The aim of this study was to investigate cognitive interference caused by tinnitus by means of a modified version of the Stroop color-word test. In a mixed-design study, the performances of tinnitus patients (n = 23) and healthy controls with normal hearing (n = 23) were compared on three versions of the Stroop test: the original version, a modified version including physical-threat words, and a tinnitus version for which tinnitus words [descriptors of tinnitus; e.g., peep) were derived empirically. Matched control conditions (words) were included for all three versions, yielding a total of six screens that were presented on a computer. Participants in the control group were matched with the patients for age and gender. Main dependent measures were performance on the Stroop tests in terms of total time for completing each test. Also included were the Tinnitus Questionnaire (TQ), the Beck Depression Inventory (BDI), the state version of the Spielberger Trait State Anxiety Inventory (STAI-S), and a subtest from the Wechsler Adult intelligence Scale measuring verbal ability Results showed that tinnitus patients performed significantly slower on all six test conditions. The classical Stroop color-word interference was replicated in both groups. Also, an effect For physical-threat words was found for both groups. Our expected tinnitus word interference could not be established. Patients scored significantly higher than controls on the BDI and the STAI, but these measures did not correlate with the Stroop results. In conclusion, the results indicate that tinnitus patients have impaired cognitive performance overall, as measured by these variations of the Stroop paradigm, but hearing impairment cannot be excluded as a possible confounder.