In Parkinson's disease (PD), basal ganglia dysfunction leads to disturbed sensorimotor integration and associated timing. Previous functional MRI and behavioural PD studies on timing indicated a specific striatal contribution to assessing spatial displacement in velocity estimation. In this computation, cerebral processing time implies demarcating discrete intervals of spatial change. To quantify these putative intervals, the threshold of perceived velocity change of a moving ball was assessed in healthy volunteers and PD patients. After rebound from the upper side of a monitor screen, the ball's velocity increased or decreased with variable magnitudes while participants indicated whether they noticed this velocity change. The threshold for detecting velocity change was around 0.014 rad/s in both groups. Moreover, velocity was perceived as equal when the ball decelerated; unchanged velocity was perceived as acceleration. This shift was 0.009 rad/s for healthy volunteers and 0.007 rad/s for PD patients, and was negatively correlated with the severity of bradykinesia. As the trajectory length before and after velocity change was the same, velocity change was also expressed as a change in stimulus duration (relative to 1 s initial duration). The temporal equivalent of a threshold for perceived velocity change was around 75 ms in both groups. The perceptual acceleration bias' is in line with the flash-lag' effect: the position of a moving stimulus is projected ahead compared with a stationary landmark. Such an extrapolation over adjacent past and predicted locations enables real-time' visuomotor control, notwithstanding delays because of intrinsic cerebral processing time. In PD, such impaired perceptual feed-forward processing may result in slow movements. (C) 2013 Wolters Kluwer Health vertical bar Lippincott Williams & Wilkins.