Review of The Engines of Cognition — LessWrong