EC² Defines a Seven-Model Democracy Simulation

The EC² project has completed a public modelling protocol based on the seven-model architecture originally set out in its proposal. The public modelling protocol defines seven connected models covering voter preferences, election design, voting behaviour, vote aggregation and political change.

The protocol provides the technical and scientific foundation for future EC² experiments. Researchers will use the framework to compare voting systems, examine how political opinions evolve and study how elections affect trust, satisfaction, participation and polarisation.

The seven-model structure originates in the original EC² proposal and represents the project’s broader research architecture. During implementation, Prof. Jan Lorenz and the Aarhus modelling team have also developed a more focused, data-driven prototype for Aarhus.

This first prototype concentrates on preference formation, candidates and parties, election configuration, individual voting and vote aggregation, while quantum-voting research is being developed separately and some functions of the original exogenous-change model are represented within the preference-formation model through news and other external inputs, with further integration possible later.

The seven-model framework and the narrower Aarhus prototype should therefore be understood as different levels of the EC² research strategy, rather than competing descriptions of the same implementation.

Both the seven-model architecture and the Aarhus implementation contribute to the development and evaluation of a common computational framework for goal-oriented democratic decision-making that is being progressively refined across the project’s quantum-theoretical and classical simulation work.

Seven Models Covering the Democratic Process

EC² is not developing one large model intended to explain all democratic behaviour. Instead, the framework combines seven specialised models:

• quantum-inspired voting concepts,
• voter preferences and beliefs,
• candidate, policy and proposal generation,
• election and ballot configuration,
• individual voting decisions,
• vote aggregation and result evaluation,
• external events affecting political behaviour.

The models can be tested separately or connected for larger experiments. This allows researchers to select the components needed for a particular question and compare different assumptions without rebuilding the entire framework.

The architecture also makes the simulation easier to analyse. Researchers can examine how voter preferences changed, how ballots were cast and how a specific electoral system converted individual choices into a collective result.

How a Simulated Election Works

The framework represents voters, groups, proposals, elections and external events.

Each virtual voter can have political preferences, demographic characteristics, social connections and beliefs about political conditions. The simulation can also track participation, satisfaction, perceived fairness and trust in election results.

Before an election, voters may receive information from campaigns, opinion polls, political events and other people in their social network. Their preferences can change in response.

When voting begins, each voter assesses the available options and submits a ballot according to the selected rules. The votes are then aggregated, and the system calculates indicators such as:

• perceived fairness,
• trust in the result,
• participation,
• voter satisfaction,
• representation quality,
• political polarisation.

The election result is then communicated back to the virtual voters. This allows the next simulation cycle to examine whether previous outcomes influence future preferences, trust or voting behaviour.

Comparing Voting Systems and Political Shocks

The EC² framework is designed to compare voting systems using the same virtual population and starting conditions.

The target protocol initially covers single-choice and quadratic voting. It also represents electoral arrangements including plurality and open-list proportional representation. Ranked-choice voting, approval voting and other methods may be added later.

Quadratic voting gives participants a budget of voting credits that they may allocate across different options, with additional votes for the same option becoming progressively more costly. This allows voters to express not only what they support, but also how strongly they care about a particular issue.

Future EC² experiments will examine how quadratic voting compares with traditional approaches. The research will focus on possible differences in representation, strategic voting, satisfaction and collective outcomes.

The framework can also introduce external events such as political scandals, policy announcements, economic crises or environmental shocks. Researchers can change the timing, strength and target group of an event to study how it affects preferences, trust and polarisation.

This will help EC² examine not only how democratic systems work in stable conditions, but also how they respond under pressure.

Real Data from Aarhus

The framework can support abstract experiments involving virtual populations, but it will also be tested using data from Aarhus, Denmark.

A voter survey conducted in March 2026 collected information about political preferences, voting behaviour, turnout, attitudes towards the political system and demographic characteristics.

The survey also compared how respondents would vote under the current Danish open-list proportional representation system and under a quadratic voting alternative.

Candidate positions were collected from Danish candidate-test platforms, including Altinget, DR.dk and TV 2. Demographic data from Statistics Denmark and Aarhus Open Data will support the creation of a synthetic population representing the wider Aarhus electorate.

The target framework can support scenarios representing approximately 180,000 voters. The synthetic population will reflect wider demographic and political patterns rather than recreate individual citizens.

Planned laboratory and field experiments developed with Aarhus Municipality will provide additional information about how people respond to different voting rules.

What Happens Next

The completed protocol defines the full target structure of the EC² simulation, but the framework is being implemented progressively.

The individual models are initially being tested separately. The preference-formation model will be assessed against preference patterns in the Aarhus survey, individual simulated voting decisions against survey-reported choices, and aggregate simulation results against Danish election data.

The project will then connect selected models and examine which assumptions have the greatest influence on results.

The protocol also defines the role of quantum-inspired research. EC² will not run large elections on a live quantum computer. Instead, selected mathematical concepts and results from small-scale quantum experiments may be incorporated into the classical simulation for comparison.

With the original architecture now specified in greater operational detail, EC² has a clear foundation for testing how alternative voting systems may influence fairness, trust, representation and political behaviour.