How initial representations shape coupled learning processes

Coupled learning processes, in which specialists from different domains learn how to make interdependent choices among alternatives, are common in organizations. We explore the role played by initial representations held by the learners in coupled learning processes using a formal agent-based model. We find that initial representations have important consequences for the success of the coupled learning process, particularly when communication is constrained and individual rates of learning are high. Under these conditions, initial representations that generate incorrect beliefs can outperform one that does not discriminate among alternatives, or even a mix of correct and incorrect representations among the learners. We draw implications for the design of coupled learning processes in organizations.