Modelling language acquisition in children using network theory

Abstract

Research in children’s language acquisition has recently benefited from the application of network theory to large sets of empirical data, which has illuminated interesting patterns and trends. Network theory is an extremely powerful modelling and analysis tool, and its full potential in terms of extracting useful information from raw data has yet to be exploited. In the present paper, we argue that well-established network analysis techniques can, and should be applied to the study of language acquisition, in order to reveal otherwise invisible patterns. We show that a key network parameter – the ranked frequency distribution of the links – provides useful information about the data, even though it had been previously neglected in this domain.