Turns out there's an entire academic field dedicated to this question! And a word for the concept: historiography! And over centuries, tens of thousands of academic historians have worked out any number of methods for assessing this sort of thing... there's even a history of historiography on the Wikipedia page about it! But you know, who knows if anything reported there is accurate?

If they're consistent, you'd have to explain how can multiple actors imagine a similar story (assuming those witnesses didn't interact). You can use probabilistic reasoning to determine whether a large portion of the story is close to reality.

Right, just like with bigfoot (?)

I feel like there is a difference between "we saw something between the trees! It must have been the bigfoot!" and "we gave this guy a cat and he ate it whole in front of us".