Why do posts on a particular topic become viral more easily than others? Why do the comments under one of your posts suddenly escalate into a “shitstorm”? What effect would it have if a specific message received significantly greater reach?  Why do certain topics receive more attention on one platform than on another one?  What effect on online debates do different ranking algorithms have? How strong is the impact of downranking disinformation?

The TWON tool allows us to explore how communication on online social networks is shaped by specific platform mechanisms. Our digital twin of online social networks allows testing the effects of alternative design options and policy interventions in counterfactual scenarios, without real-world harm. The simulation can be customized to your specific question and topic of interest to support you in policy design.

What 

Online platforms are shaped by design choices such as ranking systems, recommendation logics, engagement incentives, and interface mechanics. These choices influence what users see, how information spreads, and how public debates evolve over time. These questions gain particular urgency at a time when global geopolitical tensions, disinformation, and the rise of populist forces in many democracies worldwide increasingly shape digital infrastructures. TWON as a Service builds on TWON’s core innovation: a digital twin of an online social network that is empirically calibrated. This simulation environment allows for the testing of various platform design choices, such as ranking, personalisation, or moderation mechanisms, through counterfactual means. It also enables users to explore how these interventions affect information diffusion and democratic debate. Should you have any specific questions regarding the effects of individual platform mechanisms, we can investigate them in a platform-specific manner using TWON’s digital twin approach. Mechanisms to explore include: 
  • Ranking and recommendation systems
  • Bots and automated amplification
  • Keyword-based restrictions
  • Network topology (one-to-many broadcasting vs. peer-to-peer interactions)
  • Pre-post warnings and friction mechanisms
  • Feed mechanics
  • Interaction design

Why: 

At a time when geopolitical tensions increasingly shape digital platforms, understanding and shaping platform mechanisms is key. Establishing European sovereignty over communication platforms requires evidence-based decisions about alternatives to the current status quo. Many of the most pertinent questions in platform governance, however, cannot be answered through real-world experimentation. Testing changes to ranking algorithms, moderation rules, or personalisation options, directly in social networks is usually not feasible due to limited access, the risk of unintended societal consequences, and ethical constraints. TWON addresses this challenge by providing a testing environment that can generate counterfactual evidence at scale made for policy-advice. This allows decision-makers to evaluate “what-if” scenarios in a controlled environment, independent from platform operators.

Who is this for?

TWON as a Service has been developed for public institutions, regulators, and civil society organisations, working at the intersection of digital platforms, public debate, and democratic resilience. We offer custom-made social simulations, tailored to your specific research question and designed for the platform environment relevant to your case. If you are interested in TWON as a Service, please get in touch with us.

 

 

Achim Rettinger

holds the chair for Computational Linguistics at Trier
University and is also Director at FZI Forschungszentrum Informatik. He started his research career on agent communication and computational trust learning in Multiagent Systems in 2004. Since then he worked primarily on machine learning for graph and textual data. Currently, his research team is working on simulating human capabilities and behavior, e.g. for researching opinion formation in social networks.

Website
E-mail: rettinger@uni-trier.de

 

 

Dr. Jonas Fegert

heads the House of Participation (HoP) at the FZI Research Center for Information Technology and leads the Digital Democracy & Participation research group at the Karlsruhe Institute of Technology (KIT). He completed his PhD at KIT in Information Systems. Previously, he studied Political Science, Governance, and Public Policy. His research combines Information Systems and computational social science; it addresses disinformation, deepfakes, platform governance, polarization, and digital citizen participation. AI is a cross-cutting theme, both as a research object and a methodological instrument. He regularly brings research findings into policy and public administration.

Website
Email: fegert@fzi.de

 

 

Cosima Pfannschmidt

is scientific researcher at the Berlin branch of FZI Forschungszentrum Informatik and part of the House of Participation (HoP) sinde May 2024. She coordinates communication & dissemination activities in the project “TWON – Twin of Online Social Networks”. Her work focuses on translating scientific results into actionable policy recommendations and creating impact.

Previously, she worked in the German Bundestag. She holds an M.A. in Sociology and a B.Sc. in Economics from Ludwig-Maximilians-University München.

Website
E-mail: pfannschmidt@fzi.de