Poster #199 - Infant responses to foreign lullabies differ from responses to other foreign vocal music

• In Event: 2-163 - Poster Session 08
  In Poster Session: PS 08 Section - Social Cognition

Integrative Statement

Adult listeners accurately identify the social functions of songs on the basis of their musical forms, even when unfamiliar with their cultures of origin (Mehr et al., 2018, Current Biology). This could be because adults have a great deal of shared musical knowledge, especially from the widespread availability of music on the internet, or it could be an innate feature of the human mind. If the latter explanation is true, then young infants, who have far less musical experience than adults, should also be able to detect a link between form and function in vocal music.

Here, we tested the responses of infants to lullabies and non-lullabies recorded in 16 small-scale societies from around the world. We predicted that infants would respond differently to infant-directed songs, despite their relative unfamiliarity with world music. Example stimuli are presented in Fig. 1. First, infants (N = 38, age 2.3–13.3 months) viewed two animated characters, presented simultaneously on opposite sides of the screen in silence. Next, each character lip-synced to an audio recording of a lullaby or non-lullaby. The songs were matched on the basis of adults’ ratings of the perceived infant-directedness of the song and the perceived gender of the singer. After hearing both songs, the characters were presented once again, side-by-side and in silence, to evaluate infants’ selective attention to each character. We continuously measured infants' heart rate and gaze to each character. The study and analysis plan were preregistered.

While each character was singing, we found a significant reduction in heart rate during lullabies relative to non-lullabies (d = .31, p < .05). Despite this difference in response to foreign lullabies vs. non-lullabies, infants did not display any selective attention toward one singer or the other (all selective attention trial ps > .5). These findings suggest that infants do hear a difference between the forms of infant-directed and non-infant-directed songs, causing a detectable physiological relaxation response. However, they did not use this information to make a social inference about the characters producing those songs, as measured by an adapted version of a common selective attention measure (Kinzler, Dupoux, & Spelke, 2007, PNAS) that has previously been implemented with musical stimuli (Mehr, Song, & Spelke, 2016, Psychological Science).

In a separate task, we showed parents the same animated characters that infants viewed and asked them to choose the character whose song they would prefer to sing to their own infant if he or she were fussy. Parents exhibited a strong preference for the character who sang infant-directed songs (median 4 of 4 songs, IQR: [3, 4], p < .0001 relative to chance), conceptually replicating previous adult results in a sample of parents.

These findings show that both parents and infants respond differentially to infant-directed songs from foreign cultures relative to vocal music that is not infant-directed, providing initial support for the innateness of form-function links in song previously reported in adults.

Authors

• Constance May Bainbridge, Harvard University
  Non-Presenting Author

• S Atwood, Harvard University
  Non-Presenting Author

• Feng Xing, Harvard University
  Non-Presenting Author

• Mila Bertolo, Harvard University
  Non-Presenting Author

• Kelsie Lopez, Harvard University
  Non-Presenting Author

• Alma Bitran, Harvard University
  Non-Presenting Author

• Julie Youngers, Harvard University
  Non-Presenting Author

• Samuel Mehr, Harvard
  Presenting Author