---
title: "For GPT-4 as with Humans — paper by Cuneo et al. (2025)"
description: "Whether large language models grasp subtle form–function links in language is still contested. 37 references, cited by 2 papers here."
authors: ["Nicole Cuneo", "Eleanor Graves", "Supantho Rakshit", "Adele E. Goldberg"]
year: 2025
venue: "arXiv:2505.09005"
source: "https://arxiv.org/abs/2505.09005"
doi: "10.48550/arxiv.2505.09005"
arxiv: "2505.09005"
licence: "CC BY 4.0 (arXiv)"
licence_url: "http://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/cuneo-2025-gpt-4-humans-information-structure/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/cuneo-2025-gpt-4-humans-information-structure/)

[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/14/) / For GPT-4 as with Humans: Information Structure Predicts Acceptability of Long-Distance Dependencies

# For GPT-4 as with Humans: Information Structure Predicts Acceptability of Long-Distance Dependencies

**Authors:** Nicole Cuneo, Eleanor Graves, Supantho Rakshit and Adele E. Goldberg · **Year:** 2025 · **Venue:** arXiv:2505.09005 · **Licence:** [CC BY 4.0 (arXiv)](http://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://arxiv.org/abs/2505.09005) · [PDF](https://arxiv.org/pdf/2505.09005) · [DOI 10.48550/arxiv.2505.09005](https://doi.org/10.48550/arxiv.2505.09005) · [arXiv 2505.09005](https://arxiv.org/abs/2505.09005) · [OpenAlex](https://openalex.org/W4414938052)

**Citations:** 0 citations (OpenAlex, 2026-10-14) · [2](https://md.research.elamur.ai/free/en/papers/cuneo-2025-gpt-4-humans-information-structure/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
Whether large language models grasp subtle form–function links in language is still contested. This paper asks whether GPT-4 reproduces a pattern found in humans: the more backgrounded a constituent is in a plain sentence, the less acceptable it is as the target of a long-distance dependency (LDD), such as a wh-question, a relative clause or a cleft. This idea is the Backgrounded Constructions are Islands (BCI) account. The authors gave GPT-4 the same zero-shot Likert tasks used with human participants. One task measured backgroundedness through a negation test. The other collected acceptability ratings on LDD versions of the sentences and on the base sentences.

GPT-4's backgroundedness judgments predicted its LDD ratings but had little effect on its base-sentence ratings, which is the interaction seen in people. Its responses also tracked the human data. The result held on fresh stimuli, which addresses data contamination. A third study raised a constituent's prominence in a context sentence, and acceptability of the matching wh-question went up, which points to a causal effect. The work suggests that island effects can be partly explained by information structure, and that LLMs can serve as tools for probing linguistic theory.

## Key points
- The human benchmark came from a large study with 144 base sentences covering many construction types: various adjuncts, manner-of-speaking verb complements, double objects, to-datives, main and relative clauses, and parasitic and non-parasitic gaps. Each base sentence was hand-converted into wh-Qs, discourse-linked questions and relative clauses.
- Backgroundedness was measured with a negation task. The model was told to assume a negated sentence was true, then asked how likely it was that the embedded content still held. Content that survives negation is treated as more backgrounded.
- **Study 1a** used the original stimuli and replicated the human interaction: backgroundedness predicted LDD ratings but not declarative ratings.
- **Study 1b** used new stimuli to reduce contamination risk. It also added a fourth LDD type, it-clefts, which had not been tested on humans, and found converging results.
- **Study 2** manipulated emphasis in a preceding context sentence, and wh-question acceptability rose for the emphasized constituent. All probing was done on GPT-4 (Jan 2025) with zero-shot, minimal prompts, and requests were throttled to avoid caching artifacts.

