Some people are saying the mask reversal is an indication that reputation heuristics can't work, because the authorities are flawed. Honestly, the opposite is true.

A fundamental heuristic is to zoom out and see what others "in the know" say. At a scan you look at whether there is broad consensus or significant disagreement. For example, earth is heating up = broad consensus, geoengineering is part of a solution to that = very disputed.

And when you do that with masks you can see that there's been broad disagreement for a while. There are entire countries that recommend the practice, often countries that have a lot of experience in this stuff. And there's other authorities that say, nope, doesn't work.

In a way, it's a perfect example of why reputation heuristics around claims are so valuable. The lack of consensus around masks is easily distinguishable from consensus truth (washing your hands helps) and theories rejected by experts (novel coronavirus is man-made)

Did it hurt that some public figures expressed opinions on this without a nod to the debate? Yes. But that's precisely why "zooming out" is such a crucial skill.

As far as the credibility issue, I'd say, at least in this case, that any person with some basic training in how to think through such things would realize the mask claim always had a different status than the hand-washing claim.

If you get that, it's not going to feel like the rug was pulled out from under you by this reversal.

Not all disputed/undisputed claims are this easy to spot, there's lots of examples that are more problematic. But this particular one is an example of why these heuristics work.