Humans find reasoning about uncertainty difficult. In decision support systems and software for intelligence analysis, graphical representations are commonly used to display uncertainty. Nevertheless, our understanding of how people use the information presented in graphs displaying uncertainty to make decisions is limited. As many artificial intelligent systems require a human-in-the-loop who is able to actively take part in the analysis process, the understanding of high-level cognition involved in human-graph interaction is essential in the design of better tools for analysis. In this research, we investigate the visual behaviour that is associated with participants responses to problems testing probabilistic reasoning represented through two different visualizations (tree and Venn diagrams). Using the data from visual fixations and transitions, we present a description of different reasoning strategies covering both accurate and inaccurate reasoning for different visualization formats. The results show that gaze behaviour is related to reasoning accuracy. Moreover, this study shows that different graphs representing the same problem evoke different reasoning strategies, suggesting that higher level cognition is influenced by the graphical representation in which uncertainty is encoded.

No abstract

The use of desktop technologies, opposed to their mobile counterparts, is still predominant in the business domain and daily corporate activities. For complex business tasks, the usability of mobile applications is still perceived as inferior. Nevertheless, mobile computing could support work activities in different ways. This research seeks to explore whether mobile applications, especially sophisticated real-time groupware such as Group Decision Support Systems can match or even outperform their desktop-based counterparts regarding task performance and user satisfaction. To provide a more comprehensive comparison, two widely used web metric frameworks, HEART and PULSE (the former focusing on human behaviour and the latter focusing on technology), have been used to evaluate and compare a web-based mobile group decision support app with its desktop counterpart on a chair-led multi-step decision-making process in lab settings and real-world contexts. Moreover, data from web tracking tools and system logs of these two apps have been analyzed to deepen the level of analysis. Therefore, an innovative methodology for usability research combining lab experiments with long-term web observations is proposed. The results have shown that users interacting with the mobile version of the system had a steeper learning curve than users interacting with the desktop one. After a short practice period, participants using the mobile version of the groupware could perform as efficiently as participants using its desktop counterpart. Not only both the mobile and the desktop version of the app were used effectively for the group decision tasks chosen for this study, but also the two versions of the application were rated similarly in terms of user satisfaction. Furthermore, the mobile version had a faster adoption, better engagement, and better retention than its desktop counterpart. Despite an additional groupware middleware layer and a sophisticated user interface rendering layer, both groupware versions have yielded a user-to-user response time sufficient for real-time group interactions. These results convey important implications; as society is quickly moving towards mobile computing, web-based mobile technologies can now support multi-step group decisionmaking tasks.

In recent years, quantifying the impacts of detrimental air quality has become a global priority for researchers and policy makers. At present, the systems and methodologies supporting the collection and manipulation of this data are difficult to access. To support studies quantifying the interplay between common gaseous and particulate pollutants with meteorology and biological particles, this paper presents a comprehensive data-set containing daily air quality readings from the Automatic Urban and Rural Network, and pollen and weather data from Met Office monitoring stations, in the years 2016 to 2019 inclusive, for the United Kingdom. We describe (1) the sources from which the data were collected, (2) the methods used for the data cleaning process and (3) how issues related to missing values and sparse regional coverage were addressed. The resulting data-set is designed to be used ‘as is’ by those using air quality data for research; we also describe and provide open access to the methods used for curating the data to allow modification of or addition to the data-set.