IDEAS home Printed from https://ideas.repec.org/p/kue/epaper/e-25-006-v2.html

Comparing Risk Preferences and Reference Dependence in Humans and AI: A Persona-Based Approach with Fine-Tuning

Author

Listed:
  • Ryota IWAMOTO
  • Takunori ISHIHARA
  • Takanori IDA

Abstract

This study empirically investigates the differences in risk preferences and reference dependence between humans and generative AI. We conduct a nationwide online survey of 4,838 individuals and generate AI responses under identical conditions by using personas constructed from demographic attributes. The results show that in gain domains, both humans and the AI select risk-averse options and exhibit similar preference patterns. However, in loss domains, AI shows a stronger risk-loving tendency and responds more sharply to individual attributes such as gender, age, and income. We retrain the AI by fine-tuning it based on human choice data. After fine-tuning, the AI’s risk preference tendency moves closer to that of humans, with loss-related decisions showing the greatest improvement. Using Brier score and log loss, we suggest that fine-tuning partly reduces the gap in risk preference and reference dependence between AI and humans.

Suggested Citation

  • Ryota IWAMOTO & Takunori ISHIHARA & Takanori IDA, 2026. "Comparing Risk Preferences and Reference Dependence in Humans and AI: A Persona-Based Approach with Fine-Tuning," Discussion papers e-25-006-v2, Graduate School of Economics , Kyoto University.
    • Handle: RePEc:kue:epaper:e-25-006-v2
    as

    Download full text from publisher

    File URL: http://www.econ.kyoto-u.ac.jp/dp/papers/e-25-006-V2.pdf
    Download Restriction: no
    ---><---

    References listed on IDEAS

    as
    1. Tversky, Amos & Kahneman, Daniel, 1992. "Advances in Prospect Theory: Cumulative Representation of Uncertainty," Journal of Risk and Uncertainty, Springer, vol. 5(4), pages 297-323, October.
    2. Fulin Guo, 2023. "GPT in Game Theory Experiments," Papers 2305.05516, arXiv.org, revised Dec 2023.
    3. John J. Horton & Apostolos Filippas & Benjamin S. Manning, 2023. "Large Language Models as Simulated Economic Agents: What Can We Learn from Homo Silicus?," NBER Working Papers 31122, National Bureau of Economic Research, Inc.
    4. Peiyao Li & Noah Castelo & Zsolt Katona & Miklos Sarvary, 2024. "Frontiers: Determining the Validity of Large Language Models for Automated Perceptual Analysis," Marketing Science, INFORMS, vol. 43(2), pages 254-266, March.
    5. Elif Akata & Lion Schulz & Julian Coda-Forno & Seong Joon Oh & Matthias Bethge & Eric Schulz, 2025. "Playing repeated games with large language models," Nature Human Behaviour, Nature, vol. 9(7), pages 1380-1390, July.
