A new model of information seeking stopping behavior
MetadataShow full item record
Web search engines play an important role in peoples daily life. Widespread usage of search engine poses continuous challenges for designing information search systems that can bring people best user experience. To address this challenges, it is particularly important to understand how people seek information. In spite of a large number of studies on human information seeking, the reasons of when and why users terminate information seeking are uncertain and many proposed theories have a limited capability for predicting this type of behavior. In our study, we conducted lab-based experiments, where participants performed assigned information search tasks on Wikipedia pages. Inspired by theories and methods from cognitive science, we captured participants information search behavior such as query usage, search engine result page visits, Wikipedia page visits, and task duration. Additionally, we used eye-tracking techniques to examine the number of people's eye fixations. Using exploratory factor analysis (EFA), we have confirmed exploratory and validation processes can be distinguished based on different types of costs associated with each of them. Based on the findings of the regression tree model, evaluating the cost and gain in the validation process provide important feedback to people for controlling and monitoring their information search.