Abstract
Tinnitus has been considered an auditory phantom percept. Recently a theoretical multiphase compensation mechanism at a cortical level has been hypothesized linking auditory deafferentation to tinnitus. This Bayesian brain model predicts that two very different kinds of tinnitus should exist, depending on the amount of hearing loss: an auditory cortex related form of tinnitus not associated with hearing loss, and a (para)hippocampal form associated with hearing loss, in which the auditory cortex might be of little relevance. In order to verify this model, resting state source analyzed EEG recordings were made in 129 tinnitus patients, and correlated to the mean hearing loss, the range of the hearing loss and the hearing loss at the tinnitus frequency. Results demonstrate that tinnitus can be linked to 2 very different mechanisms. In patients with little or no hearing loss, the tinnitus seems to be more related to auditory cortex activity, but not to (para)hippocampal memory related activity, whereas in tinnitus patients with more severe hearing loss, tinnitus seems to be related to (para)hippocampal mechanisms. Furthermore hearing loss seems to drive the communication between the auditory cortex and the parahippocampus, as measured by functional and effective connectivity.