Gallery Walk Station Map (PDF)
Return to the CIRCLS’23 Agenda

1. Generative AI Methods for Education

Andrew Lan
In this poster, we will provide an overview of numerous works from the PI’s group on developing/using (generative) language models for educational applications, including automated scoring, question/feedback generation, and next generation item response models for open-ended questions.

2. Characterizing productive engagement during online teamwork sessions

Alejandra Magana
Effectively facilitating teamwork experiences, particularly in the context of large-size courses, is difficult to implement. Research suggests that instructional strategies, such as collaborative and cooperative pedagogy and introducing team training tools for students, may be effective. However, scant research reporting on the implementation of teamwork pedagogies in large classes has also reported a lack of success in their implementations. This study integrated teamwork pedagogy to facilitate a semester-long project in the context of a large-size class with 118 students organized into 26 teams. Findings describe the different ways team members engaged in team dynamics processes during two working sessions at different points in the semester.

3. Precision-guided Feedback for Deliberate Practice Using Multimodal Analytics in a Multi-user Virtual Reality Simulation

Vitaliy Popov
Existing healthcare simulation training, including VR and manikin–based learning, requires constant and real-time human observation. This limitation results in learners receiving feedback that is of variable quality – often generalized, inconsistent, and highly dependent on simulation instructors. With feedback being essential for learning and development, this limits the effectiveness and scalability of simulation training. To fill this gap, our project aims to develop and evaluate a novel debriefing system that aims to capture and visualize multimodal data streams that evaluate learners’ cognitive (e.g., clinical decision-making) and behavioral (e.g., situational awareness, communication) processes to provide data-informed feedback focused on improving team-based care of patients who suffer sudden medical emergencies. Through this new multimodal debriefing system, instructors will be able to provide new insights and personalized feedback to clinicians during post-simulation debriefing sessions to allow for more meaningful reflection, targeted intervention, and rapid development of these complex skills.”

4. NeuroVivid: A BCI Maker Experience for Neurodivergent Youth

Ibrahim Dahlstrom-Hakki
The NeuroVivid project is developing an innovative maker experience aimed at middle-school aged neurodiverse (neurodivergent and neurotypical) students. Participants build their own simple Electroencephalogram (EEG) headsets to understand and interact with their brain activity. The project team is co-designing all content with neurodivergent youth. Activities include assembly of simple headsets, block coding activities, demos and games, and an introduction to neuroscience and cognition. In this demo, we will share information about this project and demo some of the technology being developed.

5. Evaluating learning of motion graphs with a LiDAR-based smartphone application

Rebecca Vieyra
Engage with a new, free LiDAR-based smartphone game as you use embodiment to build an understanding of mathematics and physics with motion graphs. Play with a borrowed smartphone, and learn about our most recent research studies at our poster!

Data modeling and graphing skill sets are foundational to science learning and careers, yet students regularly struggle to master these basic competencies. Further, although educational researchers have uncovered numerous approaches to support sense-making with mathematical models of motion, teachers sometimes struggle to enact them due to a variety of reasons, including limited time and materials for lab-based teaching opportunities and a lack of awareness of student learning difficulties. In this paper, we introduce a free smartphone application that uses LiDAR data to support motion-based physics learning with an emphasis on graphing and mathematical modeling. We tested the embodied technology, called LiDAR Motion, with 106 students in a non-major, undergraduate physics classroom at a mid-sized, private university on the U.S. East Coast. In identical learning assessments issued both before and after the study, students working with LiDAR Motion improved their scores by a more significant margin than those using standard issue sonic rangers. Further, per a voluntary survey, students who used both technologies expressed a preference for LiDAR Motion. This mobile application holds potential for improving student learning in the classroom, at home, and in alternative learning environments.

6. Converse to learn: Using conversational AI to support children’s learning

Ying Xu
In this Gallery Walk presentation, I will showcase a series of studies focused on the use of conversational AI to enhance children’s learning across various media platforms, including television shows and e-books. Specifically, for television, we have partnered with PBS KIDS to enable children to interact contingently with a beloved character from a popular science animation show. In the realm of e-books, we have collabroated with Sesame Workshop to incorporate a popular Sesame character that serve as children’s reading companion, engaging children in dialogues as they read. A poster will be utilized to present the findings related to the feasibility and effectiveness of integrating conversational AI into these media platforms. Furthermore, we will provide a demo allowing the audience to experience first-hand the AI-based media developed within our project.

7. Application of emerging technologies perceived by and used by teenagers in international collaborations

Eric Hamilton
Application of emerging technologies, including AI, as it is perceived by and used by teenagers in international collaborations, including their perceptions and reflections about the future of generative AI. Based on our work with middle and secondary school students in such international collaborations.

