Exp Brain Res (1995) 105:163-174 9Springer-Verlag 1995 C. J. Winstein - R S. Pohl Effects of unilateral brain damage on the control of goal-directed hand movements Received: 17 January 1995 / Accepted: 16 March 1995 Abstract Insight into the functional neural substrates associated with the control of goal-directed purposive movements can be obtained through the study of the per- formance of individuals with brain damage. The control of rapid reciprocal aiming was investigated by compar- ing ipsilateral limb performance of subjects with unilat- eral brain damage to that of controls performing with the same limb. Thirty right-hand-dominant individuals, ten with right hemisphere stroke, ten with left hemisphere stroke, and ten age-matched controls performed uncon- strained alternating tapping movements under three con- ditions of task complexity. The path of the stylus was re- corded by video using two-dimensional kinematic tech- niques. Key kinematic features of the vertical and hori- zontal components of the trajectories were analyzed us- ing both quantitative and qualitative methods. All sub- jects with brain damage showed prolonged movement times; however, the locus of the slowing depended on le- sion side. Specifically, subjects with left stroke showed deficits in the open-loop component of the movement across all three conditions of task complexity, and a pro- longed reversal phase surrounding target impact, particu- larly in the most complex condition. In contrast, subjects with right stroke showed deficits in the closed-loop phase of the movement prior to target impact, particular- ly in the most complex condition when visual informa- tion was necessary for accuracy. Together, these results suggest that for the control of rapid goal-directed aiming movements, the left hemisphere is dominant for task-rel- evant aspects of processing associated with the ballistic component and the timing or triggering of sequential movements. In contrast, the right hemisphere is domi- nant for processing associated with rapid, on-line visual information even when target location is known and di- rection is certain. C. J. Winstein ( ~ ) 9 R S. Pohl Department of Biokinesiology and Physical Therapy, University of Southern California, 1540 E. Alcazar St., CHP 155, Los Angeles, CA 90033, USA; e-mail: Winstein@hsc.usc.edu, Fax no.: (213) 342-1515 Key words Kinematics 9 Hemispheric asymmetry - Stroke - Arm movements - Human Introduction Since the landmark work of Paul Fitts (1954), it has been known that the time to complete a rapid goal-di- rected aiming movement is dependent on task complexi- ty. In Fitts' (1954) formulation, task complexity or in- dex of difficulty [ID, (log 2 2A/W)] increases with larger movement distances (A) and smaller target widths (W). The control of rapid aiming movements performed in conditions where the ID is low (less than 3.58 bits) is governed primarily by open-loop, ballistic processes that do not rely on visual feedback for accuracy (Wal- lace and Newell 1983). In contrast, the control of these kinds of movements in high ID conditions (greater than 4.58 bits) is governed by at least two kinds of processes: (1) an open-loop process controlling the ballistic trans- port phase that brings the effector close to the target (Klapp 1975; Schmidt et al. 1979), and (2) a closed-loop process requiring feedback-based adjustments as the performer approaches the target endpoint (Woodworth 1899; Keele 1968; Crossman and Goodeve 1963; MacKenzie et al. 1987; Milner and Ijaz 1990; Milner 1992). Insight into the neural substrate associated with the control of goal-directed movements comes from investi- gations of the performance of individuals with unilateral brain damage (e.g., Goodale 1988). The observation of a deficit in motor performance after such injury provides evidence that the damaged area may be necessary for that component of processing underlying the deficit. To control for hand asymmetries and effector problems caused by hemiparesis and hemisensory loss, perfor- mance of the limb ipsilateral to the lesion is compared to that of controls using the same limb (e.g., Haaland et al. 1987). Studies using this approach have provided evi- dence that each hemisphere has a specialized role in the control of goal-directed voluntary movements.