So Cloze yet so Far: N400 Amplitude is Better Predicted by Distributional Information than Human Predictability Judgements
This challenges assumptions about the role of prediction in language comprehension, suggesting it may rely more on surface-level statistics, which is incremental for cognitive neuroscience and NLP.
The study compared how well computational language models (GPT-3, RoBERTa, ALBERT) versus human predictions align with N400 amplitude, a neural signal of processing difficulty, finding that the models matched the N400 more closely than human predictions.
More predictable words are easier to process - they are read faster and elicit smaller neural signals associated with processing difficulty, most notably, the N400 component of the event-related brain potential. Thus, it has been argued that prediction of upcoming words is a key component of language comprehension, and that studying the amplitude of the N400 is a valuable way to investigate the predictions we make. In this study, we investigate whether the linguistic predictions of computational language models or humans better reflect the way in which natural language stimuli modulate the amplitude of the N400. One important difference in the linguistic predictions of humans versus computational language models is that while language models base their predictions exclusively on the preceding linguistic context, humans may rely on other factors. We find that the predictions of three top-of-the-line contemporary language models - GPT-3, RoBERTa, and ALBERT - match the N400 more closely than human predictions. This suggests that the predictive processes underlying the N400 may be more sensitive to the surface-level statistics of language than previously thought.