The purpose of handwriting is to produce a legible trace quickly and fluently. Its motor control and learning therefore rely on an efficient integration of visual and proprioceptive feedback, with a transition from a control based on the written trace in beginner to a control based mainly on writing movements in expert writer. The aim of this study was to test the effect of a partial deletion of the written trace and an increase of visual information on the kinematics after the performance in a learning task. Twenty-four adults learned to write six new pseudoletters with a stylus controlled by non-dominant hand on a touch screen digital tablet. Half of the pseudoletters were trained in the modified visual FB conditions and the other half without modification (control condition). Results revealed that, in the short-term, the pseudoletters trained with the modified visual feedback were written faster and more fluently than those trained in the control condition, without spatial accuracy reduction. This method seems to be efficient, at least in proficient adults and is currently tested in children with dysgraphia.