Small-world scale-free brain graphs from EEG

Developing individualized spatial models that capture the complex dynamics of multi-electrode EEG data is essential for accurately decoding global neural activity. A widely used approach is network modeling, where electrodes are represented as nodes. A key challenge lies in defining the network edges and weights, as precise connectivity estimation is critical for enhancing neural characterization and extracting discriminative features, such as those needed for task decoding. In this work, we propose a method for inferring subject-specific brain graphs from EEG data, explicitly designed to exhibit small-world and scale-free network properties. Our approach begins by computing phase-locking values between EEG channel pairs to build a backbone graph, which is then refined into an individualized small-world and scale-free network. To reduce computational complexity while preserving subject-specific characteristics, we apply Kron reduction to the resulting graph. We evaluated the proposed method on motor imagery decoding and brain fingerprinting tasks using two EEG datasets. Results show that our model consistently outperforms other benchmark graph models. Furthermore, we show that integrating classical EEG features with those derived using graph signal processing principles significantly improves performance. Overall, our findings highlight the potential of the proposed graph construction framework to enhance EEG analysis, with promising implications for a wide range of applications in cognitive neuroscience and brain-computer interface research.