Studying microevolutionary processes in cognitive traits: a comment on Rowe and Healy