Role and modulation of various spinal pathways for human upper limb control in different gravity conditions

Alice Bruel, Lina Bacha, Emma Boehly, Constance De Trogoff, Luca Represa, Grégoire Courtine, Auke Jan Ijspeert

TL;DR

This paper explores how the human spinal cord (SC) contributes to upper limb control, particularly in adapting to different gravity conditions. Using a sophisticated neuromusculoskeletal model, the researchers found that both the brain and the spinal cord play crucial roles, with the SC offering a significant mechanism for gravity compensation.

What Problem Does This Paper Address?

Controlling our limbs to interact with the environment is a complex feat, requiring seamless coordination between our brain, spinal cord, and musculoskeletal system. While the brain is known for motor planning, the spinal cord's role in fine-tuning movements, especially in response to changing external conditions like gravity, remains less clear. Specifically, it's debated whether adaptations to gravity primarily occur at the brain level (re-planning commands) or at the spinal level (modulating existing reflexes). This study aims to disentangle these contributions by investigating how different spinal pathways influence upper limb control under varying gravitational forces and directions.

Methods

The researchers developed a computational model that integrates a modular spinal cord (SC) model with a musculoskeletal upper limb model. This integrated system receives supraspinal commands (representing brain signals) and simulates muscle sensory feedback.

Here's a breakdown of their approach:

  • Neuromusculoskeletal Model: A two-degree-of-freedom (DOF) musculoskeletal model of the upper limb was used, actuated by seven Hill-based muscles. This model was scaled to a participant's morphology.
  • Modular Spinal Cord Model: This model incorporated several key spinal pathways known to be involved in motor control, including:
  • Ia stretch reflex (monosynaptic excitatory)

    • Ia reciprocal inhibition (disynaptic inhibitory)
    • Ia heteronymous stretch reflex (monosynaptic excitatory)
    • II static stretch reflex (disynaptic excitatory)
    • Ib autogenic inhibition (disynaptic inhibitory)
    • Renshaw cell recurrent inhibition (monosynaptic inhibitory) The strength of these pathways could be modulated.
  • Supraspinal Commands: These were modeled as sinusoidal signals, representing simplified motor commands from the brain for elbow flexion and extension.

  • Scenarios and Optimization: The study explored four main scenarios:
  • Role of spinal pathways in voluntary movements: Evaluating the effect of individual spinal pathways with increasing synaptic strength on movement smoothness and muscle recruitment.
  • Role of spinal pathways in perturbed environments: Assessing how spinal pathways contribute to robustness against external force perturbations.
  • Modulation of supraspinal commands in various gravity environments: Optimizing supraspinal commands to achieve target trajectories under different gravity magnitudes and directions, both with and without a simplified SC model.
  • Modulation of spinal pathways in various gravity environments: Optimizing the synaptic strengths of the SC model to achieve target trajectories under different gravity conditions, while keeping supraspinal commands fixed.
  • Optimization Framework: A Covariance Matrix Adaptation Evolution Strategy (CMA-ES) was used to optimize parameters (supraspinal command parameters or spinal synaptic strengths) by minimizing a loss function that included trajectory error and muscle activation penalties.
  • Performance Metrics: Kinematic performance was measured using Root Mean Square Error (RMSE) and elbow speed arc length (SAL) for smoothness. Robustness against perturbations was assessed by the average deviation from the unperturbed trajectory. Muscle activation was compared to recorded EMG signals.

Datasets

The study utilized recorded upper limb kinematic and EMG data from a human participant performing repetitive planar elbow flexion-extension and circular movements. These recordings served as the "target trajectories" for the model's simulations and optimizations. No new benchmarks were introduced; rather, existing human movement data was used for validation.

Key Insights and Ideas

  • Complementary Roles of Brain and Spinal Cord: The research strongly suggests that both supraspinal (brain) commands and spinal cord (SC) reflexes are essential for adapting motor control to different gravity conditions.
  • Spinal Cord as a Gravity Compensator: A key finding is that the SC, through its reflex pathways, can significantly assist in compensating for gravity. This reduces the burden on the brain to constantly adjust motor commands.
  • Spinal Pathways Enhance Movement Quality: Increasing the strength of certain spinal pathways, particularly the Ia-MN stretch reflex and Ia-MNs heteronymous connections, led to smoother movements and improved robustness against perturbations.
  • Two Paths to Adaptation: The study demonstrated that motor control adaptation to gravity can be achieved through two primary strategies: modulating supraspinal commands or modulating spinal synaptic strengths.
  • "Minimal SC" Effectiveness: A simplified SC model, incorporating the Ia-MN stretch reflex and Ia-INa reciprocal inhibition, was sufficient to reproduce target trajectories in various gravity conditions and improve optimization efficiency.

