If someone has modified their appearance in a way that appears foolish and their intent was to signal intelligence then I my prior would be that they are extra foolish.

Instead if they know Latin then my prior assumption would be that they are intelligent.

Acquiring information is costly (multi armed bandit) which is why signaling is important. Perhaps in the future with AI and attestations we can replace expensive signals with cheaper ones that are more difficult to fake and thus more honest.

With the improvements in machine intelligence I have become much less concerned with human intelligence and only persist with disagreeing with netizens out of a quickly fading residual sentimentality.