Abstract
Personalized interactions have been discussed as beneficial for learning for decades. Now, with the rise of generative artificial intelligence (GenAI), personalized artificial human-like conversations may impact the quality of learning. Manipulating system prompts to design personalities has the potential to enhance the quality of conversation with Large Language Model (LLM)-based AI. However, it is still uncertain exactly to what extent the emotional tone of a generative AI chatbot is relevant for learning. Hence, the current study evaluates the impact of a chat-based conversation with an LLM-based AI on relevant affective (empathy, compassion, distress) and cognitive (perspective-taking, reflection, knowledge) processes in education for sustainable development. Here, the focus is on both the general impact and the particular impact of two different system prompts that assign the AI’s specific personality traits (empathic vs. compassionate). Comparing these two groups and one control group reading a text (N = 122) indicates that chatting with an empathic AI can elicit stronger emotions (e.g., empathy, compassion, distress) compared to chatting with a compassionate AI, and compared to the control. Although all groups gained knowledge, we found no group differences. Further research is necessary to ensure reliable and contextually appropriate conversations in the context of education.
Data availability
As additional source data, we have uploaded a zip-file containing the code used, a read-me file with requirements and an installation guide, and the icons used for the chat interface. The code consists of a simple python script and is includes comments to ensure easy accessibility. We made this code available via a public osf-link (https://osf.io/u3vm4/overview) and uploaded it as supplementary material 2.The source data used in this research is currently unavailable for public sharing due to strict data safety and confidentiality protocols mandated by our university. These restrictions are in place to ensure compliance with ethical standards, privacy regulations, and institutional policies.Access to the data may be granted under specific circumstances, subject to appropriate data use agreements and ethical approvals. For more information about the data, please contact the corresponding author.
References
Sheikh, H., Prins, C. & Schrijvers, E. Mission AI: the New System Technology (Springer International Publishing, 2023). https://doi.org/10.1007/978-3-031-21448-6
Brown, T. et al. Language models are few-shot learners. In Advances in Neural Information Processing Systems Vol. 33 (eds Larochelle, H., Ranzato, M., Hadsell, R., Balcan, M. F., Lin, H. et al.) 1877–1901 (Curran Associates, Inc., 2020).
Pataranutaporn, P., Liu, R., Finn, E. & Maes, P. Influencing human–AI interaction by priming beliefs about AI can increase perceived trustworthiness, empathy and effectiveness. Nat. Mach. Intell. 5, 1076–1086 (2023). https://doi.org/10.1038/s42256-023-00720-7
Milano, S., McGrane, J. A. & Leonelli, S. Large Language models challenge the future of higher education. Nat. Mach. Intell. 5, 333–334 (2023). https://doi.org/10.1038/s42256-023-00644-2
Pataranutaporn, P. et al. AI-generated characters for supporting personalized learning and well-being. Nat. Mach. Intell. 3, 1013–1022 (2021). https://doi.org/10.1038/s42256-021-00417-9
Ji, H., Han, I. & Ko, Y. A systematic review of conversational AI in Language education: focusing on the collaboration with human teachers. J. Res. Technol. Educ. 55, 48–63 (2023). https://doi.org/10.1080/15391523.2022.2142873
Steinmaurer, A. et al. Immersive learning in history education: exploring the capabilities of virtual avatars and large Language models. In Immersive Learning Research Network Vol. 2271 (eds Krüger, J. M. et al.) 363–374 (Springer Nature Switzerland, 2025).
Ginns, P., Martin, A. J. & Marsh, H. W. Designing instructional text in a conversational style: A Meta-analysis. Educ. Psychol. Rev. 25, 445–472 (2013). https://doi.org/10.1007/s10648-013-9228-0
Mayer, R. E. & Moreno, R. A cognitive theory of multimedia learning: implications for design principles. J. Educ. Psychol. 91, 358–368 (1998).
Rogers, T. B., Kuiper, N. A. & Kirker, W. S. Self-reference and the encoding of personal information. J. Pers. Soc. Psychol. 35, 677–688 (1977).
