Evaluating the value proposition of adopting virtual reality (VR) for construction safety training

The construction industry, known for its hazardous work environments, faces significant challenges in ensuring the safety and well-being of its workforce. Traditional safety training methods, while foundational, have reached a point of diminishing returns, underscoring the urgent need for innovative training solutions. As part of this quest for innovation, Virtual Reality (VR) has emerged as a promising tool, offering immersive experiences that could potentially transform safety training. However, the effective integration of VR with traditional passive training methods remains underexplored, particularly in terms of how additional features, such as haptic feedback, should be incorporated, and how these technologies influence learners' emotions and situational interest, which are known antecedents of learning.This dissertation seeks to fill these research gaps, examining the role of VR (with and without haptic feedback) in enhancing traditional safety training within the construction industry. The study investigated how haptic feedback—a technology designed to simulate tactile sensations—could augment VR training and delved into exploring how emotions and situational interests contribute to or detract from learning outcomes.Adopting a quasi-experimental research design, the study involved 221 construction workers across the United States, encompassing a variety of roles and construction domains. Participants were exposed to interventions involving passive (video-based) and active (VR-based) training, with and without the addition of haptic feedback. This approach aimed to provide a comprehensive understanding of how these modern training methods, when adopted along with traditional passive training methods, influenced the competencies required for construction workers to make safe decisions at work.The results of the study presented a nuanced picture: VR training, even when enhanced with haptic feedback, did not significantly improve learning outcomes. More importantly, the statistical analysis revealed that the learning outcomes (i.e., hazard recognition, risk tolerance, risk perception, and decision-making behavior) were influenced by the assessment strategy adopted by the experiment, which prevents this dissertation from definitively concluding the effectiveness of VR in safety training. These findings challenge prevailing assumptions about the inherent advantages of VR in safety training and call for further exploration into how these technologies can effectively be evaluated and designed to better support the learning outcomes of construction workers.Interestingly, the study found that VR training successfully induced emotional responses among participants and significantly influenced their situational interest, suggesting that VR has the potential to engage learners on emotional and motivational levels. However, these emotional and motivational engagements did not translate into improved learning outcomes. This discrepancy highlights a complex relationship between emotional engagement, interest, and learning effectiveness within VR-based safety training environments.