A single formant explicates the ubiquity of “meow”

Across languages, the species–typical vocalization by domestic cats (Felis catus silvestris) is transcribed similarly, typically corresponding to [miau:] or [wau:]. Such consistent and ubiquitous cross-linguistic transcription is apparently onomatopoetic. However, in humans, these qualities make unique use of the tongue; in comparison, most nonhuman mammals do not appear to employ their tongues while vocalizing. The purpose of this work was to explore whether tube models modeled after the buccolabial oral tract morphology of the domestic cat, may be used to reverse engineer the apparent diphthong-like quality typically perceived in cat meows (the “au” in meow). For cats specifically, the short vocal tract is likely a causal factor, as the contribution of higher formants to vowel quality in the front-to–back dimension is significantly reduced. Results of computational models and perception tests suggest that the shift in apparent vowel quality may be driven by F1, corresponding in our model to raising of the mandible.