Modulation of arm stiffness in relation to instability at the beginning or the end of goal-directed movements

In reaching to a target, stability near the target may be more critical for success than stability far from the target. Consequently, we postulated that high instability near the start would evoke less compensation than high instability near the target. Three stability conditions were implemented using a robot manipulandum: neutral stability everywhere (null field); high instability along the first half of the trajectory decreasing as the target was approached (start unstable); and instability increasing along the first half of the trajectory and remaining high as the target was approached (end unstable). Under the start unstable condition, the stiffness of the arm in the region of highest instability was significantly less than under the end unstable condition. Furthermore, the stability of the system (manipulandum plus arm) was much lower under the start unstable condition than under the null field condition whereas it was similar under the end unstable and null field conditions.