Are you curious about what exactly you are going through during a complex search process? Our research team aims to model users’ search interactions in a complex search task, to systematically, comprehensively and granularly reveal the behavioral characteristics and patterns at different search stages, and the dynamic behavioral evolution mechanism during the whole task completion process. The research would serve for the design and optimization of search system, promote users’ learning skills, and facilitate the effectiveness and efficiency of task completion.

We conducted a user experiment with 32 participants. They were asked to complete a series of daily health information related search tasks. Students’ searching, reading, writing activities were identified to be three main interaction types. We characterized the patterns and evolution of searching, reading, writing activities during this process and then finally proposed the Search Pace Model (SPM) to describe how users proceed their search interactions among different stages. As shown in Figure 1, SPM consists of three sub-processes: exploration, accumulation and composition/reporting.

In the Exploration sub-process, searching is the dominant activity. Searchers issue many queries and visit many search results pages to collect information. And the most frequent activity transitions are switching between searching and reading, and searchers rarely write during this sub-process.

When the searchers move to the Accumulation sub-process, their searching activity decrease significantly, while reading activity become dominated. Searchers visit more unique content pages and have more revisits per content page, and they often switch between reading and writing activities frequently. During this process, searchers try to understand and assimilate the acquired information, as well as start to write the search output while reading.

The last sub-process is Composition/Reporting, during which searchers spend most efforts in writing activity. Searchers still switch between writing and reading activities frequently. Besides that, searchers also often turn to search results pages while they are writing to confirm whether there is any missing information. During this process, searchers devote most of their efforts to producing their own understanding or new thoughts based on accumulated information.

Apparently, searchers would experience different sub-processes during the task completion process, from Exploration to Accumulation, and finally Composition/Reporting. However, it is possible that this process is iterative, and searchers are likely to move back and forth in different sub-processes, especially for complex tasks. More details can be found in our JIS paper: Liu, C., Song, X., & Hansen, P. (2021). Characterising users’ task completion process in learning-related tasks: A search pace model. Journal of Information Science. https://doi.org/10.1177/01655515211060527

Search Pace Model connects both the macro- and micro- levels to describe the complex nature of the information searching process. On the macro-level, it demonstrates how searching, reading and writing activities proceed during the task completion process. On the micro-level, it illustrates in detail searchers’ activity and activity transition patterns within each sub-process. This model would help the search system to monitor searchers’ behavioral and cognitive processes and have implications for the design of search systems to support users’ searching and learning during different sub-processes and achieve better learning outcomes.