Search for: All records

Abstract Exoskeletons assist and augment human movement, but their effects on proprioceptive feedback remain poorly understood. We examined how parallel exoskeleton stiffness influences primary muscle spindle firing. In an anesthetized rat preparation, controlled stretches of the medial gastrocnemius were applied with springs (0–0.5 N/mm) attached in parallel to the muscle-tendon unit (MTU) to simulate passive exoskeleton assistance. Fascicle length was measured with sonomicrometry, force and MTU length with a servo motor, and spindle instantaneous firing rate (IFR) with dorsal root recordings. Increasing exoskeleton stiffness decreased biological muscle force (3.1 ± 0.6 N to 1.6 ± 0.6 N, p < 0.001) and stiffness (4.4 ± 1.5 N/mm to 2.3 ± 1.3 N/mm, p < 0.01), while fascicle length increased (7.9 ± 1.3 mm to 8.3 ± 1.5 mm, p < 0.005). Despite these altered mechanics, spindle firing did not significantly change, and showed weak correlations with muscle length, velocity, force, and yank (R²≤ 0.14). These results indicate that exoskeleton stiffness modifies fascicle dynamics without altering spindle firing. Previously proposed models of primary afferent firing did not sufficiently explain these results. This is the first in situ investigation of exoskeleton effects on primary afferent feedback during active contractions. 

Abstract Proprioceptive sensory feedback is crucial for the control of movement. In many ways, sensorimotor control loops in the neuromuscular system act as state feedback controllers. These controllers combine input commands and sensory feedback regarding the mechanical state of the muscle, joint or limb to modulate the mechanical output of the muscles. To understand how these control circuits function, it is necessary to understand fully the mechanical state variables that are signalled by proprioceptive sensory (propriosensory) afferents. Using new computational approaches, we demonstrate how combinations of group Ia and II muscle spindle afferent feedback can allow for tuned responses to force and the rate of force (or length and velocity) and how combinations of muscle spindle and Golgi tendon organ feedback can parse external and internal (self‐generated) force. These models suggest that muscle spindle feedback might be used to monitor and control muscle forces in addition to length and velocity and, when combined with tendon organ feedback, can distinguish self‐generated from externally imposed forces. Given that these models combine feedback from different sensory afferent types, they emphasize the utility of analysing muscle propriosensors as an integrated population, rather than independently, to gain a better understanding of propriosensory–motor control. Furthermore, these models propose a framework that links neural connectivity in the spinal cord with neuromechanical control. Although considerable work has been done on propriosensory–motor pathways in the CNS, our aim is to build upon this work by emphasizing the mechanical context.