Thursday January 25th, 2024
Radboud University
Elinor Oostrom building
Room EOS N 01.150
Nijmegen, the Netherlands

Human language is studied by linguists, psychologists, and neuroscientists alike. However, language researchers do not always look for ways to bridge between these disciplines. How can we use computational modelling to link levels of description of the language system? This one-day workshop is of interest to all researchers who apply computational methods to investigate the structure, use, or neurobiology of language — and in particular to those who aim to bridge between the fields of neuroscience, psychology, and linguistics.

10.00	Welcome and coffee
10.10	Alessio Quaresima	Dendrites support the formation and recall of lexical memories
10.35	Martijn Bentum	Listening with great expectations: An investigation of word form anticipations in naturalistic speech
11.00	Benedict Schneider	The predictive power of language model surprisal for N400 and P600 effects
11.25	Coffee break
11.40	Anna Langedijk	(How) can Transformers learn generalizable logic from data?
12.05	Sandro Pezzelle	A psycholinguistic analysis of BERT's representations of compounds
12.30	Lunch
13.30	Olivia Guest	Keynote: On Logical inference over brains, behaviour, and artificial neural networks
14.15	Claire Stevenson	Do large language models solve analogies like children do?
14.40	Marianne de Heer Kloots	Do self-supervised neural speech models show human-like generalization patterns in speech perception?
15.05	Coffee break
15.20	Raquel Fernández	Measuring utterance predictability as distance from plausible alternatives
15.45	Iza Škrjanec	Adapting language models to readers' expertise improves the fit to reading times
16.10	Jakub Szewczyk	Looking for prediction error representations in visuo-linguistic integration
16.35	Hugo Weissbart	Linear forward models for the analysis of cortical dynamics in response to naturalistic speech
17.00	Drinks

⟡ Michelle Suijkerbuijk
⟡ Zoë Prins
⟡ Stefan Frank