I have seen more than one PI at an R1 universities with multiple Nature publications use this heuristic. I would not call them incompetent.

Do you not notice the circularity of your reasoning here?

Also I didn't say incompetent, I said "not very". More competent researchers make journal rep only a very small factor, and it is not via the "high rep = more trustworthy" direction (which is the bad heuristic), it is "pay-to-publish journals = not trustworthy" (better heuristic).

Once you have ruled out a publication being in a trash journal, reputation is only a very minor factor in consideration, and methodological and substantive issues are what matter.