9Auditory Cortex and Resting-State Networks

FAuditory cortex: necessary but not sufficient

Resting-state functional MRI measures the slow, spontaneous co-fluctuations of the BOLD signal across the brain when a person is doing nothing in particular. Applied to tinnitus, it confirms that the auditory cortex — Heschl’s gyrus, planum temporale and the superior temporal gyrus — shows elevated spontaneous activity and altered connectivity even in silence [2010].

But the same studies make a more important point: auditory cortex involvement, while consistent, does not by itself explain the percept. The strength of a patient’s distress correlates poorly with auditory measures and far better with how strongly the auditory cortex is coupled to non-auditory networks. Tinnitus is therefore better framed as a disorder of network connectivity than as the firing of any single region [2014].

TMultiple overlapping resting-state networks

Several canonical resting-state networks are implicated. Maudoux and colleagues showed that the auditory resting-state network itself has abnormal connectivity in tinnitus, extending beyond auditory regions into limbic, attention and prefrontal areas [2012]. Schmidt and colleagues then dissected this further, reporting that the default-mode network, the dorsal attention network and the auditory network each show distinct connectivity differences in tinnitus — and that these can be partly disentangled from the effects of hearing loss alone [2013].

Graph-based analyses add a structural perspective: certain regions act as abnormal network hubs in tinnitus, with the prefrontal and parietal cortices showing changed centrality, suggesting reorganisation of how information is routed across the whole brain rather than a focal lesion [2009].

TWhy the salience and default-mode networks matter

The recurring theme is intrusion. In health, the default-mode network deactivates when attention is captured by a meaningful stimulus, and the salience network gates which stimuli deserve that attention. In tinnitus, the phantom sound appears to behave like a perpetually salient signal: the salience network repeatedly tags it as important, and the default-mode network fails to deactivate or disengage normally, so the percept keeps re-entering awareness [2013].

This network framing explains clinical realities that a purely auditory model cannot: why distraction and engaging tasks reduce tinnitus awareness, why it intrudes most in quiet and at night, and why two people with identical audiograms and tinnitus loudness can differ enormously in how disabling they find it. The difference lives in the attention–salience–limbic coupling, not in the ear [2014].

CWhy there is no single ’tinnitus locus’

Decades of imaging have failed to identify one brain region whose activity equals tinnitus, and the resting-network literature explains why. The percept emerges from the interaction of separable subnetworks: an auditory network that supplies the candidate sound, a salience network that flags it, a distress network (limbic) that colours it, and memory and attention networks that sustain it [2014]. Different patients can arrive at the same symptom through different network configurations, which is one reason single-target treatments and single-region neuromodulation give inconsistent results.

This view also reconciles the network literature with the rate, synchrony and dysrhythmia mechanisms covered earlier: those describe how aberrant activity is generated, while the resting-state networks describe how widely it is bound into a conscious, distressing experience. The clinical corollary is that effective management increasingly targets the network — attention, salience and emotional response — rather than chasing a non-existent focal generator [2016].