capogiro last night had papaya gelato and numerous people having crises of faith about their research. it's universally acknowledged in psychology that almost everyone has something to hide in their work, and the corresponding maxim that Makes It All Better. in daniel's case, it's Averages Are Your Friend -- this because of the notorious unreliability of cortisol data which spikes and plummets at the slightest perturbation, and differs from person to person by an order of magnitude. the broader version of this problem -- that virtually all measurement in psychological experiments is unbearably coarse -- has plagued me since my early days in college. and, as if self-doubt weren't bad enough, it's the main broadside of both the anthropologists and the biologists; the former want to throw the baby out with the bathwater and give up entirely on abstraction and generalizations, and the latter think we're a mess because we can't achieve anywhere near the amount of reduction they're comfortable with. i don't have a very good reply to either of those salvos except: we're doing the best we can. you try giving someone else a several-hundred-year head start and see how well you do.