Moreno, R. & Mayer, R. E. Engaging students in active learning: the case for personalized multimedia messages. J. Educ. Psychol. 92, 724–733 (2000). https://doi.org/10.1037/0022-0663.92.4.724
Schneider, S., Nebel, S., Beege, M. & Rey, G. D. Anthropomorphism in decorative pictures: benefit or harm for learning? J. Educ. Psychol. 110, 218–232 (2018). https://doi.org/10.1037/edu0000207
Schneider, S., Beege, M., Nebel, S., Schnaubert, L. & Rey, G. D. The Cognitive-Affective-Social theory of learning in digital environments (CASTLE). Educ. Psychol. Rev. 34, 1–38 (2022). https://doi.org/10.1007/s10648-021-09626-5
Nass, C., Steuer, J. & Tauber, E. R. ACM Press, Boston, Massachusetts, United States,. Computers are social actors. in Conference companion on Human factors in computing systems - CHI ’94 204 (1994). https://doi.org/10.1145/259963.260288
Nass, C. & Moon, Y. Machines and mindlessness: social responses to computers. J. Soc. Issues. 56, 81–103 (2000). https://doi.org/10.1111/0022-4537.00153
Gambino, A., Fox, J. & Ratan, R. Building a stronger CASA: extending the computers are social actors paradigm. Hum. -Mach Commun. 1, 71–86 (2020). https://doi.org/10.30658/hmc.1.5
Lee, M. K. et al. Personalization in HRI: a longitudinal field experiment. in Proceedings of the seventh annual ACM/IEEE international conference on Human-Robot Interaction 319–326 (ACM, Boston Massachusetts USA, (2012). https://doi.org/10.1145/2157689.2157804
Gkinko, L. & Elbanna, A. Hope, tolerance and empathy: employees’ emotions when using an AI-enabled chatbot in a digitalised workplace. Inf. Technol. People. 35, 1714–1743 (2022). https://doi.org/10.1108/ITP-04-2021-0328
Ma, Z., Mei, Y., Long, Y., Su, Z. & Gajos, K. Z. Evaluating the Experience of LGBTQ+ People Using Large Language Model Based Chatbots for Mental Health Support. in Proceedings of the CHI Conference on Human Factors in Computing Systems 1–15 (ACM, Honolulu HI USA, (2024). https://doi.org/10.1145/3613904.3642482
Ho, A., Hancock, J., Miner, A. S. Psychological Relational, and emotional effects of Self-Disclosure after conversations with a chatbot. J. Commun. 68, 712–733 (2018). https://doi.org/10.1093/joc/jqy026
Bartneck, C. & Keijsers, M. The morality of abusing a robot. Paladyn J. Behav. Robot. 11, 271–283 (2020). https://doi.org/10.1515/pjbr-2020-0017
Carolus, A. et al. Alexa, I feel for you!’ Observers’ Empathetic Reactions towards a Conversational Agent. Front. Comput. Sci. 3, 682982 (2021). https://doi.org/10.3389/fcomp.2021.682982
Rosenthal-von der Pütten, A.M., Krämer, N. C., Hoffmann, N. C., Sobieraj, L., Eimler, S. C. & S. & An Experimental Study on Emotional Reactions Towards a Robot. Int. J. Soc. Robot. 5, 17–34 (2013). https://doi.org/10.1007/s12369-012-0173-8
Kim, W. B. & Hur, H. J. What Makes People Feel Empathy for AI Chatbots? Assessing the Role of Competence and Warmth. Int. J. Human–Computer Interact. 40, 4674–4687 (2024). https://doi.org/10.1080/10447318.2023.2219961
Seo, W., Yang, C., Kim, Y. H. ChaCha: Leveraging Large Language Models to Prompt Children to Share Their Emotions about Personal Events. in Proceedings of the CHI Conference on Human Factors in Computing Systems 1–20 (ACM, Honolulu HI USA, (2024). https://doi.org/10.1145/3613904.3642152
Pataranutaporn, P. et al. Future You: A Conversation with an AI-Generated future self reduces anxiety, negative emotions, and increases future self-continuity. preprint at (2024). https://doi.org/10.48550/ARXIV.2405.12514
Plass, J. L. & Kaplan, U. Emotional design in digital media for learning. in emotions, technology, design, and learning 131–161 (Elsevier, 2016). https://doi.org/10.1016/B978-0-12-801856-9.00007-4
Joksimovic, S., Ifenthaler, D., Marrone, R., De Laat, M. & Siemens, G. Opportunities of artificial intelligence for supporting complex problem-solving: findings from a scoping review. Comput. Educ. Artif. Intell. 4, 100138 (2023). https://doi.org/10.1016/j.caeai.2023.100138
Wang, J. & Fan, W. The effect of ChatGPT on students’ learning performance, learning perception, and higher-order thinking: insights from a meta-analysis. Humanit. Soc. Sci. Commun. 12, 621 (2025). https://doi.org/10.1057/s41599-025-04787-y
Celik, I., Gedrimiene, E., Siklander, S. & Muukkonen, H. The affordances of artificial intelligence-based tools for supporting 21st-century skills: A systematic review of empirical research in higher education. Australas J. Educ. Technol. https://doi.org/10.14742/ajet.9069 (2024).
