That hypothesis is clear, and I agree with it. I am a fan of Ericcson's work, but I disagree with certain examples that are often cited in articles like these.

That hypothesis is clear, and I agree with it. I am a fan of Ericcson's work, but I disagree with certain examples that are often cited in articles like these. There are many reasons, the first of which I mentioned above. The rules of aptitude assessment are much different for children and adults; the returns per year is not simply accumulative. First, the curve isn't linear. Second, the curve looks different during different ages. Forgive me if this is obvious, but the reporters never seem like they know. The ten years from age 3 to 13 is a different ten years from age 23 to 33. If you learned a language in the former interval, you'd be a native speaker; in the latter, no matter how hard you work, you won't be. The "early" brain is a completely different creature from the "late" brain (which is why I think youth education is far more important than teen+ education). It is also early in life when particular natural affinities develop. If there were two TigerWoods, let one start golf at 2, trains until 18, and deprive them for 8 years. The other starts at 10, trains until 26, and bring them together to compete after a refresher for the early start, I would bet on the early starter.