Elamur Research⌕
☰
home / papers / 2026 / Appraisal Trajectories in Narratives Reveal Distinct Patterns of Emotion Evocation

Appraisal Trajectories in Narratives Reveal Distinct Patterns of Emotion Evocation

Paper · 2026 · Johannes Schäfer, Janne Wagner and Roman Klinger

Summary

Most computational work on appraisal theory scores how a person evaluates an event at a single moment, for example in one sentence or one post. This paper argues that appraisal is a sequential process and asks two questions. First, can appraisals be recovered from a narrative without knowing which emotion it evokes? Second, do the ways appraisals develop over a story differ by emotion?

The authors used the Emotional Backstories (EBS) dataset, which contains LLM-generated stories of five events, each linked to one emotion. They prompted Llama-3.3 (70B, instruction-tuned) zero-shot to rate 21 appraisal dimensions from Troiano et al. (2023) on a 1–5 Likert scale. To build trajectories, they scored growing prefixes of each story, from the first sentence up to the full narrative.

Telling the model the target emotion changed its ratings only slightly, so the text alone carries enough appraisal cues. Across emotions, the trajectories fell into four recurring shapes. This suggests that single-point appraisal analysis misses information that could help emotion classification.

Key points

  • Data and model: EBS has 13,000 narratives across 13 emotion categories (1,000 per category). Appraisals were predicted with Llama-3.3-70B using zero-shot prompts.
  • Accuracy check: On crowd-enVent, the zero-shot approach reached RMSE 1.49/1.46, compared with 1.40 for the fine-tuned model of Troiano et al. (2023).
  • Effect of telling the model the emotion:
    • Pleasantness rose by 0.2–0.8 for positive emotions and fell by 0.3–0.7 for negative ones.
    • Only the standards dimension shifted by about 1 point or more (+1.0 anger, +1.1 guilt, +1.4 shame).
  • Trajectory shapes: The four shapes were Rising, Falling, Valley-shaped and Hill-shaped.
    • Suddenness rises for every emotion.
    • Pleasantness rises for joy but forms a valley for relief.
    • Self-responsibility falls for anger but rises for trust.
  • Limitations and resources: The experiments used one LLM, a single run, and synthetic data. Code and predictions are published by the University of Bamberg.

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 of BambergBamberg · Germany5 works · 65 citations

 Fields, topics and concepts

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksPsychology · 1,552 worksExperimental and Cognitive Psychology · 870 worksSocial Psychology · 306 works

🧭 Topics 1Research subjects and languages the work is about.Sentiment & emotion · 972 works

Cite

Cite as Johannes Schäfer, Janne Wagner and Roman Klinger. 2026. Appraisal Trajectories in Narratives Reveal Distinct Patterns of Emotion Evocation. The Proceedings for the 15th Workshop on Computational Approaches to Subjectivity, Sentiment Social Media Analysis (WASSA 2026).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.wassa-1.7

@inproceedings{schafer2026appraisal,
  title = {{Appraisal Trajectories in Narratives Reveal Distinct Patterns of Emotion Evocation}},
  author = {Johannes Schäfer, Janne Wagner and Roman Klinger},
  booktitle = {The Proceedings for the 15th Workshop on Computational Approaches to Subjectivity, Sentiment Social Media Analysis (WASSA 2026)},
  year = {2026},
  doi = {10.18653/v1/2026.wassa-1.7},
  url = {https://doi.org/10.18653/v1/2026.wassa-1.7}
}

References

15 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Johannes Schäfer and Roman Klinger. 2025. Shaping event backstories to estimate potential emotion contexts. Preprint, arXiv:2508.09954. — [arXiv:2508.09954](https://arxiv.org/abs/2508.09954) find
  2. Alok Debnath, Yvette Graham, and Owen Conlan. 2025. An appraisal theoretic approach to modelling affect flow in conversation corpora. In Proceedings of the 29th Conference on Computational Natural Language Learning, pages 233–250, Vienna, Austria. Association for Computational Linguistics. find
  3. Alexander Staller and Paolo Petta. 2001. Introducing emotions into the computational study of social norms: A first evaluation. Journal of Artificial Societies and Social Simulation, 4(1). find
  4. Marco Antonio Stranisci, Simona Frenda, Eleonora Ceccaldi, Valerio Basile, Rossana Damiano, and Viviana Patti. 2022. APPReddit: a corpus of Reddit posts annotated for appraisal. In Proceedings of the Thirteenth Language Resources and Evaluation Conference, pages 3809–3818, Marseille, France. Europea… find
  5. Paul Ekman. 1972. Universals and cultural differences in facial expressions of emotion. In Nebraska symposium on motivation 1971, volume 19. Lincoln University of Nebraska Press. find
  6. Jonathan Gratch, Stacy Marsella, Ning Wang, and Brooke Stankovic. 2009. Assessing the validity of appraisal-based models of emotion. In 2009 3rd International Conference on Affective Computing and Intelligent Interaction and Workshops, pages 1–8. find
  7. Enrica Troiano, Laura Oberländer, and Roman Klinger. 2023. Dimensional modeling of emotions in text with appraisal theories: Corpus creation, annotation reliability, and prediction. Computational Linguistics, 49(1):1–72. find
  8. Jan Hofmann, Enrica Troiano, Kai Sassenberg, and Roman Klinger. 2020. Appraisal theories for emotion classification in text. In Proceedings of the 28th International Conference on Computational Linguistics, pages 125–138, Barcelona, Spain (Online). International Committee on Computational Linguistic… find
  9. Eileen Wemmer, Sofie Labat, and Roman Klinger. 2024. EmoProgress: Cumulated emotion progression analysis in dreams and customer service dialogues. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 5660–… find
  10. Marc D Lewis. 2001. Personal pathways in the development of appraisal. In Klaus R Scherer, Angela Schorr, and Tom Johnstone, editors, Appraisal processes in emotion: Theory, methods, research, pages 205–220. Oxford University Press: London. find
  11. Hongli Zhan, Desmond C. Ong, and Junyi Jessy Li. 2023. Evaluating subjective cognitive appraisals of emotions from large language models. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 14418–14446, Singapore. Association for Computational Linguistics. find
  12. June M. Liu, He Cao, Renliang Sun, Rui Wang, Yu Li, and Jiaxing Zhang. 2025. CAPE: A Chinese dataset for appraisal-based emotional generation in large language models. In Findings of the Association for Computational Linguistics: NAACL 2025, pages 6291–6309, Albuquerque, New Mexico. Association for… find
  13. Yuxiang Zhou, Hainiu Xu, Desmond C. Ong, Maria Liakata, Petr Slovak, and Yulan He. 2025. Modeling subjectivity in cognitive appraisal with language models. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 13811– 13833, Suzhou, China. Association for Computational Lingu… find
  14. Meta. 2025. Llama (model llama-3.3-70b-instruct). Large language model. find
  15. Saif Mohammad. 2012. \#emotional tweets. In *SEM 2012: The First Joint Conference on Lexical and Computational Semantics – Volume 1: Proceedings of the main conference and the shared task, and Volume 2: Proceedings of the Sixth International Workshop on Semantic Evaluation (SemEval 2012), pages 246–2…

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/schafer-2026-appraisal-trajectories-narratives-reveal-distinct/