An empirically-derived model for predicting completion time of cursor positioning tasks in dual-task environments.

摘要

Computers are increasingly being used in settings away from the desktop. Examples of such systems include in-vehicle navigation systems, maritime electronic navigation systems, personal digital assistants, and mobile telephones, among others. Working with such devices is more complex because the interactions are typically carried out in conjunction with another cognitive or manual task, such as reading, driving, or walking. In addition, the interaction frequently happens in a non-stationary environment, i.e., one where the user and the computing device are moving.; This dissertation investigates whether completion time in dual-task situations increases, and whether the activation of on-screen controls through manual input devices, such as a trackball, joystick, or touch screen, is unchanged when the targets are not stationary. The research specifically seeks to determine whether the present models for predicting task completion time still apply in mobile settings. The analysis is centered around a series of experiments conducted on a research platform developed by the author for performing movement time evaluations. Further experiments are evaluated to determine the effects of posture, walking, and distractions on cursor positioning time using either direct or indirect input.; The research arrives at a predictive model for estimating the time it takes to enter a sequence of digits on a virtual keypad. The model is empirically validated through a series of context-aware laboratory experiments.; The results of this research provide designers of mobile devices a predictive model for mean task completion time of numerical data input tasks on a soft keypad, including specific heuristics to assure that interactions can be completed in a timely and accurate manner.