Please use this identifier to cite or link to this item: https://repository.isls.org/handle/1/11282
Title: Exploring Emotion-related Individual Differences in Adolescents' Daily Lives: An Analysis Using Large Language Models
Authors: Chen, Jingjing
Liu, Xiaobo
Qi, Xuan
Zhang, Yu
Keywords: Learning Sciences
Issue Date: 2025
Publisher: International Society of the Learning Sciences
Citation: Chen, J., Liu, X., Qi, X., & Zhang, Y. (2025). Exploring Emotion-Related Individual Differences in Adolescents' Daily Lives: An Analysis Using Large Language Models. In Rajala, A., Cortez, A., Hofmann, R., Jornet, A., Lotz-Sisitka, H., & Markauskaite, L. (Eds.), Proceedings of the 19th International Conference of the Learning Sciences - ICLS 2025 (pp. 1394-1398). International Society of the Learning Sciences.
Series/Report no.: Proceedings of the 19th International Conference of the Learning Sciences - ICLS 2025
Abstract: Diary methods can help us understand how adolescents recognize and manage emotions in their daily lives, which aids in optimizing social-emotional learning practices. This exploratory study invited 30 students to write their daily entries. These unstructured texts were then automatically coded with the help of large language models, which performed comparably to human raters. Importantly, two indicators (semantic similarity and emotional event alignment) were used to assess how an individual's feelings in specific contexts align with the common understanding of those contexts. We found positive associations between diary-derived indicators for positive entries and performance on emotion perception tasks, while students with lower semantic similarity for negative entries showed higher performance on the emotion regulation task, indicating self-regulation in negative contexts. The study highlighted the potential of using large language models to characterize students' daily emotions and preliminarily demonstrated how individual differences manifest in adolescents' emotional lives.
Description: Short Paper
URI: https://doi.org/10.22318/icls2025.329393
https://repository.isls.org//handle/1/11282
Appears in Collections:ISLS Annual Meeting 2025

Files in This Item:
File SizeFormat 
ICLS2025_1394-1398.pdf398.91 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.