Confirmation Bias in the Age of AI and Educational Approaches for Moral Metacognition

Purpose : This study explores ways to foster moral metacognition in the AI era, focusing on deeper learning in the 2022 revised curriculum. As AI intensifies confirmation bias and selective perception, learners need the ability to reflect, regulate, and monitor their thinking in moral problem-solving. Method : A literature review and theoretical analysis are conducted, centering on Narvaez’s framework of moral metacognition. Flavell’s model and Brown’s principle are used as supporting perspectives, linked with the 2022 curriculum to derive implications. Results : Through tasks that involve self-monitoring, regulation, and reflection, learners develop both cognitive and emotional competencies. Moral self-reflection, in particular, enables them to recognize biases and cultivate initiative for moral agency. Moreover, deeper learning—life-related, integrative, and reflective—provides a strong foundation for fostering moral metacognition. Conclusion : Moral metacognition, when systematically integrated into elementary through high school moral education, extends beyond knowledge acquisition to nurture problem-solving, creativity, lifelong learning, and sustainability-based competencies. The study concludes that deeper learning should serve as a fundamental educational context for the practical manifestation of moral metacognition in future curriculum design and moral education courses.