Proprioception, the sense of knowing the position and movement of body parts, allows humans to move freely without constantly watching our limbs. New research findings about how the brain controls proprioception could lead to more natural and intuitive control of artificial limbs.
Proprioception involves a complex network of sensors embedded in our muscles that relay information about limb position and movement back to our brain. However, little is known about how the brain puts together the different signals it receives from muscles.
A new study led by Alexander Mathis, PhD, at Ecole Polytechnique Federale de Lausanne sheds light on the question by exploring how our brains create a cohesive sense of body position and movement. “It is widely believed that sensory systems should exploit the statistics of the world, and this theory could explain many properties of the visual and auditory system,” Mathis said. “To generalize this theory to proprioception, we used musculoskeletal simulators to compute the statistics of the distributed sensors.”
The musculoskeletal modeling generated muscle spindle signals in the upper limb to generate a collection of “large-scale, naturalistic movement repertoire.” The researchers then used this repertoire to train thousands of task-driven neural network models on sixteen computational tasks, each of which reflected a scientific hypothesis about the computations carried out by the proprioceptive pathway, which included parts of the brainstem and somatosensory cortex.
The approach allowed the team to comprehensively analyze how different neural network architectures and computational tasks influenced the development of “brain-like” representations of proprioceptive information. They found that neural network models trained on tasks that predicted limb position and velocity were most effective, suggesting that our brains prioritize integrating the distributed muscle spindle input to understand body movement and position.
Editor’s note: This story was adapted from materials provided by Ecole Polytechnique Federale de Lausanne.
The open-access study, “Task-driven neural network models predict neural dynamics of proprioception,” was published in Cell.