## Interesting ideas
- Use negation as a graded probe for presupposition: negate the whole sentence and ask whether a sub-part still holds, so that backgroundedness becomes a number rather than a yes/no label.
- Predict an interaction rather than a single effect: the predictor should change judgments on one sentence type (LDDs) but not on a matched control (base declaratives), so general noise or plausibility cannot explain the result.
- Test causality by editing the context instead of only correlating: boost a constituent's prominence in a preceding sentence and see whether later judgments shift.
- Guard against memorization by rebuilding stimuli from scratch and adding a construction type the original benchmark never covered (here it-clefts).
- Explain extraction acceptability through information structure rather than pure syntactic island rules, so it becomes a gradient score tied to how backgrounded the source phrase is.

## For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

- **I1** (frames): prag could tag each clause or adjunct with a backgroundedness score by checking what survives sentential negation, which would make presupposition and at-issue content explicit in frames.
- **I2** (learn): evaluations could pair each test with a control condition where the feature should not matter, which works as a built-in negative control.
- **I3** (world): the text world could track discourse prominence of entities across sentences and test whether emphasis changes which referents are preferred later, for example in coreference or question focus.
- **I4** (learn): when an LLM teacher or a test set is drawn from published material, add freshly written items and an unseen construction to check that agreement is not recall.
- **I5** (tree): when the parser has to attach a fronted wh-phrase or relativized element, a backgroundedness prior on candidate host clauses (e.g., causal "by" adjuncts over temporal "while" adjuncts) could rank gap sites in an explainable way.

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

📚 Shelves (6): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 17,965 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 17,450 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 17,296 works · [🧠 Neuroscience](https://md.research.elamur.ai/free/en/shelves/neuroscience/) · 186 works · [🧠 Cognitive Neuroscience](https://md.research.elamur.ai/free/en/shelves/neuroscience/cognitive-neuroscience/) · 185 works

🧭 Topics (1): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,894 works

## Cite

Cite as: Nicole Cuneo, Eleanor Graves, Supantho Rakshit and Adele E. Goldberg. 2025. For GPT-4 as with Humans: Information Structure Predicts Acceptability of Long-Distance Dependencies. arXiv:2505.09005.

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

```bibtex
@misc{cuneo2025gpt,
  title = {{For GPT-4 as with Humans: Information Structure Predicts Acceptability of Long-Distance Dependencies}},
  author = {Nicole Cuneo, Eleanor Graves, Supantho Rakshit and Adele E. Goldberg},
  year = {2025},
  doi = {10.48550/arxiv.2505.09005},
  eprint = {2505.09005},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2505.09005}
}
```

- [← previous FEAT: A Preference Feedback Dataset through a Cost-Effective Auto-Generation and Labeling Framework for English AI Tutoring](https://md.research.elamur.ai/free/en/papers/seo-2025-feat-preference-feedback-dataset-through/)
- [next → GNNinjas at BAREC Shared Task 2025: Lexicon-Enriched Graph Modeling for Arabic Document Readability Prediction](https://md.research.elamur.ai/free/en/papers/elchafei-2025-gnninjas-barec-shared-task-2025/)
- [↑ back to papers of 2025](https://md.research.elamur.ai/free/en/papers/2025/page/14/)

