Surprised that the classic "Conditions for intuitive expertise: A failure to disagree." paper by Daniel Kahneman and Gary Klein is not referenced https://psycn

Surprised that the classic "Conditions for intuitive expertise: A failure to disagree." paper by Daniel Kahneman and Gary Klein is not referenced https://psycnet.apa.org/record/2009-13007-001 https://psycnet.apa.org/record/2009-13007-001 My takeaway: without short feedback loops it is hard to build expertise. If it takes 2 years to learn the individual you paroled reoffended, you can't learn (so there are virtually no expertise in parole officers or university admissions). In these scenarios, simple linear models outperform "experts" by a wide margin. See: "In Praise of Epistemic Irresponsibility: How Lazy and Ignorant Can You Be?" by Michael Bishop https://philpapers.org/rec/BISIPO https://philpapers.org/rec/BISIPO