Key Results

  • Easier Optimization with SC: When the SC model was included, optimizing supraspinal commands for target trajectories in different gravity conditions resulted in faster convergence and better performance (lower loss).
  • Supraspinal Command Modulation: For gravity adaptation, the primary modulations in supraspinal commands were found in the flexor sinusoid's amplitude (for gravity magnitude) and frequency (for gravity direction).
  • Spinal Pathway Modulation: In the absence of supraspinal command optimization, modulating spinal synaptic strengths also successfully reproduced target trajectories, especially for mild gravity changes.
  • Specific Pathway Effects:
    • Ia-MN stretch reflex and Ia-MNs heteronymous connections improved movement smoothness and robustness against perturbations.
  • Ia-INa reciprocal inhibition resulted in the smallest RMSE to the target trajectory.
  • II stretch reflex and Ia-MNs heteronymous pathways had a notable effect on both RMSE and smoothness.
  • Gravity-Specific Spinal Adjustments: For increased gravity magnitudes, flexor synaptic strengths tended to increase, while extensor strengths decreased, facilitating elbow flexion-extension.

Limitations and Future Work

  • Simplified Supraspinal Commands: The study used sinusoidal supraspinal commands for simple elbow movements. Future work could explore more complex supraspinal commands, potentially incorporating cerebellar learning signals.
  • Limited Muscle Representation: The model included a subset of upper limb muscles. Expanding this to include more muscles, particularly shoulder muscles, would provide a more comprehensive representation.
  • Exploration of Other Spinal Pathways: While several pathways were investigated, others like Renshaw cell connections could be explored.
  • Focus on Synaptic Strength: The study primarily focused on modulating synaptic strength. Future research could investigate other neuronal dynamics, such as activation offsets and sensory feedback delays.
  • Combined Optimization: The supraspinal commands and spinal synaptic strengths were optimized separately. A combined optimization approach could offer deeper insights into how these components interact in real-time control.
  • Understanding Mechanisms: While the study revealed that the SC can assist gravity compensation, further investigation into the precise mechanisms of this modulation is warranted.

Summary updated

Abstract

Humans can perform movements in various physical environments and positions (corresponding to different experienced gravity), requiring the interaction of the musculoskeletal system, the neural system and the external environment. The neural system is itself comprised of several interactive components, from the brain mainly conducting motor planning, to the spinal cord (SC) implementing its own motor control centres through sensory reflexes. Nevertheless, it remains unclear whether similar movements in various environmental dynamics necessitate adapting modulation at the brain level, correcting modulation at the spinal level, or both. Here, we addressed this question by focusing on upper limb motor control in various gravity conditions (magnitudes and directions) and using neuromusculoskeletal simulation tools. We integrated supraspinal sinusoidal commands with a modular SC model controlling a musculoskeletal model to reproduce various recorded arm trajectories (kinematics and EMGs) in different contexts. We first studied the role of various spinal pathways (such as stretch reflexes) in movement smoothness and robustness against perturbation. Then, we optimised the supraspinal sinusoidal commands without and with a fixed SC model including stretch reflexes to reproduce a target trajectory in various gravity conditions. Inversely, we fixed the supraspinal commands and optimised the spinal synaptic strengths in the different environments. In the first optimisation context, the presence of SC resulted in easier optimisation of the supraspinal commands (faster convergence, better performance). The main supraspinal commands modulation was found in the flexor sinusoid’s amplitude, resp. frequency, to adapt to different gravity magnitudes, resp. directions. In the second optimisation context, the modulation of the spinal synaptic strengths also remarkably reproduced the target trajectory for the mild gravity changes. We highlighted that both strategies of modulation of the supraspinal commands or spinal stretch pathways can be used to control movements in different gravity environments. Our results thus support that the SC can assist gravity compensation.
BibTeX
@article{bruel2025role,
  author = {Alice Bruel and Lina Bacha and Emma Boehly and Constance De Trogoff and Luca Represa and Grégoire Courtine and Auke Jan Ijspeert},
  title = {{Role and modulation of various spinal pathways for human upper limb control in different gravity conditions}},
  journal = {PLOS Computational Biology},
  year = {2025},
  doi = {10.1371/journal.pcbi.1012069},
}

Loading related content...

Figure