I don't really see how it challenges any theory of language evolution. The bouba kiki effect is hardly necessary nor important for having language.
This is downvoted, perhaps as a 'lazy dismissal'? But I read the SciAm article and I don't think it actually explained this point.
The finding seems to be that the bouba-kiki effect is not specific to humans and does not depend on experience. And the previously-existing theory is presented like so:
> scientists have considered [the bouba-kiki effect] a clue to the origin of language, theorizing that maybe our ancestors built their first words upon these instinctive associations between sound and meaning.
The finding is supposed to undermine, or at least challenge, the theory. But why? Is the point just that, if other species also have the bouba-kiki effect but do not have language, the bouba-kiki effect probably doesn't play as important a role as we thought? That seems to be the implication (though the innate/learned distinction also seems to be relevant, and I'm not sure why that is) -- but surely the bouba-kiki effect was never believed to be anything like a sufficient condition for the development of language, was it?
I fonud the following Wikipedia section somewhat useful for understanding the context here: https://en.wikipedia.org/wiki/Bouba/kiki_effect#Implications...
Annoyingly, very few of the citations are accessible. But the gist as I understand it is that this challenges various proposed explanations for the Bouba-Kiki effect, for example that it relates to human physiology or has a human-specific neurological basis.