms less well-resourced institutions, independent scholars and impedes the spread of knowledge (particularly in areas of the world that need it most).
I wish we could reach a point where we wouldn't allow this anymore.
Sounds like they're describing ADHD.
(Side note after the important ADHD joke: there's an old sport textbook called "Periodization" that mentions focusing on breadth rather than depth of sports experience in early life is a better path to olympic-level performance than just going hard in a single sport from a young age.)
ADHD is not correlated with high career performance, sadly, and represents a real obstacle for those struggling with it. The current social media trend of equating ADHD to a superpower which propels people to focus intensely and excel is really unfortunate.
Those who face this issue can of course try to "gamify" their upcoming tasks to themselves in a way that will hopefully steer that focus in desirable directions, but that's not always easy. The monkey mind also resists ongoing habit formation, which is the tool most non-ADHD folks would generally resort to in order to effectively manage their overall schedule and just be more on-task.
Some mental illnessess are extreme versions of traits that are often useful. It's good have one person in the village who frets about dangerous stuff, for example. Anxiety is useful at times.
But as you start to diagnose the very functional people who just need a few points to get a top uni course (or people self diagnose) ... well ... maybe you're picking up far less extreme and maladaptive versions of the trait.
Also, the ungated part doesn't say how they're measure "top" high-school vs university students. It doesn't match what I've heard about the persistence and consistency of basically all standardized tests; are they using within-school rankings for this? If so, that would fit perfectly with students being sorted during university selection.
.
* https://en.wikipedia.org/wiki/Rapid_application_development
It's a relatively common theme in sciences that someone comes out of nowhere and solves a long standing problem in a field because they don't have the specialized set of biases that keeps everyone else trapped.
One of the benefits of generalism / learning multiple fields (IMHO, again) is that you realizes that special abilities / skills don't necessarily translate well from one field to another. For example, learning to play the violin is very different from, say, playing billiards, yet becoming good at either one involves learning subtle manipulations of basically similarly-shaped pieces of wood. By involvement in multiple fields, you learn to be careful NOT to bring your "everything is a nail" mentality with you from one field to the next.