Gait phase effects in mobile interaction

One problem evaluating mobile and wearable devices is that they are used in mobile settings, making it hard to collect usability data. We present a study of tap-based selection of on-screen targets whilst walking and sitting, using a PocketPC instrumented with an accelerometer to collect information about user activity at the time of each tap. From these data the user's gait can be derived, and this is then used to investigate preferred tapping behaviour relative to gait phase, and associated tap accuracy. Results showed that users were more accurate sitting than walking. When walking there were phase regions with significantly increased tap likelihood, and these regions had significantly lower error rates, and lower error variability. This work represents an example of accelerometer-instrumented mobile usability analysis, and the results give a quantitative understanding of the detailed interactions taking place when on the move, allowing us to develop better mobile interfaces. ...

• Type: Conference
• Year: 2005
• Where: CHI
• Authors: Andrew Crossan, Roderick Murray-Smith, Stephen A. Brewster, James Kelly, Bojan Musizza
• Keywords: Better Mobile Interfaces, CHI 2005, Classification Keywords Usability, Human Computer Interaction, Mobile Usability Analysis