B Braem1, J Honoré2, M Rousseaux3, A Saj4, Y Coello5. 1. EA 1059 (URECA), Université Nord-de-France, 59653 Villeneuve-d'Ascq, France; Service de rééducation neurologique, hôpital Swynghedauw, CHRU de Lille, 59037 Lille, France. Electronic address: berenger.braem@gmail.com. 2. EA 4559 (LNFP), Université Nord-de-France, 59037 Lille, France; CNRS, 59000 Lille, France. Electronic address: Jacques.Honore@chru-lille.fr. 3. Service de rééducation neurologique, hôpital Swynghedauw, CHRU de Lille, 59037 Lille, France; EA 2691, Université Nord-de-France, 59037 Lille, France. Electronic address: Marc.Rousseaux@chru-lille.fr. 4. Neurology and Cognitive Neuroimaging, Department of Neuroscience, University of Geneva, 1211 Geneva, Switzerland; Department of Neurology, University Hospital of Geneva, 1211 Geneva, Switzerland. Electronic address: Arnaud.Saj@unige.ch. 5. EA 1059 (URECA), Université Nord-de-France, 59653 Villeneuve-d'Ascq, France. Electronic address: yann.coello@univ-lille3.fr.
Abstract
OBJECTIVE: Multimodal perception raises the issue of sensory integration. The aim of this study is to assess whether the visuo-haptic subjective vertical could be predicted from the visual and haptic unimodal performances, according to a Bayesian model, which optimizes the reliability of the multimodal estimate. The limits of the model were assessed by evaluating the impact of aging, cerebral damage and spatial deficits. METHOD: Young and older healthy participants, as well as patients with a right hemisphere lesion, suffering from spatial neglect or not, had to align a rod with the gravitational vertical in the visual, haptic and visuo-haptic modalities. RESULTS: Visuo-haptic subjective vertical was correctly predicted by the model in healthy participants, as well as in patients in spite of different performances. An anticlockwise deviation of the subjective vertical was observed in the neglect group whatever the experimental conditions. By contrast, no deviation was observed in both non-neglect and healthy groups, for the visual and the visuo-haptic modalities. Nevertheless, the haptic subjective vertical was deviated, anticlockwise in the non-neglect patients and older healthy adults, and clockwise in the young healthy adults. Moreover, the variance was the smallest in the bimodal condition. CONCLUSION: The integration of visual and haptic signals appeared to obey a Bayesian model optimizing the reliability of the multimodal estimate. This holds true despite of aging, brain damage or visuospatial disorders. Regarding the perception of the vertical, multisensory integration does not seem thus to depend only on right posterior cortical areas.
OBJECTIVE: Multimodal perception raises the issue of sensory integration. The aim of this study is to assess whether the visuo-haptic subjective vertical could be predicted from the visual and haptic unimodal performances, according to a Bayesian model, which optimizes the reliability of the multimodal estimate. The limits of the model were assessed by evaluating the impact of aging, cerebral damage and spatial deficits. METHOD: Young and older healthy participants, as well as patients with a right hemisphere lesion, suffering from spatial neglect or not, had to align a rod with the gravitational vertical in the visual, haptic and visuo-haptic modalities. RESULTS: Visuo-haptic subjective vertical was correctly predicted by the model in healthy participants, as well as in patients in spite of different performances. An anticlockwise deviation of the subjective vertical was observed in the neglect group whatever the experimental conditions. By contrast, no deviation was observed in both non-neglect and healthy groups, for the visual and the visuo-haptic modalities. Nevertheless, the haptic subjective vertical was deviated, anticlockwise in the non-neglect patients and older healthy adults, and clockwise in the young healthy adults. Moreover, the variance was the smallest in the bimodal condition. CONCLUSION: The integration of visual and haptic signals appeared to obey a Bayesian model optimizing the reliability of the multimodal estimate. This holds true despite of aging, brain damage or visuospatial disorders. Regarding the perception of the vertical, multisensory integration does not seem thus to depend only on right posterior cortical areas.
Authors: Ashley Unger; Kylie H Alm; Jessica A Collins; Jacquelyn M O'Leary; Ingrid R Olson Journal: J Int Neuropsychol Soc Date: 2016-02 Impact factor: 2.892