Neural Networks Mediating Perceptual Learning in Congenital Blindness

Daniel-Robert Chebat,Maurice Ptito,Fabien C Schneider

doi:10.1038/s41598-019-57217-w

Abstract

Despite the fact that complete visual deprivation leads to volumetric reductions in brain structures associated with spatial learning, blind individuals are still able to navigate. The neural structures involved in this function are not fully understood. Our study aims to correlate the performance of congenitally blind individuals (CB) and blindfolded sighted controls (SC) in a life-size obstacle-course using a visual-to-tactile sensory substitution device, with the size of brain structures (voxel based morphometry-VBM-) measured through structural magnetic resonance Imaging (MRI). VBM was used to extract grey matter volumes within several a-priori defined brain regions in all participants. Principal component analysis was utilized to group brain regions in factors and orthogonalize brain volumes. Regression analyses were then performed to link learning abilities to these factors. We found that (1) both CB and SC were able to learn to detect and avoid obstacles; (2) their learning rates for obstacle detection and avoidance correlated significantly with the volume of brain structures known to be involved in spatial skills. There is a similar relation between regions of the dorsal stream network and avoidance for both SC and CB whereas for detection, SC rely more on medial temporal lobe structures and CB on sensorimotor areas.

Highlights

Vision is undoubtedly a great facilitator of navigational tasks[1]
Using principal component analysis (PCA), we explore the relationship between learning performances of congenitally blind individuals (CB) and sighted control (SC) individuals and some specific brain areas known to be involved in navigation
We found that there is a significant correlation between the performance of our CB participants and their ‘visual’-tactile acuity score in the training part of the experiment for obstacle detection (r(32) = 0.286; p < 0.05), and for avoidance (r(32) = 0.314; p < 0.05)

Summary

Introduction

Vision is undoubtedly a great facilitator of navigational tasks[1] (for review see[2]). Visual cues regulate foot placements by providing constant spatial updates of the distance to the obstacle[3] in order to adapt locomotor behavior[4] In sighted people this behavior is mediated by a complex network of interacting brain regions that integrates visual information and translates visual cues into appropriate behavior[5]. The hippocampal formation and the posterior parietal cortex are traditionally thought to play a pivotal role in navigation[6] These two brain regions are involved in the processing of higher order spatial-cognitive information[7], and in the registering of spatial information which is crucial for navigation[8]. Using principal component analysis (PCA), we explore the relationship between learning performances of CB and SC individuals and some specific brain areas known to be involved in navigation

Methods

Results

Discussion

Conclusion