Reproducible real-EEG workbench for exact language-model surprisal, human N400 effects, held-out alignment, and causal auditing.
-
Updated
Jul 28, 2026 - Python
Reproducible real-EEG workbench for exact language-model surprisal, human N400 effects, held-out alignment, and causal auditing.
Contextual incongruency triggers memory reinstatement and the disruption of neural stability (2023 NeuroImage, Xiongbo Wu)
Modeling the N400 ERP with LLM surprisal — GPT-2 through GPT-Neo-2.7B
To associate your repository with the n400 topic, visit your repo's landing page and select "manage topics."