Great story (LOL!) -- thanks for that.
There would, presumably, be an "evolutionary" justification for the preference that favors traditional methods for fixing certain problems or predicting certain conditions: the more often a given method of healing or observation can be "tested" -- no matter how loosely or informally -- and not found to be clearly wrong, the more credibility it acquires for continued use, regardless of our inability to prove its correctness through rigorously controlled experiments, or to explain its validity through a plausible chain of causation. We need only consider the case of aspirin to see how the initial discovery and adoption of a "folk remedy" can far predate an understanding of why it works, and using it without understanding it is not a bad thing.
In a nutshell, methods that actually fail tend not to survive -- in the worst case, their practitioners and proponents die off having produced fewer or no offspring, but more generally and more importantly, other observers are never convinced to adopt such methods, or learn to actively avoid them.
Regarding the point made by your story, it's not unreasonable to infer that you can learn something very useful by observing the behavior of a particular group, given that the group has survived centuries or millennia in conditions where the behavior of interest could make the difference between surviving and dying off. The fact that you don't know the causes, evidence or reasoning that underly the behavior does not by itself undermine the potential value of the behavior.
autotelic adj. (of an entity or event) having within itself the purpose of its existence or happening.
™ Version 4.2