---
title: "Large Language Models Are Human-Like Internally (2025)"
description: "Several recent psycholinguistic studies found that surprisal from bigger language models predicts… 99 references, cited by 23 papers here."
authors: ["Tatsuki Kuribayashi", "Yohei Oseki", "Souhaib Ben Taieb", "Kentaro Inui", "Timothy Baldwin"]
year: 2025
venue: "Transactions of the Association for Computational Linguistics"
source: "https://aclanthology.org/2025.tacl-1.78/"
doi: "10.1162/tacl.a.58"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/kuribayashi-2025-large-language-models-are-human/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/kuribayashi-2025-large-language-models-are-human/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2025](https://md.research.elamur.ai/free/en/papers/2025/page/3/) / Large Language Models Are Human-Like Internally

# Large Language Models Are Human-Like Internally

**Authors:** [Tatsuki Kuribayashi](https://md.research.elamur.ai/free/en/authors/tatsuki-kuribayashi/), [Yohei Oseki](https://md.research.elamur.ai/free/en/authors/yohei-oseki/), Souhaib Ben Taieb, [Kentaro Inui](https://md.research.elamur.ai/free/en/authors/kentaro-inui/) and [Timothy Baldwin](https://md.research.elamur.ai/free/en/authors/timothy-baldwin/) · **Year:** 2025 · **Venue:** [Transactions of the Association for Computational Linguistics](https://md.research.elamur.ai/free/en/venues/tacl/) ([Transactions of the Association for Computational Linguistics (TACL) 2025](https://md.research.elamur.ai/free/en/venues/tacl/2025/)) · **Publisher:** [The MIT Press](https://md.research.elamur.ai/free/en/publishers/the-mit-press/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2025.tacl-1.78/) · [PDF](https://aclanthology.org/2025.tacl-1.78.pdf) · [DOI 10.1162/tacl.a.58](https://doi.org/10.1162/tacl.a.58) · [OpenAlex](https://openalex.org/W7108214532)

**Citations:** 7 citations (OpenAlex, 2026-10-14) · [23](https://md.research.elamur.ai/free/en/papers/kuribayashi-2025-large-language-models-are-human/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
Several recent psycholinguistic studies found that surprisal from bigger language models predicts human reading data worse than surprisal from small models such as GPT-2 small. This was taken as evidence that LLMs are cognitively implausible. The authors argue that this conclusion comes from examining only the last layer's output. They use two mechanistic-interpretability tools, logit-lens and tuned-lens, to obtain next-word probabilities from every intermediate layer. Each layer's surprisal is then tested as a predictor of human data in regression models that also include baseline lexical features. No part of any model is trained on human data.

Four measures were tested: self-paced reading, first-pass gaze duration, MAZE times and N400 EEG responses. On all of them, the best internal layers of large models match or beat small models. A timescale pattern also appears: early layers fit fast responses, while later layers fit slower ones. This resolves the reported split in which behavior favored small LMs and neurophysiology favored large LMs. The authors recommend layer-wise analysis as standard practice in cognitive modeling.

## Key points
- Internal-layer probabilities are extracted with logit-lens, which projects a hidden state through the final LayerNorm and unembedding matrix. Tuned-lens adds a learned per-layer affine map, and the authors use its publicly released parameters.
- The human measures are SPR, FPGD, MAZE and N400. Figure 1 illustrates the layer fits on the ZuCo corpus.
- First-pass gaze durations (about 200 ms) align best with earlier layers. N400 (about 400 ms) and MAZE data align best with later layers. The best layer is usually not the final one.
- An exploratory check on the UCL corpus, covering P600 and second-pass gaze duration, did not confirm the fast/slow ↔ early/late correspondence beyond the four core measures.
- The paper appeared in TACL vol. 13 (2025), pp. 1743–1766. Code is available at github.com/kuribayashi4/surprisal_internal_layers.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

## Institutions

- [The University of Melbourne](https://md.research.elamur.ai/free/en/universities/the-university-of-melbourne/) — University, Melbourne, Australia · 100 works · 22,618 citations
- [The University of Tokyo](https://md.research.elamur.ai/free/en/universities/the-university-of-tokyo/) — University, Tokyo, Japan · 90 works · 9,156 citations
- [Mohamed bin Zayed University of Artificial Intelligence](https://md.research.elamur.ai/free/en/universities/mohamed-bin-zayed-university-of-artificial-intelligence/) — University, Abu Dhabi, United Arab Emirates · 124 works · 5,676 citations
- [Tohoku University](https://md.research.elamur.ai/free/en/universities/tohoku-university/) — University, Sendai, Japan · 20 works · 514 citations
- [RIKEN](https://md.research.elamur.ai/free/en/labs/riken/) — Research institute, Wako, Japan · 26 works · 526 citations
- [University of Mons](https://md.research.elamur.ai/free/en/universities/university-of-mons/) — University, Mons, Belgium · 3 works · 67 citations

📚 Shelves (6): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works · [🧠 Neuroscience](https://md.research.elamur.ai/free/en/shelves/neuroscience/) · 165 works · [🧠 Cognitive Neuroscience](https://md.research.elamur.ai/free/en/shelves/neuroscience/cognitive-neuroscience/) · 164 works

🧭 Topics (2): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,320 works · [🧒 Psycholinguistics & language acquisition](https://md.research.elamur.ai/free/en/topics/psycholinguistics/) · 102 works

## Cite

Cite as: Tatsuki Kuribayashi, Yohei Oseki, Souhaib Ben Taieb, Kentaro Inui and Timothy Baldwin. 2025. Large Language Models Are Human-Like Internally. Transactions of the Association for Computational Linguistics.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@article{kuribayashi2025large,
  title = {{Large Language Models Are Human-Like Internally}},
  author = {Tatsuki Kuribayashi, Yohei Oseki, Souhaib Ben Taieb, Kentaro Inui and Timothy Baldwin},
