The Touch Interaction for Data Engagement with Science on Spheres (TIDESS) program focused on designing new experiences for learners to engage with Earth science data visualizations on a touch-interactive spherical globe display in informal learning settings such as science centers and museums. Our research used frequent design iterations in a design-based research methodology that included cycles of both lab-based research and in situ museum research.

In year one, we conducted an exploratory study with a prototype running on a flatscreen tabletop in a lab setting with 11 family groups (at least one adult and at least one child age 8-13 per group). Our design helped groups to learn about science content shown in the visualizations and also to use science practices such as asking questions and refining each other?s ideas. We also found evidence of groups using language that suggested they might be employing embodied cognition as they worked with our visualizations. Embodied cognition theory suggests humans use their bodies and the surrounding environment as tools to help when thinking and learning, rather than solely using their brain. Using the language evidence we found, we also examined the touchscreen gestures people used to interact with the visualizations to help us understand how they could be using their bodies when working with data visualizations. We acquired the sphere display as well and began adapting our tabletop version of the data visualization prototype to the new technology. Our partners, the sphere developers, adapted their software by adding new interactive functionality based on our research with the tabletop as part of the process of creating our prototype experience for the sphere.

In year 2, we conducted two studies to advance our understanding of how children and adults learn with spherical displays. Our first was a combination of focus groups on ocean content and a design probe with the spherical display. The focus groups revealed that both adults and children want to learn about sea creatures and physical phenomena like rip tides. Both groups wanted to learn about human impact on the ocean as well as impacts humans may feel from the ocean, especially as it changes over time. This finding suggests that lifelong learners have a concern for information about human activities that both harm and help conserve the ocean and its inhabitants. From the design probes, we learned that multiple children will interact with the sphere simultaneously, while adults seem to prefer to let one person take charge and guide the experience. Children also often wanted to use all five fingers to contact the sphere, which new exhibit designs for spherical displays ultimately must accommodate. This led to our second study, elicitation of touchscreen gestures that people would want to use on the sphere to make certain things happen in the data visualization. We found that, unlike with flatscreen displays, all users were more likely to use multiple fingers or whole hand gestures. Our sphere developer partners continued to implement new software features based on our research, as we iteratively revised the design of our prototype experience for the sphere.  

In Year 3, we deployed our resulting prototype experience for the sphere as a standalone, unfacilitated data visualization exhibit at the Florida Museum of Natural History for a week. We videotaped users interacting with the sphere, and data analysis continues beyond the scope of the funded project. Our preliminary analysis suggests people tend to engage with the data visualizations for both open-ended and directed investigation, supported by explicit scaffolds presented in our prototype design. This deployment in a public space with a representative audience of museum-goers was part of our broader impacts of the grant, in addition to the involvement of our research participants in participatory co-design activities conducted in public areas such as our county libraries over the life of the project. We also produced a video on the project for NSF?s STEM for All Video Showcase in 2019, available here: https://stemforall2019.videohall.com/presentations/1499.

Finally, we involved a diverse set of undergraduate (11 total, including 6 REU students) and graduate research assistants (4 total) and a postdoctoral associate on the project, from both computer science and education research. Overall, TIDESS included 11 female students, and 3 students from underrepresented minority backgrounds. Two graduate students presented at international conferences. Five undergraduates (three REU students) presented at university undergraduate research conferences.

We have published and presented this interdisciplinary work in both human-computer interaction and education research venues (five papers or talks, three under review).

Our major takeaways are:
1. Children and adults use different gestures on spherical displays than they have used on flatscreen devices

2. Experiences with data visualizations can be structured to encourage talk about science content and practices

3. Experiences with data visualization show evidence of use of embodied cognition in both language and gesture

Project website, including blog: https://uftidess.wordpress.com/

Last Modified: 12/10/2019
Modified by: Kathryn A Stofer