Simulating Influence Dynamics with LLM Agents

Authors

  • Mehwish Nasim Senior Lecturer, School of Physics, Maths and Computing, Computer Science and Software Engineering, University of Western Australia https://orcid.org/0000-0003-0683-9125
  • Syed Muslim Gilani University of Western Australia
  • Amin Qasmi Lahore University of Management Sciences
  • Usman Naseem Lecturer in Computing, School of Computing, Data Horizons Research Centre, Future Communications Research Centre, Macquarie University, Australia

DOI:

https://doi.org/10.70777/si.v2i1.13971

Keywords:

opinion dynamics, belief systems, llms, large language models, agi misinformation, agi propaganda, agi simulation, agi model, agi world model, social network analysis

Abstract

This paper introduces a simulator designed for opinion dynamics researchers to model competing influences within social networks in the presence of LLM-based agents. By integrating established opinion dynamics principles with state-of-the-art LLMs, this tool enables the study of influence propagation and counter-misinformation strategies. The simulator is particularly valuable for researchers in social science, psychology, and operations research, allowing them to analyse societal phenomena without requiring extensive coding expertise. Additionally, the simulator will be openly available on GitHub, ensuring accessibility and adaptability for those who wish to extend its capabilities for their own research.

Author Biographies

Mehwish Nasim, Senior Lecturer, School of Physics, Maths and Computing, Computer Science and Software Engineering, University of Western Australia

I am a Senior Lecturer in Computer Science at The University of Western Australia. I also hold Adjunct positions at School of Mathematical Sciences at University of Adelaide as well as at Flinders University. I am also the Social Media Coordinator for the Australain Mathematical Society (AustMS).

My research interests include: social network analysis, human-centric cyber security, machine learning, medical image processing, health analytics, and human computer interaction. My current work focuses on predicting population level events in Australia, detecting fake news and misinformation and modeling polarisation on social media. I am also working on models to improve executive decision making using methods from complex systems.

 

Syed Muslim Gilani, University of Western Australia

I recently graduated from the University of Western Australia, having completed the Bachelor of Advanced Computer Science (Honours), and graduated as the top student and the only recipient of a First Class Honours in my degree stream, with an Honours year GPA of 7/7. I have also gained extensive technical experience through industry internships, such as at Woodside, where I upgraded a buoy wave forecasting system by implementing a Transformer NN model, replacing a legacy Autoencoder-based solution, and developing expertise in sequence modelling, neural networks, and natural language processing. Further internships have required large-scale data handling, GPU utilisation, and advanced regression modelling within data science. My internship at Flinders University focused on opinion dynamics using agent-based modelling to visualise and analyse opinion shifts through a red-team-blue-team approach, building an interactive user interface to aid in simulation configuration.

Usman Naseem, Lecturer in Computing, School of Computing, Data Horizons Research Centre, Future Communications Research Centre, Macquarie University, Australia

Dr Usman Naseem is currently a lecturer (~Assistant Professor) in the School of Computing at Macquarie University. Previously, he held a lecturer position in the College of CSE at James Cook University and worked as a research fellow at the University of Sydney and the University of South Australia. He earned his PhD from the University of Sydney, Australia. Before transitioning to academia, he worked in the industry for over 10 years in various technical and leadership roles.

His research interests includes natural language processing, multimodal analysis and social computing, with a specific focus on language modeling and the design of socially aware innovative methods for various applications such as:

(i)  Cyber Informatics
(ii) Online Opinion and Sarcasm mining
(iii) Responsible Code-mixed & Low Resource Language Processing 
(iv) Multimodal Content Analysis 
(v)  Health/Medical Informatics

References

[Aher et al., 2023] Gati V Aher, Rosa I Arriaga, and

Adam Tauman Kalai. Using large language models to simulate

multiple humans and replicate human subject studies.

In International Conference on Machine Learning, pages

–371. PMLR, 2023.

[Carpentras, 2023] Dino Carpentras. Why we are failing at

connecting opinion dynamics to the empirical world. Review

of Artificial Societies and Social Simulations, 2023.

[Chen et al., 2024] Kai Chen, Zihao He, Jun Yan, Taiwei

Shi, and Kristina Lerman. How susceptible are large language

models to ideological manipulation? arXiv preprint

arXiv:2402.11725, 2024.

[Deffuant et al., 2002] Guillaume Deffuant, Fr´ed´eric Amblard,

G´erard Weisbuch, and Thierry Faure. How can extremism

prevail? a study based on the relative agreement

interaction model. Journal of artificial societies and social

simulation, 5(4), 2002.

[Ecker et al., 2022] Ullrich KH Ecker, Stephan

Lewandowsky, John Cook, Philipp Schmid, Lisa K

Fazio, Nadia Brashier, Panayiota Kendeou, Emily K

Vraga, and Michelle A Amazeen. The psychological

drivers of misinformation belief and its resistance to

correction. Nature Reviews Psychology, 1(1):13–29,

[Flache et al., 2017] Andreas Flache, Michael M¨as, Thomas

Feliciani, Edmund Chattoe-Brown, Guillaume Deffuant,

Sylvie Huet, and Jan Lorenz. Models of social influence:

Towards the next frontiers. Jasss-The journal of artificial

societies and social simulation, 20(4):2, 2017.

[Friedkin and Johnsen, 2011] Noah E Friedkin and Eugene

C Johnsen. Social influence network theory: A sociological

examination of small group dynamics, volume 33.

