Adaptive Conversational Virtual Reality Laboratory for Chemistry Education: Effects on Conceptual Understanding, Presence, and Perceived Usefulness

Virtual chemistry laboratories support access and safety, but conventional virtual environments often provide limited adaptive instructional support during practical tasks. The Adaptive Conversational Virtual Reality Laboratory (ACVRL) was developed to integrate high-fidelity chemistry simulation, adaptive scaffolding, and conversational artificial intelligence into a unified instructional system. The present study examined the effects of adaptive and conversational mechanisms on learners’ conceptual understanding, perceived usefulness, and sense of presence in virtual education chemistry. The study uses a quasi-experimental design with stratified random assignment within each of three universities (N = 160). Data collection uses a chemistry conceptual test (pretest and post-test), the Perceived Usefulness Scales, and the Igroup Presence Questionnaire. The study uses ANCOVA and independent-samples t tests in IBM SPSS Statistics. The results show a significant effect of learning mode on conceptual understanding after control for pretest performance, with higher adjusted post-test scores in the ACVRL condition than in the CVRL condition. The results also show higher perceived usefulness and a significant difference in experienced realism within the presence domain for the ACVRL condition. These findings indicate that adaptive conversational interaction enhances both cognitive and affective learning dimensions in virtual laboratories. This transformation advances the development of scalable and pedagogically adaptive models.