This work proposes Dynamic Linear Epsilon-Greedy, a novel contextual multi-armed bandit algorithm that can adaptively assign personalized content to users while enabling unbiased statistical analysis. Traditional A/B testing and reinforcement learning approaches have trade-offs between empirical investigation and maximal impact on users. Our…

We sought to identify the extent to which an innovative educational technology called ELCII (Early Language Comprehension Individualized Instruction) improved inference making for various populations of kindergarten students. Analyses examined student performance based on individual-level demographic characteristics (i.e., gender, race/ethnicity,…

Students who demonstrate mathematics difficulties in the early grades are at risk for poor educational outcomes. Fortunately, strategic early mathematics intervention programs can improve academic outcomes for students with mathematics difficulties, and instructional technology has demonstrated promise in delivering targeted and individualized…

Computer-based learning environments (CBLEs) provide unprecedented opportunities for personalized learning at scale. One such system, iSTART (Interactive Strategy Training for Active Reading and Thinking) is an adaptive, game-based tutoring system for reading comprehension. This paper describes how efforts to increase personalized learning have…

We present an integrated theoretical framework guiding the use of visual narratives in educational settings. We focus specifically on the use of static and dynamic visual narratives to teach and assess inference skills in young children and discuss evidence to support the efficacy of this approach. In doing so, first we review the basis of the…