## References

37 references, 6 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Ambridge, B., & Goldberg, A. E. (2008). The island status of clausal complements: Evidence in favor of an information structure explanation. Cognitive Linguistics 19(3), 349-381. [↗](https://doi.org/10.1515/cogl.2008.014)
2. Cuneo, N., & Goldberg, A. (2022). Islands effects without extraction: The discourse functions of constructions predict island status. In Proc. of the Cog. Sci Soc: 44:44. [find](https://scholar.google.com/scholar?q=Cuneo%2C+N%2E%2C+%26+Goldberg%2C+A%2E+%282022%29%2E+Islands+effects+without+extraction%3A+The+discourse+functions+of+constructions+predict+island+status%2E+In+Proc%2E+of+the+Cog%2E+Sci+Soc%3A+44%3A44%2E)
3. Cuneo, N., & Goldberg, A. E. (2023). The discourse functions of grammatical constructions explain an enduring syntactic puzzle. Cognition, 240, 105563. [↗](https://doi.org/10.1016/j.cognition.2023.105563)
4. Basmov, V., Goldberg, Y., & Tsarfaty, R. (2023). Simple Linguistic Inferences of Large Language Models (LLMs): Blind Spots and Blinds. arXiv preprint arXiv:2305.14785. https://arxiv.org/pdf/2305.14785 [↗](https://arxiv.org/abs/2305.14785)
5. Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021, March). On the dangers of stochastic parrots: Can language models be too big? 🦜 . In Proc. of 2021 ACM conference on fairness, accountability, and transparency (pp. 610-623). [find](https://scholar.google.com/scholar?q=Bender%2C+E%2E+M%2E%2C+Gebru%2C+T%2E%2C+McMillan%2DMajor%2C+A%2E%2C+%26+Shmitchell%2C+S%2E+%282021%2C+March%29%2E+On+the+dangers+of+stochastic+parrots%3A+Can+language+models+be+too+big%3F+%F0%9F%A6%9C+%2E+In+Proc%2E+of+2021+ACM+conference+on+fairness%2C+acc)
6. Chomsky, N. (1973). Conditions on transformations. A Festschrift for Morris Halle/Holt, Rinehart, and Winston. [find](https://scholar.google.com/scholar?q=Chomsky%2C+N%2E+%281973%29%2E+Conditions+on+transformations%2E+A+Festschrift+for+Morris+Halle%2FHolt%2C+Rinehart%2C+and+Winston%2E)
7. Grimshaw, J. (1986). Subjacency and the S/S′ Parameter. Linguistic Inquiry, 364-369. [find](https://scholar.google.com/scholar?q=Grimshaw%2C+J%2E+%281986%29%2E+Subjacency+and+the+S%2FS%E2%80%B2+Parameter%2E+Linguistic+Inquiry%2C+364%2D369%2E)
8. Dąbrowska, E. (2013). Functional constraints, usage, and mental grammars: A study of speakers' intuitions about questions with long-distance dependencies. Cognitive Linguistics 24(4), 633-665. Dentella, V., Günther, F., Murphy, E., Marcus, G. and Leivada, E. “Testing AI on Language Comprehension Tasks Reveals Insensitivity to Underlying Meaning.” Scientific Reports 14, no. 1 (November 14, 2024): 28083. https://doi.org/10.1038/s41598-024-79531-8. [↗](https://doi.org/10.1038/s41598-024-79531-8)
9. Dentella, V., Günther, F., & Leivada, E. (2023). Systematic testing of three Language Models reveals low language accuracy, absence of response stability, and a yes-response bias. PNAS 120(51), e2309583120. [↗](https://doi.org/10.1073/pnas.2309583120)
10. Erteschik-Shir, N. (1973). On the nature of island constraints (PhD thesis, MIT). [find](https://scholar.google.com/scholar?q=Erteschik%2DShir%2C+N%2E+%281973%29%2E+On+the+nature+of+island+constraints+%28PhD+thesis%2C+MIT%29%2E)
11. Fergus, A., Belluck, A., Cuneo, N. and Goldberg, A. E. (2025). Islands result from clash of functions: Single-conjunct wh-Qs. Proc. of the Cog.Sci. Society. [find](https://scholar.google.com/scholar?q=Fergus%2C+A%2E%2C+Belluck%2C+A%2E%2C+Cuneo%2C+N%2E+and+Goldberg%2C+A%2E+E%2E+%282025%29%2E+Islands+result+from+clash+of+functions%3A+Single%2Dconjunct+wh%2DQs%2E+Proc%2E+of+the+Cog%2ESci%2E+Society%2E)