Yin, J., Goh, T. T. & Hu, Y. Interactions with educational chatbots: the impact of induced emotions and students’ learning motivation. Int. J. Educ. Technol. High. Educ. 21, 47 (2024). https://doi.org/10.1186/s41239-024-00480-3
Meyer, J. et al. Using LLMs to bring evidence-based feedback into the classroom: AI-generated feedback increases secondary students’ text revision, motivation, and positive emotions. Comput. Educ. Artif. Intell. 6, 100199 (2024). https://doi.org/10.1016/j.caeai.2023.100199
Urban, M. et al. ChatGPT improves creative problem-solving performance in university students: an experimental study. Comput. Educ. 215, 105031 (2024). https://doi.org/10.1016/j.compedu.2024.105031
Lai, W. Y. W. & Lee, J. S. A systematic review of conversational AI tools in ELT: publication trends, tools, research methods, learning outcomes, and antecedents. Comput. Educ. Artif. Intell. 7, 100291 (2024). https://doi.org/10.1016/j.caeai.2024.100291
Stadler, M., Bannert, M. & Sailer, M. Cognitive ease at a cost: LLMs reduce mental effort but compromise depth in student scientific inquiry. Comput. Hum. Behav. 160, 108386 (2024). https://doi.org/10.1016/j.chb.2024.108386
Spangenberger, P. et al. Compassion is key: how virtually embodying nature increases connectedness to nature. J. Environ. Psychol. , (2025). https://doi.org/10.1016/j.jenvp.2025.102521
Bianchi, G., Pisiotis, U., Cabrera, M. & GreenComp The European sustainability competence framework. (2022). https://doi.org/10.2760/13286
Grund, J., Singer-Brodowski, M. & Büssing, A. G. Emotions and transformative learning for sustainability: a systematic review. Sustain. Sci. 19, 307–324 (2024). https://doi.org/10.1007/s11625-023-01439-5
Brundiers, K. et al. Key competencies in sustainability in higher education—toward an agreed-upon reference framework. Sustain. Sci. 16, 13–29 (2021). https://doi.org/10.1007/s11625-020-00838-2
Mayer, F. S. & Frantz, C. M. The connectedness to nature scale: A measure of individuals’ feeling in community with nature. J. Environ. Psychol. 24, 503–515 (2004). https://doi.org/10.1016/j.jenvp.2004.10.001
Barragan-Jason, G., de Mazancourt, C., Parmesan, C., Singer, M. C. & Loreau, M. Human-nature connectedness as a pathway to sustainability: A global meta-analysis. Conserv. Lett. 15, e12852 (2022). https://doi.org/10.1111/conl.12852
Otto, S. & Pensini, P. Nature-based environmental education of children: environmental knowledge and connectedness to nature, together, are related to ecological behaviour. Glob Environ. Change. 47, 88–94 (2017). https://doi.org/10.1016/j.gloenvcha.2017.09.009
Jacobson, S. K. et al. Love or loss: effective message framing to promote environmental conservation. Appl. Environ. Educ. Commun. 18, 252–265 (2019). https://doi.org/10.1080/1533015X.2018.1456380
Petrocchi, N. et al. The impact of compassion-focused therapy on positive and negative mental health outcomes: results of a series of meta-analyses. Clin. Psychol. Sci. Pract. 31, 230–247 (2024). https://doi.org/10.1037/cps0000193
Bloom, P. Against Empathy: the Case for Rational Compassion (Ecco an imprint of HarperCollins, 2016).
