Proprioceptive population coding of two-dimensional limb movements in humans: II. Muscle-spindle feedback during "drawing-like" movements.

It was proposed to study the proprioceptive sensory coding of movement trajectories during the performance of two-dimensional "drawing-like" movements imposed on the tip of the foot. For this purpose, the activity of the muscle-spindle afferents from the Extensor digitorum longus, Tibialis anterior, Extensor hallucis longus, and Peroneus lateralis muscles was recorded from the lateral peroneal nerve using the microneurographic technique. The drawing movements, describing geometrical shapes such as squares, triangles, ellipses, and circles, were imposed at a constant velocity in both the clockwise and counterclockwise directions. A total number of 44 muscle-spindle afferents were tested, 36 of which were identified as primary and eight as secondary afferents. Whatever the shape of the imposed foot movement, the primary endings from one muscle never discharged throughout the whole trajectory (on average, they discharged for only 49.2% of the length of the trajectory), whereas all the secondary endings discharged for most part of the drawing trajectories (average: 84.8%). The relationship between afferent discharge rate and direction could be described with a cosine-shaped tuning function. The peak of this function corresponded to the preferred sensory direction of the receptor-bearing muscles. The whole path of a given geometrical drawing movement was found to be coded in turn by each of the primary afferents originating from each of the muscles successively stretched. The contribution of each population of muscle afferents from each ankle muscle was represented by a "population vector", whose orientation was the preferred direction of the muscle under consideration and whose length was the mean instantaneous frequency of the afferent population. The "sum vector" corresponding to the sum of all these weighted "population vectors" was found to point in the instantaneous direction of the drawing trajectory, i.e., the tangent to the trajectory. These findings suggest that trajectory information is already encoded at the peripheral level on the basis of the integrated inputs provided by sets of receptors belonging to all the muscles acting on a given joint.