知识共享与间接互惠:基于计算实验方法的研究

摘要

The return of knowledge sharing behaviors often comes from third-party rather than recipients,research of knowledge sharing in direct reciprocity perspective are not insufficient to explain this kind of phenomenon.Adopting the computational experiments based on multi-agent modeling,this study builds a model of indirect reciprocity and knowledge sharing of large teams by giving agents ability to memory information,make decision and communicate.The results show that,due to the existence of discriminators,even unconditional sharers become absolute the majority,hiders are unable to invade the group.This phenomena mean that indirect reciprocity plays a great role in binding widespread knowledge sharing behaviors.The hiders and discriminators will coexist at low level of trust climate in organization,but knowledge sharing behaviors of agents do not appear.Even if only a few individuals can observe knowledge sharing behaviors of colleagues around them,as long as reputation can spread effectively,indirect reciprocity can promote the knowledge sharing behaviors.%团队中,个体知识共享行为的回报往往来自第三方而非受助者,而以直接互惠为视角的知识共享研究却不足以解释此类现象.采用基于计算机仿真实验方法,通过赋予Agent记忆、推理、决策和沟通等能力,研究受表型背叛、组织信任氛围及声誉传播机制影响下,间接互惠机制对团队大范围知识共享行为的维系机理.研究发现,由于辨别者的存在,即使无条件共享者会成为绝对多数,隐藏者并不会大量侵入群体.这意味着间接互惠机制维系了团队大范围知识共享行为.在低的组织信任氛围水平下,辨别者和隐藏者虽然能够共存,团队内的知识共享行为却没有出现.即使只有少数成员能观察到周围同事的知识共享行为,只要声誉信息能够有效传播,间接互惠对广泛知识共享行为的维系作用就能够发挥.