Singer, T. & Klimecki, O. M. Empathy and compassion. Curr. Biol. 24, R875–R878 (2014). https://doi.org/10.1016/j.cub.2014.06.054
Almarzouki, A. F. Stress, working memory, and academic performance: a neuroscience perspective. Stress 27, 2364333 (2024). https://doi.org/10.1080/10253890.2024.2364333
Luethi, M. Stress effects on working memory, explicit memory, and implicit memory for neutral and emotional stimuli in healthy men. Front Behav. Neurosci. 2, 416 (2008). https://doi.org/10.3389/neuro.08.005.2008
Jo, E., Epstein, D. A., Jung, H. & Kim, Y. H. Understanding the Benefits and challenges of deploying conversational ai leveraging large language models for public health intervention. in Proceedings of the CHI Conference on Human Factors in Computing Systems 1–16 (ACM, Hamburg Germany, 2023). 1–16 (ACM, Hamburg Germany, 2023). (2023). https://doi.org/10.1145/3544548.3581503
Kurian, N. & ‘No, A. no!’: designing child-safe AI and protecting children from the risks of the ‘empathy gap’ in large language models. Learn. Media Technol. https://doi.org/10.1080/17439884.2024.2367052 (2024).
Shumanov, M. & Johnson, L. Making conversations with chatbots more personalized. Comput. Hum. Behav. 117, 106627 (2021). https://doi.org/10.1016/j.chb.2020.106627
Kim, H., Koh, D. Y., Lee, G., Park, J. M. & Lim, Y. Designing Personalities of Conversational Agents. in Extended Abstracts of the CHI Conference on Human Factors in Computing Systems 1–6 (ACM, Glasgow Scotland Uk, 2019). 1–6 (ACM, Glasgow Scotland Uk, 2019). (2019). https://doi.org/10.1145/3290607.3312887
Yaden, D. B. et al. Characterizing empathy and compassion using computational linguistic analysis. Emotion 24, 106–115 (2024). https://doi.org/10.1037/emo0001205
Rutishauser, E., Hérault, B., Petronelli, P. & Sist, P. Tree height reduction after selective logging in a tropical forest. Biotropica 48, 285–289 (2016).
Clark, R. E. Reconsidering research on learning from media. Rev. Educ. Res. 53, 445–459 (1983). https://doi.org/10.3102/003465430530044
Kozma, R. B. Learning with media. Rev. Educ. Res. 61, 179–211 (1991). https://doi.org/10.3102/003465430610021
Spangenberger, P., Freytag, S. C. & Geiger, S. M. Embodying nature in immersive virtual reality: are multisensory stimuli vital to affect nature connectedness and pro-environmental behaviour? Comput. Educ. 212, 104964 (2024). https://doi.org/10.1016/j.compedu.2023.104964
Faul, F., Erdfelder, E., Buchner, A. & Lang, A. G. Statistical power analyses using G*Power 3.1: tests for correlation and regression analyses. Behav. Res. Methods. 41, 1149–1160 (2009). https://doi.org/10.3758/BRM.41.4.1149
Rat Für Sozial- Und Wirtschaftsdaten (RatSWD). Forschungsethische Grundsätze und Prüfverfahren in Den Sozial- und wirtschaftswissenschaften. RatSWD Output Ser. https://doi.org/10.17620/02671.1 (2017).
Shen, L. On a scale of state empathy during message processing. West. J. Commun. 74, 504–524 (2010). https://doi.org/10.1080/10570314.2010.512278
Pfattheicher, S., Sassenrath, C. & Schindler, S. Feelings for the suffering of others and the environment: compassion fosters proenvironmental tendencies. Environ. Behav. 48, 929–945 (2016). https://doi.org/10.1177/0013916515574549
Davis, M. H. A multidimensional approach to individual differences in empathy. JSAS Catalog of Selected Documents in Psychology. 10, 85 (1980).
