Modeling language processing at multiple timescales
多時間尺度的語言加工
Presenter: Dr. Jixing Li (李吉星)
This talk focuses on language processing in both naturalistic and experimental settings, from a longer narrative to a shorter two-word timescale. The first part of the talk examines referential processing, a phenomenon that highlights long-distance dependency in language processing. Five computational models of pronoun processing were compared against fMRI BOLD signals while both English and Chinese participants listened to a story. The results support a memory-based model that incorporates frequency and recency effects during pronoun resolution. In the second part of the talk, I used controlled experiments to understand a single mental operation–meaning composition. I show a functional dissociation within the left temporal lobe for compositional and distributional semantic processing.
Bio
Dr. Jixing Li is currently a postdoc at the Neuroscience of Language Lab (NeLLab) at New York University Abu Dhabi. She will be joining the Department of Linguistics and Translation and the Department of Social and Behavioural Sciences at City University of Hong Kong as an Assistant Professor in August 2022. Jixing received her PhD in Linguistics from Cornell University in 2019. Her research aims to connect linguistic representations and computations with human neurobiology. She applies advanced statistical and machine learning techniques to test competing psycholinguistic theories against brain data. She has expertise in various neuroimaging techniques, including fMRI, MEG, and EEG, and she integrates theoretical and analytical insights from linguistics, natural language processing (NLP) and cognitive neuroscience in her work.
Moderator:Dr. Fei Gao (高飛)
(CCBS/FAH, University of Macau)
Language: English
Time: 7:00 PM, April 29th (Beijing Time)
Zoom ID: 959 6270 3011
Passcode: ccbs
Acknowledgements
Scan the QR code to sign up as Presenter! 
Or contact Mr. Gao Fei ([email protected]) for inquiries.
继续阅读
阅读原文