集中,分组,分组后集中:探究三种群体结构对群体决策能力的影响

Effects of Centralized,Decentralized and Reintegrated Group Structures on Group Performance

采用基于Agent仿真的方法,研究决策问题不可分时,集中网络、分组网络、分组后集中网络对群体决策能力的影响。三种群体结构分别对应于三种群决策模式:集中决策、分组决策、分组后集中决策。仿真结果表明,仅当问题较为简单、分组适当、个体学习速率较低时,分组决策才小幅优于集中决策。分组后集中决策并不受问题复杂度、分组数和个体学习速率的影响,总是优于集中决策和分组决策。但为了提高分组后集中决策的决策能力,面对简单问题时,应适当减少分组、维持较低的个体学习速率;面对复杂问题时,应适当增加分组、维持适中的个体学习速率。此外,分组后集中决策的集中时机与群体决策能力呈倒U关系,并不受问题复杂度、分组数和个体学习速率的影响,但相对较早的集中时机更有利于解决很复杂的问题。

With the help of an Agent-based simulation based on NK model, we have studied the effects of three different group structures on group performance when the decision problems are not independent： a centralized group in which group members make decision together; a decentralized group in which group is divided into several subgroups; and a temporarily decentralized group which starts out with a decentralized structure and later reintegrates. The results of simulation suggest that only when the problem is simple, grouping is proper and indi- vidual learning rate is low could decentralized group be slightly better than centralized group. What＇ s more, reintegrated group is always the best of these three decision models no matter how complex the problem is, how many the groups are or how fast the individual learning rate is. However, in order to improve the group perform- ance of reintegrated group, when faced with a simple problem, we should decrease the number of groups and keep the individual learning rate on a low level. But when the problem is complex, more subgroups and moderate individual learning rate will be better. Besides, there is an inverted-U relationship between centralized timing and group performance, which is unaffected by the complexity of the decision problem, the number of subgroups and individual learning rate. However, earlier centralized timing performs better when it comes to complex problems.