Title:Signatures of prediction during natural listening in MEG data?

Abstract:The brain uses contextual information and prior knowledge to anticipate upcoming content during language comprehension. Recent research has shown predictive signals can be revealed in pre-onset ECoG activity during naturalistic narrative listening, by building encoding models based on word embeddings from Large Language Models (LLMs). Similarly, evidence for long-range predictive encoding has been observed in fMRI data, where incorporating embeddings for multiple upcoming words in a narrative improves alignment with brain activity. This study examines whether similar predictive information can be detected in MEG, a technique with higher temporal resolution than fMRI but a lower signal-to-noise ratio than ECoG. Our findings indicate that MEG captures pre-onset representations up to 1 second before word onset, consistent with ECoG results. However, unlike fMRI findings, incorporating future word embeddings did not enhance MEG encoding, even for one word into the future, which suggests that the pre-onset encoding may not reflect predictive processing. This work demonstrates that MEG combined with LLMs is a valuable approach for studying language processing in naturalistic narratives and highlights the need to study further what constitutes evidence for prediction during natural listening.