Nur Kholis, Vando Gusti Al Hakim, Yuk Mui Elly Heung, Akbar Satrio Gumelar, Husni Mubarok
Introductory programming is a foundational course in engineering education, yet many students struggle to connect coding exercises with underlying theory. Standard lab instructions often lead to superficial engagement, with students copying code without deeper conceptual understanding. Socratic questioning can foster critical thinking and self-explanation, but is rarely applied consistently due to constraints in teacher availability and dialogue skill. This study introduces the Socratic Butler, a dual-role AI chatbot that guides students through question-based dialogue while providing instructors with insights into student progress. A quasi-experimental design was implemented involving 95 electrical engineering students divided into three groups: Socratic Butler, conventional chatbot (ChatGPT), and no-chatbot control. Students using the Socratic Butler showed significantly higher learning performance and metacognitive awareness than the other groups. They reported greater cognitive load, due to the reflective effort required. These findings support dual-role AI tools to foster deeper conceptual engagement and provide scalable engineering education support. © 2026 SEFI.
Department of Electrical Engineering Education, Universitas Negeri Yogyakarta, Depok, Indonesia; Department of Mathematics and Information Technology, The Education University of Hong Kong, Hong Kong; Educational Technology Study Program, Universitas Negeri Surabaya, Surabaya, Indonesia
