Nikol Rummel ( Ruhr-Universität Bochum, Germany) | From Learning Theory to the (Digital) Learning Sciences – An Introduction and Insights from Current Research
Abstract: After a brief introduction to the Learning Sciences, I will give an overview of different theoretical conceptualizations of learning and illustrate corresponding implications for designing digital learning environments. In the second part of the keynote, I will introduce a framework of support dimensions that can provide guidance when designing digital (and non-digital) learning environments. I will discuss insights on selected dimensions of the framework from my research.
Bio: Dr. Nikol Rummel is a Full Professor and head of the Educational Psychology Lab in the Institute of Educational Research at Ruhr-Universität Bochum, Germany. She is also an Adjunct Professor in the Human-Computer Interaction Institute at Carnegie Mellon University, Pittsburgh, USA. Nikol Rummel’s main research interest is on developing and evaluating adaptive instructional support for (collaborative) learning in computer-based settings. Another focus of her work is on developing methods for automated analyses of learning process data combining multiple data sources. Prof. Rummel has published over 40 journal articles in leading international research journals, as well as over 100 refereed book chapters and conference papers. She is elected member of the Board of Directors and past president of the International Society of the Learning Sciences (ISLS). She is Associate Editor of Instructional Science, and Editorial Board member of several international journals, such as the Journal of the Learning Sciences and Learning & Instruction.
Abstract: Educational robots are often considered by roboticists as simple robots, so simple to operate that a child can access them. In reality, designing a robot for formal education is an interdisciplinary process requiring to integrate aspect of robot design, but also HRI, pedagogy, respect of several rules that are applied in schools, knowledge of the curriculum, and so on. The presentation will go through some of these aspects taking the Thymio robot as example.
Bio: Prof. Francesco Mondada receives his M.Sc. in micro-engineering in 1991 and his Doctoral degree in 1997 at the Ecole Polytechnique Fédérale de Lausanne (EPFL). In the same period he participates in the development of the Khepera mobile robot, considered a standard in robotic bio-inspired research and mentioned in more than 6000 publications (google scholar). In parallel to his thesis he co-founds the company K-Team, where he takes the role of CEO for 5 years. In 2000 he returns at EPFL after a brief period at the California Institute of Technology. He leads the development of different robots, several being produced and distributed worldwide in the fields of research and education. In 2008, Francesco Mondada creates the robotics festival of EPFL. Under his direction, this annual meeting becomes the most important event of scientific communication at EPFL. In 2010 he takes the lead of the research in educational Robotics of the Switzerland national research in robotics. In 2013, he is appointed professor at EPFL and in 2018 he takes the direction of the Center for Learning Sciences. For his activity Francesco Mondada received numerous awards, including the prestigious Latsis University Prize in 2005 and the Credit Suisse Award for Best Teaching in 2011.
Gautam Biswas (Vanderbilt University, USA) | Betty’s Brain: A Learning by Teaching Environment for Middle School Science Classrooms
Abstract: Over several years, our research team has developed Betty’s Brain, a multi-agent environment that utilizes the learning-by-teaching paradigm to help middle school students learn science. In Betty’s Brain, students teach a virtual Teachable Agent (TA) called Betty using a visual causal map representation. Once taught, Betty, can answer questions, explain her answers, and when requested by the student take quizzes, which are a set of questions created and graded by a mentor agent named Mr. Davis. The TA’s quiz performance helps students indirectly assess their own knowledge, and it also motivates them to learn more and improve their TA’s quiz scores. Overall, the learning and teaching task is complex, open-ended, and choice-rich. Thus, learners must employ a number of cognitive and metacognitive strategies to succeed in their tasks. At the cognitive level, they need to identify, understand, and represent important information from online resources in the causal map format, and use the affordances of the system to assess Betty’s progress using the quiz results. In terms of strategies, they must decide when and how to acquire information, build and modify the causal map they are creating to teach Betty, check Betty’s progress, reflect on their own understanding of both the science knowledge and the evolving causal map structure, and seek help when necessary. Their cognitive and metacognitive activities are scaffolded through dialogue and feedback provided by Betty and Mr. Davis. This feedback aims to help students progress in their learning, teaching, and monitoring tasks.
Experimental studies run in middle school classrooms show that students learn science content and do develop some metacognitive strategies through interactions with Betty and Mr. Davis. However, a number of students fail to complete their teaching task because they lack an understanding of the cognitive and metacognitive skills they need to become successful learners. In this talk, I will discuss the Betty’s Brain system, the data mining techniques we have developed to analyze students’ learning behaviors, and then discuss how we translate this understanding to develop better conversational structures to help students develop the cognitive and metacognitive skills they need to achieve success in their learning task.
