Pauses in man-machine interactions: a clue to users' skill levels and their user interface requirements

Abstract

Computer users have motionless periods of time while performing computer-based tasks. Do these pauses relate to the mental and perceptual actions required to perform tasks? Do users essentially pause while they think, wait to retrieve the next step to perform, or search the location of something on the screen? How do the pauses change as the users gain experience and progress from novice to skilled? To answer these questions, we conducted user studies to investigate the link between the pauses observed in users' interactions with computer-based applications and their skill levels. In this paper, we introduce a set of pause-related attributes that can distinguish among different levels of skills in performing Graphical User Interface GUI tasks. We employ machine learning algorithms to build skill classifiers based on these attributes. These skill classifiers can be used to create skill-adaptive applications.