My primary research interests are in adult/child sentence processing and first language acquisition of morphosyntax. Sentence processing and language learning share the same problem: The target abstract structural representations must be identified despite the fact that a) input provides no direct evidence for the target syntactic representations, and b) there are many structural/grammatical hypotheses compatible with the input. Globally speaking, I'm interested in how these identification processes are constrained by linguistic knowledge as well as cognitive mechanisms like attention and memory.
To approach this problem, I focus on the development of sentence processing mechanisms, as it provides useful insights on theories of both sentence processing mechanisms and language acquisition.
1. The development of child parser and its implication to adult parsing:
My lab investigates not only how adults understand and produce language, but also how the parsing mechanism develops over time. There are several reasons why this is useful for understanding sentence processing mechanisms. First, children sometimes show much more dramatic failures in sentence comprehension, which allows us to probe the details of parsing constraints that may be masked in adults' efficient processing. Second, we may be able to inspect the relation between grammatical knowledge and parsing much more directly; given that children are learning grammatical knowledge, we can see how the development on the grammar side affects the parsing behavior (and vice versa). Finally, parser development can shed light on the nature of syntactic priming phenomena. A recent prominent theory of syntactic priming suggests that it may reflect a long-term adaptation of structure building procedures that continue throughout life. However, most of the demonstrations come from ambiguity resolution experiments with adults, and my work aims to fill this empirical gap by exploring whether adults and children use distributional information in the same way to alter their parsing behaviors.
2. Child parsing and language acquisition:
Recent research on language acquisition has emphasized that children are sophisticated 'data analysts' who can use statistical regularities in the input to infer properties of the target language. While this line of research remains important, this perspective often leaves aside the fact that children also need to be 'data collectors' by perceiving and encoding the input with their own parsers, as the input must be converted to mental representations (called 'intake') that can feed learning processes. However, children's parsers are immature and often mis-parse sentences in a non-adult-like way, which indicates that the input distribution may be skewed and not be veridically represented in their mind (Omaki & Lidz, 2014). In this sense, understanding the child parser is critical for understanding the process of language acquisition. This line of work will eventually have implications for grammatical theories, because understanding the nature of input/intake available to children should constrain theories of what needs to be innately given to children.
I investigate these questions in my language processing and development lab, primarily using behavioral experiments and eye-tracking techniques to examine listening or reading comprehension in adults and children. We have two Eyelink 1000 eye-trackers (SR Research) in the lab, one for adults readign studies, and the other is a remote eye-tracker that can be readily used with children. I also collaborate with my colleagues abroad (especially Japan and Switzerland) to conduct cross-linguistic comparative research on sentence processing and language acquisition.
Occasionally, I also work on experimental syntax, as well as second language acquisition / bilingual sentence processing.