8. CIRCLS Community Report: Partnerships for Change: Transforming Research on Emergent Learning Technologies

Wendy Martin, Shari Gardner, Ligia Esperanza Gómez, Lin Lin-Lipsmeyer, Robb Lindgren, Janice Mak
This session is an opportunity for community members to learn more about the Community Report from its editors and authors.

9. Early identification and support of low-performing students in a STEM course

Autar Kaw
One of the known challenges with the flipped format is the pre-class expectations for the students. As part of a NSF grant, lessons have been developed and implemented on an adaptive learning platform (ALP) to improve the pre-class experience and performance in a flipped course at the University of South Florida. One of the other benefits of using adaptive learning platform is the significant amount of data it collects about student behavior and engagement with the course material. We used this data to identify and support potentially lower-performing students (C or lower students) during the first few weeks of the semester instead of waiting until the sixth week when the first unit test is graded. By the sixth week of the semester, it may be too late for the student to recover from low performance on the test due to feeling discouraged or unable to make significant adjustments in their approach to academics. We present results about how we identified the students and their improvement in learning as measured by course grades.

10. Using Neural Networks and Teacher Dashboards to Provide Automated Feedback on Elementary Mathematics and Reading Instruction

Peter Youngs
This study presents the results of two different deep neural networks models used to classify activities in video-recorded lessons of elementary mathematics and reading instruction. For many activities, we have achieved classification rates (F-1 scores) of 0.5 or higher. This study also reports data from a user study in which teachers were interviewed about their perceptions/experiences with a dashboard designed for sharing neural network data,

11. May the force be with you: Haptic feedback for grounding students’ learning with visualizations

Matthew Lira
A physicist knows that a force exists when they can feel it. But in the biomolecular sciences, sub-microscopic intermolecular forces exist that remain beyond our direct senses. We aim to introduce innovations in haptic and visualization technologies that we are implementing in biology education and learning sciences research. Our goals for the demo include fostering conversation with like-minded scholars about the rationale for our guiding design principles, determining how the technologies might facilitate theory testing and building, and indentifying new domains, contexts, or populations who might benefit from haptic feedback coupled to visualizations.

12. Collaborative Interactive Data Science Academy

David Lockett
With the goal to stimulate curiosity in the cross-cutting field of data science and emerging technologies, Meharry Medical College proposed a discovery-based summer experience that implements virtual reality, augmented reality, and mixed reality control of robotic systems using NASA geospatial and extra-terrestrial big data. This program will expose high school students to NASA research and data science tools; build statistical and critical thinking skills; and inspire the next generation of explorers, researchers, and data scientists. Meharry Medical College was awarded $418,448 for its proposal.

13. Hands-on Virtual and Mixed-Reality Science Labs

Kambiz Hamadani
We invite participants to test drive the hands-on science labs of the future. We have developed and tested virtual and mixed-reality science labs that offer optional tactile authenticity and the ability to integrate dynamic molecular visualizations in-line with macroscopic work carried out at a lab bench. We’ve also created a no-code system for content development that will fuel further development activities.

14. CourseMIRROR

Muhsin Menekse
An introductory demo or poster on how the CourseMIRROR learning system works.

15. Haptic and mixed reality system simulating IV needle insertion for hand-eye skills

Jin Woo Kim and Kwangtaek Kim
The goal of this project is to develop a bimodal Haptic-mixed reality (HMR) system that simulates IV needle insertion with variable conditions. The system creates a realistic learning environment for students to master insertion tactile skills using two hands. The project also investigates whether variability in practice improves needle insertion skills. To achieve these goals, we developed a prototype of bimanual haptic simulation using two complimentary haptic devices, a haptic glove and a stylus haptic device, integrated with MR to simulate virtual patients and IV needle insertion with variable training conditions. We conducted a learning impact study with 59 nursing students randomly assigned to experience training sessions in one of the three modes (HMR-static, HMR-variable, manikin arm). Pre/post tests measuring the accuracy and success rate of nursing insertion were conducted to measure learners’ IV insertion skills. The results were compared across the three training conditions. Post-training surveys were collected in terms of the realism and the user experience (usability) used for continuously improving the HMR system.

16. New Dimensions of American Sign Language (ASL) Learning: Implementing and Testing Signing Avatars and Immersive Learning (SAIL 2)

Lorna Quandt
We will share progress on our VR-learning game, in which participants learn introductory ASL signs from a motion-capture-powered signing avatar teacher. Users can produce the signs after seeing them from the teacher, and then the system recognizes whether or not they were signed correctly. The system includes hand recognition and interactive turn-taking with the teacher avatar. We will welcome comments and feedback from the community on our work-in-progress.

17. Learnersourcing Worked-out Examples

Tianyi Li
The poster will provide a brief description of how my team and I have designed homework assignments to engage students to learn database programming concepts by creating worked-out examples.

18. Immersive Learning for Robotics Operations

Shahin Vassigh
We will share a movie clip and a poster of the virtual reality prototype environment which we have developed for robotic arm training.

19. Python Programming Education for K-12 Students with Vision Impairments

Wei Wang
We will present the lessons learned and the tools that we have used in the Python programming education to three K-12 students with vision impairments.

20. Integrating AI in K-12 STEM education

Zhen Bhai
In this poster/demo, I will showcase learning technologies developed under the support of the NSF RETTL grant, that offer novel learning experiences for basic Artificial Intelligence (AI) concepts and methods among K-12 students and teachers. The technologies combine data visualization, embodied and collaborative learning scaffoldings to help novice learners make sense of the underlying mechanism of Machine Learning (ML) such as multi-dimensional data, clustering, and classification, as well as use the technologies as new knowledge discovery tolls for K-12 science subjects (e.g., ecology). Audience of the conference will be able to try out our learning technologies (e.g., emoji-based data clustering), and engage in conversation for future technologies to increase interests and self-efficacy of AI literacy for 21st century learners and educators, building a community to promote digital equity and inclusion, and effective integration of AI literacy in K-12 education and informal learning.

21. Promoting Cross-Disciplinary Dialogue Between Experts in Argumentation and Innovative Technologies

Alina Reznitskay
The purpose of this poster is to share the results of an open-ended online survey designed to address ideas, barriers, and opportunities for cross-disciplinary collaboration among scholars in the fields of argumentation and technology. The survey was administered following a two-day online workshop designed to promote cross-disciplinary collaboration. Eighteen participants completed the survey. We conducted content analysis to reveal common themes in survey responses.

22. AR-Classroom: Augmented Reality for Learning 3D Spatial Transformations and Their Mathematical Representations

Wei Yan
Project AR-Classroom aims to enhance undergraduate students learning spatial transformations and their mathematical representations. Understanding closely allied spatial and mathematical concepts significantly contributes to STEM learning in fields of computer graphics, computer-aided design, computer vision, robotics, and many more. The technology and learning innovations of this research include novel AR features and their implications for learning. In AR-Classroom, a student can hold and manipulate a 3D physical model (a LEGO space shuttle as an example) while simultaneously interacting with AR visualization of 3D rotations. Usability tests have been conducted for AR-Classroom leading to promising results and recommendations for improvements. The project contributes to advancing our knowledge in (1) the role of interplay between physical and virtual manipulatives to engage students in embodied learning and (2) the features of AR to make difficult, invisible concepts visible for supporting an intuitive and formal understanding of spatial reasoning and mathematical formulation.

23. QubitVR: Advancing Quantum Education in Virtual Reality

Ryan McMahan
This presentation will demonstrate QubitVR, a virtual reality (VR) application developed for learning key quantum information science (QIS) concepts. QubitVR consists of a series of educational VR modules that employ 3D Bloch sphere representations of quantum bits (qubits) to introduce concepts such as superposition, measurement, gate operations, and entanglement.

The goal of our three-year RETTL project is to create and compare intelligent tutoring versions of QubitVR that can address QIS misconceptions for individual learners. This presentation demonstrates the basic version of QubitVR, which is being used to identify QIS misconceptions and collect data for future intelligent tutoring versions based on machine learning models.

24. Academical: A Dynamic Narrative Learning Environment for Innovating Online Ethics Training

Samuel Shields
Ethical dilemmas arise in all aspects of STEM research. To help mitigate the problems associated with unethical research behavior, universities and granting agencies require training in responsible conduct of research (RCR). However, while there are a variety of approaches to teaching ethical behavior, ethics training instruction is often reduced to simplified online tutorials. Ethical oversight agencies and training experts have therefore emphasized the need for additional RCR learning tools that are more engaging, scalable, and offer a hands-on approach to exploring real world problems. To address this need, we developed Academical, an online choice-based interactive narrative game for teaching RCR through role-play. Academical applies the Self-Determination Theory (SDT) of motivation through an innovative combination of four state-of-the-art AI-based narrative systems: 1) a social simulation system, 2) a pedagogy manager, 3) a drama manager, and 4) a dynamic content selection architecture (CSA) that dynamically selects texts and options based on author goals and state maintained by the first three systems. Compared to existing approaches to dynamic CSA, this approach enables interactive stories with more flexible delivery of choice-based story content. As a result, Academical can better model complex ethical contexts while facilitating learners’ feelings of autonomy, relatedness, and competence.

