Browse Theses

This research examines multitasking from the perspective of human-computer interaction (HCI). Multitasking is defined as the performance of multiple tasks concurrently. In a computer-based environment, users generally switch between multiple computer-based tasks either due to a personal decision to break from the current task or due to an external interruption, such as an electronic notification. This dissertation describes an in-depth empirical study, using a laboratory setting with different numeric, verbal, and visual computer-based tasks. Six hundred and thirty six subjects were randomly assigned into three conditions: discretionary multitasking , where participants were allowed to decide when and how often to switch tasks, forced multitasking, where participants were forced to switch tasks at certain allotted times, and non-multitasking , where participants performed the tasks sequentially and were not allowed to multitask. In order to investigate performance effectiveness (accuracy) and performance efficiency (productivity), participants' overall accuracy and productivity scores were compared across conditions. The results suggest that during difficult tasks, subjects who were forced to multitask had the lowest accuracy. In addition, those subjects in the forced multitasking condition who felt the primary task was difficult had lower accuracy than those who felt the task was easy. This was not true in the other two conditions. Receiving interruptions during a difficult task impacted not only their primary task, but their secondary tasks as well. In the discretionary multitasking condition, the more subjects decided to multitask, the lower their accuracy scores. In fact, an additional analysis revealed that high multitaskers not only performed worse than low and medium multitaskers in the discretionary condition, but actually had the worst performance than subjects in any other condition. Medium multitaskers, however, had the highest productivity scores. While multitasking in that case was considered the best in terms of efficiency, it was not true in terms of effectiveness. Therefore, discretionary multitasking gives the illusion of high performance. Furthermore, this study also explored why people chose to multitask and the impact that had on performance. The results of this study can assist HCI researchers in developing a more comprehensive understanding of user multitasking which can lead to better interface designs.