This study investigated the effects of gender and gender pairing on behaviors and social interaction during collaborative learning in a blended learning environment. Three types of pairing (all-female, all-male, and mixed) were considered in an empirical study with 44 undergraduate students. The results revealed that, first, female students were more likely to perform well in listening, talking, and integrating during class, and in the quality of posts after class, whereas male students tended to engage more in irrelevant behavior in collaboration. Second, same-gender groups were more active in collaborative discussion, but all-male groups performed worst in quality of posts after class. Third, both male and female students were more active in same-gender groups than in mixed groups. Finally, same-gender groups tended to have greater interpersonal connectedness than mixed-gender groups. The paper ends with a list of practical implications for facilitating effective collaborative learning in co-educational college classrooms.

Giving feedback is key to online collaborative discussion, but the impact of feedback on role-scripts in online discussion remains unclear. To address this research gap, a mixed factorial experiment was conducted to investigate the potential effects and interactions of feedback type (elaborated vs. motivational feedback) and scripted roles (yes vs. no) on behavioral, cognitive, and emotional engagement. Fifty participants enrolled in an undergraduate blended course were randomly assigned to a role-script group (n = 25) and a role-free group (n = 25). Both groups received elaborated and motivational feedback sequentially. Mixed ANOVA results showed that (a) scripted roles significantly facilitated behavioral engagement; (b) elaborated feedback intrigued more cognitive engagement compared to motivational feedback; (c) the role-scripting strategy and feedback type had an interaction effect on behavioral engagement; and (d) there was a significant difference in cognitive engagement among various roles. The epistemic network analysis results revealed evolving cognitive engagement patterns associated with distinct roles when exposed to two distinct types of feedback. The implications of these findings for the design and implementation of online discussions are discussed.

While virtual reality (VR) has great educational potential, current implementations predominantly emphasize audiovisual simulations, neglecting olfactory and haptic stimuli and thereby compromising the physical fidelity of VR learning environments. Meanwhile, conflicting findings on the impact of immersion level on learning outcomes, coupled with limited research exploring the interaction between physical fidelity and immersion level, have impeded the widespread adoption of VR in education. Therefore, this study aimed to examine the effects of physical fidelity (multisensory stimuli vs. no multisensory stimuli) and immersion level (immersive VR vs. desktop VR) on learning experiences (cognitive load, learning motivation, and quality of experience) and outcomes (procedural, factual, and conceptual knowledge) in VR-based fire-safety education. A two-by-two factorial experiment was conducted with 120 Chinese university students. The results showed that multisensory stimuli significantly reduced the intrinsic cognitive load of learners. Immersive VR surpassed desktop VR for bolstering learning motivation, particularly in terms of attention, relevance, and satisfaction. Physical fidelity and immersion level have significant interaction effects on long-term retention of conceptual knowledge and satisfaction. Incorporating multisensory stimuli into immersive VR enhances the quality of experience. Our findings have significant implications and value for the application of multisensory stimuli and VR technology in education.