Navigation Training for Persons With Visual Disability Through Multisensory Assistive Technology: Mixed Methods Experimental Study.

All 72 participants were healthy adults with normal or corrected vision, not people with actual visual impairment. Older adults and people with pregnancy, psychiatric, cardiac, seizure, serious medical conditions, medical devices, or binocular vision abnormalities were excluded, limiting generalizability to many people with visual impairment. The vision loss simulation represented common symptoms such as peripheral vision loss, reduced contrast, altered color, and glare, based mainly on severe glaucoma, not all causes of visual impairment. Only one audio feedback design and one haptic belt actuator arrangement were tested. Participants received no training on the haptic belt or VR, so unfamiliarity may have affected performance. Collisions in VR had no real consequence, so participants may have behaved differently than in real street navigation. The analysis removed outliers, including 92 (31.9%) of 288 collision observations, because participants may have misunderstood or ignored feedback.

One author declared conflicts of interest related to intellectual property owned by New York University, as well as related advisory positions with equity and ad hoc compensation.

Do not conclude that this system improves safe navigation for people with visual impairment. It was tested only in healthy adults with simulated vision loss, and haptic feedback reduced collisions but increased task time and path length.