Elevated EEG fractal dimension in Parkinson's during timing tasks

Abstract

Fractal Dimension (FD), a measure of signal complexity, offers unique insights into nonlinear brain dynamics in neurodegenerative disorders. While Electroencephalography (EEG)-based biomarkers for Parkinson's disease (PD) are emerging, the fractal properties of PD-related neural activity remain underexplored. Here, we introduce FD as a novel nonlinear biomarker to distinguish PD patients (n=74) from healthy controls (n=37) during interval-timing tasks (3/7 s durations). EEG recordings revealed significantly higher FD values in PD patients during response phases, particularly in frontotemporal, parietal, and motor regions, indicating disrupted neural adaptability. Our findings reveal that short intervals engaged sensorimotor areas, while long intervals implicated frontoparietal networks, which uncovered task-specific effects. These findings highlight FD's sensitivity to PD-related neural disorganization, offering a robust diagnostic tool for capturing disease-specific complexity changes. This study establishes FD as the first EEG-derived fractal biomarker for PD, advancing our understanding of its neurophysiological mechanisms.