MetaCoXAI Framework: Linking XAI, Computational Thinking, and Metacognition for Learning

Despite growing interest in artificial intelligence in education, there remains a notable research gap concerning how AI, specifically explainable artificial intelligence (XAI), can potentially support and enhance students’ metacognitive abilities and computational thinking (CT). To bridge this gap, we propose MetaCoXAI, a novel conceptual framework that integrates XAI with computational thinking instruction, offering actionable strategies for learners to develop a deeper understanding of AI processes. Grounded in interdisciplinary theoretical insights from learning technologies, human-computer interaction, machine learning, and XAI, MetaCoXAI explicitly targets the four fundamental components of computational thinking: abstraction, decomposition, algorithm design, and debugging. The framework illustrates how XAI facilitates these CT processes, thereby positively influencing learners’ metacognitive skills. To demonstrate the practical utility and application of our proposed framework, we provide research directions highlighting how learners can utilize XAI-supported computational thinking to enhance both problem-solving proficiency and AI competency.