Uncovering the Neural Fingerprint of Akinetic States in a Parkinson’s Disease Rodent Model

Parkinson’s Disease (PD) is characterized by complex motor deficits, including transient akinetic episodes during locomotion. Here, we investigated the neural correlates of akinesia in a 6-OHDA rat model of PD. We developed the open-source toolbox neurokin to analyse the kinematic and neural signal recorded during a runway task. First, we observed that compared to control, PD rats spent significantly more time in akinetic episodes, which were correlated with an increase in beta-band power. Next, we computed a set of temporally resolved kinematic and neural features capturing the evolution of locomotion states. We employed linear and non-linear machine learning models, to retrieve salient features that characterized akinetic episodes. Beyond confirming established associations with beta power, we identified Hjorth complexity and mobility as time-domain features modulated by akinetic onset. Our findings highlight novel, computationally lightweight biomarkers that might serve as targets for future state-adaptive neuromodulation therapies.