    6. Armin Falk & Anke Becker & Thomas Dohmen & Benjamin Enke & David B. Huffman & Uwe Sunde, 2017. "Global Evidence on Economic Preferences," NBER Working Papers 23943, National Bureau of Economic Research, Inc.
    7. Alexander L. Brown & Taisuke Imai & Ferdinand M. Vieider & Colin F. Camerer, 2024. "Meta-analysis of Empirical Estimates of Loss Aversion," Journal of Economic Literature, American Economic Association, vol. 62(2), pages 485-516, June.
    8. Leland Bybee, 2023. "Surveying Generative AI's Economic Expectations," Papers 2305.02823, arXiv.org, revised May 2023.
    9. Armin Falk & Anke Becker & Thomas Dohmen & Benjamin Enke & David Huffman & Uwe Sunde, 2018. "Global Evidence on Economic Preferences," The Quarterly Journal of Economics, President and Fellows of Harvard College, vol. 133(4), pages 1645-1692.
    10. Qiaozhu Mei & Yutong Xie & Walter Yuan & Matthew O. Jackson, 2024. "A Turing test of whether AI chatbots are behaviorally similar to humans," Proceedings of the National Academy of Sciences, Proceedings of the National Academy of Sciences, vol. 121(9), pages 2313925121-, February.
    11. Ayato Kitadai & Sinndy Dayana Rico Lugo & Yudai Tsurusaki & Yusuke Fukasawa & Nariaki Nishino, 2024. "Can AI with High Reasoning Ability Replicate Human-like Decision Making in Economic Experiments?," Papers 2406.11426, arXiv.org.
    12. Jingru Jia & Zehua Yuan & Junhao Pan & Paul E. McNamara & Deming Chen, 2024. "Decision-Making Behavior Evaluation Framework for LLMs under Uncertain Context," Papers 2406.05972, arXiv.org, revised Nov 2024.
    13. Philip Brookins & Jason DeBacker, 2024. "Playing games with GPT: What can we learn about a large language model from canonical strategic games?," Economics Bulletin, AccessEcon, vol. 44(1), pages 25-37.
    14. Daniel Kahneman & Amos Tversky, 2013. "Prospect Theory: An Analysis of Decision Under Risk," World Scientific Book Chapters, in: Leonard C MacLean & William T Ziemba (ed.), HANDBOOK OF THE FUNDAMENTALS OF FINANCIAL DECISION MAKING Part I, chapter 6, pages 99-127, World Scientific Publishing Co. Pte. Ltd..
    15. Yang Chen & Samuel N. Kirshner & Anton Ovchinnikov & Meena Andiappan & Tracy Jenkin, 2025. "A Manager and an AI Walk into a Bar: Does ChatGPT Make Biased Decisions Like We Do?," Manufacturing & Service Operations Management, INFORMS, vol. 27(2), pages 354-368, March.
    16. Fabio Motoki & Valdemar Pinho Neto & Victor Rodrigues, 2024. "More human than human: measuring ChatGPT political bias," Public Choice, Springer, vol. 198(1), pages 3-23, January.
    17. Bisbee, James & Clinton, Joshua D. & Dorff, Cassy & Kenkel, Brenton & Larson, Jennifer M., 2024. "Synthetic Replacements for Human Survey Data? The Perils of Large Language Models," Political Analysis, Cambridge University Press, vol. 32(4), pages 401-416, October.
    Full references (including those not matched with items on IDEAS)

