Psychologically, agents always like to consider similar opinions. Moreover, in real opinion dynamics, people’s opinions usually influence their actions. Therefore, inspired by the HK bounded confidence model, and continuous opinions and discrete action model, in this paper, we propose opinions and actions dynamics model under bounded confidence to investigate the evolution of opinions and actions in a group of agents. In this model, it is assumed that agents have continuous opinions and discrete actions for a certain issue. Each agent often can notice the discrete actions of other agents, but cannot acquire their continuous opinions. So, agents always try to estimate other agents’ opinions based on their actions. Then based on the estimation opinions and bounded confidence, agents update their opinions and actions. Simulation experiments analysis shows that more agents keep silence or undecided as the hesitation range increases. Larger bounded confidence value leads to the stronger attracting power of agents. When the opinion distribution widths of agents with an action are smaller than the bounded confidence value, the agents will be completely attracted by the adjacent agents with large opinion distribution widths and show adjacent actions in the final time.

中文翻译：

---

有限信心下的意见和行动动态

在心理上，代理人总是喜欢考虑相似的意见。此外，在真实的意见动态中，人们的意见通常会影响他们的行为。因此，受HK有界置信模型、连续意见和离散行为模型的启发，在本文中，我们提出了有限置信度下的意见和行为动力学模型来研究一组代理中意见和行为的演变。在这个模型中，假设代理人对某个问题有连续的意见和离散的行动。每个智能体通常可以注意到其他智能体的离散行为，但无法获得他们的连续意见。因此，代理人总是试图根据他们的行为来估计其他代理人的意见。然后基于估计意见和有限置信度，代理更新他们的意见和行动。模拟实验分析表明，随着犹豫范围的增加，越来越多的智能体保持沉默或未决定。有界置信值越大，代理的吸引力越强。当具有某个动作的智能体的意见分布宽度小于有界置信度值时，智能体将完全被具有较大意见分布宽度的相邻智能体所吸引，并在最后一次显示相邻的动作。