{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,17]],"date-time":"2025-12-17T13:07:50Z","timestamp":1765976870717,"version":"build-2065373602"},"reference-count":24,"publisher":"MDPI AG","issue":"9","license":[{"start":{"date-parts":[[2025,9,2]],"date-time":"2025-09-02T00:00:00Z","timestamp":1756771200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Instituto Presbiteriano Mackenzie","award":["PQ 303356\/2022-7","88881.694458\/2022-01"],"award-info":[{"award-number":["PQ 303356\/2022-7","88881.694458\/2022-01"]}]},{"name":"Brazilian agencies CNPq (Conselho Nacional de Desenvolvimento Cient\u00edfico e Tecnol\u00f3gico)","award":["PQ 303356\/2022-7","88881.694458\/2022-01"],"award-info":[{"award-number":["PQ 303356\/2022-7","88881.694458\/2022-01"]}]},{"name":"CAPES (Coordena\u00e7\u00e3o de Aperfei\u00e7oamento de Pessoal de N\u00edvel Superior)","award":["PQ 303356\/2022-7","88881.694458\/2022-01"],"award-info":[{"award-number":["PQ 303356\/2022-7","88881.694458\/2022-01"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Entropy"],"abstract":"<jats:p>Understanding and modeling the flow of information in human societies is essential for capturing phenomena such as polarization, opinion formation, and misinformation diffusion. Traditional agent-based models often rely on simplified behavioral rules that fail to capture the nuanced and context-sensitive nature of human decision-making. In this study, we explore the potential of Large Language Models (LLMs) as data-driven, high-fidelity agents capable of simulating individual opinions under varying informational conditions. Conditioning LLMs on real survey data from the 2020 American National Election Studies (ANES), we investigate their ability to predict individual-level responses across a spectrum of political and social issues in a zero-shot setting, without any training on the survey outcomes. Using Jensen\u2013Shannon distance to quantify divergence in opinion distributions and F1-score to measure predictive accuracy, we compare LLM-generated simulations to those produced by a supervised Random Forest model. While performance at the individual level is comparable, LLMs consistently produce aggregate opinion distributions closer to the empirical ground truth. These findings suggest that LLMs offer a promising new method for simulating complex opinion dynamics and modeling the probabilistic structure of belief systems in computational social science.<\/jats:p>","DOI":"10.3390\/e27090923","type":"journal-article","created":{"date-parts":[[2025,9,2]],"date-time":"2025-09-02T16:05:22Z","timestamp":1756829122000},"page":"923","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Simulating Public Opinion: Comparing Distributional and Individual-Level Predictions from LLMs and Random Forests"],"prefix":"10.3390","volume":"27","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-4542-5048","authenticated-orcid":false,"given":"Fernando","family":"Miranda","sequence":"first","affiliation":[{"name":"Programa de P\u00f3s-Gradua\u00e7\u00e3o em Engenharia El\u00e9trica e Computa\u00e7\u00e3o, Universidade Presbiteriana Mackenzie, S\u00e3o Paulo 01302-000, SP, Brazil"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6022-0270","authenticated-orcid":false,"given":"Pedro Paulo","family":"Balbi","sequence":"additional","affiliation":[{"name":"Programa de P\u00f3s-Gradua\u00e7\u00e3o em Engenharia El\u00e9trica e Computa\u00e7\u00e3o, Universidade Presbiteriana Mackenzie, S\u00e3o Paulo 01302-000, SP, Brazil"},{"name":"Faculdade de Computa\u00e7\u00e3o e Inform\u00e1tica, Universidade Presbiteriana Mackenzie, S\u00e3o Paulo 01302-000, SP, Brazil"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2025,9,2]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Feng, S., Park, C.Y., Liu, Y., and Tsvetkov, Y. (2023). From pretraining data to language models to downstream tasks: Tracking the trails of political biases leading to unfair NLP models. arXiv.","DOI":"10.18653\/v1\/2023.acl-long.656"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"258","DOI":"10.1038\/s42256-022-00458-8","article-title":"Large pre-trained language models contain human-like biases of what is right and wrong to do","volume":"4","author":"Schramowski","year":"2022","journal-title":"Nat. Mach. Intell."},{"key":"ref_3","unstructured":"Lin, L., Wang, L., Guo, J., and Wong, K.F. (2024). Investigating Bias in LLM-Based Bias Detection: Disparities between LLMs and Human Perception. arXiv."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Raj, C., Mukherjee, A., Caliskan, A., Anastasopoulos, A., and Zhu, Z. (2024, January 21\u201323). Breaking bias, building bridges: Evaluation and mitigation of social biases in llms via contact hypothesis. Proceedings of the AAAI\/ACM Conference on AI, Ethics, and Society, San Jose, CA, USA.","DOI":"10.1609\/aies.v7i1.31715"},{"key":"ref_5","unstructured":"Santurkar, S., Durmus, E., Ladhak, F., Lee, C., Liang, P., and Hashimoto, T. (2023, January 23\u201329). Whose opinions do language models reflect?. Proceedings of the International Conference on Machine Learning. PMLR, Honolulu, HI, USA."