149 Testing the Reliability of Multi-Sensory Cue Use During Navigation in Virtual Reality

People rely on their senses to participate in their daily activities. However, research has not fully explored how the senses are used in combination during navigation. In navigation, two primary sensory cues are visual landmarks and the cues that come from vestibular and proprioceptive (body-based) senses associated with people’s movements. We aim to investigate the contribution of both visual and body-based cues when people are navigating. Our project has two objectives; 1) To test how single cues (body-based or visual) are weighted and combined during navigation and 2) To test whether those measures remain stable over time. We have hypothesized that participants will perform better with access to two cues than with access to a single cue and that errors and variability will decrease with repeated trials. Twenty participants were asked to put on a virtual reality headset and to perform a “homing” task where they walked two legs of a triangle and then returned to where they remembered the starting location to be. The participant repeats the same sequence under each of the three conditions possible.  The first condition acts as a control where sensory cues are unaltered; the second condition disrupts the use of body-based cues for location by spinning the participant  before returning to the target; and the third condition limits the participants’ ability to use their vision by removing visual landmarks. With many trials and two visits, we can assess the accuracy and variability over time, as well as how accuracy and variability varies when using a single cue versus two cues. This method will help us understand the contribution of practice and learning to sensory cue combinations. We ran a 2 (test session) x 3 (condition) ANOVA on mean accuracy to assess the accuracy of returning to the home location across all three conditions and whether it improved in the second session. There was no significant effect of test session, showing that participants did not improve from the first to second session. But, there was a significant effect of cue condition showing that performance improved in the consistent cue condition compared to the self-motion only condition. We ran correlations between session 1 and session 2 for each of the 3 cue conditions, on both accuracy and variability to determine if performance remained consistent over time. We found the strongest correlations for the consistent condition, showing reliability across test sessions when two cues are available, but less reliability with only a single cue. Understanding how people navigate with multisensory information can act as a resource when developing helpful forms of intervention in cases of sensory decline. 

References