Bio: Gautam Biswas is a Cornelius Vanderbilt Professor of Engineering, a Professor of Computer Science and Engineering in the EECS Department, and a Senior Research Scientist at the Institute for Software Integrated Systems, Vanderbilt University. He conducts research in intelligent systems with primary interests in modeling and simulation, analysis of complex embedded systems, data mining, and Open-Ended Learning Environments (CBLEs) for STEM disciplines. The most notable project in this area is the Teachable Agents project, where students learn science by building causal models of scientific processes. His work on learning environments has exploited the synergy between computational thinking ideas and STEM concepts and practices to help students learn STEM topics by building simulation models. He has also developed innovative educational data mining techniques for studying students learning behaviors and linking them to metacognitive strategies. For his work in data mining for diagnosis, he received the NASA 2011 Aeronautics Research Mission Directorate Technology and Innovation Group Award for Vehicle Level Reasoning Systems.
Prof. Biswas has over 650 refereed publications, and his research has been supported by funding from ARL, NASA, NSF, DARPA, and the US Department of Education. He is an associate editor of Metacognition and Learning and the IEEE Transactions on Learning Technologies. He is a fellow of the IEEE and the Prognostics and Health Management Society, and member of the AAAI, AAAS, ACM, AIED, EDM, ISLS, LAK, and the Sigma Xi Research Societies.
Abstract: Modeling and predicting students’ knowledge and behavior is an important task in an adaptive computer-based learning environment. The teaching decisions in such a system are based on the predictions of the so-called student model. A lot of work has focused on constructing models that are able to accurately represent and predict the knowledge of the student. In this talk, I will first give an overview of the most popular approaches to student modeling. I will present two widely used techniques, which are based on probabilistic models and item response theory. I will then introduce more complex models with higher representational power. Finally, I will demonstrate, how general features can be integrated into a student model.
Bio: Tanja Käser is a postdoctoral researcher at Stanford University. Before joining Stanford, she worked as a postdoctoral researcher at ETH Zurich and as a consultant for Disney Research Zurich and Dybuster AG. Tanja received her PhD in Computer Science from ETH Zurich; her thesis was distinguished with the Fritz Kutter Award of ETH Zurich. Tanja works in the field of artificial intelligence in education and is especially interested in modeling and predicting student thinking and learning to provide optimal computer-based learning environments.
Luis P. Prieto (Tallinn University, Estonia) | How (not) to fail at digital learning: Classroom orchestration edition
Abstract : About ten years ago, a group of educational technology researchers were puzzled by an apparently simple question: We have developed so many cool, new (and often, free) technologies to help people learn… how come these are only used rarely (if at all) in your average school classroom? The answer seemed to be “classroom orchestration”: not just paying attention to whether people learn, or the plain usability of the technology, but also looking at the many other factors that come into play when you use technology in a real, everyday educational setting. In this interactive talk I will review not only the concept and history of classroom orchestration research. I will also offer concrete frameworks and methods to apply this concept to develop and enhance your own learning technology design and research. Drawing from examples of failures (and some successes) in learning technology research, I will show how making use of these ideas can help you avoid the pitfalls that come from putting your learning technology in “the real (educational) world”.
Bio: Luis P. Prieto is a Senior Research Fellow at the Center of Excellence in Educational innovation of Tallinn University (Estonia). A former Marie Curie Fellow at EPFL (Switzerland), his research in the field of technology-enhanced learning (TEL) spans areas like classroom orchestration, learning design, computer-supported collaborative learning, tangible/paper interfaces for learning, technology support for teacher professional development, or learning and teaching analytics (especially, multimodal learning analytics). In his relatively short research career he has authored more than 70 peer-reviewed academic publications. He is also interested in transversal issues in doctoral education, such as Ph.D. supervision techniques, scientific communication, or Ph.D. student productivity and wellbeing (see his upcoming blog about these topics).
Abstract: Professor Gustafson will present the work done on social robots at KTH. He will describe their efforts in developing the social robot platform Furhat and methods for interpreting and generating multimodal attention and turntaking behaviours, with a special focus on multiparty interactions. Finally, he will describe the work done on mutual gaze and joint attention in human-robot collaboration on assembly tasks.
Bio: Joakim Gustafson (KTH) is a professor in speech technology and head of the department of Speech, Music and Hearing. He has been a prolific researcher and active systems developer of spoken and multimodal dialogue systems since 1993. He has an industrial background from Telia Research where he led the research work of the EU project NICE, that developed a computer game where kids could interact with animated 3D characters using a combination of speech and gestures. He currently has three research projects where social robots act as third-hand helpers in assembly, social skills coaches for autistic children and companions for the elderly with the task of detecting early signs of dementia.