Multisensory Integration In Dynamic Distance Perception

Distance perception is a kind of perception that observers estimate the distance of targets using different cues in the environment. It means the perception of the distance of the target and the observer himself, or the perception of two target objects. Dynamic distance perception is one kind of distance perception when observers move from one place to another. It is not only a key of finishing series of actions but also a premise of the survival of animals and human being. In our daily life, individuals mainly use visual and body-based information to estimate the dynamic distance. Visual information concludes the optic flow and static visual information, while the body-based information consists with proprioceptive, vestibular and kinetics information. Studies showed that in dynamic distance judgment, observers usually use optic flow and body-based information as cues.Most previous studies usually probe the utility weighting of each cue in natural environment by either cue-availability paradigm or cue-conflict paradigm. However, there are few researches which could provide confident evident of the weighting of each cue in dynamic distance estimation because of the limit of methods and conditions. In order to discuss the relative utility weighting of visual and body-based cue, respectively, the current research used both cue-available paradigm and cue-conflict paradigm, and probed the effect of the reliability of visual cue to the weighting. Experiment probed the combination strategy of visual information and body-based information by cue-available paradigm.In Experiment 1, subjects first observed a target object for 5 seconds before they turn 180°and reproduce the distance of the object to themselves. When reproducing, there are three conditions: blindwalking-, visual only-and walking with vision condition. The results demonstrated that both visual and body-based information could be accurate cue in dynamic distance perception; it was more accurate when individuals estimated distance in the condition of blindwalking than in vision only condition did; the estimation was the most accurate when both cues were presented. Experiment 2 & 3 tried to calculate the relative weighting of each cue in dynamic distance perception via cue-conflict paradigm. The subjects were presented a target object for 5 seconds and then turned 180°to reproduce the distance they just observed. When reproducing, the subjects experienced both visual and body-based information; however, the two cues were not match, that is, the optic flow gain was not 1. In experiment 2, the optic flow gain we used was 0.75,1 and 1.33, and each optic flow gain condition occupied 33% in total number. Moreover, all conditions were presented randomly. The results showed that the visual weighting in this experiment was 12% thus the body-based information was 88%. This illustrated that in dynamic distance estimation, the weighting of body-based information was higher than visual cue. The methods in Experiment 3 was similar to Experiment 2, however, the proportion of 0.75- and 1.33 optic flow gain conditions were lower, only 8% in total. The visual weighting was 26% in Experiment 3.All results of the three experiments demonstrated that:1 both visual and body-based information could be cues of dynamic distance perception; 2 the estimation was most accurate when both visual and body-based information were presented to subjects; 3 body-based information trumped vision when estimating walked distance; 4 the weighting of a cue could be affected by the consistency and reliability of the cue.