Tinnitus is a phantom auditory percept in the absence of external acoustic stimulation.1 Its prevalence in the United States increases with age from 5% for young adults to 14% after age 65 years.2 Typically, tinnitus is viewed as having an exclusively auditory origin. However, recent brain imaging studies3,4 suggest that nonauditory brain structures could be involved in the genesis of tinnitus. In this respect, Rauschecker et al5 proposed that tinnitus might be the result of a dysfunctional neural “noise-cancellation” mechanism. They postulated that a peripheral deafferentation (eg, aging) generates a tinnitus-related activity that is normally blocked at the level of the medial geniculate nucleus via amygdalar inhibitory projections. However, no clinical evidence has supported this hypothesis to date. To clarify the role of the medial temporal lobe (MTL) structures (eg, amygdala5 and hippocampus3) in tinnitus, we compared the prevalence of tinnitus among patients who underwent unilateral MTL resection encroaching on the amygdala with that among matched controls and participants with self-reported epilepsy (SRE) but no surgery. The surgical cases were expected to have increased difficulty in inhibiting the tinnitus signal and therefore a higher prevalence of tinnitus.