Infants relax in response to unfamiliar foreign lullabies

Abstract

Music is characterized by acoustic forms that are predictive of its behavioural functions. For example, adult listeners accurately identify unfamiliar lullabies as infant-directed on the basis of their musical features alone. This property could reflect a function of listeners' experiences, the basic design of the human mind, or both. Here, we show that US infants (N = 144) relax in response to eight unfamiliar foreign lullabies, relative to matched non-lullaby songs from other foreign societies, as indexed by heart rate, pupillometry and electrodermal activity. They do so consistently throughout the first year of life, suggesting that the response is not a function of their musical experiences, which are limited relative to those of adults. The infants' parents overwhelmingly chose lullabies as the songs that they would use to calm their fussy infant, despite their unfamiliarity. Together, these findings suggest that infants may be predisposed to respond to common features of lullabies found in different cultures.

Fig. 1 |

Structure of the experiment. Infants viewed videos of animated characters who either appeared in silence (during preference trials) or who sang the songs one at a time, next to a distracting animation of slowly-moving colored boxes.

Fig. 2 |

Lullabies reduce infant heart rate. a The points depict mean trial-wise heart rates, normalized to the previous 14 s trial (regardless of its type), for each infant (N = 142), with the gray lines indicating the pairs of points that represent the same infants; the violin plots (coloured areas) are kernel density estimations; the horizontal black lines indicate the means across all participants; and the shaded white boxes indicate the 95% confidence intervals of the means. The points are jittered to improve clarity. Heart rates were reduced during lullabies (the mean z-score was negative and significantly different than 0 (in z-scores, M = −0.15, SD = 0.43, 95% CI [−0.23, −0.08]; t(140) = −4.28, p < .001, d = 0.36), denoted by the horizontal dotted line), relative to the previous trial, but no such effect was found for non-lullabies (M = −0.01, SD = 0.4, 95% CI [−0.07, 0.06]; t(139) = −0.21, p = 0.83, one-sample two-tailed t-test). Within-infants, heart rate during lullabies was significantly lower than during non-lullabies (t(138) = −2.75, p = 0.007, paired two-tailed t-test). b An analysis of heart rate over time, averaged across all trials, shows that while heart rate drops initially in all singing trials, the drop is more pronounced in lullabies, driving the overall effect. The lines and confidence bands are from a generalized additive model that does not account for nesting. ***p < .001; **p < .01

Fig. 3 |

Pupil dilation is reduced during lullabies. Collapsing across all singing trials, pupil size was lower during lullabies than non-lullabies (t(3086) = 2.507, p = 0.012, β = 0.089), in the subset of the participants studied (N = 30). The blue and red lines and confidence bands are from a LOESS regression that does not account for nesting.

Fig. 4 |

Lullabies attenuate increases in arousal. The black dotted line denotes the expected rise in electrodermal activity during a trial, from a linear model (N = 25,938 observations from 144 infants, t(25819) = 38.5, p < .001, β = 0.002). This rise is attenuated during lullaby trials but not during non-lullaby trials, such that the expected level of electrodermal activity by the end of a lullaby trial is reduced (expected difference at time = 14 s; β = 0.075, 95% CI [−0.098, −0.052], χ² = 289.2, p < .001, d = 0.25).

Fig. 5 |

Parents prefer foreign lullabies to non-lullabies for soothing their own infants. The histogram displays the distribution of parents’ choices of whose song (lullaby-singer or non-lullaby-singer) they would prefer to sing to their own infant (if the infant were fussy and if they knew how to sing both songs; N = 135 parents of our 144 infants participated in this follow-up survey). Parents made this choice four times, so the maximum number of lullaby-singer choices was 4. The dashed line indicates chance level of 2 choices. Parents almost always chose the lullaby-singer (interquartile range: 3–4; z = 9.89, p < 0.001, Wilcoxon two-tailed signed-rank test).