Date
Publisher
arXiv
This paper presents an expanded account of the Holistic Cognitive Development (HCD) framework for reflective and creative learning in computing education. The HCD framework integrates design thinking, experiential learning, and reflective practice into a unified constructivist pedagogy emphasizing autonomy, ownership, and scaffolding. It is applied across courses in game design (CS3247, CS4350), virtual reality (CS4240), and extended reality systems, where students engage in iterative cycles of thinking, creating, criticizing, and reflecting. The paper also examines how AI-augmented systems such as iReflect, ReflexAI, and Knowledge Graph-enhanced LLM feedback tools operationalize the HCD framework through scalable, personalized feedback. Empirical findings demonstrate improved reflective depth, feedback quality, and learner autonomy. The work advocates a balance of supportive autonomy in supervision, where students practice self-directed inquiry while guided through structured reflection and feedback.
Study at a glance
- Question: How can AI tools support reflective and creative learning within a holistic cognitive development framework?
- Setting and data: Undergraduate creative computing courses (Game Development and Virtual Reality Systems) at the National University of Singapore; sample size not stated.
- Method: Conceptual framework synthesis and review of findings from prior tool development studies and student surveys.
Why it matters
Designing interactive systems requires both technical skill and empathy, making learning experiential and multidisciplinary. Educators need scalable ways to provide rich, personalized feedback and support reflection in large creative computing courses without reducing student autonomy.
Key findings
- In creative media courses, average reflective writing scores improved significantly after students received AI-generated formative feedback, showing moderate-to-large gains in rubric-based measures.
- Prompt optimization and repeated evaluation increased correlation between AI scores and expert ratings to values around r = 0.7 to 0.8.
- Knowledge Graph-enhanced feedback reduced variance in scoring and generated comments that students and instructors judged to be more specific and helpful than baseline GPT-4 outputs.
Summary drafted with AI from the full paper and may contain errors. See the full paper for details.
What is the application?
Who is the user?
Who age?
Why use AI?
