# Dual Alignment Between Language Model Layers and Human Sentence Processing Source: [Dual Alignment Between Language Model Layers and Human Sentence Processing]() Collection index: [llms.txt]() Data updated: 2026-04-22T05:24:29.471Z Authors: Tatsuki Kuribayashi, Alex Warstadt, Yohei Oseki, Ethan Gotlieb Wilcox Published: 2026-04-20T17:51:26.000Z ## Abstract A recent study (Kuribayashi et al., 2025) has shown that human sentence processing behavior, typically measured on syntactically unchallenging constructions, can be effectively modeled using surprisal from early layers of large language models (LLMs). This raises the question of whether such advantages of internal layers extend to more syntactically challenging constructions, where surprisal has been reported to underestimate human cognitive effort. In this paper, we begin by exploring internal layers that better estimate human cognitive effort observed in syntactic ambiguity processing in English. Our experiments show that, in contrast to naturalistic reading, later layers better estimate such a cognitive effort, but still underestimate the human data. This dual alignment sheds light on different modes of sentence processing in humans and LMs: naturalistic reading employs a somewhat weak prediction akin to earlier layers of LMs, while syntactically challenging processing requires more fully-contextualized representations, better modeled by later layers of LMs. Motivated by these findings, we also explore several probability-update measures using shallow and deep layers of LMs, showing a complementary advantage to single-layer's surprisal in reading time modeling. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Psychometric Predictive Power on SAP ROI](): Psychometric Predictive Power; SAP ROI; MVRR: 14.1. - [Psychometric Predictive Power on SAP All](): Psychometric Predictive Power; SAP All; MVRR: 11.7. - [Human Reading Time Prediction on SAP All sentences](): Human Reading Time Prediction; SAP All sentences; MVRR: not recorded. - [Human Reading Time Prediction on SAP Region of Interest](): Human Reading Time Prediction; SAP Region of Interest; MVRR: not recorded. - [Syntactic Surprisal Update Measurement (KL-divergence) on SAP All (full)](): Syntactic Surprisal Update Measurement (KL-divergence); SAP All (full); MVRR: not recorded. ## Original sources and implementation [Original paper]() [PDF]()