> Why would being bad at one subject or skill imply they are bad at all skills? Statistically, human cognitive capabilities of all types are positively correla

> Why would being bad at one subject or skill imply they are bad at all skills? Statistically, human cognitive capabilities of all types are positively correlated, for what it's worth. If you observe someone failing badly at structured reasoning about a given topic, then you would be reasonable to expect they'd also be worse than average at surface-level talk about the same topic. However, Constantin's argument in the article was that the two capabilities -- surface-level talk vs structured understanding -- look independent in her personal experience. I'm skeptical of her conclusion. I think the two capabilities probably are linked, but it's a weak link. You just might need a lot more samples and more confidence the samples were chosen randomly before the statistical relationship became noticeable. If so then in personal experience and practical life, the capabilities might as well be independent.