Next-Generation Neurotechnologies Inspired by Motor Primitive Model for Restoring Human Natural Movement

Abstract

Advances in neuroengineering and artificial intelligence are transforming the landscape of motor rehabilitation, aiming to restore human movement as natural as possible. In recent decades, more advanced interventions are increasingly achievable via hybrid robotic systems, neuroprosthetics, and brain-computer interfaces. However, a fundamental gap of these neurotechnologies remains in modeling the complexity of neuromotor control, particularly how the central nervous system coordinates high-dimensional motor outputs in naturalistic behaviors. Rooted in theoretical neuroscience, the motor primitive (MP) model proposes an adaptable framework to deconstruct and reproduce motor tasks through low-dimensional modules. Interestingly, recent studies have indicated that the MP model may reform current-generation neurotechnologies by digitally shaping the course of human-machine interaction. In this narrative review, we will critically examine conventional target settings and identify their limitations in guiding biomimetic control in neurotechnologies. We then introduce the MP model with its machine learning and physiological scaffolds for better understanding and replicating human natural movement. Finally, we will present its potential in facilitating the next-generation neurotechnologies across kinematic, muscular, and neural domains. By modeling motor control in human and neuroengineering, we believe that the MP-inspired paradigms can initiate a new era of intelligent, patient-specific, and naturalistic motor restoration for various neurological and traumatic diseases.