Barrera Lemarchand Federico, Semeshenko Viktoriya, Navajas Joaquín, Balenzuela Pablo
Laboratorio de Neurociencia, Escuela de Negocios, Universidad Torcuato Di Tella, Av. Figueroa Alcorta 7350, 1428 Buenos Aires, Argentina.
Facultad de Ciencias Económicas, Universidad de Buenos Aires, Buenos Aires, Argentina and Instituto Interdisciplinario de Economía Política de Buenos Aires, CONICET-Universidad de Buenos Aires, Av. Córdoba 2122, C1120 AAQ Buenos Aires, Argentina.
Chaos. 2020 Jun;30(6):063141. doi: 10.1063/5.0004504.
Understanding the opinion formation dynamics in social systems is of vast relevance in diverse aspects of society. In particular, it is relevant for political deliberation and other group decision-making processes. Although previous research has reported different approaches to model social dynamics, most of them focused on interaction mechanisms where individuals modify their opinions in line with the opinions of others, without invoking a latent mechanism of argumentation. In this paper, we present a model where changes of opinion are due to explicit exchanges of arguments, and we analyze the emerging collective states in terms of simple dynamic rules. We find that, when interactions are equiprobable and symmetrical, the model only shows consensus solutions. However, when either homophily, confirmation bias, or both are included, we observe the emergence and dominance of bipolarization, which appears due to the fact that individuals are not able to accept the contrary information from their opponents during exchanges of arguments. In all cases, the predominance of each stable state depends on the relation between the number of agents and the number of available arguments in the discussion. Overall, this paper describes the dynamics and shows the conditions wherein deliberative agents are expected to construct polarized societies.
理解社会系统中的意见形成动态在社会的各个方面都具有重大意义。特别是，它与政治审议和其他群体决策过程相关。尽管先前的研究报告了不同的社会动态建模方法，但其中大多数都侧重于个体根据他人意见改变自己意见的互动机制，而没有引入潜在的论证机制。在本文中，我们提出了一个意见变化归因于明确的论点交流的模型，并根据简单的动态规则分析出现的集体状态。我们发现，当互动是等概率且对称时，该模型仅显示共识解。然而，当包含同质性、确认偏差或两者都包含时，我们观察到两极分化的出现和主导，这是由于个体在论点交流过程中无法接受对手的相反信息。在所有情况下，每个稳定状态的优势取决于讨论中主体数量与可用论点数量之间的关系。总体而言，本文描述了动态过程，并展示了审议主体构建两极分化社会的条件。