  journal = {Transactions of the Association for Computational Linguistics},
  year = {2025},
  doi = {10.1162/tacl.a.58},
  url = {https://doi.org/10.1162/tacl.a.58}
}
```

- [← previous Knowledge Boundary of Large Language Models: A Survey](https://md.research.elamur.ai/free/en/papers/li-2025-knowledge-boundary-large-language-models/)
- [next → LongDocURL: a Comprehensive Multimodal Long Document Benchmark Integrating Understanding, Reasoning, and Locating](https://md.research.elamur.ai/free/en/papers/deng-2025-longdocurl-comprehensive-multimodal-long-document/)
- [↑ back to papers of 2025](https://md.research.elamur.ai/free/en/papers/2025/page/3/)

## References

99 references, 13 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. C. Aurnhammer and S. L. Frank. 2019. Comparing gated and simple recurrent neural network architectures as models of human sentence processing. In Proceedings of Cognitive Science, pages 112–118. https://doi.org/10 .31234/osf.io/wec74 [find](https://scholar.google.com/scholar?q=C%2E+Aurnhammer+and+S%2E+L%2E+Frank%2E+2019%2E+Comparing+gated+and+simple+recurrent+neural+network+architectures+as+models+of+human+sentence+processing%2E+In+Proceedings+of+Cognitive+Science%2C+pages+112%E2%80%93118%2E+https)
2. Khai Loong Aw, Syrielle Montariol, Badr AlKhamissi, Martin Schrimpf, and Antoine Bosselut. 2024. Instruction-tuning aligns LLMs to the human brain. In COLM 2024. [find](https://scholar.google.com/scholar?q=Khai+Loong+Aw%2C+Syrielle+Montariol%2C+Badr+AlKhamissi%2C+Martin+Schrimpf%2C+and+Antoine+Bosselut%2E+2024%2E+Instruction%2Dtuning+aligns+LLMs+to+the+human+brain%2E+In+COLM+2024%2E)
3. Andrea Banino, Jan Balaguer, and Charles Blundell. 2021. PonderNet: Learning to ponder. In 8th ICML Workshop on Automated Machine Learning (AutoML). [find](https://scholar.google.com/scholar?q=Andrea+Banino%2C+Jan+Balaguer%2C+and+Charles+Blundell%2E+2021%2E+PonderNet%3A+Learning+to+ponder%2E+In+8th+ICML+Workshop+on+Automated+Machine+Learning+%28AutoML%29%2E)
4. Lisa Beinborn and Nora Hollenstein. 2024. Cognitive Plausibility in Natural Language Processing Synthesis Lectures on Human Language Technologies. Springer International Publishing, Cham. https://doi.org/10 .1007/978-3-031-43260-6 [find](https://scholar.google.com/scholar?q=Lisa+Beinborn+and+Nora+Hollenstein%2E+2024%2E+Cognitive+Plausibility+in+Natural+Language+Processing+Synthesis+Lectures+on+Human+Language+Technologies%2E+Springer+International+Publishing%2C+Cham%2E+https%3A%2F%2Fdoi%2E)
5. Nora Belrose, Zach Furman, Logan Smith, Danny Halawi, Lev McKinney, Igor Ostrovsky, Stella Biderman, and Jacob Steinhardt. 2023. [Eliciting latent predictions from transformers with the tuned lens](https://md.research.elamur.ai/free/en/papers/belrose-2023-eliciting-latent-predictions-transformers-tuned/). arXiv preprint.
6. Stella Biderman, Hailey Schoelkopf, Quentin Gregory Anthony, Herbie Bradley, Kyle O’Brien, Eric Hallahan, Mohammad Aflah Khan, Shivanshu Purohit, USVSN Sai Prashanth, Edward Raff, Aviya Skowron, Lintang Sutawika, and Oskar van der Wal. 2023. Pythia: A suite for analyzing large language models across… [find](https://scholar.google.com/scholar?q=Stella+Biderman%2C+Hailey+Schoelkopf%2C+Quentin+Gregory+Anthony%2C+Herbie+Bradley%2C+Kyle+O%E2%80%99Brien%2C+Eric+Hallahan%2C+Mohammad+Aflah+Khan%2C+Shivanshu+Purohit%2C+USVSN+Sai+Prashanth%2C+Edward+Raff%2C+Aviya+Skowron%2C+Linta)
7. Veronica Boyce and Roger Philip Levy. 2023. A-maze of natural stories: Comprehension and surprisal in the maze task. Glossa Psycholinguistics. https://doi.org/10.5070/G6011190 [↗](https://doi.org/10.5070/G6011190)
8. Gino Brunner, Yang Liu, Damian Pascual, Oliver Richter, Massimiliano Ciaramita, and Roger Wattenhofer. 2019. On identifiability in transformers. In ICLR 2019. [find](https://scholar.google.com/scholar?q=Gino+Brunner%2C+Yang+Liu%2C+Damian+Pascual%2C+Oliver+Richter%2C+Massimiliano+Ciaramita%2C+and+Roger+Wattenhofer%2E+2019%2E+On+identifiability+in+transformers%2E+In+ICLR+2019%2E)
9. Manuel G. Calvo and Enrique Meseguer. 2002. Eye movements and processing stages in reading: Relative contribution of visual, lexical, and contextual factors. The Spanish Journal of Psychology, 5(1):66–77. https://doi.org/10.1017/S1138741600005849, PubMed: 12025367 — [doi:10.1017/s1138741600005849](https://doi.org/10.1017/s1138741600005849) [↗](https://doi.org/10.1017/S1138741600005849)
10. Charlotte Caucheteux, Alexandre Gramfort, and Jean-Rémi King. 2023. Evidence of a predictive coding hierarchy in the human brain listening to speech. Nature Human Behaviour, 7(3):430–441. https://doi.org/10.1038/s41562-022-01516-2, PubMed: 36864133 [↗](https://doi.org/10.1038/s41562-022-01516-2)
11. Matthew W. Crocker. 2007. Computational psycholinguistics. The Handbook of Computational Linguistics and Natural Language Processing. https://doi.org/10.1038/s41562-022-01516-2 [↗](https://doi.org/10.1038/s41562-022-01516-2)
12. Guy Dar, Mor Geva, Ankit Gupta, and Jonathan Berant. 2023. [Analyzing transformers in embedding space](https://md.research.elamur.ai/free/en/papers/dar-2023-analyzing-transformers-embedding-space/). In Proceedings of ACL 2023, pages 16124–16170. https://doi.org/10 .18653/v1/2023.acl-long.893
13. Olaf Dimigen, Werner Sommera, Annette Hohlfeld, Arthur M. Jacobs, and Reinhold Kliegl. 2011. Coregistration of eye movements and EEG in natural reading: Analyses and review. Journal of Experimental Psychology: General, 140(4):552. https://doi.org/10 .1037/a0023885, PubMed: 21744985 [find](https://scholar.google.com/scholar?q=Olaf+Dimigen%2C+Werner+Sommera%2C+Annette+Hohlfeld%2C+Arthur+M%2E+Jacobs%2C+and+Reinhold+Kliegl%2E+2011%2E+Coregistration+of+eye+movements+and+EEG+in+natural+reading%3A+Analyses+and+review%2E+Journal+of+Experimental+Ps)
14. Chris Dyer, Adhiguna Kuncoro, Miguel Ballesteros, and Noah A. Smith. 2016. Recurrent neural network grammars. In Proceedings of NAACL. Association for Computational Linguistics. https://doi.org/10.18653/v1/N16-1024 [↗](https://doi.org/10.18653/v1/N16-1024)
