Virtual Reality (VR) has emerged as a promising pedagogical tool in STEM education, offering immersive learning experiences that can enhance student understanding of complex theoretical concepts. This study investigated the effectiveness of VR in an undergraduate thermodynamics Laboratory focusing on fugacity and Henry’s Law, by comparing a face-to-face (F2F) to VR in two student cohorts.

Our research questions focused on the impact of learning modality (F2F/VR) on:
1) perceived learning gains,
2) perceived skill development, and
3) actual learning gains.

We used a retrospective Post-Pre survey to evaluate students’ perceptions and knowledge in two academic terms (F2F: Spring 2024; n = 34; VR: Spring 2025, n = 36). Survey items asked students to evaluate their comprehension of five key concepts right after the lab (Post) and when thinking retrospectively before completing the lab (Pre). Five knowledge-based questions evaluated students’ knowledge gains from the lab. Questions were identical between the two semesters.

Pre to Post repeated measures analysis revealed improvement in students’ confidence in defining what is fugacity in both semesters [F(1,68)=17.56, p<0.001, η²=0.21], suggesting effective learning regardless of modality. Laboratory equipment usage [F(1,67)=24.31, p<0.001, η²=0.27], data collection skills [F(1,68)=22.49, p<0.001, η²=0.25], and learning about fugacity practically [F(1,68)=4.58, p=0.036, η²=0.06] also improved from Pre to Post in both semesters. Finally, F2F students reported a statistically significant preference towards a F2F lab format compared to a recorded lab. On the other hand, VR students did not show such a preference, showing a significant interaction effect [F(1,68)=3.49, p<0.07, η² = 0.05].

Knowledge assessment revealed that VR students significantly outperformed F2F students on some but not all conceptual understanding items: dominant forces when fugacity coefficient exceeds one [VR: 47% correct vs F2F: 15% correct, F(1,59)=8.35, p=0.005] and residual volume at zero pressure [VR: 55% correct vs F2F: 30% correct, F(1,59)=4.81, p=0.032]. Similar performances were seen in three other knowledge-based items.

The VR cohort also differed from the F2F cohort on Problem Solving skills [favouring the VR; t(64)=2.08, p=0.04, d=0.51] and Teamwork Collaboration scores [favoring F2F; t(64)=5.64, p<0.001, d=1.38]. The other skills evaluated – knowledge application, posing questions, systems thinking, and design thinking - showed no significant differences between cohorts.

Our results suggest that for perceived and actual learning gains, as well as skill development, VR is highly comparable to a traditional F2F format in thermodynamics. The significant improvements in equipment usage and data collection skills indicate that VR successfully simulates critical laboratory experiences and thus can serve as a teaching alternative when required to provide a rapid response, as was during COVID-19. However, the variation in teamwork between different cohorts highlights the need for careful consideration of group interaction when designing VR learning modules. This comparability allows the exploration of the use of VR in subject areas that, for technical or safety-related reasons, cannot be taught in a hands-on lab. Future work will help evaluate these possibilities in other domains of knowledge and the possibility of supporting group collaboration in VR settings.

Keywords: Virtual Reality, higher STEM education, thermodynamics, skill development.