Integrating insights from cognitive psychology into current network theory on the social capital of brokering and closed networks, we argue that cognitive style is a critical contingency explaining the relation between social network position and innovative performance. Based on a “complementary fit” argument, we posit that a social network rich in structural holes enhances the innovative performance of employees with an adaptive cognitive style; however, individuals with an innovative cognitive style are most innovative when embedded within a closed network of densely interconnected contacts. Using data on the individual cognitive styles and complete workplace social networks of all employees within a design and manufacturing firm, we show that our theorized contingency mechanism accounts for a large share of empirical variation in employee innovative performance over and above existing social network explanations.