Publication
Globally, songs and instrumental melodies are slower, higher, and use more stable pitches than speech [Stage 2 Registered Report]
Publisher:
Center for Open Science
Date:
30-11-2022
DOI:
10.31234/OSF.IO/JR9X7
Abstract: What, if any, similarities and differences between music and speech are consistent across cultures? Both music and language are found in all known human societies and are argued to share evolutionary roots and cognitive resources, yet no studies have compared similarities and differences between song, speech, and instrumental music across languages on a global scale. In this Registered Report, we analyze a novel dataset of 300 high-quality annotated audio recordings representing matched sets of singing, recitation, conversational speech, and instrumental music from our 75 coauthors whose 55 1st/heritage languages span 21 language families to find strong evidence for cross-culturally consistent differences and similarities between music and language. Of our six pre-registered predictions, five were strongly supported: relative to speech, songs use 1) higher pitch, 2) slower temporal rate, and 3) more stable pitches, while both songs and speech used similar 4) pitch interval size, and 5) timbral brightness. Our 6th prediction that song and speech would show similar pitch declination was inconclusive, with exploratory analysis suggesting that songs tend to follow an arched contour while speech contours tend to decline overall but end with a slight rise. Because our non-representative language s le and unusual design involving coauthors as participants could affect our results, we also performed robustness analyses - including a parallel reanalysis of a previously published dataset of 418 song/speech recordings from 209 in iduals whose 16 languages span 11 language families (Hilton & Moser et al., 2022, Nature Human Behaviour) - which confirmed that our conclusions are robust to these potential biases. Exploratory analyses identified additional features such as phrase length, intensity, and rhythmic/melodic regularity that also consistently distinguish song from speech, and suggest that such features also vary along a “musi-linguistic” continuum in a cross-culturally consistent manner when including instrumental melodies and recited lyrics. Further exploratory analysis suggests that pitch height is the only consistently sexually dimorphic feature (female singing/speaking is almost one octave higher than male on average), and that other factors such as musical training and recording context may also interact to influence the magnitude of song-speech differences. Our study provides strong empirical evidence for the existence of cross-cultural regularities in music and speech.