For example, picture the upward swing in pitch that our voices take toward the end of a question (“Do you want to go to the park today?”): It’s much more dramatic when we address young children than adults.
While parents may feel a bit silly using baby talk, they shouldn’t: Babies not only prefer listening to these exaggerated contours, but they also learn new words more easily from them. By highlighting the structure of speech, such as the differences between the vowels “a” and “o,” motherese helps babies translate a torrent of sound into meaningful units of language.
Although scientists know a lot about the changes in rhythm and pitch in infant-directed speech, we know much less about the role of timbre, or tone color, which includes the breathiness, roughness or nasality in a voice.
The timbre of an instrument (whether buzzy, warm or twangy) clearly affects how we experience music, but its role in language is less obvious. When my colleagues and I looked into the tone color of baby talk, we made some surprising discoveries. Mothers change their overall timbre when speaking to babies, almost as if they’re morphing their voice into a different instrument to address these unique little listeners.
Timbre is a complex acoustic feature that helps us distinguish the unique flavors of sounds around us. For example, Barry White’s silky-smooth voice sounds different from Leonard Cohen’s gravelly one or comedian Gilbert Gottfried’s nasally one even if they’re all singing the same note. Contorting the shape of your vocal tract (which goes from your vocal cords all the way up to your lips) results in different resonances, allowing celebrity impersonators and voice-over artists to change their overall timbre.
When an orchestra tunes up, all instruments play the same pitch (A440 hertz), but we can still hear their distinct textures (timbres): breathy woodwinds, buzzy brass, mellow strings, etc. As soon as you hear a snippet of music, you can easily tell whether it’s classical, rock or country mainly because of its instrumentation, or collection of timbres.
Because timbre refers to a more complex collection of features than pitch, rhythm or volume, it is a less well-understood property of sounds. But we do know that timbre provides an important pointer to different sound sources, thus helping us identify people, animals and objects based on their characteristic auditory “fingerprints.” So, we wondered whether mothers might unconsciously change their overall fingerprints when talking to their babies, perhaps to signal that an important source of speech, which is highly relevant for learning, is coming their way.
In a new study published in Current Biology, we report for the first time that mothers shift their overall vocal timbre when speaking to their infants.
In the Princeton Baby Lab, where researchers study how children learn, we recorded English-speaking mothers while they played with and read to their 7-to-12-month-old infants, and while they spoke to an adult experimenter. (To keep the overall pitch range fairly consistent across participants, we tested only mothers, but we’d like to see whether the results generalize to fathers and other caregivers in the future.)
We then came up with a mathematical formula for the timbre fingerprint of each mother’s voice and found that adult-directed and infant-directed speech had consistently different fingerprints.
Specifically, we were able to train artificial-intelligence software to distinguish infant- and adult-directed speech, even when we fed the software just one second of speech.
Most surprising, in a second sample of non-English-speaking mothers, we found that this timbre shift was also highly consistent across nine diverse languages (Spanish, French, Russian, Polish, Hungarian, German, Hebrew, Mandarin and Cantonese). This suggests that these timbre shifts may represent a universal form of communication with infants.
Being able to identify baby talk across multiple languages could give us rich information about the amount and type of language children hear at home and at preschool (for example, overheard adult conversation vs. speech directed at them) across different cultural environments. This could help researchers and educators predict and improve outcomes such as vocabulary and success in school. Our framework could also lead to new research avenues on how speakers adjust their timbre when speaking with, for example, friends, bosses, political constituents and romantic partners.
Because the mothers in our study were never specifically told we were measuring the acoustical properties of their voices (let alone their timbre) — they just knew we were broadly interested in their interactions with their babies — these shifts are highly natural and possibly unconscious. And given how well they generalized across the diverse group, it’s likely that they are an important cue used to capture babies’ attention and help them learn language.
Hopefully this is reassuring for parents feeling self-conscious about their own baby talk: However strange you may think you sound, chances are you’re helping your baby learn, and you’re in good company.
Piazza is a postdoctoral fellow at the Princeton Neuroscience Institute. Marius Catalin Iordan, a postdoctoral researcher at the institute, and Casey Lew-Williams, an assistant professor in the university’s Department of Psychology, contributed to this report.