Benchmarking cognitive functions should be much more interesting than it currently is, with data tracking progress and regression. I feel like the tests are too

Benchmarking cognitive functions should be much more interesting than it currently is, with data tracking progress and regression. I feel like the tests are too focused on low level raw hardware bandwith of simple tasks, and not enough on semantic reading comprehension and problem solving. E.g robustness against cognitive biases should be benchmarked, thinking outside of the boxes/ solutions by transpositions/free associations should too. Such a toolset would also help measure the effect of nootropics differently, vs e.g. the magnesium threonate ~9 IQ points increase