Using virtual reality to investigate physical environmental factors related to cycling in older adults : a comparison between two methodologies

Introduction: Cycling has a positive impact on physical, mental and social health, and slows the aging process. However, there is still a large potential to increase the cycling levels in Belgian older adults. In order to promote cycling for transport, safe and attractive street characteristics have previously been investigated by using cross-sectional surveys and manipulated photographs. As VR-technology is still rarely used in transportation research, the aim of this study was to develop and compare the use of two different novel VR-applications, i.e. cycling in a 3D-CAVE and cycling with a VR-headset, as experimental approaches with regard to the sense of presence, the representation of the reality, and simulator sickness. Furthermore, the moderating effects of personal characteristics and test sequence were investigated. Methods: In total, 108 older adults (>= 65 years) participated in the cross-over experiment. The participants performed two cycling tests (i.e. cycling through virtually displayed existing streets in Ghent) using both VR-applications (3D-CAVE and VR-headset) in random order. After each cycle test, participants had to complete the same questionnaire. Results: Both VR-methodologies are equally good to be used among older adults (>= 65 years) in future research, i.e. identify which characteristics in the physical environment have an impact on cycling for transport. Additionally, there are no specific requirements for particular target groups regarding the kind of VR-application. General preference was given to the test setup that was completed in second place, indicating the importance of habituation to the virtual environment. Conclusions: Both VR-applications can be used in future research. The advantage of the VR-headset in comparison to the 3D-CAVE, is that the VR-headset is more practical to use at different locations. Especially in regard to the older populations, it is more convenient to bring the test setup closer to the subjects themselves