The effect replicates in languages with other writing systems.

That seems to me like it just shifts the problem one level. Why are K's and Kikis spiky and why are B's and Boubas round. Why is it universal too across people with different writing systems and languages.

I can think of so many counterexamples on the spot: Latin, Arabic, Hebrew, Japanese syllabaries. It's really only the Greek Kappa that is spiky.

In Telugu, k is one of the smoother letters: కి (ki: the squiggle at the top is the i vowel sign).

I think it is related to the physics of the mouth producing the sound, and we do a form of synesthesia: doing b-u-b seems (to me) quite smoother of a transition than k-i-k. If I stop blowing the u sound my lips close again; when I finish the i I have released the muscles and I need to hold again for the next k. It al feels more sudden an explosive with k. Also the b sound you voice it (otherwise it would be p).

There is a pre-print which claims to have found this effect in 3 day old chicks. Given three day old baby chickens are not renowned for their literacy skills, it would point to some much more deeper origin.

https://www.biorxiv.org/content/10.1101/2024.05.17.594640v1....

https://www.science.org/content/article/scienceadviser-even-...