Elamur Research⌕
☰
home / papers / 2026 / Surprisal Minimisation over Goal-directed Alternatives Predicts Production Choice in Dialogue

Surprisal Minimisation over Goal-directed Alternatives Predicts Production Choice in Dialogue

Paper · 2026 · Thomas P. Utting, Mario Giulianelli and Arabella Sinclair

Summary

Information-theoretic measures such as surprisal and uniform information density (UID) are often used to explain why speakers pick one phrasing over another. It is usually unclear whether these measures reflect effort for the speaker or for the listener. The authors argue that this depends on the set of alternatives a cost is compared against. Alternatives that all express the speaker's fixed intent (goal-directed) yield a speaker-side reading. Alternatives constrained only by shared context (goal-agnostic) yield a listener-side reading.

They model production as cost-sensitive choice and use language models to generate both kinds of alternative sets. The case study covers matrix-verb choice points in an English open-ended dialogue corpus. Four costs are compared: surprisal, local and global uniformity, and length. The analysis uses both rank-based (deterministic) and conditional-logit (probabilistic) methods.

Surprisal minimisation over goal-directed alternatives predicts human choices best in both analyses, which supports surprisal as a speaker-side pressure. The work offers a scalable way to test pragmatic production models in naturalistic language.

Key points

  • The central idea is that a cost measure has no fixed interpretation. Goal-directed alternatives make it a speaker cost, and goal-agnostic ones make it a listener cost.
  • In the probabilistic model, the surprisal-based model has the highest log-likelihood. Its negative effect on selection is about 7× stronger for goal-directed than for goal-agnostic alternatives.
  • Local and global uniformity predict choice among goal-directed alternatives, but the effect reverses among goal-agnostic ones. Length favours shorter continuations only in the goal-agnostic condition, suggesting a listener pressure, though this model fits worst.
  • The probabilistic analysis is a conditional logit with a cost × goal-condition interaction and standard errors clustered by context.
  • Limitations include a single language (English), one choice-point type, a small sample, whole-continuation cost aggregates, and reliance on LMs. Code and data are at github.com/the-context-lab/productionchoice.

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

🎓University College LondonLondon · United Kingdom107 works · 22,923 citations🎓University of AberdeenAberdeen · United Kingdom23 works · 1,583 citations

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 18,614 worksArtificial Intelligence · 17,965 worksArts and Humanities · 17,450 worksLanguage and Linguistics · 17,296 works

🧭 Topics 2Research subjects and languages the work is about.Dialogue systems · 654 worksPsycholinguistics & language acquisition · 108 works

Cite

Cite as Thomas P. Utting, Mario Giulianelli and Arabella Sinclair. 2026. Surprisal Minimisation over Goal-directed Alternatives Predicts Production Choice in Dialogue. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.acl-long.1814

@inproceedings{utting2026surprisal,
  title = {{Surprisal Minimisation over Goal-directed Alternatives Predicts Production Choice in Dialogue}},
  author = {Thomas P. Utting, Mario Giulianelli and Arabella Sinclair},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-long.1814},
  url = {https://doi.org/10.18653/v1/2026.acl-long.1814}
}