Cambridge University Press, 2011. DOI: https://doi.org/10.4074/S2113520711012138

[Hegselmann and Krause, 2015] Rainer Hegselmann and

Ulrich Krause. Opinion dynamics under the influence

of radical groups, charismatic leaders, and other constant

signals: A simple unifying model. Networks and

Heterogeneous Media, 10(3):477–509, 2015. DOI: https://doi.org/10.3934/nhm.2015.10.477

[Horton, 2023] John J Horton. Large language models as

simulated economic agents: What can we learn from homo

silicus? Technical report, National Bureau of Economic

Research, 2023.

[Kandel, 1986] Denise B Kandel. Processes of peer influences

in adolescence. In Development as action in context:

Problem behavior and normal youth development, pages

–227. Springer, 1986.

[Kelman, 1958] Herbert C Kelman. Compliance, identification,

and internalization three processes of attitude change.

Journal of conflict resolution, 2(1):51–60, 1958. DOI: https://doi.org/10.1177/002200275800200106

[Kelman, 1961] Herbert C Kelman. American association

for public opinion research. The Public Opinion Quarterly,

(1):57–78, 1961.

[Kudiabor, 2024] Helena Kudiabor. Virtual lab powered

by ‘AI scientists’ super-charges biomedical research

[preprint]. Nature, December 2024.

[Lan et al., 2024] Yihuai Lan, Zhiqiang Hu, LeiWang, Yang

Wang, Deheng Ye, Peilin Zhao, Ee-Peng Lim, Hui Xiong,

and Hao Wang. LLM-based agent society investigation:

Collaboration and confrontation in avalon gameplay. In

Proceedings of the 2024 Conference on Empirical Methods

in Natural Language Processing, pages 128–145, Miami,

Florida, USA, November 2024. Association for Computational

Linguistics.

[Leng and Yuan, 2023] Yan Leng and Yuan Yuan. Do

llm agents exhibit social behavior? arXiv preprint

arXiv:2312.15198, 2023.

[Mathias et al., 2016] Jean-Denis Mathias, Sylvie Huet, and

Guillaume Deffuant. Bounded confidence model with

fixed uncertainties and extremists: The opinions can keep

fluctuating indefinitely. Journal of Artificial Societies and

Social Simulation, 19(1):6, 2016.

[Nasim et al., 2022] Mehwish Nasim, Richard Stearne, and

Tim Legrand. A framework for modelling social influence

in a wargame setting, May 2022. Completion Report Submitted

to the Funding Agency.

[National Institute of Standards and Technology (NIST), 2015]

National Institute of Standards and Technology (NIST).

Red team, 2015.

[OpenAI, 2023] OpenAI. Gpt-4 technical report. https:

//openai.com/research/gpt-4, 2023. Accessed: 2024-12-

[Papachristou and Yuan, 2024] Marios Papachristou and

Yuan Yuan. Network formation and dynamics among

multi-llms. arXiv preprint arXiv:2402.10659, 2024.

[Qu and Wang, 2024] Yao Qu and Jue Wang. Performance

and biases of large language models in public opinion simulation.

Humanities and Social Sciences Communications,

(1):1–13, 2024. DOI: https://doi.org/10.1080/14432471.2024.2396666

[Tan et al., 2024] Eugene Tan, Thomas Stemler, and Michael

Small. Cognitive dissonance and introversion effects on

opinion dynamics and echo chamber formation. Physica

A: Statistical Mechanics and its Applications, 2024.

[Tang et al., 2024] Jiakai Tang, Heyang Gao, Xuchen Pan,

Lei Wang, Haoran Tan, Dawei Gao, Yushuo Chen,

Xu Chen, Yankai Lin, Yaliang Li, et al. Gensim: A general

social simulation platform with large language model

based agents. arXiv preprint arXiv:2410.04360, 2024.

[Wang et al., 2024] Lei Wang, Chen Ma, Xueyang Feng,

Zeyu Zhang, Hao Yang, Jingsen Zhang, Zhiyuan Chen,

Jiakai Tang, Xu Chen, Yankai Lin, et al. A survey on large

language model based autonomous agents. Frontiers of

Computer Science, 18(6):186345, 2024.

[Yu, 2024] Chen Yu. How will ai steal our elections? Center

for Open Science, 2024.

[Zhao et al., 2024] Andrew Zhao, Daniel Huang, Quentin

Xu, Matthieu Lin, Yong-Jin Liu, and Gao Huang. Expel:

Llm agents are experiential learners. In Proceedings of

the AAAI Conference on Artificial Intelligence, volume 38,

pages 19632–19642, 2024.

[Zucker, 1978] Harold Gene Zucker. The variable nature of

news media influence. Annals of the International Communication

Association, 2(1):225–240, 1978. DOI: https://doi.org/10.1080/23808985.1978.11923728

Schematic of AGI influence on humans-Fig 1

Downloads

Published

2025-03-16

How to Cite

Nasim, M., Gilani, S. M., Qasmi, A., & Naseem, U. (2025). Simulating Influence Dynamics with LLM Agents. SuperIntelligence - Robotics - Safety & Alignment, 2(1). https://doi.org/10.70777/si.v2i1.13971

Issue

Vol. 2 No. 1 (2025): Large Language Models I

Section

Articles

Categories

License

Copyright (c) 2025 Mehwish Nasim, Syed Muslim Gilani, Amin Qasmi, Usman Naseem

Creative Commons License

This work is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.