Very young children can acquire a foreign language faster with trained native speakers using ‘parentese’ in a highly social context, a study suggests.
The research is the latest in a series of studies looking at how small children learn by picking up the separate sounds of language. The findings also raise hope that all infants, regardless of socioeconomic status, can acquire a language given the right conditions.
‘Our study has direct implications for designing programmes for kids who need to learn English’, co-author Naja Ferjan Ramirez told the Gazette. ‘We are very excited from that perspective’.
For a period of eighteen weeks, the study compared the progress in Spanish and English of two groups of children (from 7 to 33 months old) in an infant bilingual programme in Madrid, Spain.
The control group followed the centre’s usual English language development programme, which included two hours of weekly instruction with a bilingual teacher. The experimental group experienced hour-long daily sessions with a native speaker tutor, where the focus was on developing a highly social and interactive learning context. Tutors involved children in games and addressed them using parentese, infant-directed speech with higher pitch and slower tempo. Tutors were undergraduates or recent graduates from the University of Washington who underwent a two-week training period.
By the end of the eighteen weeks, the intervention group scored significantly higher in English. From a baseline of 12 English vocalisations recorded per hour for both groups, the control group averaged at 13 after six weeks, while the experimental group jumped to 74. The complexity of the language used by the children of the experimental group also increased: their phrases lengthened from 2.2 words to 2.8, whereas the control group stayed at 2.2. In the meantime, both groups developed their first language, Spanish, at the same rate. Re-tested eighteen weeks later, the experimental group still retained what they had learnt. The results showed this was entirely independent of the children’s socio-economic background.
When asked why the study employed native-speaker tutors, Dr Ferjan Ramirez said: ‘We are looking at very young children, and there is quite a bit of research showing that children at this age are very sensitive and detect the subtle phonetic differences between native and non-native speakers.’ Learning a second language at this age, she explained, unfolds similarily to the learning of your first language: it starts with individual sounds, then it goes to syllables, then words. The study also reports that parentese from late bilinguals can be more inconsistent across phonetic contexts than parentese of native speakers – and babies pick up on those differences.
In this study, this was also a way of keeping all conditions equal by exposing children to just one variety of English . ‘It was only an initial exploration, and we wanted to keep the study environment as consistent as possible,’ Dr Ferjan Ramirez said. But how much accent is too much for children at this age? Dr Ferjan Ramirez said that further research is needed, but she recommends that parents speak to their children in their native language, as long as this is the language in which they are the most fluent and feel at ease. ‘Speaking a language in which parents are comfortable allows them to have the best and most natural interactions with their children’. There are studies, she explained, that show that speaking English to your children even if you have an accent and are not fluent does not necessarily benefit the child’s English development, but can slow down the development of the other language.
‘The situation may be different for older children who already have a first language and can use it as a base to scaffold their second language learning, but in this case the learning process is different’. However, she adds, ‘this study was not designed to test how crucial the use of native speech is.’
‘We did not set out to answer that question in this initial study, and there were other factors at play, such as parentese, that were not equal between the two groups’.
Another factor was social interaction. In a previous study by the same team, 9-month-old babies were exposed to twelve sessions of Mandarin Chinese. One group was taught by a tutor, while the second group received the same input through television and the third through radio.
After the intervention, the babies in the radio and television group were just as good as the baseline group who didn’t receive any input. The other group could discriminate Chinese sounds in the same way as babies growing up in China.
But when does this ‘unique window of opportunity for language learning’ close? ‘We don’t have an answer’, said Dr Ferjan Ramirez. ‘The critical period for language learning is a well-known phenomenon and it has been studied for years. In terms of learning about the sounds of language, it really is the very early infancy’.