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"337","DOI":"10.1017\/pan.2023.2","article-title":"Out of one, many: Using language models to simulate human samples","volume":"31","author":"Argyle","year":"2023","journal-title":"Political Anal."},{"key":"ref_7","first-page":"11785","article-title":"Capturing failures of large language models via human cognitive biases","volume":"35","author":"Jones","year":"2022","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_8","unstructured":"Dasgupta, I., Lampinen, A.K., Chan, S.C., Sheahan, H.R., Creswell, A., Kumaran, D., McClelland, J.L., and Hill, F. (2022). Language models show human-like content effects on reasoning tasks. arXiv."},{"key":"ref_9","first-page":"10622","article-title":"Evaluating and inducing personality in pre-trained language models","volume":"36","author":"Jiang","year":"2023","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_10","unstructured":"Park, J.S., O\u2019Brien, J., Cai, C.J., Morris, M.R., Liang, P., and Bernstein, M.S. (November, January 29). Generative agents: Interactive simulacra of human behavior. Proceedings of the 36th Annual ACM Symposium on User Interface Software and Technology, Francisco, CA, USA."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Wang, Z., Zhang, D., Agrawal, I., Gao, S., Song, L., and Chen, X. (2025). Beyond Profile: From Surface-Level Facts to Deep Persona Simulation in LLMs. arXiv.","DOI":"10.18653\/v1\/2025.findings-acl.1094"},{"key":"ref_12","unstructured":"Xu, R., Wang, X., Chen, J., Yuan, S., Yuan, X., Liang, J., Chen, Z., Dong, X., and Xiao, Y. (2024). Character is Destiny: Can Large Language Models Simulate Persona-Driven Decisions in Role-Playing?. arXiv."},{"key":"ref_13","unstructured":"Liu, Y., Sharma, P., Oswal, M.J., Xia, H., and Huang, Y. (2024). Personaflow: Boosting research ideation with llm-simulated expert personas. arXiv."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Cheng, M., Piccardi, T., and Yang, D. (2023). CoMPosT: Characterizing and evaluating caricature in LLM simulations. arXiv.","DOI":"10.18653\/v1\/2023.emnlp-main.669"},{"key":"ref_15","unstructured":"Aher, G.V., Arriaga, R.I., and Kalai, A.T. (2023, January 23\u201329). Using large language models to simulate multiple humans and replicate human subject studies. Proceedings of the International Conference on Machine Learning. PMLR, Honolulu, HI, USA."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Chen, Y., Hu, Y., and Lu, Y. (2025). Predicting Field Experiments with Large Language Models. arXiv.","DOI":"10.2139\/ssrn.5389144"},{"key":"ref_17","unstructured":"Zhang, X., Lin, J., Mou, X., Yang, S., Liu, X., Sun, L., Lyu, H., Yang, Y., Qi, W., and Chen, Y. (2025). SocioVerse: A World Model for Social Simulation Powered by LLM Agents and A Pool of 10 Million Real-World Users. arXiv."},{"key":"ref_18","unstructured":"Anthis, J.R., Liu, R., Richardson, S.M., Kozlowski, A.C., Koch, B., Evans, J., Brynjolfsson, E., and Bernstein, M. (2025). LLM Social Simulations Are a Promising Research Method. arXiv."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1057\/s41599-024-03609-x","article-title":"Performance and biases of Large Language Models in public opinion simulation","volume":"11","author":"Qu","year":"2024","journal-title":"Humanit. Soc. Sci. Commun."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Jiang, S., Wei, L., and Zhang, C. (2024). Donald Trumps in the Virtual Polls: Simulating and Predicting Public Opinions in Surveys Using Large Language Models. arXiv.","DOI":"10.31219\/osf.io\/k9eyh"},{"key":"ref_21","first-page":"126","article-title":"Exploring the potential and limitations of large language models as virtual respondents for social science research","volume":"10","author":"Rakovics","year":"2024","journal-title":"Intersect. East Eur. J. Soc. Politics"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Jeong, H.J., and Lee, W.C. (2016). The level of collapse we are allowed: Comparison of different response scales in Safety Attitudes Questionnaire. Biom. Biostat. Int. J., 4.","DOI":"10.15406\/bbij.2016.04.00100"},{"key":"ref_23","unstructured":"Team, G., Kamath, A., Ferret, J., Pathak, S., Vieillard, N., Merhej, R., Perrin, S., Matejovicova, T., Ram\u00e9, A., and Rivi\u00e8re, M. (2025). Gemma 3 technical report. arXiv."},{"key":"ref_24","unstructured":"Yang, A., Yu, B., Li, C., Liu, D., Huang, F., Huang, H., Jiang, J., Tu, J., Zhang, J., and Zhou, J. (2025). Qwen2. 5-1m technical report. arXiv."}],"container-title":["Entropy"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1099-4300\/27\/9\/923\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,9]],"date-time":"2025-10-09T18:38:20Z","timestamp":1760035100000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1099-4300\/27\/9\/923"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,9,2]]},"references-count":24,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2025,9]]}},"alternative-id":["e27090923"],"URL":"https:\/\/doi.org\/10.3390\/e27090923","relation":{},"ISSN":["1099-4300"],"issn-type":[{"type":"electronic","value":"1099-4300"}],"subject":[],"published":{"date-parts":[[2025,9,2]]}}}