By giving these things to random people, we don't do a very good job of evaluating their effects. If something works on 10% of the population, randomly selecting participants in an experiment will never give a significant result. If something is generally safe, it seems better to experiment on yourself.
The problem this solves is that some things work on some people. Neurotransmitters are blunt instruments that do many things- and are subject to adaptation effects where you get an effect for a while then the system adjusts to the new baseline. Maybe someone out there has a deficiency and adding more helps them for a bit, other people don't have a deficiency and get no results. Other people have a deficiency, something works for a little bit, then the brain adjusts to the new levels and the effect slowly disappears.
Until we can measure the level of the physiological effect we are targeting, the n=1 experiment seems like the only real way to see what group you fall into.
replies(1):