This study aims to determine whether, in the light of motor control theories, a tablet can be used as a mouse substitute to interact with graphical objects displayed on another screen. It was found that a lack of information needed to plan and control movement affected pointing movement performance to varying degrees, depending on the device used. Three distinct conditions were studied in order to quantify performance loss for pointing movements that are visually controlled through a video screen. Participants were instructed to perform aiming movements. These were carried out directly on the tablet, either through a video screen capable of displaying feedback on hand movements, or through a video screen which was not capable of displaying feedback on hand movements. The data reveal that pointing movements through a video display only seem to be usable if feedback on arm location is displayed. With feedback, the error rates for pointing movements are similar to those carried out directly on a tablet. However, the movement time is doubled. These findings complement those reported in previous studies (Smith &?Smith, 1990) which suggest that decorrelating the visuo-motor loop in pointing movements leads to a decrease in performance. Furthermore, for the first time, a Fitts' Law protocol is used in an attempt to quantify this phenomenon.