Language acquisition

I'm interested in first language acquisition. Children can learn language surprisingly easily and efficiently. I'm devoted to exploring how children learn, with a dynamic perspective combining statistical learning, social interaction, and the development of language-related cognition. 

Currently my primary research topic is:

  • Children's development of early multiword utterances 

Children’s first utterances are short and often ungrammatical. The growth in utterance length bears on fundamental issues in language acquisition, such as when children develop genuinely syntactic representations and what roles are played by cognitive and linguistic constraints on development. I investigate the quantitative and qualitative development of children’s early multiword utterances. I use CHILDES corpus and computational modeling to explore the underlying mechanism that motivates utterance length development. With language model and human subjects experiment, I also study the syntactic structure of children's very early utterances. 

Secondary research topic includes:

  • Number word learning.

  • The impact of parent-child interaction on children's development on combinatorial speech.

Language processing through the lens of network science

Network science has become an important domain and has been increasingly applied to understanding language representation and processing. The methodology is powerful in capturing not only the global architecture of a complex system as a whole, but also the detailed interaction patterns between different pieces of information. I use network science approach to understand semantic organization as well as language processing. 

Currently my primary research topic includes:

  • Modeling bilingual lexicon through code-switching speech

   Bilingual speakers could frequently switch between two languages with any cognitive cost. How do bilinguals organize semantic lexicons of the two languages during the spontaneous code-switching speech? What drives the language choice during code-switching? I use network science analysis to address those questions and highlight the unique contributions of the interconnections and the interactions between words to speech production. 

  • The large-scale structure of semantic networks across different languages