25. Algorithmic literacy: what, so what, now what

Yianna Vovides
This poster lays out a curriculum for enhancing instructors’ algorithmic literacy. The curriculum is grounded in inquiry-based learning methods and aims to engage instructors in deepening their understanding of how algorithms make decisions and the implication of such decisions in education and the workplace.

26. Using Collaborative Agent-Based Computer Modeling to Explore Tri-Trophic Cascades with Elementary School Science Students

Anthony Petrosino
This poster investigates an in-service teacher and her student’s abilities to utilize, implement, and enact a participatory agent-based modeling program, developed as part of the group-based cloud computing (GbCC) for STEM Education Project funded by the National Science Foundation. In this first cycle of design-based implementation research with an in-service teacher and her 300 students, we examine student participatory learning and teacher experience. By implementing models with teachers, we intend to 1) improve iteratively the GbCC learning technologies and 2) develop more informed and aligned pedagogies for teaching in socially mediated and generative learning environments.

27. Exploring A Metaverse System for Social Learning in Collaborative Augmented Reality with Virtual and Human Pedagogical Agents

Marjorie Zielke
We will present results from our NSF study on developing an AR system that enables social learning with virtual and live teachers, peers and patients.

28. Investigating the Role of Interest in Middle Grade Science with a Multimodal Affect-Sensitive Learning Environment

Jonathan Rowe
This project focuses on the design, development, and investigation of a multimodal affect-sensitive learning environment for investigating and triggering student interest in middle school science.

29. Fellows Tier of the AI CIRCLS Mock Review Panels presentation

Janice Mak, Matt Matilla, Armanda Lewis, and Richard Kalunga

In response to a need to build capacity and representation in education research involving emerging technology, we present a model through which groups of practitioners and early career researchers acquired firsthand experience with reading, reviewing, and evaluating grants aligned with National Science Foundation’s review criteria. This model of holding a mock review panel is a promising approach to equipping diverse groups of potential grant applicants to engage in dialogue to calibrate around defining what constitutes a high quality proposal based primarily on whether or not the sample proposals exhibit intellectual merit and broader impacts. The elements needed to implement a successful mock review process and suggested protocol are outlined in this poster and potential spaces to apply this mock review process. Through focused efforts to demystify the grant application process for early career researchers and practitioners, it is possible to expand the diversity of research teams and the research done in the field.

30. Exploring AI hardware applications in experiential learning environments

Andrea Ramirez-Salgado
A matter of national security for the US is supporting the development of the semiconductor and microchip workforce to address the chip shortage and its negative consequences. Consequently, higher education institutions are establishing programs to train the necessary workforce and address the CHIPS Act passed by US legislature in 2022. However, students pursuing engineering degrees are more inclined to pursue software design rather than focus on hardware, primarily due to the perceived complexities and the time-consuming nature of electronic manufacturing processes. To address this issue, we have undertaken a project funded by the NSF IUSE program to explore the usability and feasibility of AI-focused hardware activities specifically for first-year undergraduate students across various engineering disciplines. These activities utilize an inexpensive custom hardware board equipped with motion, weather, ultrasonic, and light sensors to collect environmental data, which is then processed using machine learning algorithms to make predictions about different conditions. We aim to demystify AI, cultivate situational interest in computer hardware design and support hardware engineering identity development. This poster presentation will provide an overview of the curriculum’s scope, implementation, and preliminary results.

31. Participatory Design For Human Well-Being

Sanaz Ahmadzadeh Siyahrood
AI systems are impacting society in profound ways. They have the potential to bring significant benefits, but they can also pose risks – including discrimination, human rights abuse, psychological harm, privacy violations, and employment disruptions. Because of this, it is crucial that AI systems are designed in a responsible and ethical manner. At present, however, there is a lack of tested, effective practice recommendations for ensuring that AI systems are designed in this way. Without such recommendations, AI developers will, at best, struggle to implement processes that ensure responsible and ethical design; at worst, the public could be misled by promises of responsible design. Therefore, the goal of our project is to shape the design of all technologies that AI-ALOE will develop during its period of funding based on an ethical analysis that follows the “IEEE Recommended Practice for Assessing the Impact of Autonomous and Intelligent Systems on Human Well-Being’ (2020).” The IEEE Standard requires the engagement of “stakeholders” to determine the possible effects of AI technologies on their well-being. The participatory design team uses three different methods to engage the stakeholders in the design process as follows: (1) Focus group discussions, (2) Individual interviews and (3) Surveys.

32. Supporting high school students’ skill development through co-design of a hybrid documentation toolkit for self-directed learning experiences

Marti Louw, Talia Stol, and Daragh Byrne
This poster summarizes a case study of the co-design and evaluation of learning support tools resulting from a design-based research engagement with facilitators of a high school self-directed learning (SDL) experience. We developed an integrated set of physical and digital documentation tools – the SDL documentation kit – to support student reflection on their learning and articulation of their skills while building an evidence base that they could use to apply for competency-based micro-credentials. The documentation kit consisted of a QR-linked manipulable poster, an evidence upload form, a structured cloud folder, and a timeline visualization function. The kit was made available to over 60 students, with 14 students and two educators taking part in an exploratory study of tool deployment. Mixed methods findings suggest that the kit was successful in supporting key documentation-related learning processes of reflection and articulation, and that students linked the documentation kit affordances of organization, tracking progress, and making their thinking visible to their confidence in completing their SDL experience and applying for selected micro-credentials. This work has implications for the design of documentation technology for learning, including the importance of centering pedagogical values and practices.

33. Constraints and Affordances of Mobile Learning Experience Platforms for Children with Autism

Aaron Kline
Autism Spectrum Disorder (ASD) affects an estimated 214 million children worldwide, including 1 million children under the age of 10 in the United States. Children with ASD often experience social communication challenges, including difficulty with language, attention, facial expressions, and social interactions, and delays in access to therapy magnify these challenges greatly. To address this, we have developed two learning aid platforms, Superpower Glass and GuessWhat, that leverage mobile technologies and artificial intelligence to provide opportunities for early interventions at home for children with autism. Leveraging smartphone and wearable-based technologies allows us to provide structured interactions focusing on social and emotional awareness broadly, including to geographic regions that are particularly underserved by autism care providers, while simultaneously collecting video and other data that allows for evaluation of engagement and progress tracking. However, utilizing these technologies remotely also presents challenges to both interaction design and automated analysis of the interaction data. Design iteration based on these affordances and constraints is critical to ensuring that these platforms can provide a wide variety of learning experiences remotely for millions of children at home, and provide the feedback necessary to improve and personalize the impact of those experiences for children in the future.

34. Hui: A case of (re)creating Indigenous stories with emerging technologies

Breanne Litts and Rogelio E. Cardona-Rivera
In this poster/demo, we will share our process of (re)creating) an Indigenous story from oral storytelling tradition to physical prototype to digital prototype. The name of the experience is: “Hui,”an‘O ̄lelo Hawai‘i Hawaiian language word translated as: to band together, assemble, organize. Hui serves as an illustrative example of how to steward stories across media with culturally sustaining/revitalizing approach.

35. Measuring Students’ Attentional States During Online Physics Learning: Initial Results

Lester Loschky
Computer-Assisted Instruction (CAI) will continue to remain ubiquitous in the post COVID-19 world. A key problem for CAI is students’ attention during instruction. Compared to a teacher looking at their students during an in-person class, it is much harder to assess the attentiveness of online students. Yet, CAI is an ideal context within which to investigate these issues, because it allows every element of instruction to be controlled and measured, and it also allows detailed and comprehensive measurements of learners’ behavior during naturalistic learning activities. Here, we report on the initial results of a study of 100 students’ attentional states while studying a multimodal Physics module and their subsequent learning outcomes. To measure students’ attentional states, we combined information from a webcam, an eye tracker, an egocentric camera (showing what students looked at), and a computer mouse. We also included mind-wandering probes at regular intervals to measure self-reported mind-wandering, and we used a retrospective recall procedure to ask students about their attentional states during times they were looking away from the materials for several seconds. We use the above data sources to divide students’ attentional and cognitive states into a 2 x 2 matrix, in which the two rows are whether they are thinking about the learning materials or not, and the two columns are whether they are looking at the learning materials or not (D’Mello, 2016). This yields four quadrants for attentional and cognitive states: Q1: looking at and thinking about the learning materials; Q2: not looking at the learning materials, but thinking about them (e.g., note taking, using a calculator, or thinking hard while looking elsewhere); Q3: looking at the learning materials, but not thinking about them (mind-wandering); Q4: neither looking at the learning materials, nor thinking about them (off-task). We measured students learning with a 26 item pre-test, post-test, and 1-week retention test. We also assessed students’ self-rated engagement, and executive working memory. We report on the initial results of this study, in terms of how time spent in each of the four quadrants was associated with learning outcomes, and how this was mediated by their executive working memory capacity, and their level of engagement.

36. Understanding Student Practices in Collaborative Science Inquiry in a Scaffolded Game-Based CSCL Environment

Cindy Hmelo-Silver
The study aims to examine the design of scaffolding embedded in a collaborative inquiry-based game environment based on an understanding of group-level collaboration patterns and disciplinary learning in middle school life science. Regarding data sources, we utilized pre- and post-assessment scores, students’ written responses, individual trace log data, and survey responses concerning their satisfaction with collaboration. For analysis, we conducted descriptive statistics, paired t-tests, and K-means clustering. Subsequently, we compared learning outcomes and engagement patterns between groups. The results offer insights into an additional layer of scaffolding and trigger conditions for supporting students’ collaborative science inquiry.

37. Professional-development for Emerging Education Researchers: PEER Institute materials and activities

Scott Franklin
We have developed activities that help emerging STEM Education researchers and scholars develop research interests into actionable questions, identify promising data, methods and theories, and facilitate communication to peers, funders, and other audiences.

38. An Intelligent Assistant to Support Teachers and Students in Simulation-Based Science Learning

Shubhra Kanti Karmaker Santu
This project will develop an artificial intelligence-based conversational framework (iLab) to create dialog-based interactive laboratory experiences for middle school science students and teachers in the context of simulation-based science experiments. A key component of the framework is an intelligent conversational agent (SimPal) that will engage with teachers in a dialog to solicit their instructional goals associated with simulation experiments and store them using a computational representation. The agent will then use this representation to facilitate and mediate an interactive dialog (powered by state-of-the-art large language models) with students as they run experiments to enhance their learning experience. The agent will proactively ask students reflection questions, provide them with real-time customized feedback, track students’ progress, and then analyze their responses and report back to the teacher. Unlike existing intelligent tutoring systems and pedagogical conversational agents, the framework will work with any off-the-shelf third-party simulations in any domain and be used by any teacher or U.S. school district, a unique feature of this project. Further, teachers will work as partners in developing and deploying this technology. As such, the project is expected to make unique contributions to benefit student learning and, thereby, have a broad reach and appeal in U.S. schools.

39. Teachers’ Perceptions of AI-supported Writing in the Engineering Design Process

Roxanne Moore and Gennie Mansi
This study uses teacher interviews to understand teachers’ perceptions around using artificial intelligence (AI) technology as a part high school engineering pedagogy.

40. Exploring Challenges and Opportunities to Support Designers in Learning to Co-create with AI-based Manufacturing Design Tools

Nikolas Martelaro
AI-based design tools are proliferating in professional software to assist engineering and industrial designers in complex manufacturing and design tasks. These tools take on more agentic roles than traditional computer-aided design tools and are often portrayed as “co-creators.” Yet, working effectively with such systems requires different skills than working with complex CAD tools alone. To date, we know little about how engineering designers learn to work with AI-based design tools. In this study, we observed trained designers as they learned to work with two AI-based tools on a realistic design task. We find that designers face many challenges in learning to effectively co-create with current systems, including challenges in understanding and adjusting AI outputs and in communicating their design goals. Based on our findings, we highlight several design opportunities to support designer-AI co-creation better.

41. Using a Simulated Classroom to Reduce Educator Bias

Rhonda Christensen
Improving teaching strategies through a simulated teaching environment has been shown to improve teacher self-efficacy, teaching skills, classroom management and multicultural awareness. The current study is using the simulation program simSchool to help educators recognize possible implicit bias with the goal of recognizing, reflecting and reducing any biases that may exist. Framing effect bias was used to detect possible bias due to expectations for students who were differing in gender, ethnicity and English language learner status, but underlying characteristic and capabilities were the same. Simulation-captured data are used to understand the changes that occur as educators have the opportunity over multiple sessions to adjust their teaching strategies based on objective performance and feedback data provided by the simulated system.

42. Embodied Code: Flow-based Visual Programming in Virtual Reality

Ying Wu
The Embodied Coding Environment (ECE) is a flow-based visual programming system for creative coding in virtual reality (VR). It is designed to increase sensorimotor engagement with programming, to lower the barrier of entry for novice programmers, and to scaffold the learning of computational concepts through physical metaphor and gestural interaction. Spatial representations of code, outputs of the code, and physical gestures are co-located in a virtual 3D space. The design of this system was motivated by outcomes from an initial need finding study conducted with computer science (CS) teachers and CS education experts in San Diego County. This work revealed the importance of visualization and body movement in the process of developing code. Visual representations and embodied gestures can support planning and design thinking during the initial prototyping phases. They can also support problem solving during debugging, and learning and communication throughout the process. This hands-on demo will afford the opportunity to explore coding in 3D space through self-guided tutorials and exercises, as well as creative, student-driven projects built by means of the platform. Participants will have the option of trying out these activities firsthand in the Oculus Quest 2 or watching a video cast of other users in the environment.

43. Doing Experiments at Scale is EASI!

John Stamper
In this research we introduce the Experiments As a Service Infrastructure that allows educational research to be implemented at large scales. We have designed the system as an API that can connect with almost any existing LMS or edtech system. We envision the system running experiments with thousands of students in hundreds of conditions by taking advantage of our adaptive experimental framework. The basis of this work was fundamental in our winning the Digital Learning XPRIZE in May 2023.