12. Francis, E. J., & Michaelis, L. A. (2017). When relative clause extraposition is the right choice, it’s easier. Language and Cognition, 9(2), 332-370. [↗](https://doi.org/10.1017/langcog.2016.21)
13. Gauthier, J., Hu, J., Wilcox, E., Qian, P., & Levy, R. (2020, July). [SyntaxGym: An online platform for targeted evaluation of language models](https://md.research.elamur.ai/free/en/papers/gauthier-2020-syntaxgym-online-platform-targeted-evaluation/). In Proc. of the 58th Annual Meeting of the ACL Givón, T. (Ed.). (2020). Discourse and syntax. Brill.
14. Goldberg, A. E. (2006). Constructions at Work: the nature of generalization in language. OUP: Oxford. [find](https://scholar.google.com/scholar?q=Goldberg%2C+A%2E+E%2E+%282006%29%2E+Constructions+at+Work%3A+the+nature+of+generalization+in+language%2E+OUP%3A+Oxford%2E)
15. Goldberg, A. E. (2013) Backgrounded constituents cannot be “extracted”. In J. Sprouse (Ed). Syntactic Islands. OUP: Oxford. 221-238. [↗](https://doi.org/10.1017/cbo9781139035309.012)
16. Halliday, M. A. K. (2013). Halliday in the 21st Century: Volume 11. A&C Black. [find](https://scholar.google.com/scholar?q=Halliday%2C+M%2E+A%2E+K%2E+%282013%29%2E+Halliday+in+the+21st+Century%3A+Volume+11%2E+A%26C+Black%2E)
17. Hofmeister, P., & Sag, I. A. (2010). Cognitive constraints and island effects. Language, 86(2), 366. Howitt, K., Sathvik N., Dods, A., and Hopkins, R. M. “Generalizations across Filler-Gap Dependencies in Neural Language Models.” arXiv, October 23, 2024. http://arxiv.org/abs/2410.18225. [↗](https://arxiv.org/abs/2410.18225)
18. Hu, J., & Levy, R. (2023). [Prompting is not a substitute for probability measurements in large language models](https://md.research.elamur.ai/free/en/papers/hu-2023-prompting-is-not-substitute-probability/). arXiv preprint arXiv:2305.13264. [↗](https://arxiv.org/abs/2305.13264)
19. Hu, J., Mahowald, K., Lupyan, G., Ivanova, A., & Levy, R. (2024). Language models align with human judgments on key grammatical constructions. PNAS. 121(36), e2400917121. [↗](https://doi.org/10.1073/pnas.2400917121)
20. Jeretic, P., Warstadt, A., Bhooshan, S., & Williams, A. (2020). [Are natural language inference models IMPPRESsive? Learning IMPlicature and PRESupposition](https://md.research.elamur.ai/free/en/papers/jeretic-2020-are-natural-language-inference-models/). arXiv preprint arXiv:2004.03066. [↗](https://arxiv.org/abs/2004.03066)
21. Kallini, J., Papadimitriou, I., Futrell, R., Mahowald, K, and Potts, C. 2024. “[Mission: Impossible Language Models](https://md.research.elamur.ai/free/en/papers/kallini-2024-mission-impossible-language-models/).” http://arxiv.org/abs/2401.06416. [↗](https://arxiv.org/abs/2401.06416)
22. Karttunen, L. (1974). Presupposition and Linguistic Context. Theoretical Linguistics, 1(1-2), 182-194. [↗](https://doi.org/10.1515/thli.1974.1.1-3.181)
23. Kobzeva, A., Arehalli, S., Linzen, T., & Kush, D. (2023, June). Neural Networks Can Learn Patterns of Island-insensitivity in Norwegian. In Proc. of the Soc. for Computation in Linguistics 2023 (pp. 175-185). [↗](https://doi.org/10.31234/osf.io/4grqf)
24. Lambrecht, K. (1994). Information structure and sentence form: Topic, focus, and the mental representations of discourse referents. Cambridge University Press. [find](https://scholar.google.com/scholar?q=Lambrecht%2C+K%2E+%281994%29%2E+Information+structure+and+sentence+form%3A+Topic%2C+focus%2C+and+the+mental+representations+of+discourse+referents%2E+Cambridge+University+Press%2E)
