Modifying upper-limb inter-joint coordination in healthy subjects by training with a robotic exoskeleton

Area of Science:

• Robotics
• Neuroscience
• Biomechanics

Background:

• Modifying pathological upper-limb coordination in stroke survivors is crucial for neurorehabilitation.
• Robot-assisted training may enhance motor recovery in hemiparetic patients.
• Existing research often focuses on planar end-effector adaptation, neglecting 3D movement and inter-joint coordination.

Purpose of the Study:

• To investigate how the human central nervous system adapts to altered upper-limb coordination during 3D movements using a robotic exoskeleton.
• To explore the effects of joint-level, velocity-dependent force fields on pointing and tracking tasks.
• To analyze the adaptation and generalization of motor strategies in response to perturbations.

Main Methods:

• Twenty healthy participants performed 3D pointing and tracking tasks with a 4-DOF robotic arm exoskeleton.
• Inter-joint velocity-dependent force fields were applied at the joint level, perturbing coordination without directly constraining end-effector movement.
• Kinematic analysis (end-point and joint levels), final posture measurements, and principal component analysis (PCA) of inter-joint coordination were used.

Main Results:

• Perturbing force fields induced kinematic modifications at joint, end-effector, and inter-joint coordination levels.
• Adaptation to force fields occurred in only 21% of movements, but post-effects (persistence of modified coordination) were observed in 86% of cases during wash-out and follow-up.
• Generalization of altered coordination was observed for unexposed targets, with significant inter-individual differences.

Conclusions:

• This study provides the first quantified characterization of modified upper-limb coordination in healthy subjects using joint-level viscous force fields.
• The observed persistence and generalization of altered coordination suggest potential for robot-assisted neurorehabilitation strategies.
• These findings could inform the development of robotic interventions to retrain pathological arm synergies in stroke survivors.