Parameter optimization for applying the prepulse gap paradigm to humans.

Background and objectives: Turner and colleagues introduced a new method that can detect tinnitus in animals. The stimulus is composed of a small background noise that is identical to the pitch of the tinnitus and a large pulse noise that can evoke a startle response. In normal rats, the gap decreases the startle reflex. However, in tinnitus rats, the gap does not decrease the startle reflex. The goal of this study was to optimize the stimulation paradigm so that the prepulse inhibition of N1-P2 amplitude would be maximized in the normal human subjects.

Subjects and methods: Seven normal control subjects without tinnitus were recruited. The stimulus was composed of two different sounds: the softer background noise and the louder pulse noise. A 50 msec silent gap was inserted before the pulse noise as the gap condition (G condition) but not in the no-gap condition (N condition). The averaged amplitude of the N1-P2 cortical response was recorded for the G and N conditions.

Results: The G/N ratio was the smallest when the gap was 20 msec prior to the pulse noise. The G/N ratio was 84.8±16.8% with the Hanning window and 78.5±5.9% without the window. The G/N ratio was 91.1±24.9%, 78.0±5.4%, and 79.0±18.1% when the intensity of the background noise was 10, 20, and 32 dB SL, respectively. When the intensity of the background noise was 20 and 32 dB SL, the N1-P2 amplitude of the G condition was significantly smaller than that of the N condition.

Conclusions: The optimal stimulus should be composed of the 1 kHz pulse noise without Hanning window. The intensity of the background noise should be 20 dB HL and the location of the gap should be 20 msec prior to the pulse noise. It seems that with these optimized parameters we could expect a 78.0% inhibition of N1-P2 amplitude in normal subjects without tinnitus.