Immersive virtual worlds: Multi-sensory virtual environments for health and safety training

10/01/2019 ∙ by Glyn Lawson, et al. ∙ 0

Virtual environments (VEs) offer potential benefits to health and safety training: exposure to dangerous (virtual) environments; the opportunity for experiential learning; and a high level of control over the training, in that aspects can be repeated or reviewed based on the trainee's performance. However, VEs are typically presented as audiovisual (AV) systems, whereas engagement of other senses could increase the immersion in the virtual experience. Moreover, other senses play a key role in certain health and safety contexts, for example the feel of heat and smell in a fire or smell in a fuel leak. A multisensory (MS) VE was developed, which provided simulated heat and smell in accordance with events in a virtual world. As users approached a virtual fire, they felt heat from three 2 kW heaters and smelled smoke from a scent diffuser. Behaviours in the MS VE demonstrated higher validity than those in a comparable AV VE, which ratings and verbatim responses indicated was down to a greater belief that participants were in a real fire. However, a study of the effectiveness of the MS VE as a training tool demonstrated that it did not offer benefits over AV as measured by a written knowledge test and subjective ratings of engagement, attitude towards health and safety and desire to repeat. However, the study found further evidence for the use of AV VEs in health and safety training, particularly as the subjective ratings were generally better than for PowerPoint based training. Despite the lack of evidence for MS simulation on traditional measures of training, the different attitudes and experiences of users suggest that it may have value as a system for changing trainees' attitudes towards their personal safety and awareness. This view was supported by feedback from industrial partners.



There are no comments yet.


page 1

page 31

page 33

page 36

page 37

page 38

page 39

page 40

This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.