Abstract: Abstract Contrasting views exist on how network structure predict knowledge creation. Basing on theoretical analysis this paper argues that network usually is viewed as opportunity agency and information agency in previous research. In consideration of the difference in seizing the opportunity and the essence of innovation, this problem may be resolved by introducing past performance and network heterogeneity. Using 11 years longitudinal data from patent collaboration in US biology technology, the paper investigates the effect of central network position and moderating effects of past performance and network heterogeneity. The results suggest network centrality has positive effects on knowledge creation. As past performance increases, the effect of centrality on knowledge creation increases. Network heterogeneity has a negative effect on this moderating relation. This paper contributes to a better understanding of actor’s knowledge creation mechanism, and provides some suggestions to social network theory.

摘要： 摘要 基于前人文献中对社会网络影响个体知识创新的研究争论，该文通过理论分析认为，以往研究通常把网络作为了机会的代理和信息的代理。为了深入了解网络影响知识创新的机理，同时考虑到个体在利用网络机会上的差异，该研究将个体以往绩效与网络异质性构建理论模型。并基于此理论模型，对美国生物高科技行业11年间的专利合作数据进行实证研究。研究结果表明，网络中心性与知识创新正相关，但这种正相关关系同时受到个体以往绩效的正向调节作用；而这种调节作用同时又受到网络异质性的负向调节。研究结论有助于我们更好地理解个体特征与网络对知识创新的交互机理，并为推动社会网络理论的发展做出贡献。

关键词: 知识创新, 合作网络, 中心性, 网络异质性, Knowledge Creation, Collaborative Network, Centrality, Network Heterogeneity