Fig. 4.

Trajectory following (slow) task for the assessment of sensorimotor impairments. a Subjects need to actively follow with their index finger a target trajectory displayed on the tablet screen. Control subjects can follow the target more accurately, while stroke subjects are further away from the target and their movement is more fragmented. b There is a high agreement (70%) in impairment classification between this task metric and the clinical measure of proprioception (kUDT). c The task has good reliability on both the less affected and on the more affected side (ICC $>0.70$). d The box plot indicates increasing tracking error RMSE with increasing proprioceptive impairment (according to kUDT). There is a significant difference between controls (N=62) and all stroke sub-groups on the affected side (kUDT=2: N=14, kUDT=1: N=8, kUDT=0: N=8), as well as between less affected side of stroke subjects (N=30) and the group with the most severely impaired proprioception (N=8). Acronyms - D: dominant, LA: less affected, A: affected, kUDT: kinesthetic Up-Down Test, ICC: intraclass correlation coefficient, SRD: smallest real difference. Statistical significance: p-value<0.05: *, p-value<0.01: **, p-value<0.001: ***