n the clinical community there is a need for assessment tools that allow for objective, quantitative and valid measures of motor dysfunction. In this paper, we report on the design and evaluation of a serious game that engages patients with Parkinson's disease in upper extremity (hand/arm) movements. The game employs augmented reality to show virtual movement targets, i.e. candies falling from a conveyor belt, and a haptic game controller to catch the candies, that is able to acquire quantitative data about the patient's movement. This paper first describes the design process of the game and the system components. Secondly, we present results of our small quantitative evaluation study (N11, age: 26–60, healthy persons) regarding the usability of the system, the task load and user experience of the game. Our findings show that the system has a relatively good usability and the game is engaging, but there is still need for technical improvement with regard to tracking the controller in 3D space.