Arm movement control in gravity

In this study, we investigated whether gravity influences goal-directed planar arm movements by having participants perform point-to-point outward and inward movements in a reclinable chair, varying orientation relative to gravity from horizontal (0°) to vertical (90°) in steps of 15°. Strict temporal (400 ± 100 ms) and spatial constraints ensured comparable movement conditions across orientations. Consistent with the literature, we found significant differences between inward and outward movements in terms of movement curvature ( x dev ) and time-to-peak-velocity ( t pv ), which were consistent across all chair reclining angles. However, when comparing inward movements at different angles and outward movements at different angles, we found no significant effect of gravity on both x dev and t pv . Using an optimal control musculoskeletal model of the human arm, we predicted movement trajectories for several commonly proposed cost functions. We found that only jerk-based (kinematic) cost functions reproduced the experimentally observed movement kinematics, while cost functions incorporating muscle activation (e.g. control effort) or muscle output (e.g. muscle force, muscle torque) predicted clear gravity-dependent effects that were not present in the experimental data. In conclusion, we found no effect of body orientation relative to gravity on the kinematics of goal-directed arm movements; observed differences were due only to differences in movement direction. In addition, only kinematics-based cost functions generated movements similar to those observed experimentally.