Emergence of Hierarchical Emotion Organization in Large Language Models

Maya Okawa, Bo Zhao, Eric Bigelow, Rose Yu, Tomer Ullman, Ekdeep Singh Lubana, Hidenori Tanaka
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:94528-94549, 2026.

Abstract

As large language models (LLMs) increasingly power conversational agents, understanding how they model users’ emotional states is critical for ethical deployment. Inspired by emotion wheels, i.e., a psychological framework that argues emotions organize hierarchically, we analyze probabilistic dependencies between emotional states in model outputs. We find that LLMs naturally form hierarchical emotion trees that align with human psychological models, and larger models develop more complex hierarchies. We also uncover systematic biases in emotion recognition across socioeconomic personas, with compounding misclassifications for intersectional, underrepresented groups. Human studies reveal striking parallels, suggesting that LLMs internalize aspects of social perception. Beyond highlighting emergent emotional reasoning in LLMs, our results hint at the potential of using cognitively-grounded theories for developing better model evaluations.

Cite this Paper


BibTeX
@InProceedings{pmlr-v306-okawa26b, title = {Emergence of Hierarchical Emotion Organization in Large Language Models}, author = {Okawa, Maya and Zhao, Bo and Bigelow, Eric and Yu, Rose and Ullman, Tomer and Lubana, Ekdeep Singh and Tanaka, Hidenori}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {94528--94549}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/okawa26b/okawa26b.pdf}, url = {https://proceedings.mlr.press/v306/okawa26b.html}, abstract = {As large language models (LLMs) increasingly power conversational agents, understanding how they model users’ emotional states is critical for ethical deployment. Inspired by emotion wheels, i.e., a psychological framework that argues emotions organize hierarchically, we analyze probabilistic dependencies between emotional states in model outputs. We find that LLMs naturally form hierarchical emotion trees that align with human psychological models, and larger models develop more complex hierarchies. We also uncover systematic biases in emotion recognition across socioeconomic personas, with compounding misclassifications for intersectional, underrepresented groups. Human studies reveal striking parallels, suggesting that LLMs internalize aspects of social perception. Beyond highlighting emergent emotional reasoning in LLMs, our results hint at the potential of using cognitively-grounded theories for developing better model evaluations.} }
Endnote
%0 Conference Paper %T Emergence of Hierarchical Emotion Organization in Large Language Models %A Maya Okawa %A Bo Zhao %A Eric Bigelow %A Rose Yu %A Tomer Ullman %A Ekdeep Singh Lubana %A Hidenori Tanaka %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-okawa26b %I PMLR %P 94528--94549 %U https://proceedings.mlr.press/v306/okawa26b.html %V 306 %X As large language models (LLMs) increasingly power conversational agents, understanding how they model users’ emotional states is critical for ethical deployment. Inspired by emotion wheels, i.e., a psychological framework that argues emotions organize hierarchically, we analyze probabilistic dependencies between emotional states in model outputs. We find that LLMs naturally form hierarchical emotion trees that align with human psychological models, and larger models develop more complex hierarchies. We also uncover systematic biases in emotion recognition across socioeconomic personas, with compounding misclassifications for intersectional, underrepresented groups. Human studies reveal striking parallels, suggesting that LLMs internalize aspects of social perception. Beyond highlighting emergent emotional reasoning in LLMs, our results hint at the potential of using cognitively-grounded theories for developing better model evaluations.
APA
Okawa, M., Zhao, B., Bigelow, E., Yu, R., Ullman, T., Lubana, E.S. & Tanaka, H.. (2026). Emergence of Hierarchical Emotion Organization in Large Language Models. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:94528-94549 Available from https://proceedings.mlr.press/v306/okawa26b.html.

Related Material