Fig 1.
Neurophysiological models of tinnitus.
(A) Thalamocortical dysrhythmia model [8,9]. In the deafferented region of the auditory cortex, thalamic inputs induce theta activity. Due to decreased lateral inhibition, this region is surrounded by an area of abnormal gamma activity (“edge effect”) leading to the TI percept. (B) Global Brain Model [13]. Decreased inhibition and increased excitability in the auditory cortices result from reduced sensory input. Tinnitus processing involves a globally extended fronto-parieto-cingulate network which amplifies the auditory neural activity by top-down influence. The level of tinnitus-related distress correlates with the extent of top-down amplification.
Table 1.
Characteristics of participant groups.
Fig 2.
Individual (black lines) and median (coloured lines) audiograms for all ears in each group.
(A) Non-TI controls (n = 19), (B) tinnitus participants (n = 28). The median thresholds are shown in blue and red, respectively.
Fig 3.
Spatial hierarchies used in source analysis.
Diagram shows constituent Brodmann areas (BAs) of ROIs. Anterior and posterior cingulate cortex contain the listed BAs from both left and right hemisphere. Connectivity between two BAs is estimated by averaging over ten random pairs of voxels drawn from BAs, connectivity between ROIs and top-level component is obtained by averaging over respective lower levels. Abbreviations of ROIs used in the text: (l/r)AC: (left/right) auditory cortex, ACC: anterior cingulate cortex, DLPFC: dorsolateral prefrontal cortex, Ins: insula, OFC: orbitofrontal cortex, PCC: posterior cingulate cortex, PHC: parahippocampus.
Fig 4.
Power spectra for sensory (A) and global components (B).
Error bars show standard error.
Fig 5.
Functional connectivity for sensory and global components.
Imaginary coherency is shown (A) within ACs, (B) within global network, (C) between ACs and global network. Frequency regions with significant differences (p<0.05, green) were determined with a cluster-based permutation test, regions with 0.05≤p≤0.1 are marked in yellow.
Fig 6.
Effective connectivity for sensory and global components.
PDC is shown (A) within ACs, (B) within the global network, and as (C) outflow from and (D) inflow into ACs from global network. Frequency regions with significant differences (p<0.05, green) were determined with a cluster-based permutation test, regions with 0.05≤p≤0.1 are marked in yellow.
Fig 7.
Correlations between TI handicap score and functional connectivity.
Correlations for TI participants are shown (A) within the ACs, (B) within the global network, (C) between the ACs and the global network.
Fig 8.
Tinnitus network as obtained from exploratory PDC analysis.
Red arrows indicate connections with significant differences (p<0.05) between TI and controls in some frequency window. Abbreviations: AC: auditory cortex, ACC: anterior cingulate cortex, DLPFC: dorsolateral prefrontal cortex, Ins: insula, OFC: orbitofrontal cortex, PCC: posterior cingulate cortex, PHC: parahippocampus; l/r: left/right.