The purpose of this paper is to investigate the influence on electroencephalogram (EEG) at the prefrontal cortex due to tinnitus and sounds. Tinnitus is a phenomenon that a person hears sounds even though there is no sound source. Conventional methods of investigating tinnitus are subjective and need long time. In this paper, EEG at the prefrontal cortex was used because the relationship between tinnitus and prefrontal cortex has been revealed. Tinnitus patients and healthy controls were classified and the distress level was estimated using EEG in researches so far, and identified the necessary features for these analyses. However, there is a problem that it is not certain whether features used in these analyses arise similarly in tinnitus and other sounds. The purpose of this paper is to verify which EEG features identified in previous researches arise from tinnitus or sounds. EEG data of tinnitus patients and healthy controls were measured and a questionnaire was conducted on tinnitus distress to the tinnitus patients. EEG measurements of healthy controls at resting with no stimulation and at listening to sounds which simulated tinnitus were performed and a questionnaire was conducted to see which sound they felt distress. Differences of EEG between tinnitus patients and healthy controls and differences of EEG between controls at resting with no stimulation and at listening to sound were different. It is thought that EEG changes of tinnitus patients at the prefrontal cortex arise from only tinnitus.