15. Kawin Ethayarajh. 2019. How contextual are contextualized word representations? Comparing the geometry of BERT, ELMo, and GPT-2 embeddings. In Proceedings of EMNLP-IJCNLP 2019, pages 55–65. https://doi.org/10 .18653/v1/D19-1006 [find](https://scholar.google.com/scholar?q=Kawin+Ethayarajh%2E+2019%2E+How+contextual+are+contextualized+word+representations%3F+Comparing+the+geometry+of+BERT%2C+ELMo%2C+and+GPT%2D2+embeddings%2E+In+Proceedings+of+EMNLP%2DIJCNLP+2019%2C+pages+55%E2%80%9365%2E+https%3A%2F%2Fdo)
16. Kara D. Federmeier, Edward W. Wlotko, Esmeralda De Ochoa-Dewald, and Marta Kutas. 2007. Multiple effects of sentential constraint on word processing. Brain Research, 1146:75–84. https://doi.org/10.1016/j.brainres.2006.06.101, PubMed: 16901469 — [doi:10.1016/j.brainres.2006.06.101](https://doi.org/10.1016/j.brainres.2006.06.101) [↗](https://doi.org/10.1016/j.brainres.2006.06.101)
17. Kenneth I. Forster, Christine Guerrera, and Lisa Elliot. 2009. The maze task: Measuring forced incremental sentence processing time. Behavior Research Methods, 41(1):163–171. https://doi.org/10.3758/BRM.41.1 .163, PubMed: 19182136 — [doi:10.3758/brm.41.1](https://doi.org/10.3758/brm.41.1) [↗](https://doi.org/10.3758/BRM.41.1)
18. Stefan L. Frank and Rens Bod. 2011. Insensitivity of the human sentence-processing system to hierarchical structure. Psychological Science, 22(6):829–834. https://doi.org/10.1177/0956797611409589, PubMed: 21586764 [↗](https://doi.org/10.1177/0956797611409589)
19. Stefan L. Frank, Irene Fernandez Monsalve, Robin L. Thompson, and Gabriella Vigliocco. 2013. Reading time data for evaluating broad-coverage models of english sentence processing. Behavior Research Methods, 45:1182–1190. https://doi.org/10.3758/s13428-012-0313-y, PubMed: 23404612 [↗](https://doi.org/10.3758/s13428-012-0313-y)
20. Stefan L. Frank, Leun J. Otten, Giulia Galli, and Gabriella Vigliocco. 2015. The ERP response to the amount of information conveyed by words in sentences. Brain and Language, 140:1–11. https://doi.org/10.1016/j.bandl .2014.10.006, PubMed: 25461915 — [doi:10.1016/j.bandl](https://doi.org/10.1016/j.bandl) [↗](https://doi.org/10.1016/j.bandl)
21. Richard Futrell, Edward Gibson, and Roger P. Levy. 2020. Lossy-context surprisal: An information-theoretic model of memory effects in sentence processing. Journal of Cognitive Science. https://doi.org/10.1111/cogs.12814, PubMed: 32100918 [↗](https://doi.org/10.1111/cogs.12814)
22. Richard Futrell, Edward Gibson, Harry J. Tily, Idan Blank, Anastasia Vishnevetsky, Steven T. [find](https://scholar.google.com/scholar?q=Richard+Futrell%2C+Edward+Gibson%2C+Harry+J%2E+Tily%2C+Idan+Blank%2C+Anastasia+Vishnevetsky%2C+Steven+T%2E)
23. Piantadosi, and Evelina Fedorenko. 2021. The natural stories corpus: A reading-time corpus of english texts containing rare syntactic constructions. Language Resource and Evaluation, 55(1):63–77. https://doi.org/10 .1007/s10579-020-09503-7, PubMed: 34720781 [find](https://scholar.google.com/scholar?q=Piantadosi%2C+and+Evelina+Fedorenko%2E+2021%2E+The+natural+stories+corpus%3A+A+reading%2Dtime+corpus+of+english+texts+containing+rare+syntactic+constructions%2E+Language+Resource+and+Evaluation%2C+55%281%29%3A63%E2%80%9377%2E+http)
24. Mario Giulianelli, Luca Malagutti, Juan Luis Gastaldi, Brian DuSell, Tim Vieira, and Ryan Cotterell. 2024a. [On the proper treatment of tokenization in psycholinguistics](https://md.research.elamur.ai/free/en/papers/giulianelli-2024-proper-treatment-tokenization-psycholinguistics/). In Proceedings of EMNLP 2024, pages 18556–18572. https://doi.org/10.18653/v1/2024 .emnlp-main.1032 [↗](https://doi.org/10.18653/v1/2024)
25. Mario Giulianelli, Andreas Opedal, and Ryan Cotterell. 2024b. Generalized measures of anticipation and responsivity in online language processing. In Findings of EMNLP 2024, pages 11648–11669. https://doi.org/10 .18653/v1/2024.findings-emnlp.682 [find](https://scholar.google.com/scholar?q=Mario+Giulianelli%2C+Andreas+Opedal%2C+and+Ryan+Cotterell%2E+2024b%2E+Generalized+measures+of+anticipation+and+responsivity+in+online+language+processing%2E+In+Findings+of+EMNLP+2024%2C+pages+11648%E2%80%9311669%2E+https%3A%2F)
26. Aaron Gokaslan and Vanya Cohen. 2019. [Openwebtext corpus](https://md.research.elamur.ai/free/en/papers/goodkind-2018-predictive-power-word-surprisal-reading/). [http://Skylion007.github.io/OpenWebTextCorpus](http://Skylion007.github.io/OpenWebTextCorpus) Adam Goodkind and Klinton Bicknell. 2018. Predictive power of word surprisal for reading times is a linear function of language model quality. In Proceedings of CMCL, pages 10–18. https://doi.org… [↗](http://Skylion007.github.io/OpenWebTextCorpus)
27. Alex Graves. 2016. Adaptive computation time for recurrent neural networks. arXiv preprint. [find](https://scholar.google.com/scholar?q=Alex+Graves%2E+2016%2E+Adaptive+computation+time+for+recurrent+neural+networks%2E+arXiv+preprint%2E)
28. Michael Hahn, Richard Futrell, Roger Levy, and Edward Gibson. 2022. A resource-rational model of human processing of recursive linguistic structure. Proceedings of the National Academy of Sciences of the United States of America, 119(43):e2122602119. https://doi.org/10.1073/pnas.2122602119, PubMed:… — [doi:10.1073/pnas.2122602119](https://doi.org/10.1073/pnas.2122602119) [↗](https://doi.org/10.1073/pnas.2122602119)
29. John Hale. 2001. [A probabilistic Earley parser as a psycholinguistic model](https://md.research.elamur.ai/free/en/papers/hale-2001-probabilistic-earley-parser-as-psycholinguistic/). In Proceedings of NAACL 2001, pages 159–166. https://doi.org/10.3115/1073336.1073357 [↗](https://doi.org/10.3115/1073336.1073357)
30. John Hale. 2016. Information-theoretical complexity metrics. Language and Linguistics Compass, 10(9):397–412. https://doi.org/10.1109/TIT.2016.2568202 — [doi:10.1109/tit.2016.2568202](https://doi.org/10.1109/tit.2016.2568202) [↗](https://doi.org/10.1109/TIT.2016.2568202)
31. Kenneth Heafield. 2011. [KenLM: Faster and smaller language model queries](https://md.research.elamur.ai/free/en/papers/heafield-2011-kenlm-faster-smaller-language-model/). In Proceedings of the Sixth Workshop on Statistical machine Translation, pages 187–197.
