What we can learn depends on what we already know; a child who can't count cannot learn arithmetic, and a child who can't segment words cannot identify properties of verbs in her language. Language acquisition, like learning in general, is incremental. How do children draw the right generalizations about their language using incomplete and noisy representations of their linguistic input? In this talk, I'll examine some of the first steps of syntax acquisition in one-year-old infants, using behavioral methods to probe their linguistic representations, and computational methods to ask how they learn from those representations. Taking argument structure as my case study, I will show: (1) that infants differentiate core clause arguments like 'subject' and 'object' when learning verbs, (2) that infants develop the ability to recognize when arguments have been displaced in "non-basic" clauses like wh-questions, but do so only after learning local argument dependencies, and (3) that it is possible for infants to draw accurate generalizations even when they cannot yet recognize displaced arguments, by learning to treat some of their data as signal and some of their data as noise for early grammar learning. I will argue that the approach I take for studying this particular learning problem will generalize widely, allowing us to build new models for understanding the role of development in grammar learning.
January 29, 2020
Guest speaker: Laurel Perkins (École normale supérieure)
Our department is pleased to welcome back Laurel Perkins (BA 2010), now a postdoc at l'École normale supérieure in Paris. Laurel earned her Ph.D. from the University of Maryland in 2019, focusing on the earliest stages of the L1 acquisition process, especially with respect to syntax and semantics. Her talk, "How to grow a grammar: Syntactic development in one-year-olds", will be taking place on Friday, January 31, at 3 PM in SS 560A.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment