A key challenge in education is designing course content that keeps students engaged. To do this, we need ways to design learning experiences so that they can be tailored to the needs of each student and optimized with user data. In this talk, I will discuss lessons learned from the development of Refraction, a free online video game for learning fractions that has attracted one million players. I will show how we can automatically generate progressions of practice problems for teaching a procedural skill through analysis of that procedure. Furthermore, I will present a programming-by-demonstration framework capable of synthesizing programs that reproduce 28 different systematic errors that were demonstrated by real students across 9 procedures in K-12 mathematics. I will also demonstrate data-driven optimization of engagement in two online games, through a multivariate test with 27,000 players that examined the impact of secondary game objectives on player behavior. Future directions include designing games to teach complex topics such as reading comprehension, foreign language, and critical thinking, restructuring content to match learner skills and strategies, and discovering optimal learning pathways.
Erik Andersen is a Ph.D. candidate in Computer Science and Engineering at the University of Washington. He is a member of the Center for Game Science and a developer at Engaged Learning. Andersen’s research aims to make education more engaging, automated, and adaptive. He develops fully-automated tools that synthesize pragmatic, principled learning pathways for target concepts. Andersen is a co-creator of two educational video games, Refraction and Treefrog Treasure, which have won major awardsand attracted millions of players. He has also worked on large-scale studies of engagement in video games, visual data mining, character animation and path planning.