Davis, M. H. Measuring individual differences in empathy: evidence for a multidimensional approach. J. Pers. Soc. Psychol. 44, 113–126 (1983). https://doi.org/10.1037/0022-3514.44.1.113
Paulus, C. Empathie und die Big Five. Preprint at (2016). https://doi.org/10.22028/D291-23416
Kleespies, M. W., Braun, T., Dierkes, P. W. & Wenzel, V. Measuring connection to nature—A illustrated extension of the inclusion of nature in self scale. Sustainability 13, 1761 (2021). https://doi.org/10.3390/su13041761
Gibbs, G. Learning by doing: a guide to teaching and learning methods (FEU, 1988).
Hatton, N. & Smith, D. Reflection in teacher education: towards definition and implementation. Teach. Teach. Educ. 11, 33–49 (1995). https://doi.org/10.1016/0742-051X(94)00012-U
Rivera, R. The reflective writing continuum: re-conceptualizing Hatton & Smith’s types of reflective writing. Int J. Res. Stud. Educ. 6(2), 49–67 (2016). https://doi.org/10.5861/ijrse.2016.1559
Heynen, E. J. E., Van Der Helm, G. H. P., Stams, G. J. J. M. & Korebrits, A. M. Measuring empathy in a German youth prison: A validation of the German version of the basic empathy scale (BES) in a sample of incarcerated juvenile offenders. J. Forensic Psychol. Pract. 16, 336–346 (2016). https://doi.org/10.1080/15228932.2016.1219217
Jolliffe, D. & Farrington, D. P. Development and validation of the basic empathy scale. J. Adolesc. 29, 589–611 (2006). https://doi.org/10.1016/j.adolescence.2005.08.010
Dornhoff, M., Sothmann, J. N., Fiebelkorn, F. & Menzel, S. Nature relatedness and environmental concern of young people in Ecuador and Germany. Front. Psychol. 10, 453 (2019). https://doi.org/10.3389/fpsyg.2019.00453
Spangenberger, P., Geiger, S. M. & Freytag, S. C. Becoming nature: effects of embodying a tree in immersive virtual reality on nature relatedness. Sci. Rep. 12, 1–11 (2022). https://doi.org/10.1038/s41598-022-05184-0
Pitt, C., Norris, K. & Pecl, G. Informing future directions for climate anxiety interventions: a mixed-method study of professional perspectives. J. Outdoor Environ. Educ. 27, 209–234 (2024). https://doi.org/10.1007/s42322-023-00156-y
Funding
Open Access funding enabled and organized by Projekt DEAL.
Author information
Authors and Affiliations
Contributions
Pia Spangenberger was responsible for concept and design of the work, statistical analysis, the interpretation of data as well as the first draft. Georg F. Reuth programmed the LLM-based AIs , and contributed to the theoretical background, and revised the first draft. Jule M. Krüger contributed substantively to the statistical analysis, and revised the first drafts. Lena Baumann contributed to the theoretical background, and provided feedback to the first draft. Pia Spangenberger and Georg F. Reuth did the acquisition of the original data. Steve Nebel revised the design of the study, and contributed to the first draft.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Ethics
The experiment was part of a teaching unit on Artificial Intelligence in Education at a German university, which did not extend beyond standard educational practices. We conducted the experiment in line with the Declaration of Helsinki (2013). All experimental protocols were carried out in accordance with the guidelines of the German Educational Research Association Society provided as a self-assessment via the German Data Forum (2024), which is endorsed by the German Research Foundation (DFG). All experimental protocols were reviewed and approved internally by the participating lecturers and researchers.
All participants were informed about the purpose of the experiment, and we obtained informed consent from all participants. The informed consent was obtained electronically. The dataset was fully anonymized, containing only non-identifying demographic variables (age and gender). Participation in the study and allowing the data to be used for the analysis was entirely voluntary, and there were no disadvantages for those who chose not to participate or who decided to drop out at any point during the study.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Spangenberger, P., Reuth, G.F., Krüger, J.M. et al. Chatting with an LLM-based AI elicits affective and cognitive processes in education for sustainable development. Sci Rep (2026). https://doi.org/10.1038/s41598-026-39317-6
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-026-39317-6