Is there an optimal arm posture? Deterioration of finger localization precision and comfort sensation in extreme arm-joint postures.

Abstract

Processing of joint redundancy is one of the most important problems in motor control. For instance, gaze orientation can be obtained with an infinite number of eye and head combinations. It has been proposed that a solution to this problem might be the minimization of eye and head position-signal errors. For arm movements, where the excess of degrees of freedom is even higher, cost function was proposed as a criterion for movement selection, reflecting some comfort variable evoked from the peripheral inputs, e.g. optimal muscular energy cost or glucose consumption. However, no biological implication of comfort on motor control has yet been demonstrated. We have further investigated this approach by hypothesizing that arm posture choice also relies on a minimization of position-signal errors arising from individual joints. The prediction is that accuracy of fingertip localization by pointing made by the contralateral hand would be enhanced for comfortable postures of the target arm and degraded for uncomfortable postures using extreme joint positions. Results show an increase in pointing variability when extreme joint postures are used (wrist flexion, shoulder elevation, or both). This increase in pointing variability is proportional to the increase in subjective discomfort rating. Individual joint effects can be added arithmetically into a whole arm value for both discomfort rating and pointing variable and constant error. These results suggest that the choice of comfortable postures for the arm corresponds to an optimization of arm position-signal reliability. This new constraint might be a useful tool for further investigation on posture or trajectory formation.