Computational Motor Control of Human Movements

Due to the numerous degrees of freedom in the human motor system, there actually exist an infinite number of possible movements for any given task. Unfortunately, it is currently unknown how the human brain chooses one movement of the plethora of possible movements to solve the task at hand. This lack of understanding about the human brain is the starting point of this thesis. Two studies were conducted. Both of them are based on an experiment in which 20 subjects had to point five times to four different targets. All movements were tracked with an IR-tracking system. The calculation of inverse kinematics and inverse dynamics was done using a biomechanical model (MKD-Tools). The purpose of the first study was to analyze the human pointing gestures with respect to invariant movement features that occur across different movement tasks and different subjects. A comparison of the pointing movements in extrinsic and intrinsic coordinates revealed that the trajectories of the hand in extrinsic coordinates were much simpler than in intrinsic joint coordinates. Furthermore, in contrast to the joint angle trajectories, the trajectories of the hand were highly invariant across different movement tasks. Furthermore, the results indicate that the brain may use a compensatory control strategy on the joint level to assure the planned trajectory is achieved. In the second study a computational model of human movement planning was developed to unravel the principles the brain might use to choose one movement from plethora of possible movements in a given situation. Thereby, different optimization principles were examined. The comparison of predicted and measured movements exhibited that a minimum jerk strategy on joint level yielded the closest fit to the human data