32. Nora Hollenstein, Jonathan Rotsztejn, Marius Troendle, Andreas Pedroni, Ce Zhang, and Nicolas Langer. 2018. ZuCo, a simultaneous EEG and eye-tracking resource for natural sentence reading. Scientific Data, 5:180291. https://doi.org/10.1038/sdata.2018 .291, PubMed: 30531985 — [doi:10.1038/sdata.2018](https://doi.org/10.1038/sdata.2018) [↗](https://doi.org/10.1038/sdata.2018)
33. Eghbal A. Hosseini, Martin Schrimpf, Yian Zhang, Samuel Bowman, Noga Zaslavsky, and Evelina Fedorenko. 2024. Artificial neural network language models predict human brain responses to language even after a developmentally realistic amount of training. Neurobiology of Language, 5(1):43–63. https://do… — [doi:10.1162/nol](https://doi.org/10.1162/nol) [find](https://scholar.google.com/scholar?q=Eghbal+A%2E+Hosseini%2C+Martin+Schrimpf%2C+Yian+Zhang%2C+Samuel+Bowman%2C+Noga+Zaslavsky%2C+and+Evelina+Fedorenko%2E+2024%2E+Artificial+neural+network+language+models+predict+human+brain+responses+to+language+even+af)
34. Jennifer Hu, Michael A. Lepori, and Michael Franke. 2025. Signatures of human-like processing in transformer forward passes. arXiv preprint arXiv:2504.14107. — [arXiv:2504.14107](https://arxiv.org/abs/2504.14107) [find](https://scholar.google.com/scholar?q=Jennifer+Hu%2C+Michael+A%2E+Lepori%2C+and+Michael+Franke%2E+2025%2E+Signatures+of+human%2Dlike+processing+in+transformer+forward+passes%2E+arXiv+preprint+arXiv%3A2504%2E14107%2E+%E2%80%94+%5BarXiv%3A2504%2E14107%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs)
35. Kuan-Jung Huang, Suhas Arehalli, Mari Kugemoto, Christian Muxica, Grusha Prasad, Brian Dillon, and Tal Linzen. 2024. Large-scale benchmark yields no evidence that language model surprisal explains syntactic disambiguation difficulty. Journal of Memory and Language, 137:104510. https://doi .org/10.10… — [doi:10.1016/j.jml.2024.104510](https://doi.org/10.1016/j.jml.2024.104510) [find](https://scholar.google.com/scholar?q=Kuan%2DJung+Huang%2C+Suhas+Arehalli%2C+Mari+Kugemoto%2C+Christian+Muxica%2C+Grusha+Prasad%2C+Brian+Dillon%2C+and+Tal+Linzen%2E+2024%2E+Large%2Dscale+benchmark+yields+no+evidence+that+language+model+surprisal+explains+syn)
36. Ryan J. Hubbard, Joost Rommers, Cassandra L. Jacobs, and Kara D. Federmeier. 2019. Downstream behavioral and electrophysiological consequences of word prediction on recognition memory. Frontiers in Human Neuroscience, 13:291. https://doi.org/10.3389/fnhum .2019.00291, PubMed: 31555111 — [doi:10.3389/fnhum](https://doi.org/10.3389/fnhum) [↗](https://doi.org/10.3389/fnhum)
37. Marcel A. Just and Patricia A. Carpenter. 1980. A theory of reading: From eye fixations to comprehension. Journal of Psychological Review. https://doi.org/10.1037/0033 -295X.87.4.329, PubMed: 7413885 — [doi:10.1037/0033](https://doi.org/10.1037/0033) [↗](https://doi.org/10.1037/0033)
38. Nikita Kitaev, Steven Cao, and Dan Klein. 2019. [Multilingual constituency parsing with self-attention and pre-training](https://md.research.elamur.ai/free/en/papers/kitaev-2019-multilingual-constituency-parsing-self-attention/). In Proceedings of ACL 2019, pages 3499–3505. https://doi.org/10.18653/v1/P19-1340 [↗](https://doi.org/10.18653/v1/P19-1340)
39. Nikita Kitaev and Dan Klein. 2018. [Constituency parsing with a self-attentive encoder](https://md.research.elamur.ai/free/en/papers/kitaev-2018-constituency-parsing-self-attentive-encoder/). In Proceedings of ACL 2018, pages 2676–2686. https://doi.org/10.18653/v1/P18 -1249 [↗](https://doi.org/10.18653/v1/P18)
40. Tatsuki Kuribayashi, Yohei Oseki, and Timothy Baldwin. 2024. Psychometric predictive power of large language models. In Findings of NAACL 2024, pages 1983–2005. https://doi.org/10.18653/v1/2024.findings -naacl.129 [↗](https://doi.org/10.18653/v1/2024.findings)
41. Tatsuki Kuribayashi, Yohei Oseki, Ana Brassard, and Kentaro Inui. 2022. Context limitations make neural language models more human-like. In Proceedings of EMNLP 2022, pages 10421–10436. https://doi.org/10 .18653/v1/2022.emnlp-main.712 [find](https://scholar.google.com/scholar?q=Tatsuki+Kuribayashi%2C+Yohei+Oseki%2C+Ana+Brassard%2C+and+Kentaro+Inui%2E+2022%2E+Context+limitations+make+neural+language+models+more+human%2Dlike%2E+In+Proceedings+of+EMNLP+2022%2C+pages+10421%E2%80%9310436%2E+https%3A%2F%2Fdoi%2Eor)
42. Tatsuki Kuribayashi, Yohei Oseki, Takumi Ito, Ryo Yoshida, Masayuki Asahara, and Kentaro Inui. 2021. Lower perplexity is not always human-like. In Proceedings of ACL-IJCNLP 2021, pages 5203–5217. https://doi.org/10.18653/v1/2021.acl-long.405 [↗](https://doi.org/10.18653/v1/2021.acl-long.405)
43. Marta Kutas and Kara D. Federmeier. 2011. Thirty years and counting: finding meaning in the N400 component of the event-related brain potential (ERP). Annual review of psychology, 62(1):621–647. https://doi.org/10 .1146/annurev.psych.093008.131123, PubMed: 20809790 [find](https://scholar.google.com/scholar?q=Marta+Kutas+and+Kara+D%2E+Federmeier%2E+2011%2E+Thirty+years+and+counting%3A+finding+meaning+in+the+N400+component+of+the+event%2Drelated+brain+potential+%28ERP%29%2E+Annual+review+of+psychology%2C+62%281%29%3A621%E2%80%93647%2E+https)
