Exploring the neural activity underlying tinnitus perception and its therapeutic exploitation

Tinnitus, the perception of sound in the absence of an external sound source, occurs in ten to fifteen percent of the general adult population. The phantom sound can be accompanied by several nonspecific symptoms, including hearing loss, anxiety, depression, and concentration difficulties. There currently is no cure for tinnitus, partially due to an incomplete understanding of its pathophysiology. Although tinnitus is often associated with hearing loss, there is no straightforward causal relationship between both symptoms. A cascade of neural alterations may lead to a pattern of modified activity and connectivity in several brain areas, ultimately leading to the perception of a phantom sound. So far, it has proven difficult to target this neural activity, and a lack of biomarkers hampers the search for successful treatments. Therefore, this thesis explores neurobehavioral measures that may function as biomarkers for the perception of tinnitus, and examines whether the neural activity underlying tinnitus can be targeted therapeutically. In a first chapter, we investigate whether tinnitus can be characterized using auditory evoked potentials, electrophysiological responses to auditory stimuli. A meta-analysis revealed that the P300 component, a response to unexpected but relevant stimuli, is diminished in patients with tinnitus. The second chapter of this thesis exposes subtle deficits in cognitive performance, confirming the impact of tinnitus on the executive control of attention. These neural and behavioral measures are combined in a third chapter, which uses novel analytic techniques to further illuminate the neural alterations underlying tinnitus. This chapter culminates in the development of a model to detect the presence of tinnitus, based on an integration of neural and neurobehavioral measures. A second part of this thesis focuses on the use of high-definition transcranial direct current stimulation (HD-tDCS) as a treatment for tinnitus. Although tinnitus severity decreased over time in both the active treatment and the sham control group, we report no beneficial effect of active HD-tDCS over sham treatment. In a final chapter, we present a machine learning model that is able to predict treatment response. The work presented in this dissertation confirms the role of the P300 response as an objective read-out of tinnitus-related cortical activity. Novel measures can provide insight into the pathophysiology of tinnitus and may also be used as biomarkers to facilitate research into experimental treatments. Furthermore, we offer recommendations for future randomized controlled trials in tinnitus populations. Lastly, this thesis demonstrates the appeal of machine learning models for treatment prediction.

ISBN:978-94-6458-247-5

Jaar van publicatie:2022

Trefwoorden:Doctoral thesis

Toegankelijkheid:Open