If it helps, geniuses in very narrow fields with immediate feedback, like chess, aren't the smartest most of the time. They are just optimised for solving a particular problem with a closed feedback loop, much like an artificial neural network in machine learning.