Global resting-state functional connectivity of neural oscillations in tinnitus with and without hearing loss

Abstract

This study examined global resting-state functional connectivity of neural oscillations in individuals with chronic tinnitus and normal and impaired hearing. We tested the hypothesis that distinct neural oscillatory networks are engaged in tinnitus with and without hearing loss. In both tinnitus groups, with and without hearing loss, we identified multiple frequency band-dependent regions of increased and decreased global functional connectivity. We also found that the auditory domain of tinnitus severity, assayed by the Tinnitus Functional Index, was associated with global functional connectivity in both auditory and nonauditory regions. These findings provide candidate biomarkers to target and monitor treatments for tinnitus with and without hearing loss.

Keywords: biomarkers; electroencephalography; functional brain imaging; hearing loss; magnetoencephalography; tinnitus.

Figure 1

Whole brain global functional connectivity in the theta band. Images show regions of significant group differences following 5% FDR and 5‐voxel cluster correction. Boxplots below each image illustrate differences across all groups in the voxel with the maximum group differences within each cluster. Groups with normal hearing (HL−) are shaded in dark gray and hearing loss groups (HL+) are shaded in light gray. Global functional connectivity is quantified by the mean imaginary coherence values of each voxel and represented in the boxplots by the black diamond and the median is represented by the white line. FDR, false discovery rate

Figure 2

Whole brain contrasts in alpha connectivity. Images depict regions of significant group differences following 5% FDR and 5‐voxel cluster correction. Boxplots below each image illustrate differences across all groups in the voxel with the maximum group differences within each cluster. Groups with normal hearing (HL−) are shaded in dark gray and hearing loss groups (HL+) are shaded in light gray. Mean imaginary coherence values are represented in the boxplots by the black diamond, and the median is represented by the white line. FDR, false discovery rate

Figure 3

Whole brain contrasts in beta connectivity. Images depict regions of significant group differences following 5% FDR and 5‐voxel cluster correction. Boxplots below each image illustrate differences across all groups in the voxel with the maximum group differences within each cluster. Groups with normal hearing (HL−) are shaded in dark gray and hearing loss groups (HL+) are shaded in light gray. Mean imaginary coherence values are represented in the boxplots by the black diamond and the median is represented by the white line. FDR, false discovery rate

Figure 4

Whole brain contrasts in low gamma connectivity. Images depict regions of significant group differences following 5% FDR and 5‐voxel cluster correction. Boxplots below each image illustrate differences across all groups in the voxel with the maximum group differences within each cluster. Groups with normal hearing (HL−) are shaded in dark gray and hearing loss groups (HL+) are shaded in light gray. Mean imaginary coherence values are represented in the boxplots by the black diamond and the median is represented by the white line. FDR, false discovery rate

Figure 5

Whole brain correlations between tinnitus symptoms and theta connectivity. Images depict regions of significant group differences following 5% FDR and 5‐voxel cluster correction. Scatterplots below each image illustrate associations for the voxel with the maximum group correlation within each cluster. FDR, false discovery rate

Figure 6

Whole brain correlations between tinnitus symptoms and alpha connectivity. Images depict regions of significant group differences following 5% FDR and 5‐voxel cluster correction. Scatterplots below each image illustrate associations for the voxel with the maximum group correlation within each cluster. FDR, false discovery rate

Figure 7

Whole brain correlations between tinnitus symptoms and beta connectivity. Images depict regions of significant group differences following 5% FDR and 5‐voxel cluster correction. Scatterplots below each image illustrate associations for the voxel with the maximum group correlation within each cluster

Figure 8

Whole brain correlations between tinnitus symptoms and low gamma (30‐50 Hz) connectivity. Images depict regions of significant group differences following 5% FDR and 5‐voxel cluster correction. Scatterplots below each image illustrate associations for the voxel with the maximum group correlation within each cluster