Large Language Model Tools for Enhancing Student Learning Processes in Computing Education

Large language model (LLM) tools are increasingly being integrated into computing education, offering personalized explanations, examples, and feedback to support student learning. However, these tools raise concerns that students may use these tools to complete tasks rather than engage in deeper processes of problem solving, reflection, and comprehension. This dissertation explores how LLMs can be used as tools to foster deep learning and meaningful engagement in computing education. Through a systematic literature review, surveys, and classroom interventions, this research investigates how learners currently use LLMs, the challenges they face in productive engagement, and how strategies such as metacognitive prompting can encourage more reflective and equitable learning processes.