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Ryota IWAMOTO & Takunori ISHIHARA & Takanori IDA, 2025. "Comparing Risk Preferences and Reference Dependence in Humans and AI: A Persona-Based Approach with Fine-Tuning「Revised in July 2026.」," Discussion papers e-25-006, Graduate School of Economics , Kyoto University.
    2. Tianjia Dong & Nadav Kunievsky & James A. Evans, 2026. "Measuring Behavior Portability in Large Language Models," Papers 2606.22797, arXiv.org.
    3. Huong Trang Kim & Quang Nguyen, 2024. "Linking top managers’ behavioural traits with business practices and firm performance," Review of Managerial Science, Springer, vol. 18(11), pages 3253-3296, November.
    4. Ding, Yihong & Robinson, Elizabeth & Balcombe, Kelvin, 2025. "The role of experience in climate adaptation: Evidence from a field experiment in China," China Economic Review, Elsevier, vol. 94(PB).
    5. Julia Ihli, Hanna & Chiputwa, Brian & Winter, Etti & Gassner, Anja, 2022. "Risk and time preferences for participating in forest landscape restoration: The case of coffee farmers in Uganda," World Development, Elsevier, vol. 150(C).
    6. Jonathan Chapman & Erik Snowberg & Stephanie Wang & Colin Camerer, 2018. "Loss Attitudes in the U.S. Population: Evidence from Dynamically Optimized Sequential Experimentation (DOSE)," CESifo Working Paper Series 7262, CESifo.
    7. Slosse, Wannes & Mutuku, Kennedy Vaati & de Clercq, Michaël & D’Haese, Marijke & Schoors, Koen & Buysse, Jeroen, 2025. "Uncertainty and inequality preferences in the Ultimatum Game: A case study on Kenyan smallholder farmers," Journal of Behavioral and Experimental Economics (formerly The Journal of Socio-Economics), Elsevier, vol. 118(C).
    8. Thomas Dohmen & Simone Quercia & Jana Willrodt, 2023. "On the psychology of the relation between optimism and risk taking," Journal of Risk and Uncertainty, Springer, vol. 67(2), pages 193-214, October.
    9. Schaewitz, Johannes & Wang, Mei & Rieger, Marc Oliver, 2022. "Culture and Institutions: Long-lasting effects of communism on risk and time preferences of individuals in Europe," Journal of Economic Behavior & Organization, Elsevier, vol. 202(C), pages 785-829.
    10. Martin G. Kocher, 2025. "Measuring uncertainty preferences: what we know and what we need," The Geneva Risk and Insurance Review, Palgrave Macmillan;International Association for the Study of Insurance Economics (The Geneva Association), vol. 50(2), pages 119-135, September.
    11. Inhwa Kim & Keith J. Gamble, 2022. "Too much or too little information: how unknown uncertainty fuels time inconsistency," SN Business & Economics, Springer, vol. 2(2), pages 1-33, February.
    12. Upravitelev, A., 2023. "Neoclassical roots of behavioral economics," Journal of the New Economic Association, New Economic Association, vol. 58(1), pages 110-140.
    13. Yingnan Yan & Tianming Liu & Yafeng Yin, 2025. "Valuing Time in Silicon: Can Large Language Models Replicate Human Value of Travel Time," Papers 2507.22244, arXiv.org, revised Dec 2025.
    14. Pietro Bini & Lin William Cong & Xing Huang & Lawrence J. Jin, 2026. "Behavioral Economics of AI: LLM Biases and Corrections," Papers 2602.09362, arXiv.org.
    15. Cappelen, Alexander W. & Sørensen, Erik Ø. & Tungodden, Bertil & Xu, Xiaogeng, 2025. "Risk taking on behalf of others: Does the timing of uncertainty revelation matter?," Discussion Paper Series in Economics 13/2025, Norwegian School of Economics, Department of Economics.
    16. Shenkman, E. & Parshakov, P. & Matkin , N., 2026. "Comparative analysis of large language models and human behavior in the Dictator Game," Journal of the New Economic Association, New Economic Association, vol. 71(2), pages 177-203.
    17. Mikhail Freer & Daniel Friedman & Christian Ghiglino & Elke Weidenholzer, 2026. "The Joneses Visit an Economics Lab," Papers 2607.07353, arXiv.org.
    18. Brandts, Jordi & El Baroudi, Sabrine & Huber, Stefanie J. & Rott, Christina, 2021. "Gender differences in private and public goal setting," Journal of Economic Behavior & Organization, Elsevier, vol. 192(C), pages 222-247.
    19. Erdem, Orhan & Ashok, Ragavi Pobbathi, 2026. "LLMs are not weird: Comparing AI and human financial decision-making," Journal of Behavioral and Experimental Economics (formerly The Journal of Socio-Economics), Elsevier, vol. 120(C).
    20. Heckman, James J. & Galaty, Bridget & Tian, Haihan, 2023. "The Economic Approach to Personality, Character and Virtue," IZA Discussion Papers 16133, IZA Network @ LISER.

    More about this item

    Keywords

    ;
    ;
    ;
    ;
    ;
    ;
    ;

    JEL classification:

    • D91 - Microeconomics - - Micro-Based Behavioral Economics - - - Role and Effects of Psychological, Emotional, Social, and Cognitive Factors on Decision Making
    • C91 - Mathematical and Quantitative Methods - - Design of Experiments - - - Laboratory, Individual Behavior

    NEP fields

    This paper has been announced in the following NEP Reports:

    Statistics

    Access and download statistics

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:kue:epaper:e-25-006-v2. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Graduate School of Economics Project Center (email available below). General contact details of provider: https://edirc.repec.org/data/fekyojp.html .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.