Uh I hate to be the bearer of bad news, but "hard" science has a shitload of problems with reproducibility as well.
It's actually a big problem with modern research in general.
I often dream of Brainiac Olympics. A month to challenge ideas, theories, find new ones, reproduce previous results in old or novel ways.
Not a quantitative hard science for sure, but that people try to understand psyche is not useless. The mind is complex and a lot of things that appeared as BS to me 4 years ago started to become a lot more plausible as I went through my own share of hurdles. I'm still eager for neuroscience, psychology and AI to overlap.
This is pretty much going to be my thesis, but you will have to wait a few years :)