44. Ellen F. Lau, Colin Phillips, and David Poeppel. 2008. A cortical network for semantics: (de)constructing the N400. Nature Reviews Neuroscience, 9(12):920–933. https://doi.org/10.1038/nrn2532, PubMed: 19020511 — [doi:10.1038/nrn2532](https://doi.org/10.1038/nrn2532) [↗](https://doi.org/10.1038/nrn2532)
45. Yiğitcan Kaya, Sanghyun Hong, and Tudor Dumitras. 2019. Shallow-Deep Networks: Understanding and mitigating network overthinking. In Proceedings of ICML 2019. [find](https://scholar.google.com/scholar?q=Yig%CC%86itcan+Kaya%2C+Sanghyun+Hong%2C+and+Tudor+Dumitras%2E+2019%2E+Shallow%2DDeep+Networks%3A+Understanding+and+mitigating+network+overthinking%2E+In+Proceedings+of+ICML+2019%2E)
46. Roger Levy. 2008. Expectation-based syntactic comprehension. Journal of Cognition, 106(3):1126–1177. https://doi.org/10.1016/j.cognition.2007.05.006, PubMed: 17662975 [↗](https://doi.org/10.1016/j.cognition.2007.05.006)
47. Alan Kennedy, Robin Hill, and Joël Pynte. 2003. The dundee corpus. In Proceedings of the 12th European conference on eye movement. [find](https://scholar.google.com/scholar?q=Alan+Kennedy%2C+Robin+Hill%2C+and+Joe%CC%88l+Pynte%2E+2003%2E+The+dundee+corpus%2E+In+Proceedings+of+the+12th+European+conference+on+eye+movement%2E)
48. Richard L. Lewis and Shravan Vasishth. 2005. An activation-based model of sentence processing as skilled memory retrieval. Cognitive Sciences, 29(3):375–419. https://doi.org/10.1207/s15516709cog0000 25, PubMed: 21702779 — [doi:10.1207/s15516709cog0000](https://doi.org/10.1207/s15516709cog0000) [↗](https://doi.org/10.1207/s15516709cog0000)
49. Falk Lieder and Thomas L. Griffiths. 2019. Resource-rational analysis: Understanding human cognition as the optimal use of limited computational resources. Behavioral and Brain Sciences, 43(e1):e1. https://doi.org/10.1017/S0140525X1900061X, PubMed: 30714890 [↗](https://doi.org/10.1017/S0140525X1900061X)
50. Xi Victoria Lin, Todor Mihaylov, Mikel Artetxe, Tianlu Wang, Shuohui Chen, Daniel Simig, Myle Ott, Naman Goyal, Shruti Bhosale, Jingfei Du, Ramakanth Pasunuru, Sam Shleifer, Punit Singh Koura, Vishrav Chaudhary, Brian O’Horo, Jeff Wang, Luke Zettlemoyer, Zornitsa Kozareva, Mona Diab, Veselin Stoyano… — [Few-shot Learning with Multilingual Language Models](https://md.research.elamur.ai/free/en/papers/lin-2021-few-shot-learning-multilingual-language/)
51. David Marr. 1982. Henry Holt and Co., Inc., USA. [find](https://scholar.google.com/scholar?q=David+Marr%2E+1982%2E+Henry+Holt+and+Co%2E%2C+Inc%2E%2C+USA%2E)
52. Kate McCurdy and Michael Hahn. 2024. Lossy context surprisal predicts task-dependent patterns in relative clause processing. In Proceedings of CoNLL 2024, pages 36–45. https://doi.org/10.18653/v1/2024.conll-1.4 [↗](https://doi.org/10.18653/v1/2024.conll-1.4)
53. Clara Meister, Mario Giulianelli, and Tiago Pimentel. 2024. Towards a similarity-adjusted surprisal theory. In Proceedings of EMNLP 2024, pages 16485–16498. https://doi.org/10.18653/v1/2024.emnlp-main.921 [↗](https://doi.org/10.18653/v1/2024.emnlp-main.921)
54. Clara Meister, Tiago Pimentel, Thomas Clark, Ryan Cotterell, and Roger Levy. 2022. Analyzing wrap-up effects through an information-theoretic lens. In Proceedings of ACL 2022, pages 20–28. https://doi.org/10.18653/v1/2022.acl-short.3 [↗](https://doi.org/10.18653/v1/2022.acl-short.3)
55. Danny Merkx and Stefan L. Frank. 2021. Human sentence processing: Recurrence or attention? In Proceedings of CMCL, pages 12–22. https://doi.org/10.18653/v1/2021 .cmcl-1.2 [↗](https://doi.org/10.18653/v1/2021)
56. James Michaelov, Catherine Arnett, and Ben Bergen. 2024a. Revenge of the fallen? Recurrent models match transformers at predicting human language comprehension metrics. In COLM 2024. [find](https://scholar.google.com/scholar?q=James+Michaelov%2C+Catherine+Arnett%2C+and+Ben+Bergen%2E+2024a%2E+Revenge+of+the+fallen%3F+Recurrent+models+match+transformers+at+predicting+human+language+comprehension+metrics%2E+In+COLM+2024%2E)
57. James A. Michaelov, Megan D. Bardolph, Cyma K. Van Petten, Benjamin K. Bergen, and Seana Coulson. 2024b. Strong prediction: Language model surprisal explains multiple N400 effects. Neurobiology of Language, 5(1):107–135. https://doi.org/10.1162/nol a 00105, PubMed: 38645623 — [doi:10.1162/nol](https://doi.org/10.1162/nol) [↗](https://doi.org/10.1162/nol)
58. Sathvik Nair and Philip Resnik. 2023. Words, subwords, and morphemes: What really matters in the surprisal-reading time relationship? In Findings of EMNLP 2023. https://doi.org/10.18653/v1/2023.findings -emnlp.752 nostalgebraist. 2020. Interpreting GPT: The logit lens. Blog post, retrieved 20 June,… [↗](https://doi.org/10.18653/v1/2023.findings)
59. Samer Nour Eddine, Trevor Brothers, Lin Wang, Michael Spratling, and Gina R. Kuperberg. 2024. A predictive coding model of the N400. Cognition, 246(105755):105755. https://doi.org/10.1016/j.cognition.2024 .105755, PubMed: 38428168 — [doi:10.1016/j.cognition.2024](https://doi.org/10.1016/j.cognition.2024) [↗](https://doi.org/10.1016/j.cognition.2024)
