Exactly. I think there's a difference between "epistemic rationality" (avoiding cognitive biases, attempting to achieve accurate, rigorous beliefs) and "instrum

Exactly. I think there's a difference between "epistemic rationality" (avoiding cognitive biases, attempting to achieve accurate, rigorous beliefs) and "instrumental rationality" (holding beliefs that lead to a result, even if they are not strictly speaking, correct -- this would be heuristics, bias toward action and exploration). A focus on eliminating cognitive biases seems like it should lead us to the right solutions, but outside of certain academic fields, it often doesn't because the focus is more on not having incorrect thoughts instead of doing something, even if that something is wrong (at first). This means a mastery of all the material on "LessWrong" likely isn't going to make you better at innovating. I don't think rationality is overvalued so much as "epistemic rationality" is overvalued.