Dual-tasking reveals severity-dependent reorganization of cortical beta energy landscapes in Parkinson's disease

Dual-task impairment is a hallmark of Parkinson's disease (PD), yet the large-scale neural mechanisms underlying postural–motor interference remain poorly understood. In particular, it is unclear how cortical network dynamics reorganize across disease severity when postural control competes with concurrent task demands. This study investigated EEG-derived beta-band cortical energy landscapes in healthy older adults, early-stage PD, and mid-stage PD during single- and dual-task conditions. Dual-task behavioral cost increased with disease severity for concurrent manual performance ( p < 0.001), whereas a quadratic pattern was observed for postural performance. Energy landscape analysis revealed severity-dependent reconfiguration of cortical beta dynamics. Dual-task–related landscape changes in effective network flexibility (ΔN eff ), landscape geometry (ΔE var and ΔE mag ), and dominant low-energy attractor organization (ΔLow mass and ΔLow area) showed significant monotonic trends ( p < 0.05), reflecting progressive constrained cortical network dynamics with advancing PD severity. In addition, dual-task–related landscape alterations were associated with clinical severity, as indexed by Hoehn and Yahr stage ( |r| = 0.353–0.423, p = 0.016–0.048), and showed associations with motor impairment, as measured by MDS-UPDRS part III scores (| r | = 0.333–0.455, p = 0.009–0.063). These findings demonstrate that dual-task demands induce severity-dependent reconfiguration of cortical beta energy landscapes in PD. Energy landscape geometry may capture systems-level neural constraints associated with dual-task susceptibility in PD, providing a physiologically grounded framework to characterize disease–related functional vulnerability.