# Altered parietal multisensory integration in chronic tinnitus during closed-loop real-time fMRI auditory downregulation

## TL;DR

This study shows that real-time fMRI neurofeedback can reduce auditory cortex activity in chronic tinnitus patients, with changes in parietal multisensory integration.

## Contribution

The study identifies parietal operculum 3 (OP3) as a key region in multisensory integration during tinnitus and shows its connectivity changes during auditory downregulation.

## Key findings

- Participants successfully reduced auditory cortex activity through real-time fMRI neurofeedback.
- Bilateral auditory cortices decreased connectivity with OP3 during closed-loop downregulation.
- OP3 showed reduced connectivity with a multisensory integration network during the intervention.

## Abstract

Chronic subjective tinnitus is the most common form of tinnitus and refers to an internal persistent phantom auditory perception. Evidence from neuroimaging studies has established tinnitus and its associated distress as a brain network disorder, with multiple brain regions displaying dysregulated activity or connectivity extending way beyond the auditory pathway. Hyperactivity in the auditory cortex has been associated with perceived tinnitus loudness, and somatosensory-auditory interactions have recently been involved in promising treatment avenues through their disruption using non-invasive or bimodal neuromodulation techniques, albeit with low mechanistic evidence.

In this study, we evaluated the neural effects of prolonged downregulation of bilateral auditory cortex activity mediated by real-time fMRI neurofeedback in individuals with moderate to severe (Tinnitus Handicap Inventory [THI] scores ≥ 48) chronic tinnitus. Twenty-one participants (aged 49 ± 11.4 years old, 16 males, 5 females) completed 15 fMRI neurofeedback sessions over 3–4 months each, as part of a randomized clinical trial (ClinicalTrials.gov: NCT05737888) comparing neurofeedback interventions over the current standard of care, cognitive behavioral therapy (CBT). We performed whole-brain general linear modeling analyses to delineate regulated brain areas, accounting for age, gender, and THI scores at baseline. Task-modulated functional connectivity analyses were carried out using psychophysiological interactions to unveil associated connectivity patterns emerging during cognitively demanding tinnitus defocalization.

Most participants succeeded at reducing the average activity of their auditory cortex throughout the training. Whole brain analyses additionally revealed a strong downregulation of parietal operculum 3 (OP3), a region previously reported to be activated in the right hemisphere during experimentally induced transient phantom percepts. In accordance with the hypothesis that OP3 may mediate the integration of multisensory inputs in tinnitus, we have shown that both left and right auditory cortices decrease their connectivity with OP3 during closed-loop auditory downregulation. Moreover, we observed a reduced connectivity of bilateral OP3 with a functional multisensory integration network that was previously found to be engaged by the primary and secondary auditory cortices during audio-tactile integration.

These findings support the hypothesis of OP3 having a key role in multisensory integration stemming from altered somatosensory-auditory crosstalk in chronic tinnitus. Targeted neuromodulation to desynchronize the connectivity between OP3 and the auditory cortex could further inform our understanding of the mechanisms behind recent successful bimodal interventions for reducing tinnitus.