I doubt that the premise is true. Self-reporting is not a reliable indicator of actual performance in this case. It seems plausible that an empirical study woul

I doubt that the premise is true. Self-reporting is not a reliable indicator of actual performance in this case. It seems plausible that an empirical study would find mathematicians way above the average in elementary arithmetic, though possibly trumped by waiters and cashiers who often have to calculate without aid. Of course, without an actual study this is just speculation.