25. Liu, Y., Winckel, E., Abeillé, A., Hemforth, B., & Gibson, E. (2022). Structural, functional, and processing perspectives on linguistic island effects. Annual Review of Linguistics, 8(1), 495-525. [↗](https://doi.org/10.1146/annurev-linguistics-011619-030319)
26. Lu, J., Pan, D., & Degen, J. (2024). Evidence for a discourse account of manner-of-speaking islands. LSA presentation. [↗](https://doi.org/10.1353/lan.2025.a978271)
27. Mandelkern, M., & Linzen, T. (2024). Do Language Models’ Words Refer? Computational Linguistics, 1-10. Mao, R., Hemforth, B., and Abeille e, A. (2024) The discourse function explains “islands”. AMLaP. Online hal-04859279 Mulders, I., & Ruys, E. G. (2024). ChatGPT as an informant. Nota Bene, 1(2), 242-260. [↗](https://doi.org/10.1162/coli_a_00522)
28. Namboodiripad, S., Cuneo, N., Kramer, M. A., Sedarous, Y., Bisnath, F., Sugimoto, Y., & Goldberg, A. E. (2022). Backgroundedness predicts island status of non-finite adjuncts in English. Proc. of the Cog. Sci. Soc. [find](https://scholar.google.com/scholar?q=Namboodiripad%2C+S%2E%2C+Cuneo%2C+N%2E%2C+Kramer%2C+M%2E+A%2E%2C+Sedarous%2C+Y%2E%2C+Bisnath%2C+F%2E%2C+Sugimoto%2C+Y%2E%2C+%26+Goldberg%2C+A%2E+E%2E+%282022%29%2E+Backgroundedness+predicts+island+status+of+non%2Dfinite+adjuncts+in+English%2E+Proc%2E+of+the+)
29. Phillips, C. (2006). The real-time status of island phenomena. Language, 795-823. [↗](https://doi.org/10.1353/lan.2006.0217)
30. Potts, C. (2004). The logic of conventional implicatures. OUP. [↗](https://doi.org/10.1093/acprof:oso/9780199273829.001.0001)
31. Potts, C. (2023). Characterizing English Preposing in PP constructions. LingBuzz lingbuzz/007495. [↗](https://doi.org/10.1017/s0022226724000227)
32. Ross, J. R. (1967). Constraints on variables in syntax. Ph.D. Thesis, Massachusetts Institute of Technology. [find](https://scholar.google.com/scholar?q=Ross%2C+J%2E+R%2E+%281967%29%2E+Constraints+on+variables+in+syntax%2E+Ph%2ED%2E+Thesis%2C+Massachusetts+Institute+of+Technology%2E)
33. Sieker, J., & Zarrieß, S. (2023, December). [When Your Language Model Cannot Even Do Determiners Right: Probing for Anti-Presuppositions and the Maximize Presupposition! Principle](https://md.research.elamur.ai/free/en/papers/sieker-2023-when-your-language-model-cannot/). In Proc. of the 6th blackboxnlp workshop: Analyzing and interpreting neural networks for NLP (pp. 180-198).
34. Sprouse, J. (2013) (Ed). Experimental syntax and island effects. Cambridge University Press. [↗](https://doi.org/10.5281/zenodo.8100574)
35. Sravanthi, S. L., Doshi, M., Kalyan, T. P., Murthy, R., Bhattacharyya, P., & Dabre, R. (2024). PUB: A Pragmatics Understanding Benchmark for Assessing LLMs' Pragmatics Capabilities. arXiv preprint arXiv:2401.07078. Warstadt, A., Parrish, A., Liu, H., Mohananey, A., Peng, E., Wang, S-F., and Bowman, S.R. “[BLiMP: The Benchmark of Linguistic Minimal Pairs for English](https://md.research.elamur.ai/free/en/papers/warstadt-2020-blimp-benchmark-linguistic-minimal-pairs/).” Transactions of the ACL 8 (December 2020): 377–92. //doi.org/10.1162/tacl_a_00321. [↗](https://doi.org/10.1162/tacl_a_00321)
36. Ward, G., & Birner, B. J. (2019). 12 Discourse effects of word order variation. Semantics: Sentence and Information Structure, 413. [↗](https://doi.org/10.1515/9783110589863-012)
37. Winckel, E., Abeillé, A., Hemforth, B., & Gibson, E. (2025). Discourse-based constraints on long-distance dependencies generalize across constructions in English and French. Cognition, 254, 105950. [↗](https://doi.org/10.1016/j.cognition.2024.105950)

*Updated 1 November 2026*