References

66 references, 16 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Matthew Aylett and Alice Turk. 2004. The smooth signal redundancy hypothesis: A functional explanation for relationships between redundancy, prosodic prominence, and duration in spontaneous speech. Language and Speech, 47(1):31–56. PMID: 15298329. find
  2. E.G. Bard, A.H. Anderson, Y. Chen, H.B.M. Nicholson, C. Havard, and S. Dalzel-Job. 2007. Let’s you do that: Sharing the cognitive burdens of dialogue. Journal of Memory and Language, 57(4):616–641. Language-Vision Interaction. find
  3. Leon Bergen, Roger Levy, and Noah Goodman. 2016. Pragmatic reasoning through semantic inference. Semantics and Pragmatics, 9:20–1. find
  4. Simon Betz, Nataliya Bryhadyr, Olcay Türk, and Petra Wagner. 2023. Cognitive load increases spoken and gestural hesitation frequency. Languages, 8(1). find
  5. Kathryn Bock and Willem JM Levelt. 1994. Language production: Grammatical encoding. In Handbook of psycholinguistics, pages 945–984. Academic Press. find
  6. Hyung Won Chung, Le Hou, Shayne Longpre, Barret Zoph, Yi Tay, William Fedus, Yunxuan Li, Xuezhi Wang, Mostafa Dehghani, Siddhartha Brahma, Albert Webson, Shixiang Shane Gu, Zhuyun Dai, Mirac Suzgun, Xinyun Chen, Aakanksha Chowdhery, Alex Castro-Ros, Marie Pellat, Kevin Robinson, and 16 others. 2022.… find
  7. Reuben Cohn-Gordon, Noah Goodman, and Christopher Potts. 2019. An incremental iterated response model of pragmatics. In Proceedings of the Society for Computation in Linguistics (SCiL) 2019, pages 81–90.
  8. Michael Xavier Collins. 2014. Information density and dependency length as complementary cognitive models. Journal of psycholinguistic research, 43:651– 681. find
  9. Christophe Coupé, Yoon Mi Oh, Dan Dediu, and François Pellegrino. 2019. Different languages, similar encoding efficiency: Comparable information rates across the human communicative niche. Science Advances, 5(9):eaaw2594. find
  10. Judith Degen. 2023. The rational speech act framework. Annual Review of Linguistics, 9(Volume 9, 2023):519–540. find
  11. Judith Degen, Michael Franke, and Gerhard Jager. 2013. Cost-based pragmatic inference about referential expressions. In Proceedings of the Annual Meeting of the Cognitive Science Society, volume 35. find
  12. Sebastian Farquhar, Jannik Kossen, Lorenz Kuhn, and Yarin Gal. 2024. Detecting hallucinations in large language models using semantic entropy. Nature, 630(8017):625–630. find
  13. August Fenk and Gertraud Fenk. 1980. Konstanz im Kurzzeitgedächtnis –Konstanz im sprachlichen Informationsfluß? Zeitschrift für experimentelle und angewandte Psychologie, 27(3):400–414. find
  14. Michael C Frank and Noah D Goodman. 2012. Predicting pragmatic reasoning in language games. Science, 336(6084):998–998.
  15. Michael Franke. 2014. Typical use of quantifiers: A probabilistic speaker model. In Proceedings of the Annual Meeting of the Cognitive Science Society, volume 36. find
  16. Richard Futrell. 2023. Information-theoretic principles in incremental language production. Proceedings of the National Academy of Sciences, 120(39):e2220593120. find
  17. Richard Futrell. 2024. An information-theoretic account of availability effects in language production. Topics in Cognitive Science, 16(1):38–53. find
  18. Matteo Gay, Coleman Haley, Mario Giulianelli, and Edoardo Ponti. 2026. Is information density uniform when utterances are grounded on perception and discourse? In Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers), page… find
  19. Dmitriy Genzel and Eugene Charniak. 2002. Entropy rate constancy in text. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, pages 199–206, Philadelphia, Pennsylvania, USA. Association for Computational Linguistics.
  20. Dmitriy Genzel and Eugene Charniak. 2003. Variation of entropy and parse trees of sentences as a function of the sentence number. In Proceedings of the 2003 Conference on Empirical Methods in Natural Language Processing, pages 65–72. find
  21. Mario Giulianelli. 2022. Towards pragmatic production strategies for natural language generation tasks. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 7978–7984, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
  22. Mario Giulianelli, Joris Baan, Wilker Aziz, Raquel Fernández, and Barbara Plank. 2023a. What comes next? Evaluating uncertainty in neural text generators against human production variability. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 14349–14371… find
  23. Mario Giulianelli and Raquel Fernández. 2021. Analysing human strategies of information transmission as a function of discourse context. In Proceedings of the 25th Conference on Computational Natural Language Learning, pages 647–660, Online. Association for Computational Linguistics. find