44. Problem Solving Process Visualization

Magy Seif El-Nasr
A visualization system that allows learners to walk through their own solutions and solutions of others using process mining as an underlying algorithm.

45. AI at the Wheel: Navigating the Future of Autonomous Robotics in Duckietown

Matt Matilla
This poster/demo summarizes the essence of the Duckietown platform. Duckietown is a modular, state-of-the-art platform that seamlessly integrates hardware, software, simulation, datasets, and learning materials. It serves as a versatile tool for teaching, learning, and research, catering to a wide audience, including learners as young as 14. This title reflects Duckietown’s commitment to advancing knowledge in robotics and AI, particularly in the context of autonomous systems, making it a suitable platform to explore the fundamentals of computer science and automation.

46. Inq-ITS for real-time AI-based assessment and instruction

Janice Gobert and Mike Sao Pedro
We will describe Inq-ITS and Inq-Blotter, which use patented AI to do real-time assessment, scaffolding, and alerting on science practices to support both teachers and students.

47. Project CAST: Coaching At Scale for data sTorytelling

Jiaqi Gong
The burgeoning demand for data storytelling underscores the need for more than just strong analytical and programming abilities – it requires a specialized skillset to effectively communicate data narratives. Despite its critical role in the success of data-centric projects, a deficit exists in scalable educational methods and research for these skills. To bridge this gap, we must champion cross-disciplinary educational research to facilitate scalable teaching and learning in data storytelling. Our project combines expertise from various scientific and design disciplines to develop an intelligent coaching platform, using computational models and tools to enhance data storytelling education.

48. Immersive learning experiences for augmented reality-enhanced computational thinking education

Kyungbin Kwon
My team developed mixed-reality learning environments and examined the preliminary evidence of their effectiveness on students’ computational thinking (CT) learning. We also examined how multimodal data might reveal the immersive learning experiences the students had. We collaborated closely with two teachers from a local school to design engaging learning activities that leveraged the mixed-reality learning environments. This intervention involved a total of 47 students from first and second-grade classrooms over a four-day period of classroom activities. Additionally, the team extended their reach to a school situated in a rural area, broadening participation and implementing the learning environments in individual learning contexts.

The examination of learning outcomes indicated a significant improvement in CT skills and a positive attitude toward the learning environments. We are currently analyzing video recordings of students’ learning activities to understand how students interacted with the learning environments and how this interaction influenced their learning process from embodied learning perspectives.

At the conference, I will demonstrate the learning environment, discuss design principles embedded in it, and present findings related to student learning.

49. Teaching Collaborative Problem-Solving Skills Using Intelligent Tutoring Systems

Emmanuel Johnson
Students entering the modern workforce must possess more than technical abilities to be successful. They must know how to resolve conflicts and solve problems collaborative with managers and teammates. However, we find that acquiring these skills is costly, and there often don’t exist personalized solutions. Personalized learning systems such as Intelligent tutoring systems have made great strides in teaching technical skills, but we are still at the onset of training systems for interpersonal skills. In this talk, I will focus on my work in building AI systems to teach interpersonal skills that are critical to working in groups. I will discuss my work on developing intelligent tutoring techniques for teaching negotiation skills. I will begin by presenting models of negotiation that can be used by these intelligent tutoring systems to teach, as well as by agent-based systems that enable students to practice their skills. I will then highlight the metrics I’ve developed for assessing students’ negotiation abilities and show that these metrics can be used to provide personalized feedback. Next, I will demonstrate through user studies that this personalized feedback leads to improved outcomes for student negotiators. I will discuss potential applications and how I am leveraging the lessons learned from building AI systems to teach negotiation for a collection of other collaborative problem-solving skills.

50. i-Learn: Empowering Engineering Learners Using Visualizations in Mixed Reality and Machine Learning

Ivan Mutis
This research explores learning with technologies involving the influence of individual cognitive function (perceptual, attentional, and cognitive skills). There remains a gap in knowledge on how or why learners arrive at different results in the learning process. This research project bridges that divide by considering the learner’s individual characteristics as they execute problem-solving tasks while interacting with the advanced technology-enabled environment’s machine-learning (ML) and mixed-reality (MX) technology. By focusing on components of learning cognition (e.g., working memory and sustained attention) as well as mental simulation and situational awareness, this project (1) uses ML predictive modelling to explore the effects of individual differences on learners’ performance in problem-solving tasks modelling, (2) studies learners’ moments of impasse in problem-solving tasks that demand spatial and cognitive ability, and (3) develops and assesses the effectiveness of an interactive and adaptive MX platform for learning when used by students with different cognitive and attentional abilities. Research outcomes will inform the design of adaptive learning technologies and the customization of instruction in engineering education.

