Hi there! You have found my (Benedikt V Meylahn) webpage. I recently joined the Dutch Institue for Emergent Phenomena
as Postdoctoral Fellow on Polarisation, Segregation and Inequality. There my goal is build an understanding of society as complex system by connecting modelling and data driven
approaches, and to use this understanding to guide policy toward meaningful change.
RESEARCH
I am interested in modelling multi-agent systems (typically societies or communities), often with an element of learning. In my PhD we started with a project in institutional trust modelled with learning and social influence.
This grew into a project about opinion dynamics in which agents learn to how reliable their opinion is (how much evidence there is for it).
Towards the end of my PhD we started working on interpersonal trust: trust between agents, which we model as a coordination game.
Now, in my Postdoc, I have been challenging myself to connect more to empirical data.
Peer-reviewed research articles and conference proceedings:
- Meylahn, BV, 2025. `Multi-agent reinforcement learning in the all-or-nothing public goods game on networks.' In: Proceedings of the 24th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2025): 1492-1500 (arxiv.org, ifaamas.org)
- Meylahn, BV, De Turck, K & Mandjes, M, 2025. `Trust in society: A stochastic compartmental model,' Physica A: Statistical Mechanics and its Applications, 668: 130563. (arxiv.org, j.physa)
- Meylahn, BV & Meylahn, JM, 2024. `How social reinforcement learning can lead to metastable polarisation and the voter model.' PLoS ONE, 19(12): e0313951. (plos.org)
- Meylahn, BV & Searle, C, 2024. `Opinion dynamics beyond social influence.' Network Science, 12(4): 339-365. (cambridge.org)
- Meylahn, BV, den Boer, AV & Mandjes, M, 2024. `Interpersonal trust: Asymptotic analysis of a stochastic coordination game with multi-agent learning,' Chaos, 34(6): 063119. (aip.org)
- Meylahn, BV, den Boer, AV & Mandjes, M, 2024. `Trusting: Alone and together,' The Journal of Mathematical Sociology, 48(4): 424–478. (tandfonline.com)
- Meylahn, BV & van Vuuren, JH, 2021. `The evolutionary spatial snowdrift game on a cycle: An asymptotic analysis,' Open Journal of Discrete Applied Mathematics, 4(3): 36-59. (PSIR.org)
Ongoing research projects:
- I am working on a project about opinion dynamics with Frank Pijpers and Michel Mandjes. We try to understand the effect of the common assumption that extremes exist in the opinion space. The preprint is available on the arXiv: arxiv.org.
- With others in the DIEP group on polarisation, segregatio, and inequality we are trying to understand the relationship between anti-immigrant sentiment and the exposure to immigration in the Netherlands bringing together
a variety of data-sources.
- With two other postdocs I am looking at the effect of different social networks in the Ising-model for opinion dynamics in the context of influence maximisation.
I am open to start new collaborations; get in touch (contact details at the bottom). I would very much like to take some of the ideas from multi-agent learning, trust and opinion dynamics,
and see how/if they apply in the context of user choices. This could be about traffic users, climate actions, or voting behaviour for example.
TEACHING
Since I was a Masters student I have taught as teaching assistent for various courses at Stellenbosch University and later also at the University of Amsterdam. Here is an overview of the courses I have been involved in.
At the Stellenbosch University (2018-2021):
- Operations research 345, BEng (Industrial y3)
- Industrial programming 244, BEng (Industrial y2)
I have written a short, non-technical blog post on Game Theory (particularly the Trust Game)
and modelling human behaviour for The Networks Pages. I presented an overview of the work conducted during my PhD at the DIEP seminar which was recored and is
available on youtube: Multi-agent Learning Models for Social Dynamics.
SHORT BIO
I was born in South Africa where I studied for a Bachelor and Master of Industrial Engineering (Stellenbosch University).
My final year research project (skripsie) was completed (in 2018) under the supervision of Dr Christa Searle.
We replicated and extended an agent-based simulation on (counter-)insurgency. Then I completed (in 2021) a research Master of Industrial Engineering under the supervision
of Prof Jan H van Vuuren. We studied the evolutionary spatial snowdrift game on a cycle under imitation based learning.
During both my Bachelor and Master research projects I was a part of the Stellenbosch Unit for Operations Research in Engineering (SUnORE) research group.
Until May (2025) I was PhD student at the University of Amsterdam in the
Korteweg-de Vries Institute for Mathematics. My project was part of
the Networks consortium which spans four institutions across the Netherlands.
My promoter was Michel Mandjes and my co-promoter was Arnoud den Boer. On the 26th of November 2025 I succesfully defended my dissertation titled `Stochastic models of trust'.
I have enjoyed singing in various ensembles during my studies (Kraaie van Wilgenhof, Stellenbosch University Chamber Choir, Vrije Universiteit (Amsterdam) Kamerkoor, and Studentenkoor Amsterdam). Recently I have discovered the wonderful world of tabletop role-playing games.
The banner at the top depicts my dog Ada (named after Ada Lovelace) on a beach in South Africa. She's a good dog.
CONTACT
Ben Meylahn, Korteweg-de Vries Institute, University of Amsterdam
Mailing address: Ben Meylahn, Science Park 105-107, 1098 XG Amsterdam, The Netherlands.
e-mail: b dot v dot meylahn at uva dot nl