Chain free association, creativity, and the default mode network.

Research on creativity shows that creative thinking entails both executive (controlled) and associative (spontaneous) processes. Yet standard creativity tasks cannot reliably isolate these two types of cognitive processes, making it difficult to understand the relation between the two and the roles of their corresponding brain networks in creative cognition. In this study we used a behavioral and neuroimaging approach in an effort to establish chain free association (FA) tasks as a relevant method for directly investigating spontaneous associative thinking and its role in creative cognition. We further examined the relation between performance on such tasks and intelligence. Participants completed common creativity tasks and then underwent fMRI scanning while producing FA chains. Instructions to participants that emphasized the spontaneous nature of the task, coupled with proper control conditions that were balanced for difficulty, enabled us to uncover spontaneous (as opposed to controlled) processes. To examine whether behavioral measures that can be derived from FA chains (associative fluency, associative flexibility and semantic remoteness between associations) are indicative of unconstrained spontaneous associative processing and are related to different aspects of verbal creativity and intelligence, scores on these measures were correlated with scores on creativity tasks and on an intelligence task, and with brain activity. We found that: (1) the Default Mode Network (DMN), a network involved in self-generated and internally-directed thought, was more involved in chain FA than in other tasks expected to reflect more controlled forms of internally-directed thought, suggesting that the DMN involvement might be related to the unconstrained spontaneous nature of chain FA. Higher involvement of the left IFG, SFG, MFG under chain FA was also revealed; (2) higher scores on different behavioral measures from FA chains were related to higher activation of the DMN and to reduced activation of the left IFG, a major node in the executive function network; (3) behavioral measures from FA chains were correlated with different aspects of creative performance but not with intelligence. Taken together, these findings lend support to the hypothesis that chain FA involves associative spontaneous thinking. They further suggest that behavioral measures derived from chain FA could indicate patterns of unconstrained associative thinking, related to reduced cognitive control, that are relevant for creative ideation, and might be able to serve as a measure of these patterns.