Objective:

 The aim of this study is to uncover large-scale resting-state networks related to real-life creativity using high-density electroencephalography (HD-EEG) and to test whether the strength of functional connectivity within these networks could predict individual creativity in novel subjects.

Background:

Creativity is a fundamental ability for human progress. Cognitive neuroscience research has shown that specific functional brain patterns can be related to creativity during multiple tasks but also at rest. Nevertheless, the electrophysiological correlates of a highly creative brain remain largely unexplored.

Design/Methods:

We acquired resting state HD-EEG data from 90 healthy participants who completed the «Inventory of Creative Activities and Achievements» questionnaire. We then employed connectome-based predictive modeling (CPM); a machine-learning technique that predicts behavioral measures from brain connectivity features.

Results:

Using a support vector regression, our results revealed patterns of functional connectivity related to both high and low creativity, in the gamma frequency band (30-45Hz). In leave-one-out cross-validation, the combined model of high and low creativity networks predicted individual creativity scores with very good accuracy (r= 0.34, p= 0.0009). Furthermore, the predictive power of the model was established by an external validation on a second independent resting state EEG dataset (N= 41), where we found a statistically significant relationship between the observed and predicted creativity scores (r= 0.37, p= 0.01).

Conclusions:

These findings reveal large-scale electrophysiological networks that could predict individual real-life creativity at rest, providing a crucial foundation for developing EEG network-based markers of creativity.