Acquiring motor skills, such as dance, is considered a difficult task, as novice learners frequently encounter challenges in promptly comprehending and adjusting their movements in learning contexts. In this study, we present a virtual environment training system that incorporates haptic feedback and generative AI to support novices’ learning of body movements in dance. The system guides the user through haptic feedback in VR, generated from dance sequence data captured from professional choreographers. User studies with novice participants were conducted to 1) establish effective feedback for guidance and 2) evaluate the user’s training experiences using the system. We also conducted comprehensive interviews with participants and experts to confirm the design strategies and seek insights regarding the perception and utilization of virtual environments, along with haptic and visual feedback in dance learning processes. Based on the outcomes, we established the design guidelines for virtual environment training systems that facilitate learning bodily movement regulation, including in dance education.

Cybersickness is a major obstacle to immersive VR experiences, and various strategies have been proposed to overcome it. In particular, providing artificial sensory stimuli is considered a prominent approach, which improves presence and real-virtual congruence, thereby helping the situation. In this study, we evaluated the effectiveness of artificial sensory stimuli in inducing virtual walking sensations in mitigating cybersickness. We conducted two experiments involving a total of 80 participants: (1) a static user experiment ($\mathrm{N}=40$), in which participants used controller-based steering without physical movement, and (2) a dynamic user experiment ($\mathrm{N}=40$), in which participants walked on an omnidirectional treadmill, enabling virtual movement similar to real walking. In both experiments, participants were divided into two groups based on the presence or absence of vibrotactile feedback. We collected heart rate and subjective discomfort data during exposure to a VR environment and then evaluated cybersickness and presence. Across both experiments, the results indicated that virtual walking sensations induced through vibrotactile feedback significantly reduced cybersickness without negatively affecting the perceived task load or presence. Additionally, in the dynamic user locomotion experiment, a greater reduction in cybersickness was observed when vibrotactile feedback was present. These findings suggest that even if the real-virtual sensations are synchronized, vibrotactile feedback may induce cognitive distraction that further mitigates cybersickness.