60. Byung-Doh Oh and William Schuler. 2023a. Transformer-based language model surprisal predicts human reading times best with about two billion training tokens. In Findings of EMNLP 2023, pages 1915–1921. [https://10.18653/v1/2023.findings-emnlp.128](https://10.18653/v1/2023.findings-emnlp.128) [↗](https://10.18653/v1/2023.findings-emnlp.128)
61. Byung-Doh Oh and William Schuler. 2023b. [Why does surprisal from larger transformer-based language models provide a poorer fit to human reading times?](https://md.research.elamur.ai/free/en/papers/oh-2023-why-does-surprisal-larger-transformer/) TACL, 11:336–350. https://doi.org/10.1162/tacl a 00548 [↗](https://doi.org/10.1162/tacl)
62. Byung-Doh Oh and William Schuler. 2024. Leading whitespaces of language models’ subword vocabulary pose a confound for calculating word probabilities. In Proceedings of EMNLP 2024, pages 3464–3472. https://doi.org/10.18653/v1/2024.emnlp-main .202 [↗](https://doi.org/10.18653/v1/2024.emnlp-main)
63. Byung-Doh Oh, Shisen Yue, and William Schuler. 2024. Frequency explains the inverse correlation of large language models’ size, training data amount, and surprisal’s fit to reading times. In Proceedings of EACL 2024, pages 2644–2663. https://doi.org/10 .18653/v1/2024.eacl-long.162 [find](https://scholar.google.com/scholar?q=Byung%2DDoh+Oh%2C+Shisen+Yue%2C+and+William+Schuler%2E+2024%2E+Frequency+explains+the+inverse+correlation+of+large+language+models%E2%80%99+size%2C+training+data+amount%2C+and+surprisal%E2%80%99s+fit+to+reading+times%2E+In+Proceedin)
64. Andreas Opedal, Eleanor Chodroff, Ryan Cotterell, and Ethan Wilcox. 2024. On the role of context in reading time prediction. In Proceedings of EMNLP 2024, pages 3042–3058. eling with Python. In 9th Python in Science Conference. https://doi.org/10 .25080/Majora-92bf1922-011 https://doi.org/10.18653/v… [find](https://scholar.google.com/scholar?q=Andreas+Opedal%2C+Eleanor+Chodroff%2C+Ryan+Cotterell%2C+and+Ethan+Wilcox%2E+2024%2E+On+the+role+of+context+in+reading+time+prediction%2E+In+Proceedings+of+EMNLP+2024%2C+pages+3042%E2%80%933058%2E+eling+with+Python%2E+In+9th+Py)
65. Tiago Pimentel, Clara Meister, Ethan G. Wilcox, Roger P. Levy, and Ryan Cotterell. 2023. On the effect of anticipation on reading times. TACL, 11:1624–1642. https://doi.org/10.1162/tacl_a_00603 — [doi:10.1162/tacl_a_00603](https://doi.org/10.1162/tacl_a_00603) [↗](https://doi.org/10.1162/tacl_a_00603)
66. Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever. 2019. Language models are unsupervised multitask learners. OpenAI blog. [find](https://scholar.google.com/scholar?q=Alec+Radford%2C+Jeffrey+Wu%2C+Rewon+Child%2C+David+Luan%2C+Dario+Amodei%2C+and+Ilya+Sutskever%2E+2019%2E+Language+models+are+unsupervised+multitask+learners%2E+OpenAI+blog%2E)
67. Keith Rayner. 1998. Eye movements in reading and information processing: 20 years of research. Psychological Bulletin, 124(3):372–422. https://doi.org/10.1037/0033-2909 .124.3.372, PubMed: 9849112 — [doi:10.1037/0033-2909](https://doi.org/10.1037/0033-2909) [↗](https://doi.org/10.1037/0033-2909)
68. Keith Rayner and Charles Clifton Jr. 2009. Language processing in reading and speech perception is fast and incremental: Implications for event-related potential research. Biological Psychology, 80(1):4–9. https://doi.org/10.1016/j.biopsycho.2008.05.002, PubMed: 18565638 — [doi:10.1016/j.biopsycho.2008.05.002](https://doi.org/10.1016/j.biopsycho.2008.05.002) [↗](https://doi.org/10.1016/j.biopsycho.2008.05.002)
69. Keith Rayner, Gretchen Kambe, and Susan A. Duffy. 2000. The effect of clause wrap-up on eye movements during reading. Quarterly Journal of Experimental Psychology Section A: Human Experimental Psychology, 53(4):1061–1080. https://doi.org/10 .1080/713755934, PubMed: 11131813 [find](https://scholar.google.com/scholar?q=Keith+Rayner%2C+Gretchen+Kambe%2C+and+Susan+A%2E+Duffy%2E+2000%2E+The+effect+of+clause+wrap%2Dup+on+eye+movements+during+reading%2E+Quarterly+Journal+of+Experimental+Psychology+Section+A%3A+Human+Experimental+Psychol)
70. Marten van Schijndel and Tal Linzen. 2021. Single-stage prediction models do not explain the magnitude of syntactic disambiguation difficulty. Cognitive Science, 45(6):e12988. https://doi.org/10.1111/cogs.12988, PubMed: 34170031 — [doi:10.1111/cogs.12988](https://doi.org/10.1111/cogs.12988) [↗](https://doi.org/10.1111/cogs.12988)
71. Martin Schrimpfa, Idan Asher Blank, Greta Tuckute, Carina Kauf, Eghbal A. Hosseini, Nancy Kanwisher, Joshua B. Tenenbaum, and Evelina Fedorenko. 2021. The neural architecture of language: Integrative modeling converges on predictive processing. Proceedings of the National Academy of Sciences, 118(45… — [doi:10.1073/pnas.2105646118](https://doi.org/10.1073/pnas.2105646118) [find](https://scholar.google.com/scholar?q=Martin+Schrimpfa%2C+Idan+Asher+Blank%2C+Greta+Tuckute%2C+Carina+Kauf%2C+Eghbal+A%2E+Hosseini%2C+Nancy+Kanwisher%2C+Joshua+B%2E+Tenenbaum%2C+and+Evelina+Fedorenko%2E+2021%2E+The+neural+architecture+of+language%3A+Integrative+)
