Zaller-Deffuant Model of Mass Opinion

Abstract

Recent formulation of the Zaller model of mass opinion is generalized to include the interaction between agents. The mechanism of interaction is close to the bounded confidence model. The outcome of the simulation is the probability distribution of opinions on a given issue as dependent on the mental capacity of agents. Former result was that a small capacity leads to a strong belief. Here we show that an intensive interaction between agents also leads to a consensus, accepted without doubts.

Bounded confidence model: addressed information maintain diversity of opinions

“…Note that while for Ref. [14], Fig. 10 one simulation the state of unanimity is characterized by a small variance, the variance averaged over populations with unanimous and extreme opinions is large.…”

“…This generalisation mate the approach even more close to the bounded confidence model, then it is justified to term it as the Zaller-Deffuant model. In both works [12,14], a community was subject to a stream of messages, where positive and negative opinions on a given issue appear with equal frequencies. The proper outcome of opinions should then be "I don't know", and the quality of the social performance can be measured by an evaluation of the variance σ 2 of the opinion distribution.…”

“…In [12], where the communication between agents was absent, the variance was found to be anti-correlated with the tolerance parameter µ, which measured an individual ability to receive/accept distant messages. In [14], an intensive exchange of messages between agents was demonstrated to lead to unanimity. This option was found to deteriorate the collective performance, by producing agents who are strongly convinced in their opinions.…”

Opinion transmission in organizations: an agent-based modeling approach

“…In the agent-based literature, this attribute expresses agents' bounded confidence, or the idea that an agent can be influenced by another only if their opinions are close enough (Deffuant et al, 2001;Deffuant et al, 2002;Deffuant, 2006;Fortunato, 2004;Hegselmann & Krause, 2002;Kułakowski, 2009;Malarz, Gronek & Kułakowski, 2011). Heterogeneous levels of latitude of acceptance are known to substantially affect results (Rouchier & Tanimura, 2012).…”

“…Agentbased computer simulation has been extensively used to model dynamic processes in which consensus, polarization or fragmentation appear as emergent properties of the social system, owing to endogenous feedback effects arising from multiple, repeated interactions between individuals who influence one another. With heterogeneous populations, influences may spread from one subset of agents to others, under given conditions (Deffuant, 2006;Deffuant, Amblard, Weisbuch & Faure, 2002;Deffuant, Neau, Amblard & Weisbuch, 2001;Gargiulo & Mazzoni, 2008;Hegselmann & Krause, 2002;Jager & Amblard, 2004;Kułakowski, 2009;Malarz, Gronek & Kułakowski, 2011).…”