The role of parental speech on infant language development: Insights from Norwegian language

When talking to their infants, parents, spontaneously, use Infant-Directed Speech (IDS): they lengthen sounds, exaggerate and vary their voice pitch, and hyper-articulate sounds. Previous research has claimed that IDS is universal and parents use it to facilitate early language development in their infants. Yet, a number of limitations in previous research (e.g., uncontrolled acoustic processing of sounds, poor audio sample sizes) and the results of recent advanced acoustic analyses of IDS challenge these interpretations and raise doubt about facilitating role of IDS in language development. One of the main limitations is that the overwhelming majority of the studies on IDS was done on American English parents, who tend to produce most extreme voice exaggerations, as compared to other languages, and does not use pitch or sound lengthening to distinguish words, meaning that parents can vary these two parameters in IDS without affecting word meaning. What about other languages, e.g., Norwegian, that do vary pitch (single tone hender-hands versus double tone hender-happens) and lengthening (short vowel takk versus long vowel tak) to convey meaning? Would Norwegian-speaking parents enhance them more while talking to infants? How does variation in IDS relate to infants' language skills? Finally, if parents do not exaggerate pitch or hyper-articulate while talking to infants, what other cues facilitate language learning in Norwegian infants? The current study addresses all of the above questions and examines, longitudinally, the acoustic properties of Norwegian IDS and their role in speech sound discrimination, word comprehension and word production in 6 to 18-month-old Norwegian-learning infants. Infants' language skills will be tested, in an Eye-tracking paradigm, at the ages of 6, 9, 12, 15 and 18 months. The project represents a unique opportunity to examine, longitudinally, infant-directed speech in Norwegian (so far understudied) and to examine its role in early language development. We are happy to announce that the longitudinal data collection over a period of one year is over! In July 2022 we finished the last bunch of tests with 18-month-old infants. We are very glad to report that out of 70 infants in the initial sample - when infants were 6 months of age - we were able to retain 62 in the final sample - when infants were 18 months of age. The drop rate, therefore, was extremely low in our study. We are very thankful to the participating families. Given that the data collection has been over, we invited all project members (including our expert in developmental research Alex Cristia from l'ENS, Paris and a statistician Roger Mundry, Germany) for a one week intensive meeting, where we preregistered two big studies that will come up as papers from the project. One will be on the longitudinal changes in the acoustics of Norwegian parents talking to their 6-18-month-old infants and another on the effects of parental speech on infants' lexical development over a period from 9 to 18 months of age. Finally, in September, we organised a Workshop on the Role of Input in less studied languages, where researchers from the University of Oslo, but also many other Norwegian Universities gave talks on how infants and children acquire language in non-WIERD countries. There were more than 60 attendants. By the end of 2022, the data that we have collected is being processed and analysed with the help of a speech engineer (Post-Doc on the project) and 2 research assistants.

When talking to infants, parents, intuitively, modify their adult speech: they slow it down, exaggerate and vary more the pitch of their voice, but also they hyper-articulate the corner vowels /i/, /a/ and /u/ (as in sheep, shark, shoe). Commonly called INFANT-DIRECTED SPEECH (IDS), it is believed to be universal and is claimed to attract infants’ attention and to boost early language learning. However, a number of limitations in previous (mostly in English-learning infants) research and the results of recent advanced acoustic analyses of IDS challenge these interpretations and raise doubt about the facilitating role of IDS in language development. The current project carefully examines all these claims, in a longitudinal design, and assesses, for the first time, the role of IDS in language development of Norwegian-learning infants. Norwegian, unlike English, uses prosody (hender "hands" vs hender "happens") and lengthening (takk "roof" vs tak "thank") to convey meaning, suggesting that remarkable variation is already present in adult-directed speech. First, we will assess the 'universality' hypothesis by examining the acoustic quality of speech in Norwegian-speaking parents interacting with their 6 to 18 month old infants. Second, we will examine the 'preference' hypothesis by assessing, longitudinally, infants' preference for IDS over the adult speech, at the ages of 6 and 12 months. Third, we will perform state-of-the-art acoustic analyses of parental speech and will identity those parameters that predict infant performance in sound discrimination and word comprehension tasks (from 6 to 12 months of age). Finally, we will examine what specific parameters of parental speech predict the size of the production vocabulary when infants are 15 and 18 months of age. The project represents a unique opportunity to examine, longitudinally, infant-directed speech in Norwegian (so far unstudied) and to examine its role in early language development.