| Abstract [eng] |
Purpose: This study presents the design and pilot evaluation of a virtual learning environment (VLE) integrated with an in-person digital escape room (DER) for learning derivatives in a mathematics course. It addresses three research questions: RQ1: How can a scaffolded VLE help students connect the derivative-as-limit concept with rule-based differentiation fluency? RQ2: How can responsible AI-assisted reflection be embedded without replacing learner reasoning? RQ3: What feasibility, engagement, perception, and formative performance patterns emerge when the VLE is used as flipped preparation for an in-person DER? Methodology: The VLE combines concept priming, first-principles derivations, procedural differentiation tasks, logarithmic differentiation, tiered hints, worked-solution reveal logic, and AI-assisted reflection prompts. A pilot study with first-year undergraduate students (N = 48) collected data on completion, session duration, hint use, AI-reflection use, formative quiz results, item-level diagnostics, and post-DER perceptions. Findings: 94% of students completed the full VLE sequence, with a median duration of 56 minutes (IQR: 47–64); 62% used at least one hint, and 68% used AI-assisted reflection. Post-DER ratings were positive for Interest/Enjoyment (M = 3.45, SD = 0.52), Perceived Competence (M = 3.18, SD = 0.58), Relatedness/Teamwork (M = 3.36, SD = 0.55), and Perceived Educational Value (M = 3.42, SD = 0.49). The formative quiz average was 4.1/5 and revealed interpretable misconceptions in first-principles derivation, chain-rule use, and logarithmic differentiation. Originality and Impact: The VLE–DER blueprint links calculus learning objectives, scaffolded room design, validator-based feedback, hint tiers, AI-supported metacognition, telemetry, and item-level diagnostics. The model offers transferable feasibility and design evidence for serious-game-based mathematics learning. |
