AI for active learners: Applying insights from the learning sciences and human-centered AI to educational contexts

Applying AI in education requires balancing the potential beneficial learning outcomes with the risk of undermining the learning process. To balance these benefits and risks, my research draws on insights from the learning sciences and human-centered AI to develop applications for educational contexts that promote effective learning environments. The shift in the learning sciences towards a pedagogy in which students more actively engage in interaction with knowledge, their peers, and their teachers has also increased the focus on student motivation, ultimately leading to deeper conceptual understanding. Human-centered AI is placing human control more centrally, making it “AI-in-the-loop” rather than “human-in-the-loop”. As control is closely linked to transparency, we employ various explainable AI paradigms in our research. Building on these overarching principles, we aim to develop two educational AI systems that offer personalized experiences, tailored to individual needs.