Modelling enduring institutions: The complementarity of evolutionary and agent-based approaches

Abstract

Empirical work has shown that societies can sometimes avoid antisocial outcomes, such as the Tragedy of the Commons, by establishing institutional rules that govern their interactions. Moreover, groups are more likely to avoid antisocial outcomes when they design and enforce their own rules. But this raises the question: when will group members put effort into maintaining their institution so that it continues to provide socially beneficial outcomes? Ostrom derived a set of empirical principles that predict when institutions will endure, which have subsequently been formalised in agent-based models that are based on an executable description of the content of an individual's behaviour. Here we show how these models can be complemented by evolutionary game theory, which focuses on the value or payoff of different behaviours, rather than on the mechanistic content of the behaviour. Using such a value-based model, we determine exactly when individuals will be incentivised to maintain their institution and enforce its rules, including the critical amount that a group must invest into incentivising agents to monitor rule compliance. We highlight the complementarity of content-based and value-based modelling approaches, and therefore provide a step towards unifying theoretical and empirical approaches to understanding enduring institutions and other social phenomena.

The evolution of deception

“…We define and run extensive simulations of agent-based models of PGGs by applying mechanism design and game theory to study the evolution of behaviour of populations of agents of a fixed size N in six different PGGs, each with a different set of strategies. Our method can be considered a content-based approach to study the evolution of social systems [ 34 ]. We chose to follow this type of approach due to the fact that deception and deception detection implied the modelling of complex cognitive aspects of agent behaviour.…”

“…We chose to follow this type of approach due to the fact that deception and deception detection implied the modelling of complex cognitive aspects of agent behaviour. As noted by the authors in [ 34 ], cognitive aspects increase the complexity of the model and make a value-based approach intractable. Ideally, in situations where cognitive aspects are not modelled, both content-based and value-based approaches are desirable to study the evolution of agent behaviour.…”

Cost efficiency of institutional incentives in finite populations

“…They include kin and group selection [Hamilton, 1964, Traulsen and Nowak, 2006], direct and indirect reciprocities [Ohtsuki and Iwasa, 2006, Krellner and Han, 2020, Nowak and Sigmund, 2005, Han et al, 2012, Okada, 2020], spatial networks [Santos et al, 2006, Perc et al, 2013, Antonioni and Cardillo, 2017, Peña et al, 2016], reward and punishment [Fehr and Gachter, 2000, Boyd et al, 2003, Sigmund et al, 2001, Herrmann et al, 2008, Hauert et al, 2007a, Boyd et al, 2010], and pre-commitments [Nesse, 2001, Han et al, 2013, Martinez-Vaquero et al, 2017, Han et al, 2016, Sasaki et al, 2015]. Institutional incentives, namely, rewards for cooperation and punishment of wrongdoing, are among the most important ones [Wang et al, 2019, Sigmund et al, 2001, Han and Tran-Thanh, 2018, Sigmund et al, 2010, Vasconcelos et al, 2013, Chen et al, 2015, Wu et al, 2014, García and Traulsen, 2019, Góis et al, 2019, Powers et al, 2018]. Differently from other mechanisms, in order to carry out institutional incentives, it is assumed that there exists an external decision maker (e.g.…”

A Local Search for Numerical Optimisation Based on Covariance Matrix Diagonalisation

“…This idea has been extensively applied over the past three decades in both continuous and combinatorial domains. Some of the keywords associated with this idea are Metaheuristics [9,25], Hyper-heuristics [2], Memetic Computing [19], Genetic Programming [11], Agent Systems [24], and Algorithm's Portfolio [33,22].…”