The messier the system, the more rigorous the methodology should be. Pre-registering trials to fight p-hacking, and open data should be the norm. That's decidedly not what we've gotten from this field, and some prominent figures have even been fighting such measures and consider the reproducibility crisis to not be a big deal. That's the problem.
In order to make definitive statements about said system, you need to start with the most basic behaviours and characteristics.
As you work your way up, you’ll mountains of data to capture every permutation of human behaviour and emotion.
We’re so, so far from any of that.