Empirical and simulation studies are described which assess the form of central control signals for human limb movement. Recent claims about the complexity of control signals are addressed using a mathematical model of two-joint planar arm movement. The model is based on the lambda version of the equilibrium point hypothesis and includes neural control signals, reflexes, reflex delays, muscle mechanical properties, realistic musculo-skeletal geometry, limb dynamics and external loads. It is shown that ostensibly "complex" features of limb motion such as the "2/3 power law" and non-monotonic patterns of limb impedance are predicted using simple constant-rate equilibrium shifts which do not explicitly encode these movement properties. In addition it is shown that simulated limb impedance using the A model both during movement and in statics matches empirically estimated values of impedance reported in the literature. / Empirical studies assessed the control of muscle coactivation by measuring tonic levels of electromyographic activity in shoulder and elbow muscles at the end of reaching movements in a horizontal plane. Shoulder muscle coactivation was related to the amplitude and velocity of shoulder motion, and unrelated to elbow motion whereas elbow and double-joint muscle activity was related to elbow motion, and unrelated to shoulder motion. It is suggested that muscle coactivation at the shoulder and elbow may be controlled independently. / Other experiments addressed the extent to which control signals are adjusted to account for interaction torques---torques arising at one joint due to the motion of limb segments about adjacent joints. Electromyographic activity of limb muscles was measured during single- and multi-joint movements in which the magnitude or direction of interaction torque was systematically varied. During single-joint movements, phasic activity which preceded motion and varied in magnitude with interaction torque was observed in muscles acting at the stationary joint. During multi-joint movement EMG activity in muscles at one joint was modulated to offset interaction torque arising from limb motion about an adjacent joint. It is suggested that control signals to muscles are adjusted to offset interaction torques arising from limb dynamics. Schemes for incorporating information about dynamics into the position control framework proposed in the equilibrium point hypothesis are discussed.
Identifer | oai:union.ndltd.org:LACETR/oai:collectionscanada.gc.ca:QMM.37574 |
Date | January 1999 |
Creators | Gribble, Paul L. |
Contributors | Ostry, David J. (advisor) |
Publisher | McGill University |
Source Sets | Library and Archives Canada ETDs Repository / Centre d'archives des thèses électroniques de Bibliothèque et Archives Canada |
Language | English |
Detected Language | English |
Type | Electronic Thesis or Dissertation |
Format | application/pdf |
Coverage | Doctor of Philosophy (Department of Psychology.) |
Rights | All items in eScholarship@McGill are protected by copyright with all rights reserved unless otherwise indicated. |
Relation | alephsysno: 001687269, proquestno: NQ55337, Theses scanned by UMI/ProQuest. |
Page generated in 0.0154 seconds