72. Skipper Seabold and Josef Perktold. 2010. Statsmodels: Econometric and statistical mod-Cory Shain, Idan A. Blank, Evelina Fedorenko, Edward Gibson, and William Schuler. 2022. Robust effects of working memory demand during naturalistic language comprehension in language-selective cortex. Journal of N… — [doi:10.1523/jneurosci.1894-21](https://doi.org/10.1523/jneurosci.1894-21) [find](https://scholar.google.com/scholar?q=Skipper+Seabold+and+Josef+Perktold%2E+2010%2E+Statsmodels%3A+Econometric+and+statistical+mod%2DCory+Shain%2C+Idan+A%2E+Blank%2C+Evelina+Fedorenko%2C+Edward+Gibson%2C+and+William+Schuler%2E+2022%2E+Robust+effects+of+working)
73. Cory Shain, Clara Meister, Tiago Pimentel, Ryan Cotterell, and Roger Levy. 2024. Large-scale evidence for logarithmic effects of word predictability on reading time. Proceedings of the National Academy of Sciences, 121(10):e2307876121. https://doi.org/10.1073/pnas.2307876121, PubMed: 38422017 — [doi:10.1073/pnas.2307876121](https://doi.org/10.1073/pnas.2307876121) [↗](https://doi.org/10.1073/pnas.2307876121)
74. Noam Siegelman, Sascha Schroeder, Cengiz Acartürk, Hee-Don Ahn, Svetlana Alexeeva, Simona Amenta, Raymond Bertram, Rolando Bonandrini, Marc Brysbaert, Daria Chernova, Sara Maria Da Fonseca, Nicolas Dirix, Wouter Duyck, Argyro Fella, Ram Frost, Carolina A. Gattei, Areti Kalaitzi, Nayoung Kwon, Kaidi… [find](https://scholar.google.com/scholar?q=Noam+Siegelman%2C+Sascha+Schroeder%2C+Cengiz+Acartu%CC%88rk%2C+Hee%2DDon+Ahn%2C+Svetlana+Alexeeva%2C+Simona+Amenta%2C+Raymond+Bertram%2C+Rolando+Bonandrini%2C+Marc+Brysbaert%2C+Daria+Chernova%2C+Sara+Maria+Da+Fonseca%2C+Nicolas+D)
75. Nathaniel J. Smith and Roger Levy. 2013. The effect of word predictability on reading time is logarithmic. Cognition, 128(3):302–319. https://doi.org/10.1016/j.cognition .2013.02.013, PubMed: 23747651 — [doi:10.1016/j.cognition](https://doi.org/10.1016/j.cognition) [↗](https://doi.org/10.1016/j.cognition)
76. Robyn Speer. 2022. rspeer/wordfreq: v3.0. [find](https://scholar.google.com/scholar?q=Robyn+Speer%2E+2022%2E+rspeer%2Fwordfreq%3A+v3%2E0%2E)
77. Jakub M. Szewczyk and Kara D. Federmeier. 2022. Context-based facilitation of semantic access follows both logarithmic and linear functions of stimulus probability. Journal of Memory and Language, 123(104311):104311. [find](https://scholar.google.com/scholar?q=Jakub+M%2E+Szewczyk+and+Kara+D%2E+Federmeier%2E+2022%2E+Context%2Dbased+facilitation+of+semantic+access+follows+both+logarithmic+and+linear+functions+of+stimulus+probability%2E+Journal+of+Memory+and+Language%2C+123)
78. Holmes a benchmark to assess the linguistic competence of language models. TACL, 12:1616–1647. https://doi.org/10.1162/tacl a 00718 https://doi.org/10.1016/j.jml.2021 .104311, PubMed: 36337731 — [doi:10.1016/j.jml.2021](https://doi.org/10.1016/j.jml.2021) [↗](https://doi.org/10.1162/tacl)
79. Jakub M. Szewczyk, Emily N. Mech, and Kara D. Federmeier. 2022. The power of ‘‘good’’: Can adjectives rapidly decrease as well as increase the availability of the upcoming noun? Journal of Experimental Psychology: Learning, Memory, and Cognition, 48(6):856–875. https://doi.org/10.1037/xlm0001091, Pu… — [doi:10.1037/xlm0001091](https://doi.org/10.1037/xlm0001091) [↗](https://doi.org/10.1037/xlm0001091)
80. Ian Tenney, Dipanjan Das, and Ellie Pavlick. 2019. [BERT rediscovers the classical NLP pipeline](https://md.research.elamur.ai/free/en/papers/tenney-2019-bert-rediscovers-classical-nlp-pipeline/). In Proceedings of ACL 2019, pages 4593–4601. https://doi.org/10 .18653/v1/P19-1452
81. Mariya Toneva and Leila Wehbe. 2019. Interpreting and improving natural-language processing (in machines) with natural language-processing (in the brain). In NeurIPS 2019, pages 14928–14938. [find](https://scholar.google.com/scholar?q=Mariya+Toneva+and+Leila+Wehbe%2E+2019%2E+Interpreting+and+improving+natural%2Dlanguage+processing+%28in+machines%29+with+natural+language%2Dprocessing+%28in+the+brain%29%2E+In+NeurIPS+2019%2C+pages+14928%E2%80%9314938%2E)
82. Pranali Vani, Ethan Gotlieb Wilcox, and Roger Levy. 2021. Using the interpolated maze task to assess incremental processing in english relative clauses. Proceedings of the Annual Meeting of the Cognitive Science Society, 43(43). [find](https://scholar.google.com/scholar?q=Pranali+Vani%2C+Ethan+Gotlieb+Wilcox%2C+and+Roger+Levy%2E+2021%2E+Using+the+interpolated+maze+task+to+assess+incremental+processing+in+english+relative+clauses%2E+Proceedings+of+the+Annual+Meeting+of+the+Cognit)
83. Andrea de Varda and Marco Marelli. 2023. Scaling in cognitive modelling: A multilingual approach to human reading times. In Proceedings of ACL 2023, pages 139–149. https://doi.org/10.18653/v1/2023 .acl-short.14 [↗](https://doi.org/10.18653/v1/2023)
84. Andrea Gregor de Varda, Marco Marelli, and Simona Amenta. 2024. Cloze probability, predictability ratings, and computational estimates for 205 english sentences, aligned with existing EEG and reading time data. Behavior Research Methods, 56(5):5190–5213. https://doi.org/10.3758/s13428-023 -02261-8,… — [doi:10.3758/s13428-023](https://doi.org/10.3758/s13428-023) [↗](https://doi.org/10.3758/s13428-023)