  24. Mario Giulianelli, Andreas Opedal, and Ryan Cotterell. 2024. Generalized measures of anticipation and re-sponsivity in online language processing. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 11648–11669, Miami, Florida, USA. Association for Computational Linguisti… find
  25. Mario Giulianelli, Arabella Sinclair, and Raquel Fernández. 2021. Is information density uniform in task-oriented dialogues? In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 8271–8283, Online and Punta Cana, Dominican Republic. Association for Computat… find
  26. Mario Giulianelli, Arabella Sinclair, and Raquel Fernández. 2022. Construction repetition reduces information rate in dialogue. In Proceedings of the 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Lan… find
  27. Mario Giulianelli, Sarenne Wallbridge, Ryan Cotterell, and Raquel Fernández. 2026. Incremental alternative sampling as a lens into the temporal and representational resolution of linguistic prediction. Journal of Memory and Language, 148. find
  28. Mario Giulianelli, Sarenne Wallbridge, and Raquel Fernández. 2023b. Information value: Measuring utterance predictability as distance from plausible alternatives. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 5633–5653, Singapore. Association for Co… find
  29. Noah D. Goodman and Daniel Lassiter. 2015. Probabilistic semantics and pragmatics uncertainty in language and thought. In The Handbook of Contemporary Semantic Theory, chapter 21, pages 655–686. John Wiley & Sons, Ltd. find
  30. Britt Hadar, Joshua Skrzypek, Arthur Wingfield, and Boaz Ben-David. 2016. Working memory load affects processing time in spoken word recognition: Evidence from eye-movements. Frontiers in Neuroscience, 10. find
  31. John Hale. 2001. A probabilistic Earley parser as a psycholinguistic model. In Second Meeting of the North American Chapter of the Association for Computational Linguistics.
  32. Barbara Howarth and Anne H. Anderson. 2007. Introducing objects in spoken dialogue: The influence of conversational setting and cognitive load on the articulation and use of referring expressions. Language and Cognitive Processes, 22(2):272–296. find
  33. Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. 2019. Roberta: A robustly optimized bert pretraining approach. Preprint, arXiv:1907.11692.
  34. Jennifer Hu, Roger Levy, Judith Degen, and Sebastian Schuster. 2023. Expectations over unspoken alternatives predict pragmatic inferences. Preprint, arXiv:2304.04758.
  35. R Duncan Luce. 1959. Individual choice behavior, volume 4. Wiley New York. find
  36. Jennifer Hu, Roger Levy, and Sebastian Schuster. 2022. Predicting scalar diversity with context-driven uncertainty over alternatives. In Proceedings of the Workshop on Cognitive Modeling and Computational Linguistics, pages 68–74, Dublin, Ireland. Association for Computational Linguistics.
  37. Iva Ivanova and Victor S. Ferreira. 2019. The role of working memory for syntactic formulation in language production. Journal of Experimental Psychology: Learning, Memory, and Cognition, 45(10):1791–1814. find
  38. T. Florian Jaeger. 2010. Redundancy and reduction: Speakers manage syntactic information density. Cognitive Psychology, 61(1):23–62. find
  39. Ayush Jain, Vishal Singh, Sidharth Ranjan, Rajakrishnan Rajkumar, and Sumeet Agarwal. 2018. Uniform Information Density effects on syntactic choice in Hindi. In Proceedings of the Workshop on Linguistic Complexity and Natural Language Processing, pages 38–48, Santa Fe, New-Mexico. Association for Co… find
  40. Jaap Jumelet, Willem Zuidema, and Arabella Sinclair. 2024. Do language models exhibit human-like structural priming effects? Findings of the Association for Computational Linguistics: ACL 2024, pages 14727– 14742. find
  41. Frank Keller. 2004. The entropy rate principle as a predictor of processing effort: An evaluation against eye-tracking data. In Proceedings of the 2004 Conference on Empirical Methods in Natural Language Processing, pages 317–324, Barcelona, Spain. Association for Computational Linguistics. find
  42. Tatsuki Kuribayashi, Yohei Oseki, and Timothy Baldwin. 2024. Psychometric predictive power of large language models. In Findings of the Association for Computational Linguistics: NAACL 2024, pages 1983–2005, Mexico City, Mexico. Association for Computational Linguistics. find
  43. Roger Levy. 2008. Expectation-based syntactic comprehension. Cognition, 106(3):1126–1177. find
  44. Roger Levy and T. Florian Jaeger. 2006. Speakers optimize information density through syntactic reduction. In Advances in Neural Information Processing Systems, volume 19. MIT Press. find
