T. Gross et al. (Eds.): INTERACT 2009, Part I, LNCS 5726, pp. 274–287, 2009. © IFIP International Federation for Information Processing 2009 Insight into Goal-Directed Movements: Beyond Fitts’ Law Karin Nieuwenhuizen 1 , Dzmitry Aliakseyeu 2 , and Jean-Bernard Martens 1 1 Eindhoven University of Technology, Department of Industrial Design, Den Dolech 2, 5600 MB Eindhoven, The Netherlands {C.J.H.Nieuwenhuizen,J.B.O.S.Martens}@tue.nl 2 Philips Research, Media Interaction, High Tech Campus 5, 5656 AE Eindhoven, The Netherlands Dzmitry.Aliakseyeu@philips.com Abstract. Various methods and measures have been developed to assess the quality of input devices and interaction techniques. One approach to investigat- ing the performance of input devices and interaction techniques is to focus on the quality of the produced movements. The current paper proposes a new method of analyzing goal-directed movements by dividing them into meaning- ful phases. In addition to the proposed analysis method a selection of measures is suggested to assess different aspects of rapidly aimed movements. In order to evaluate the added value of the proposed analysis method an experiment has been conducted to compare two input devices (mouse versus stylus with tablet) with respect to their performance on a multi-directional pointing task. The re- sults show that the analysis into several phases reveals clear differences in the movement strategy. Keywords: Input devices, interaction techniques, movement analysis, perform- ance measures. 1 Introduction A key aspect of human-computer interaction is the use of input devices, such as key- board, mouse, or tablet with stylus. Since millions of people use computers on a daily basis it is important that input devices are well tailored to users’ needs. This is espe- cially true when devices are designed for people with disabilities [9, 11], or when devices are used within challenging environments, such as virtual reality. To improve existing input devices or in the process of developing new ones it is necessary to iden- tify the reasons for variations in performance. Several studies have compared input devices or interaction techniques by observ- ing characteristics of movement paths during the execution of basic tasks such as pointing, selecting or steering [4, 10, 15, 14, 9, 8]. From these comparative studies two different approaches towards movement analysis can be discerned. The first ap- proach considers characteristics of the overall movement, such as total time or throughput (Fitts’ Law). The second approach, mostly found in human movement research, assumes that a movement consists of several submovements and that this