Multijoint grasping movements. Simulated and observed effects of object location, object size, and initial aperture.

Studies of human prehension have revealed characteristic patterns of grasping kinematics. We sought to gain insight into the determinants of those patterns by means of a computer simulation and accompanying behavioral experiment concerning multijoint, planar grasping behavior. The simulation was based on a recent theory of posture-based motion planning which hypothesizes that movement preparation entails time-limited, multiple task-constraint satisfaction. Prehension was modeled with a stick-figure animation involving 12 series of 81 grasping movements. Circular objects to be grasped were located at three angles (45 degrees, 90 degrees, and 135 degrees) and at three distances (20 cm, 30 cm, and 40 cm) relative to the initial location of the hand in the workplane. Additionally, three object sizes (2 cm, 4 cm, and 6 cm in diameter) and three initial aperture sizes (0.3 cm, 3.3 cm, and 7.0 cm) were used. Analyses of the simulated grasping movements focused on the time course of the hand opening, the tangential velocity of the wrist, and the rotations of the joints in the arm, hand, and fingers. The results showed that the model accurately mimicked detailed kinematics of prehension observed in earlier studies. With respect to the frequently reported relationship between object size and hand opening, the simulations further revealed an effect of initial aperture. This predicted effect was confirmed in an experiment in which four participants performed analogous planar grasping tasks. An analysis of the time course of the opening of the hand showed that maximum aperture covaried with initial aperture. A conclusion of this work is that a major determinant of grasping kinematics is avoidance of collisions with objects that are to be grasped.