We created an environment called Ubiquitous Collaboration Support (UbiCoS) encompassing three different digital platforms. The aim of the system is to support collaboration and help-giving (the way students provide help to each other), and improve/increase students’ collaborative interactions across different contexts. We also designed the system to better understand whether students’ help-giving and/or learning benefits from the addition of multiple contexts. 

Our system is novel in its multi-platform approach. Computer-Supported Collaborative Learning (CSCL) environments are often designed to support student collaboration within a single digital platform. However, with the growth of technology in classrooms and the additional variable of a global pandemic, students often find themselves working in multiple contexts (i.e., a student might work face-to-face with a peer on one task and then move to engaging in an online discussion for homework). 

In our research with this system, we aimed to implement ways to support students’ collaboration that adapted to their motivational  needs. Motivation is a key component contributing to students’ collaboration and learning. We used several motivational factors to provide support to the students at the beginning of each collaborative activity. The support was designed in the form of prompts to facilitate collaboration and badges as a way to provide positive feedback.

In addition to the collaborative support, we also designed and implemented a tool which students used to report their motivation within and between class activities. The tool was designed using the concept Personas [Cooper, 1999] and is described in more detail below. 

This system was designed to be paired with an 8th grade math curriculum that promotes productive interactions in both face-to-face and digital environments following Modeling Pedagogy. For more information on this curriculum, see the curriculum page of our website. 


To complement the curriculum and the small-group and whole-class discussions, students interact in the following three digital contexts.


The first was a digital textbook called ModelBook that we developed to contain the curricular materials (e.g., question prompts, homework assignments), allow students to log their work (e.g., students could upload photos of their whiteboards), and enable students to interact digitally with their classmates (via a chat system). 

In ModelBook, students can see two windows: instructions text on the left, and one of two  interactive tools on the right as follows. 

During gallery discussions, students would get support based on their reported motivational values. Relevant sentence starters would be displayed and students could use them to formulate their responses.

Figure 1: Gallery Discussion Interface

Khan Academy 

The other collaborative platform used is Khan Academy. While most known for its instructional videos, Khan Academy also allows asynchronous collaboration with geographically distributed learners in a question and answer environment under each video ( In this platform students watch short videos and participate in Q&A forums. By answering questions, students are encouraged to articulate their understanding and engage in help-giving behavior with a broader group than just their immediate classroom. Figure 2 shows questions and answers posted by the students in the Khan Academy interface. 

Figure 2: Student discussion in Khan Academy

Cobi Virtual Teachable Agent

The third digital context was a Virtual Teachable Agent adapted from Lubold, et al. (2019). In this system, students work individually with a virtual agent on a desktop/laptop to help it solve mathematics problems. Students follow a provided worked example and explain each step to the agent. The agent responds in spoken dialogue and was designed to respond to different answers that a student might provide. For example, if the student provides the correct answer but not an explanation, the agent prompts the students to elaborate. 

The digital agent provides yet another type of context to practice collaborating and help-giving. Through these interactions, students practice help-giving skills in a context where domain knowledge is less of a factor (because of the provided worked examples) and social anxiety is lessened since students are interacting with a digital agent rather than a peer.

Persona Tool

To help us understand student motivation in learning and to help us build better adaptive support for students, we developed a method for dynamically assessing motivation within the context of digital adaptive learning environments.

We used an interactive tool inspired by personas to dynamically assess motivation, allowing students to update their own motivation within context. We designed this interactive tool to improve upon typical assessments of motivation like self-report or behavioral assessment by external observers. Self-report questionnaires do not take contextual and dynamic aspects of motivation into account, while behavior-based assessments can leave students with a lack of agency. 

We embedded four personas [Ahmed et al, 2021] into an interactive tool in the digital textbook interface (ModelBook). Each persona included a name, a picture, and a short narrative as is typical in persona methods. Once the students reported their motivational value using a survey at the beginning of the study, we matched students with a persona, then, before each of the digital activities, asked students to modify the characteristics of their matched persona. The design of this persona tool was iteratively designed over the course of pilot study, and personas were designed through a two-stage co-design process.  The following image demonstrates the design of the persona tool on the right hand side of the digital textbook. 

Figure 3: Dynamic persona tool on the right side of Modelbook

Cooper, A. (1999). The inmates are running the asylum. Macmillan.

Lubold, N., Walker, E., Pon-Barry, H., & Ogan, A. (2019, June). Comfort with robots influences rapport
with a social, entraining teachable robot. In International Conference on Artificial Intelligence in
Education (pp. 231-243). Springer, Cham.

Ahmed, I., Clark, A., Metzger, S., Wylie, R., Bergner, Y., & Walker, E. (2021, June). Interactive Personas: Towards the Dynamic Assessment of Student Motivation within ITS. In International Conference on Artificial Intelligence in Education (pp. 43-47). Springer, Cham.