The difference would be that the laws of physics are much more rigorous than the 'laws' of psychology/sociology. Outliers in the latter aren't necessarily indi

The difference would be that the laws of physics are much more rigorous than the 'laws' of psychology/sociology. Outliers in the latter aren't necessarily indicative of anything wrong with theory in general, while even a single outlier in anything in the former is indicative of an incomplete model. There are various results in physics which should be predictable via both GR and QM independently. The results should be the same as both models are supposed to be describing the same thing, so it follows that there should be some sort of gradual transition as one set of effects gradually comes to dominate over the other. Otherwise we'd see a single point in the data where QM stops being accurate and GR takes over, but despite investigating so many different scales, we have not seen any such cutoff point.