Measuring creativity : associative thinking in semantic networks

Identifying creative ability and its determinants is crucial in understanding artistic and innovative achievements. Previous work has shown that performance across established creativity tasks does not correlate within participants. A potential reason for this finding is that most creativity tasks lack well-defined performance criteria. In this paper, we develop a novel tool for measuring creative ability and assess its performance through experimental tests. We construct a semantic network serving as the underlying structure of our tool. Based on this network, participants perform two associative thinking tasks, Local Search and Depth Search. We characterise each task by relating it to an established measure of creativity, finding that performance in our proposed tasks is significantly related to their matched creativity task across several dimensions. Our new tool improves on established creativity tasks by utilising a predefined solution space. While capturing key features of established methodologies, it substantially increases on the ease of implementation and interpretation. In addition we also provide causal evidence on the effect of incentives on our tool.