  45. Clara Meister, Mario Giulianelli, and Tiago Pimentel. 2024. Towards a similarity-adjusted surprisal theory. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 16485–16498, Miami, Florida, USA. Association for Computational Linguistics. find
  46. Clara Meister, Tiago Pimentel, Patrick Haller, Lena Jäger, Ryan Cotterell, and Roger Levy. 2021. Revisiting the Uniform Information Density hypothesis. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 963– 980, Online and Punta Cana, Dominican Republic…
  47. Antje Meyer. 2023. Timing in conversation. Journal of Cognition, 6. find
  48. Iman Mirzadeh, Keivan Alizadeh, Hooman Shahrokhi, Oncel Tuzel, Samy Bengio, and Mehrdad Farajtabar. 2025. Gsm-symbolic: Understanding the limitations of mathematical reasoning in large language models. Preprint, arXiv:2410.05229.
  49. Aron Molnar, Jaap Jumelet, Mario Giulianelli, and Arabella Sinclair. 2023. Attribution and alignment: Effects of local context repetition on utterance production and comprehension in dialogue. In Proceedings of the 27th Conference on Computational Natural Language Learning (CoNLL), pages 254–273, Si… find
  50. Byung-Doh Oh and William Schuler. 2023. Why does surprisal from larger transformer-based language models provide a poorer fit to human reading times? Transactions of the Association for Computational Linguistics, 11.
  51. Jonathan Peelle. 2017. Listening effort: How the cognitive consequences of acoustic challenge are reflected in brain and behavior. Ear and Hearing, 39:1. find
  52. Tiago Pimentel, Clara Meister, Elizabeth Salesky, Simone Teufel, Damián Blasi, and Ryan Cotterell. 2021. A surprisal–duration trade-off across and within the world’s languages. Preprint, arXiv:2109.15000. find
  53. Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever. 2019. Language models are unsupervised multitask learners. OpenAI. Accessed: 2024-11-15. find
  54. 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. find
  55. Arabella Sinclair, Jaap Jumelet, Willem Zuidema, and Raquel Fernández. 2022. Structural persistence in language models: Priming as a window into abstract language representations. Transactions of the Association for Computational Linguistics, 10:1031–1050.
  56. Arabella Sinclair, Anastasia Klimovich-Gray, Jaap Jumelet, Nika Adamian, and Agnieszka Konopka. 2026. Structural priming in humans and large language models. Journal of Memory and Language, 149:104713. find
  57. Nathaniel J. Smith and Roger Levy. 2013. The effect of word predictability on reading time is logarithmic. Cognition, 128(3):302–319. find
  58. Andreas Stolcke, Klaus Ries, Noah Coccaro, Elizabeth Shriberg, Rebecca Bates, Daniel Jurafsky, Paul Taylor, Rachel Martin, Carol Van Ess-Dykema, and Marie Meteer. 2000. Dialogue act modeling for automatic tagging and recognition of conversational speech. Computational Linguistics, 26(3):339–374.
  59. Eleftheria Tsipidi, Samuel Kiegeland, Franz Nowak, Tianyang Xu, Ethan Wilcox, Alex Warstadt, Ryan Cotterell, and Mario Giulianelli. 2025. The harmonic structure of information contours. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)… find
  60. Eleftheria Tsipidi, Franz Nowak, Ryan Cotterell, Ethan Wilcox, Mario Giulianelli, and Alex Warstadt. 2024. Surprise! Uniform Information Density isn’t the whole story: Predicting surprisal contours in long-form discourse. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language… find
  61. Julia White, Jesse Mu, and Noah D Goodman. 2020. Learning to refer informatively by amortizing pragmatic reasoning. In Proceedings of the Annual Meeting of the Cognitive Science Society, volume 42. find
  62. Ethan G. Wilcox, Tiago Pimentel, Clara Meister, Ryan Cotterell, and Roger P. Levy. 2023. Testing the predictions of surprisal theory in 11 languages. Transactions of the Association for Computational Linguistics, 11:1451–1470.
  63. Adina Williams, Nikita Nangia, and Samuel R. Bowman. 2018. A broad-coverage challenge corpus for sentence understanding through inference. Preprint, arXiv:1704.05426.
  64. Yang Xu and David Reitter. 2018. Information density converges in dialogue: Towards an information-theoretic model. Cognition, 170:147–163. find
  65. Jun Sen Yee, Mario Giulianelli, and Arabella J. Sinclair. 2024. Efficiency and effectiveness in task-oriented dialogue: On construction repetition, information rate, and task success. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evalu… find
  66. Lianmin Zheng, Wei-Lin Chiang, Ying Sheng, Siyuan Zhuang, Zhanghao Wu, Yonghao Zhuang, Zi Lin, Zhuohan Li, Dacheng Li, Eric Xing, Hao Zhang, Joseph E. Gonzalez, and Ion Stoica. 2023. Judging LLM-as-a-judge with MT-Bench and Chatbot Arena. In Thirty-seventh Conference on Neural Information Processing…

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/utting-2026-surprisal-minimisation-over-goal-directed/