Welcome! Thank you for checking out the CEASAR project. We are really proud of this partnership between the University of Illinois Urbana-Champaign and Concord Consortium. CEASAR is a multi-device simulation platform (AR headsets and tablet computers) that allows groups of students to work together on astronomy problem-solving tasks. We have brought CEASAR to numerous undergraduate classrooms so far, and we are currently analyzing video and logfile data to make some inferences about how this configuration of technology supports collaborative learning. Please let us know if you have any questions about the research or technology aspects of the project. We are excited to engage and think with you about other application areas for these kinds of digital learning tools. 

Hi Robb! I hope you are doing well! Nice to "see" you here. I am curious to learn more about how you make the CEASAR content accessible both in AR and also on tablets, etc? What are some of the biggest challenges in making the content compatible with multiple platforms, and how have you addressed those challenges so far?

I see that in some of your demos, the students are in small groups where 1-2 students have AR headsets, and the others don't. How does this work in practical terms? I am curious if the students share the headset, or if there are any structured instructions for the students about how to share information from AR with the students who aren't seeing the AR content? 

As a last point--I just love seeing how much more the students are gesturing when they are using (or have just used?) the AR glasses! The possibilities there are really profound. I am excited to see more work from you and I hope we can be in touch in the future! 

Hello Lorna, Let me jump in as some of your questions are about the technical development of the CEASAR cross-platform program as that work was done at Concord Consortium. The CEASAR program was developed in Unity that, in principle, eases cross-platform work. We found that Unity was very useful for holding the "model" of the solar system--the celestial sphere and the sun and moon that all move according the the correct date, time, and location on the Earth's surface. However, because the user interface elements are so different between the HoloLens 2 and the web implementation that, within Unity, there needs to be independent versions of the program for each platform that tap into the solar system model. During the course of the project, owning to the difficulty of obtaining the HoloLens 2, we also developed a VR version (that ran on a Quest) that advanced our knowledge of developing for a 3D "surround" model and bridging between 3D and flat devices. The difficulty of juggling the code between the several platforms required that when the HoloLens became available, we just focused on the two platforms.  In addition to the UI and keeping the visible elements in step, there is a server program that manages the communication and state between the platforms. All this made for some challenging dev work. 

Loved this chance to learn more about the CEASAR project! Really exciting work. Hoping to float a question past the team about movement/gesture. In the video clips, one student wearing the headset is gesturing/interacting with the simulation while other students (not wearing the headset) are watching without access to the visual simulation. Am I reading that correctly? If so, how do you see students collaborating in these moments? Is it perhaps the case that the partial access to the visual simulation is generative (e.g., the student wearing the headset puts into words what they are seeing in ways that might be beneficial to other members of their group)? Or does this create a chasm between members of the group that is difficult to cross? I'm curious about the conversational dynamics here because they would have big implications for how many AR headsets you would need in a classroom. Would love to hear any of your thoughts on this. 

Thanks for sharing. Looks like a great project. I'm curious about the differences between the VR and AR versions, and any possible conclusions about what kinds of collaboration they facilitate. Do you think if the VR version had everyone else in VR along with some representation of themselves (e.g. avatars) you'd see as much interaction and collaboration as with AR + tablet?

Also, did all the students get the same information? We did a cross-platform project (VR + tablets) and noticed the early versions didn't have much collaboration because the VR players could solve the puzzle on their own. We then separated which roles get which information so they depend on each other, and that prompted more communication between roles. Maybe an approach like that could increase collaboration in your application even in a VR + tablet case? For reference, this is the project I'm talking about: https://education.mit.edu/project/clevr/.

Do you think if everyone were in AR the communication would change a lot? How?

Hi Dan. Thanks for adding your questions. It seems that you have done some very related work. A complete collaborative VR version is intriguing and especially offers viable remote collaborative possibilities with avatars as you suggest. It also raises some challenges if some specific learning objectives are required, say, for a course. In that case, ways of structuring tasks and collaboratively assembling information or outcomes requires specific design and programming. (Yes, an intriguing possibility.) Going into this project, we imagined all users with AR headsets, but the reality of our test subjects--labs in intro astronomy classes--required that all students participate so we went with the cross-platform approach. This serves a very real practical purpose. We made the decision that the experience was a source of information to be drawn on, a tool--it is simply the stars, sun, and moon and methods to change day, time, and location. Beyond the initial "play" time, getting to know the affordances, the tasks and recording of outcomes were on paper which provided data on the effectiveness of the intervention.

As James has noted above, VR users often went into their own worlds, and did not communicate their observations, whereas in AR verbal communication was far more frequent and useful to the group. So, for face-to-face groupings, this seems more effective. We did consider giving users on the two platform different affordances based on strengths of each device, but it did not seem necessary, and given the constraints of the project this was not done. Yes, if this were done, I do see a differentiation of roles, but I'm not sure that would be desirable in a rather time-constrained classroom situation. 

Thanks again. 

Thank you for sharing.  I find your comment on AR requiring communication important.  It is difficult for everyone to have headsets, especially when there are a class of students.

Really enjoyed this video and seeing the students in action. In addition to looking at their gestures, I curious if you all are looking at their spatial language or use of multimodal gestures (connections between what they say and the gestures that they make).

Really great project. I see so much appeal and potential in the application of this technology, even beyond the STEM classroom. My question is about collaboration and teamwork. We know how challenging it can be to set students up to be able to collaborate constructively (sometimes someone "does all the work," others are reticent to work on a team, student internalization of individual accountability and rewards in school may undermine the idea of a team culture, or assignments do not logically lend themselves to collaboration). Do you see the use of this technology and the way assignments are designed as facilitating collaboration, or do you still have to provide other supports for teachers and students to be able to collaborate well?

Thanks for the question Patrik. I enjoyed your CPR2 project video as well! Facilitating collaboration was the explicit goal of CEASAR, even though the technologies that we were using (AR headsets, tablets) are more frequently associated with individual activities. Particularly with AR, it is not going to be feasible to provide all individual students with high-end tech, so we felt it was important to figure out how to configure the activities such that it distributed the task across devices and across people. We think we made great headway on this by creating a shared simulated night sky environment with multiple entry points. This created a space where students could see the effects of other students' actions, but often still necessitated verbal communication to make the interpretations of those actions clear. The other key to engendering collaboration was task design. When students were simply exploring the night sky using the various tech, there was unsurprisingly not much evidence of collaboration. But given the parameters of the "lost at sea" navigation task they were given, working with each other and communicating across the different perspectives was really essential. To be sure, not all groups collaborated successfully, but we feel like we're learning a lot about what was happening when collaboration was achieved.  

This is incredibly exciting work. Well done to you all.