Smoothness is a hallmark of healthy movement and has the potential to be used as a marker of recovery in rehabilitation settings. While much past research has focused on shoulder and elbow movements (reaching), little is known about movements of the wrist despite its importance in everyday life and its impairment in many neurological and biomechanical disorders. Our current lack of knowledge regarding wrist movement prevents us from improving current models, diagnosis, and treatment of wrist disorders. In particular, while movement smoothness is a well-known characteristic of reaching movements and may potentially be used to diagnose and monitor recovery from neurological impairments, little is known about the smoothness of wrist rotations. Therefore, because the smoothness of wrist rotations has not been characterized, it cannot be used as a marker for diagnosis and evaluation. This study examines the smoothness of wrist rotations in comparison to the known baseline of reaching movements. Subjects were asked to perform wrist and reaching movements under a variety of conditions, including different speed and direction. To measure movement smoothness, this study used an established metric of speed profile number of maxima and presents a novel method for characterizing smoothness by fitting a minimum-jerk trajectory to real movement data.The results show that 1) wrist rotations are significantly less smooth than reaching movements (p≤0.0016), 2) smoothness decreases significantly as speed decreases (p<0.0001), and 3) wrist movements exhibit a pattern of smoothness that varies significantly between targets and outbound/inbound movement directions (p<0.0001). Potential causes for results 1 and 3 are presented and tested by simulation or reference to prior studies, because these findings were previously unknown. The decrease in smoothness with speed (result 2) has been found in prior studies of smoothness in reaching and finger movements. The reasoning behind the first result is explored by testing whether the difference in smoothness between wrist and reaching movements was due to differences in mechanical, muscular, neural, or protocol-related properties. The reasoning behind the third result is explored by testing whether the difference in wrist direction was due to anisotropy in musculoskeletal dynamics or anisotropy in movement duration. The simulations show that the wrist’s bandwidth is greater than that of the arm, and that there is nonvoluntary power in the bandwidth of the wrist that would be low-pass filtered in reaching movements, indicating that at least some of the difference in smoothness between wrist and reaching movements is due to differences in mechanical properties. Differences in muscular, neural, or protocol-related properties (signal-dependent noise, proprioceptive acuity, and the speed requirements of the task, respectively) do not appear to be the cause of the difference in smoothness between wrist and reaching movements. Differences in wrist smoothness between movement directions appears to be related to differences in movement duration between directions.



College and Department

Ira A. Fulton College of Engineering and Technology; Mechanical Engineering



Date Submitted


Document Type





movement, kinematics, smoothness, minimum jerk, motor control