The wisdom of the crowd refers to the phenomenon in which a group of individuals, each making independent decisions, can collectively arrive at highly accurate solutions—often more accurate than any individual within the group. This principle relies heavily on independence: if individual opinions are unbiased and uncorrelated, their errors tend to cancel out when averaged, reducing overall bias. However, in real-world social networks, individuals are often influenced by their neighbors, introducing correlations between decisions. Such social influence can amplify biases, disrupting the benefits of independent voting. This trade-off between independence and interdependence has striking parallels to ensemble learning methods in machine learning. Bagging (bootstrap aggregating) improves classification performance by combining independently trained weak learners, reducing bias. Boosting, on the other hand, explicitly introduces sequential dependence among learners, where each learner focuses on correcting the errors of its predecessors. This process can reinforce biases present in the data even if it reduces variance. Here, we introduce a new meta-algorithm, casting, which captures this biological and computational trade-off. Casting forms partially connected groups (“castes”) of weak learners that are internally linked through boosting, while the castes themselves remain independent and are aggregated using bagging. This creates a continuum between full independence (i.e., bagging) and full dependence (i.e., boosting). This method allows for the testing of model capabilities across values of the hyperparameter which controls connectedness. We specifically investigate classification tasks, but the method can be used for regression tasks as well. Ultimately, casting can provide insights for how real systems contend with classification problems.

This model proposes a new approach analyzing to the doctrinal paradox by considering a deliberative process (which can be represented by an agent-based model) in comparison with classical (binary) majority voting and an aggregation of (continuous) degrees of belief prior to majority voting. This model is a multivariate extension of the Hegselmann–Krause opinion dynamics model.

Micro-targeted vs stochastic political campaigning agent-based model simulation. Written by Toby D. Pilditch (University of Oxford, University College London), in collaboration with Jens K. Madsen (University of Oxford, London School of Economics)

The purpose of the model is to explore the various impacts on voting intention among a population sample, when both stochastic (traditional) and Micto-targeted campaigns (MTCs) are in play. There are several stages of the model: initialization (setup), campaigning (active running protocols) and vote-casting (end of simulation). The campaigning stage consists of update cycles in which “voters” are targeted and “persuaded” - updating their beliefs in the campaign candidate / policies.

The objective of building a social simulation in the Populism and Civic Engagement (PaCE) project is to study the phenomenon of populism by mapping individual level political behaviour and explain the influence of agents on, and their interdependence with the respective political parties. Voters, political parties and – to some extent – the media can be viewed as forming a complex adaptive system, in which parties compete for citizens’ votes, voters decide on which party to vote for based on their respective positions with regard to particular issues, and the media may influence the salience of issues in the public debate.

This is the first version of a model exploring voting behaviour in Austria. It focusses on modelling the interaction of voters and parties in a political landscape; the effects of the media are not yet represented. Austria was chosen as a case study because it has an established populist party (the “Freedom Party” FPO), which has even been part of the government over the years.

PoliSEA represents a continuous policy process cycle, integrated with the dynamics of a fishery social-ecological system. The policy process in the model is represented by interactions between policymakers and interest groups and subsequent voting during which policymaker decide to increase or decrease the fishing quota for the next season. Policymakers’ positions can be influenced by lobbying of interest groups or interest group coalitions. The quota adopted through the policy process determines the amount of fish that can be harvested from the fish population during the season.

Simulation Software for Random-Subset Voting with Borda, approval, plurality and Condorcet.

This software simulates the Random-Subset Voting method for Borda, plurality, approval and Condorcet.

In this paper we introduce an agent-based model of elections and government formation where voters do not have perfect knowledge about the parties’ ideological position. Although voters are boundedly rational, they are forward-looking in that they try to assess the likely impact of the different parties over the resulting government. Thus, their decision rules combine sincere and strategic voting: they form preferences about the different parties but deem some of them as inadmissible and try to block them from office. We find that the most stable and durable coalition governments emerge at intermediate levels of informational ambiguity. When voters have very poor information about the parties, their votes are scattered too widely, preventing the emergence of robust majorities. But also, voters with highly precise perceptions about the parties will cluster around tiny electoral niches with a similar aggregate effect.

This Netlogo replication of Kollman, K., J.H. Miller and S.E. Page (1997) Political Institutions and Sorting in a Tiebout Model, American Economic Review 87(5): 977-992. The model consists of of citizens who can vote for partie and move to other jurisdictions if they expect their preferences are better served. Parties adjust their positions to increase their share in the elections.

This is a simplified version of a Complex Model of Voter Turnout by Edmonds et al.(2014). It was developed to better understand the mechanisms at play on that complex model.