Neuroplastic changes in functional wiring in sensory cortices of the congenitally deaf: A network analysis

Abstract

Congenital sensory deprivation induces significant changes in the structural and functional organisation of the brain. These are well-characterised by cross-modal plasticity, in which deprived cortical areas are recruited to process information from non-affected sensory modalities, as well as by other neuroplastic alterations within regions dedicated to the remaining senses. Here, we analysed visual and auditory networks of congenitally deaf and hearing individuals during different visual tasks to assess changes in network community structure and connectivity patterns due to congenital deafness. In the hearing group, the nodes are clearly divided into three communities (visual, auditory and subcortical), whereas in the deaf group a fourth community consisting mainly of bilateral superior temporal sulcus and temporo-insular regions is present. Perhaps more importantly, the right lateral geniculate body, as well as bilateral thalamus and pulvinar joined the auditory community of the deaf. Moreover, there is stronger connectivity between bilateral thalamic and pulvinar and auditory areas in the deaf group, when compared to the hearing group. No differences were found in the number of connections of these nodes to visual areas. Our findings reveal substantial neuroplastic changes occurring within the auditory and visual networks caused by deafness, emphasising the dynamic nature of the sensory systems in response to congenital deafness. Specifically, these results indicate that in the deaf but not the hearing group, subcortical thalamic nuclei are highly connected to auditory areas during processing of visual information, suggesting that these relay areas may be responsible for rerouting visual information to the auditory cortex under congenital deafness.

Keywords: auditory cortex; congenitally deaf; functional magnetic resonance imaging; network analysis; visual perception.

FIGURE 1

Stimuli used in the fMRI experiments. Wedges (a) appeared alternately in each quadrant of the screen, and annuli (b) appeared at four different sizes in the centre of the screen.

FIGURE 2

Location of the regions of interest (ROI) analysed. Coloured in light blue are all ROIs used for analyses according to centre coordinates given in Table 1. L/R denotes left and right hemisphere, respectively.

FIGURE 3

Changes in community structure. Within the two groups and the two conditions (DeafA, DeafW, HearA and HearW), resulting community structures are shown. Colours denote different communities according to colour bar. Angle of vision similar to Figure 2. Location of the spheres visualised according to centre coordinates given in Table 1.

FIGURE 4

Distribution of mean‐divided partial correlation coefficients. The number of correlation coefficients per bin is shown for each of the two groups and the two conditions (DeafA, DeafW, HearA and HearW). The bin centres are displayed, the bin width is 0.25. The total number of correlations coefficients across all groups/conditions is almost identical (mean = 1202 ± 19).

FIGURE 5

Circular graph plots of strongest correlations. All mean‐divided partial correlation coefficients >1.75 are plotted for each of the two groups and the two conditions (DeafA, DeafW, HearA and HearW). Colour coding according to community structure, darker lines indicate higher correlation. Node notation according to labels given in Table 1, ‐L/‐R denotes left and right hemisphere, respectively.

FIGURE 6

Number of connections to auditory and visual areas. Independent of experimental condition, the number of connections to auditory areas is higher in the deaf group compared to the hearing group. The number of connections to visual areas does not differ between groups.

FIGURE 7

Circular graph plot of overall correlations as shown in Figure 5. The correlations of the left thalamus (Thal‐L—blue), the right thalamus (Thal‐R—green), the left pulvinar (Pul‐L—red) and right pulvinar (Pul‐R—grey) are outlined for each of the two groups and the two experimental conditions (DeafA, DeafW, HearA and HearW). There is no difference in HearA and HearW, while in DeafA and DeafW the number of correlations of these regions to auditory areas increases. Darker lines indicate higher correlation, ‐L/‐R denotes left and right hemisphere, respectively.