85. Shravan Vasishth, Katja Suckow, Richard L. Lewis, and Sabine Kern. 2010. Short-term forgetting in sentence comprehension: Crosslinguistic evidence from verb-final structures. Language and Cognitive Processes, 25(4):533–567. https://doi.org/10.1080/01690960903310587 [↗](https://doi.org/10.1080/01690960903310587)
86. Andreas Waldis, Yotam Perlitz, Leshem Choshen, Yufang Hou, and Iryna Gurevych. 2024. [find](https://scholar.google.com/scholar?q=Andreas+Waldis%2C+Yotam+Perlitz%2C+Leshem+Choshen%2C+Yufang+Hou%2C+and+Iryna+Gurevych%2E+2024%2E)
87. Daphne Wang, Mehrnoosh Sadrzadeh, Miloš Stanojević, Wing-Yee Chow, and Richard Breheny. 2024. How can large language models become more human? In Proceedings of CMCL 2024, pages 166–176. https://doi.org/10.18653/v1/2024.cmcl-1.14 [↗](https://doi.org/10.18653/v1/2024.cmcl-1.14)
88. Chris Wendler, Veniamin Veselovsky, Giovanni Monea, and Robert West. 2024. [Do LlaMas work in English? On the latent language of multilingual transformers](https://md.research.elamur.ai/free/en/papers/wendler-2024-do-llamas-work-english-latent/). In Proceedings of ACL 2024, pages 15366–15394. https://doi.org/10.18653/v1/2024.acl-long.820 [↗](https://doi.org/10.18653/v1/2024.acl-long.820)
89. Ethan Wilcox, Clara Meister, Ryan Cotterell, and Tiago Pimentel. 2023a. Language model quality correlates with psychometric predictive power in multiple languages. In Proceedings of EMNLP 2023, pages 7503–7511. https://doi.org/10.18653/v1/2023 .emnlp-main.466 [↗](https://doi.org/10.18653/v1/2023)
90. Ethan Wilcox, Clara Meister, Ryan Cotterell, and Tiago Pimentel. 2023b. Language model quality correlates with psychometric predictive power in multiple languages. In Proceedings of EMNLP 2023, pages 7503–7511. https://doi.org/10.18653/v1/2023 .emnlp-main.466 [↗](https://doi.org/10.18653/v1/2023)
91. Ethan Wilcox, Pranali Vani, and Roger Levy. 2021. A targeted assessment of incremental processing in neural language models and humans. In Proceedings of ACL, pages 939–952. https://doi.org/10.18653/v1/2021 .acl-long.76 [↗](https://doi.org/10.18653/v1/2021)
92. Ethan Gotlieb Wilcox, Jon Gauthier, Jennifer Hu, Peng Qian, and Roger Levy. 2020. On the predictive power of neural language models for human real-time comprehension behavior. In Proceedings of Cognitive Science, pages 1707–1713. [find](https://scholar.google.com/scholar?q=Ethan+Gotlieb+Wilcox%2C+Jon+Gauthier%2C+Jennifer+Hu%2C+Peng+Qian%2C+and+Roger+Levy%2E+2020%2E+On+the+predictive+power+of+neural+language+models+for+human+real%2Dtime+comprehension+behavior%2E+In+Proceedings+of+Cognit)
93. Ethan Gotlieb Wilcox, Michael Hu, Aaron Mueller, Tal Linzen, Alex Warstadt, Leshem Choshen, Chengxu Zhuang, Ryan Cotterell, and Adina Williams. 2024. Bigger is not always better: The importance of human-scale language modeling for psycholinguistics. [find](https://scholar.google.com/scholar?q=Ethan+Gotlieb+Wilcox%2C+Michael+Hu%2C+Aaron+Mueller%2C+Tal+Linzen%2C+Alex+Warstadt%2C+Leshem+Choshen%2C+Chengxu+Zhuang%2C+Ryan+Cotterell%2C+and+Adina+Williams%2E+2024%2E+Bigger+is+not+always+better%3A+The+importance+of+hum)
94. PsyArXiv. https://doi.org/10.31234/osf.io/rfwgd_v1 [↗](https://doi.org/10.31234/osf.io/rfwgd_v1)
95. Naoko Witzel, Jeffrey Witzel, and Kenneth Forster. 2012. Comparisons of online reading paradigms: Eye tracking, moving-window, and maze. Journal of Psycholinguistic Research, 41(2):105–128. https://doi.org/10.1007/s10936-011-9179-x, PubMed: 22002037 [↗](https://doi.org/10.1007/s10936-011-9179-x)
96. Edward W. Wlotko and Kara D. Federmeier. 2012. So that’s what you meant! Event-related potentials reveal multiple aspects of context use during construction of message-level meaning. Neuroimage, 62(1):356–366. https://doi.org/10.1016/j.neuroimage.2012 .04.054, PubMed: 22565202 — [doi:10.1016/j.neuroimage.2012](https://doi.org/10.1016/j.neuroimage.2012) [↗](https://doi.org/10.1016/j.neuroimage.2012)
97. Ryo Yoshida, Hiroshi Noji, and Yohei Oseki. 2021. Modeling human sentence processing with left-corner recurrent neural network grammars. In Proceedings of EMNLP 2021, pages 2964–2973. https://doi.org/10 .18653/v1/2021.emnlp-main.235 [find](https://scholar.google.com/scholar?q=Ryo+Yoshida%2C+Hiroshi+Noji%2C+and+Yohei+Oseki%2E+2021%2E+Modeling+human+sentence+processing+with+left%2Dcorner+recurrent+neural+network+grammars%2E+In+Proceedings+of+EMNLP+2021%2C+pages+2964%E2%80%932973%2E+https%3A%2F%2Fdoi%2Eorg%2F)
98. Susan Zhang, Stephen Roller, Naman Goyal, Mikel Artetxe, Moya Chen, Shuohui Chen, Christopher Dewan, Mona T. Diab, Xian Li, Xi Victoria Lin, Todor Mihaylov, Myle Ott, Sam Shleifer, Kurt Shuster, Daniel Simig, Punit Singh Koura, Anjali Sridhar, Tianlu Wang, and Luke Zettlemoyer. 2022. [OPT: Open pre-t…](https://md.research.elamur.ai/free/en/papers/zhang-2022-opt-open-pre-trained-transformer/)
99. Wangchunshu Zhou, Canwen Xu, Tao Ge, Julian McAuley, Ke Xu, and Furu Wei. 2020. BERT loses patience: Fast and robust inference with early exit. [find](https://scholar.google.com/scholar?q=Wangchunshu+Zhou%2C+Canwen+Xu%2C+Tao+Ge%2C+Julian+McAuley%2C+Ke+Xu%2C+and+Furu+Wei%2E+2020%2E+BERT+loses+patience%3A+Fast+and+robust+inference+with+early+exit%2E)

*Updated 30 October 2026*