Uncovering the spatiotemporal structure of neural avalanches through optimal transport and dynamic time warping
bioRxiv (Cold Spring Harbor Laboratory) July 16, 2026 Filip Novický, Nikola Jajcay, Jaroslav Hlinka
Neural avalanches—bursts of coordinated brain activity—are typically studied using scale-free statistics, but their detailed spatiotemporal structure and recurrence have been difficult to analyze due to variability in duration and spatial extent. A new method using flexible alignment with unbalanced optimal transport and subsequence dynamic time warping allows comparison of events of different lengths and configurations. Applied to 64-channel EEG data from 63 participants in the PsiConnect psilocybin study, hierarchical clustering revealed 12 recurring propagation patterns. Under psilocybin, oscillating sequences—those with alternating polarity and spatial propagation—were reduced relative to stable sequences, shifting the balance toward stable patterns. This effect was confirmed by a permutation test and largely driven by one cluster, indicating specific neural dynamics temporally affected by psilocybin.