Gintare Grigonyte
Gintare Grigonyte

Gintare Grigonyte is a researcher at the Department of Linguistics, Stockholm University. Her research aims at providing insight into the nature and structure of natural human language by applying computational techniques. Her research up to date has focused on several topics: automatic extraction of domain terminologies and semantic relationships, multilingual lexicography through information retrieval, biomedical NLP and language evolution, and NLP for first and second language acquisition.

Using computational linguistic models is beneficial for descriptive linguistics and psycholinguistics. We will look into models applied to various English genres and learner language: 1) surprisal and 2) a syntactic parser, allowing to investigate the role of ambiguity and the interplay between idiom and syntax principles.

The findings show that surprisal and ambiguity are higher for learner language, while parser scores and model fit are lower. The main goal is to show that these statistical models are global models of language processing. They model, on the one hand, the unexpectedness of the continuation of the discourse, and, on the other hand, the amount of ambiguity which needs to be resolved for analysis and interpretation. Failures to generate optimal sequences in language production, such as nonnative-like utterances by language learners exhibit, increase processing load, both for human and automatic processors.