<![CDATA[To address excessive cognitive load and the lack of subject-based psychomotor evaluation in chemistry education, this study evaluates ViLabS, a deep-learning and computer vision website, for improving students' cognitive and psychomotor abilities. Using an explanatory sequential mixed-methods design, a quasi-experimental study involved 40 Grade X students selected via purposive sampling. Data were collected through cognitive tests, AI system logs, video-observation rubrics, and practicality questionnaires, followed by in-depth interviews. Results revealed ViLabS is highly feasible and practical (>92%). Cognitively, the experimental group achieved a significantly higher N-Gain (0.71) than the control (0.41) (p<0.05). Autonomously monitored psychomotor accuracy also significantly outperformed classical manual demonstrations (p<0.05). Thematic analysis confirmed the AI's instant corrective feedback minimized cognitive load and fostered precise muscle memory. Despite technical constraints such as internet and lighting dependency, ViLabS shows strong efficacy in accelerating theoretical understanding and kinesthetic proficiency. Future research should expand datasets and develop offline mobile applications. Ultimately, this study contributes a proactive, multimodal AI framework that advances the paradigm of AI-assisted laboratory learning.]]>
Copyrights © 2026