51. Exploring AIFORGOOD Summer Camp Curriculum to Foster Middle School Students’ Understanding of Artificial Intelligence (AI)

Keunjae Kim
This study explores the structure of a summer camp curriculum and the development of middle school students’ understanding of machine learning (ML). Pre-post tests and surveys measuring AI knowledge and attitudes toward AI were analyzed by two-way repeated measure ANOVA. Thematic analysis was used to analyze over 30 hours of video footage as well as student artifacts to examine the integration of curriculum and the development of student’s understanding. The ML curriculum consisted of an introductory module, the creation of ML-based artifacts, and a mini-project addressing community issues. The findings revealed that students’ AI knowledge and attitudes toward AI significantly improved after intervention. Additionally, their understanding on ML encompassed recognizing the importance of data quality and quantity, differentiating ML from conventional programming, and understanding prediction and classification as ML outcomes. These findings highlight the effectiveness of ML curriculum with hands-on experiences, scalable to benefit rural districts by the democratization of AI.

52. Design of XR technology for psychomotor skills learning (PSL): Integrating layers of feedback in instruction to prompt deep PSL

Hemalathaa Kasiviswanath Yuvaraja
Current research involving psychomotor skill learning uses sophisticated features of XR technologies (VR/AR), with promising implications towards accelerating learning process and enhancing learning outcome. A prototype, functioning as a model is presented that replicates the potential representation of underlying PSL mechanism for upper body hand exercise, encompassing both the mental processing and physical body movements involved in generating meaning. Many researchers have prioritized overt behavior to enhance PSL, overlooking the significance of mental processing needed to achieve deeper levels of PSL. Mental processing involves visualization, mental rehearsal and self-regulation that enables learners to completely immerse in the psychomotor task leading to more effective and efficient PSL. The organization of external sensory information accompanied by muscular action (i.e., physical body movement) is based on mental processing that is essential for PSL. The prototype developed is evaluated with demonstration of a proof-of-concept that describes, explains and measures the underlying PSL phenomenon. Issues concerning learner focus on psychomotor task, learning, environment, technology, and implications of the proposed prototype are discussed.

53. A Teacher Dashboard tool for English Language Arts teachers’ learning about collaborative argumentation

Terrence Zhang
The presentation will be a demo of the web-based teacher dashboard focusing on the UI design and functions.

54. Visual Behavior Analysis for Collaborative Learning and Play Therapy

R. Leila Barmaki
In this work, we introduce innovative visual analytical methods to evaluate and enhance the effectiveness of collaborative learning and play therapy interventions for children and students. Departing from traditional approaches, this research employs computer vision techniques to automatically assess students’ collaboration or children’s social interaction competencies based on Physical Proximity, Movement Synchrony, and Mutual Gaze. These measures are closely linked to attention and social intelligence, providing a nuanced understanding of group dynamics in educational and therapeutic settings. By integrating visual behavior analysis, this approach not only offers a more objective assessment but also has the potential to significantly improve the outcomes of collaborative learning and play therapy, ultimately fostering better social and cognitive development in children.

55. Ethical Emotion AI in Online Learning for People of Mixed Abilities

Yun Huang
Emotion AI, also known as affective computing, encompasses the recognition, interpretation, simulation, and response to human emotions and cues. Despite its potential, there has been a lack of systematic investigations into the effective design and utilization of Emotion AI for individuals of mixed abilities, especially within the context of online learning. This project investigates innovative Emotion AI solutions that foster self-reflection and enhance social interactions among two adult learner populations: those who are hearing, and those who are deaf or hard of hearing (DHH). While this study primarily focuses on designing Emotion AI for video-based learning, its contributions, including the promotion of the social inclusiveness of online learning, and the understanding of human-AI interaction among minority learner populations, have both theoretical and practical implications in broader application domains.

56. Emerging Scholars CIRCLS

Arun Balajiee and Yeonji Jung
Emerging Scholars CIRCLS is a community of early career scholars, doctoral students, and postdoctoral researchers engaged or interested in interdisciplinary computer and learning sciences research. Visit our Gallery Walk booth and engage with our community members to learn more.

57. Unpacking NSF RITEL Grant Solicitation: A User-Friendly Rubric with DEI Emphasis

Carmen Ana Ramos-Pizarro, Aleshia Hayes, Yingjie Liu, and Natalie Ottey

This project aims to demystify grant solicitation by unpacking it into a user-friendly rubric format that includes essential items researchers should address and valuable best practices recommendations. Suggested guidelines for Diversity, Equity, and Inclusion (DEI) are highlighted for consideration in all project phases.