Three postulates of the prehension control are introduced: (1) prehension synergies are organized as hierarchical, two-step computational processes; (2) to control the finger forces, the CNS uses mode variables; and (3) a torque constraint is a core constraint affecting finger forces in grasping. Based on the postulates and the preliminary data a set of testable hypotheses is formulated: (A) The virtual finger forces obey the principle of superposition. (B) The antagonist moments, when they are mechanically unnecessary, are due to the mode control and the enslaving effects resulting from it. (C) The CNS responds to changes in the object geometry by chain reactions. (D) The mode commands to the fingers generating agonist moments are proportional to the fingers' mechanical advantage, i.e. the distance from the axis of rotation. (E) Grasp stability is under stiffness control. In particular: (a) a small (<10 ram) increase or decrease in the handle width brings about proportional force increments/decrements; and (b) the force increments/decrements in the involved fingers are equal. We will perform four experiments. The experiments will respectively address hypotheses A and B, hypothesis C, hypothesis D, and hypothesis E. In experiment 1, the subjects will be asked to hold a handle at equilibrium while resisting different torques and loads. In experiment 2, in addition to changes in the torque, the handle geometry will be varied. In experiment 3, the handle will be fixed to a shaft. The subjects will generate a moment of force about the shaft as a pivot while the shaft location will vary among the trials in a systematic manner. In experiment 4, a motorized handle that allows for changing the grip width will be used. The handle will either expand or collapse while the subjects resist a prescribed torque and load.
Showing the most recent 10 out of 138 publications
