中文摘要：

多智能体模型的构建可分为若干环节,其中状态更新规则往往被设定为同步状态更新规则。然而,现实自然和社会系统中普遍存在的是异步状态更新规则。已有文献发现异步状态更新规则下和同步状态更新规则下的系统动力学过程有着显著区别,但目前尚实证或实验研究探索异步性的性质。为此,基于Pearson相关系数定义了状态更新序列间一致性度量指标,然后基于由随机异步状态更新规则产生的序列间一致性频率分布,构建了判定状态更新规则有序性的方法。最后,将所构建方法应用于一组二阶段任务更新实验数据,探索性分析了现实系统的异步状态更新规则性质。结果发现,大部分系统的状态更新规则具有随机有序性。研究结论可帮助理解社会系统动力学过程,同时对多智能体建模具有重要启示。

英文摘要：

The construction of multi-agent-based model consists of several components,in which the state update rule is mostly set as synchronous.In real life,however,agreat portion of natural and social systems follow asynchronous state update rules.Furthermore,a few studies have revealed that the pattern of dynamics under different state update rules might be totally different.Which emphasizes the importance of discriminating or deducing the properties of state update rule that is not explored in exist literature.This paper defines an index for depicting the ordering between two sequences by using Pearson correlation coefficient,and develops a method for discriminating the property of state update rule based on the distribution of this index.A two-stage state update experiment is designed to generate the state update sequences.These sequences are analyzed by the proposed discriminating method,and thus their orderings of state update rule are revealed.The result shows that state update rule in most social systems are random ordering.This finding would help understand the dynamics of social systems,and also enlighten agent-based modeling.