Children learn languages from their environment. One of the central topics of this research unit is to investigate how children extract patterns from their ambient language. We know from research in various areas of acquisition – such as, for instance, vocabulary learning, acquisition of morphology, and of complex syntax – that the distribution of the input has a direct impact on the learning speed of a child. In a comparative study of Russian, German and English child directed speech (Stoll, Abbot-Smith & Lieven, 2009), we looked at the initial strings of the utterances that children hear in order to investigate how much repetition they are exposed to. We found a very high degree of repeated word-strings in child-directed speech in all three languages, which differ significantly in their word order patterns and morphology. This suggests that there is much more regularity and repetition in the input than previously assumed. Currently, we are testing how far these repetitive patterns extend to other parts of utterances in a variety of typologically different languages.
So far, most research has focused on the language directed to children. Child-directed speech, however, constitutes only a small percentage of the utterances that a child hears every day, at least in most cultures. Usually, most of the utterances that a child is exposed to are addressed to other people, with the child overhearing these interactions. The impact of this linguistic input has hardly ever been investigated and, as in the great majority of available corpora the recording situations were confined to mother-child interactions, it can often not be studied at all.
Our research focuses on natural recordings of children learning Chintang. These recordings include all the people surrounding a child in their natural environment. This research is conducted in cooperation with Elena Lieven from the Department of Developmental and Comparative Psychology at the Max Planck Institute for Evolutionary Anthropology, Leipzig.