The effect of proprioceptive training on multisensory perception under visual uncertainty.

Most previous studies on multisensory integration concern mandatory integration. Moreover, no study has evaluated the effect of modality training on the result of integration. The purpose of this study is to evaluate the effect of training on visual-proprioceptive integration; i.e., we investigate the effect of proprioceptive modality training on the hand location estimation, when visual feedback exists. To achieve this goal, the effect of visual uncertainty on the estimation of hand position in visual-proprioceptive integration is studied in two groups: trained proprioception and untrained proprioception. The visual uncertainty is provided by an unpredictable spatial shift between visual and proprioceptive sensory feedbacks. The experiment was performed on nine subjects with trained proprioception and 11 subjects without proprioception training. The experiment had three phases: "familiarization", to draw participant's attention to a modality, "proprioception test", to estimate the hand position using only proprioception, and "vision-proprioception test", to investigate the effect of the visual uncertainty (bias) on hand position estimation. Our results indicate that: (i) modality training increases the subject reliance on the proprioceptive sensory information (i.e., bias decrement in sensory integration); and (ii) increasing the discrepancy between the modalities leads to more uncertainty (i.e., variance) in the estimation of hand position, but the variance of the final estimate is less than the variance of the proprioceptive estimate. This result confirms the theory that both senses contribute to the multisensory perception and in contrast to some studies, the dominant sense does not fully override the non-dominant one in the range of applied shift between the sensory sources.