Rhythm and modes of maternal vocal stimulation to 3-month-old infants: A comparison across cultures and immigration experience

Infant-directed (ID) registers are universally used by caregivers both when talking and singing to infants. Cross-cultural studies on vocal patterns of early mother-infant interaction have shown a main pattern of protoconversation (diachronic exchanges characterized by a considerable amount of ID-speech) in Western urban middle-class families, and a main pattern of protosongs (through which infants are geared to maternal rhythm) in traditional rural societies where relatedness is emphasized more than autonomy. However, these studies have not explored acoustic properties of maternal vocal stimulation, namely vocal rhythm and rhythmic micro-variations (‘expressive timing’), which have been indicated as fundamental aspects to sustain infant’s attention and emotional engagement, given the infants’ sensitivity to the rhythmic structure of acoustic stimuli. Still little is known about vocal rhythm in mother-infant interaction in different sociocultural contexts; to our knowledge, only two studies (Gratier, 1999, 2003) examined this in an immigration context. The present study used acoustic analysis to examine both temporal characteristics and modes (speech vs. singing) of maternal vocal stimulation during spontaneous interaction with 3-month-old infants in three sociocultural contexts: Italian middle-class families, West African (Cameroonian, Ghanaian, Nigerian) immigrant families in Italy, and Cameroonian/Nso rural families. Twenty mother-infant dyads from each cultural context were videotaped longitudinally during spontaneous interaction, from 4 to 12 weeks. For the present study, acoustic analysis was performed on audio segments of 60 s of continuous interaction with 12-week-old infants. Spectrograms and pitch plots were obtained for each segment using the software PRAAT, and maternal vocal rhythm measured in ‘beat units’ (Mdur) and ‘tempo’ (beats per min), ‘expressive timing’ (SD of beat units Mdur), durations of ID-speech and ID-singing, and within- and between-speaker pauses were assessed. Results reveal similarities across groups for both measures of maternal vocal rhythm, but wider ‘expressive timing’ for Nso mothers than for Italian and immigrant mothers. As expected, Italian mothers showed significantly longer durations of ID-speech than Nso and immigrant mothers. By contrast, both the latter showed longer durations of ID-singing than the former, although the difference was significant only between immigrant and Italian mothers. Indeed, immigrant mothers used playsongs much more than mothers in the other groups, showing to maintain the protosong mode of interacting with infants typical of their original culture. Interestingly, however, the occurrence duration of ID-singing in the immigrant group was significantly longer than in the other groups -similar to that of ID-speech used by Italian mothers- and frequently followed by between-speakers pauses, suggesting an adaptation of protosongs to the structure of protoconversations. Finally, Nso mothers showed significantly longer durations of silent time (within- and between-speaker pauses) than Italian and immigrant mothers. Overall, these findings indicate that mothers from different cultural contexts share similar rhythmic structures of vocal stimulation, supporting the hypothesis that ID-vocal rhythm is biologically rooted. By contrast, the modes of vocal stimulation (ID-speech, ID-singing, and pauses) appear to be clearly culturally specific. Finally, findings concerning immigrant mothers uncover an active acculturation process showing how these mothers synthesize vocal patterns from two different sociocultural backgrounds into a creative blend