Multi-Aspect Information Use Task Performance: The Roles of Topic Knowledge, Task Structure, and Task Stage

Information search, quite often, is not an isolated activity, but is accompanied by the use of the located information to generate some outcome. Frequently seen are “complex” tasks that consist of multiple aspects and can be divided into sub-tasks and/or finished in multiple sessions. This paper explores how search systems may help users with their multi-aspect tasks by examining whether, and how user knowledge and task structure play roles in leading to better work task performance. A 3-session lab experiment was conducted with 24 participants, each coming 3 times to work on 3 subtasks of a general task, couched either as the “parallel ” or the “dependent ” structure type. The overall task was to write a report on the general topic, with interim documents produced for each subtask. Results show that users ’ previous knowledge of task topics positively correlated with task performance, but the evolving and the exit knowledge did not. Neither task session nor task structure significantly affected session task performance. However, there was a tendency that those users with different levels of task knowledge performed differently in different tasks: higher knowledge users tended to perform better in the parallel task than in the dependent; lower knowledge users tended to perform better in the dependent task than in the parallel. These results can possibly be explained by users ’ searching and writing behaviors. Our findings help understand factors influencing information use task performance, and have implications on designing personalized systems that support information use task accomplishment. Keywords Information use task performance, topic knowledge, multi-aspect task, task structure