Regional variability of stretch reflex amplitude in the cat medial gastrocnemius muscle during a postural task.

The relationship between local fibre stretch velocity (mechanical input) and the corresponding local reflex electromyographic (EMG) amplitude (a measure of the neural output) was assessed to determine the contribution of muscle spindle feedback in postural control. We hypothesized that traditionally measured input variables (e.g., the velocity of an external movement or whole muscle velocity) may not accurately represent the mechanical input to the muscle spindles, especially when the background forces are small. Three cats were trained to stand on pedestals while ankle rotations were applied to the left hindlimb. EMG and fiber movement in both proximal and distal regions of the muscle were recorded in addition to muscle length and tendon force. We found that local muscle velocity was correlated poorly with whole muscle velocity, demonstrating that internal and external muscle movements are often dissimilar, particularly during tasks that involve modest levels of muscle activation. Local EMG reflex amplitudes were correlated well with the corresponding local fiber stretch velocities (R values ranging from 0.5 to 0.8) but not with muscle stretch velocity. The lack of crossed correlations between fiber stretch velocities and reflex EMG amplitudes measured in proximal versus distal regions of the muscle suggests the presence of a local reflex component. It is concluded that changes in local muscle fiber length represent the mechanical input to spindles better than changes in the total muscle length. Additionally, spindles have a specific role in the reflex activation of nearby muscle fibers.