Neural network modelling of brain responses during language comprehension (B09*)

• Subject Area: Mathematics
• Term: since 2021
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 318763901

Deep neural network (DNN) models perform increasingly well on complex tasks including language comprehension. The proposed project aims to investigate whether DNN models of language provide good models of human language comprehension. This will be tested by comparing internal representations and dynamics in DNN models of language comprehension with internal representations and dynamics in humans as inferred from brain activity during comprehension. We will also investigate the dependency of the DNN model performance on different approaches to their training, i.e. assimilation of language data.

• DFG Programme: Collaborative Research Centres
• Subproject of: SFB 1294: Data Assimilation – The Seamless Integration of Data and Models
• Applicant Institution: Universität Potsdam
• Project Heads: Professorin Dr. Milena Rabovsky, since 7/2021; Professor Dr.-Ing. Sebastian Reich, since 7/2021