Impact of interaction techniques on performance in virtual reality-based drone training for construction

Purpose – This study aims to examine how different input modalities in virtual reality (VR) affect operator performance during drone-assisted tasks in construction. As drones become increasingly integrated into workflows, the demand for effective operator training solutions continues to grow. While prior efforts have demonstrated the utility of VR for simulating drone operations, they often overlook how specific input modalities, both physical controller and interface design, shape the user's ability to perform spatial navigation and operational tasks. This research addresses that gap by examining the impact of input modalities on operator performance, offering insights into designing training systems that align with the cognitive and physical demands of real-world drone piloting. Design/methodology/approach – A within-subject experimental study was conducted in a simulated construction site environment. Thirty-eight participants completed hazard inspection tasks under two input modalities and their associated control interfaces: (1) joystick-based control via game controller; and (2) slider-based control using standard VR motion controllers. Performance was measured by the number of hazards identified and the time required to complete the task in both conditions. Findings – Wilcoxon signed-rank test results revealed that joystick-based control yielded a statistically significant 12.3% increase in hazard identification score and a 16% reduction in task time. These findings suggest that realism of input modality can enhance task efficiency and effectiveness within VR-based environments. Originality/value – This study contributes to the growing literature on VR-based training for drone operators by providing empirical evidence on the effect of input modalities on performance outcomes offering practical guidance for developing user-centered VR training frameworks that replicate real-world operational scenarios.