Infant physiological and behavioural responses to parental song

• In Event: The influence of maternal communication on infants’ biobehavioral state

Abstract

Caregivers around the world sing to their infants, and adopt a specific infant-directed style when doing so. Infant-directed song is more emotional, higher and more variable in pitch, and slower in tempo than self- or adult-directed song. Infants respond with interest and attention to such vocalizations. In everyday listening an infant hears song in a variety of contexts – they hear soothing songs at nap time, and exciting songs during play. Sometimes songs maintain their positive mood, and sometimes songs cheer them up when they are distressed. Songs may be highly familiar and socially meaningful, or completely novel. In short, the infants’ musical world is highly variable, yet little is known about how these contexts influence infant attention, emotional responses, and physiological responses. This talk will present results from a series of experiments exploring these questions by analyzing both infant behaviour (i.e. looking time, affect) and physiology (skin conductance).
In experiment 1, we asked 30 mothers to sing Twinkle Twinkle Little Star to their infant, alternating between singing in a soothing way and in a playful way. While playful renditions captured infant attention (p < .001) and resulted in stable skin conductance levels, soothing song reduced infant arousal (p = .03) and correlated positively with maternal arousal levels (p = .004).
In experiment 2, 140 parents completed an adapted Still-Face procedure. They would first play with their infants for 2 minutes, and then when cued, became unresponsive for up to 1 minute or until infants displayed distress for 10 seconds. Following this, mothers were asked to use speech, a song familiar to their infant, or a song unfamiliar to cheer their infant up. While song (familiar or unfamiliar) was generally more effective than speech at capturing infant attention (p < .001) and reducing arousal (p = .003) and negative distress (p < .001), familiar songs were the most effective.
In experiment 3, we used a head-turn preference procedure to explore infant responses to recordings of their own mother and a stranger (another participants’ mother) singing both familiar and unfamiliar songs (n = 28). Infants listened longer to familiar songs regardless of singer identity (p = .003, see Figure 1). This robust preference for familiar songs is especially interesting given the wide range of tempos and musical keys in the recordings. Post-hoc exploratory analyses reveal preliminary evidence that infants may prefer their own mother’s rendition of familiar songs when the stranger rendition deviates greatly in tempo, p = .044 (no similar effects for pitch deviations). Infant physiological responses do not differ across song and singer familiarity conditions, but may be related to the mean fundamental frequency of the song recordings (p = .002).
Together, these studies highlight the diversity of infant responses to song, depending on song features and social context. This diversity in responses is consistent with the multifaceted functions of music throughout the lifespan, including emotion regulation, social cuing, and group bonding.

Authors

• Laura K. Cirelli, University of Toronto
  Presenting Author

• Haley Kragness, University of Toronto
  Non-Presenting Author