Auditory evoked responses to tone-pips in notched-noise provide frequency specific estimations of thresholds. Most often, test frequencies are 0.5-4 kHz. Thresholds are expected to match with a high degree to behavioral thresholds. However, only few studies are available containing data on stimulus and averaging parameters, especially at a test frequency of 0.25 kHz. In order to find "optimal" parameters for a widely used device (Nicolet Spirit Version 1.6), we designed five experiments on stimulus polarity (Exp.I), sweep frequency (Exp.II), stimulus duration (Exp.III), notched noise level (Exp.IV), and threshold estimation (Exp.V). The experiments also included a low test frequency of 0.25 kHz. These experiments were applied to 29 healthy volunteers (n=29, aged 20-41 years with normal hearing and no history of illness of the ear). We found that alternating stimulus polarity and a sweep frequency of 43.5/s evoked responses that were highest with respect to amplitudes (0.2-0.4 mV) and correlations (0.49-0.91). A novel finding of the study was that the highest amplitudes and correlations could be achieved if stimulus durations and notched-noise-levels were specifically adjusted to the test frequencies (stimulus durations 0.25 kHz: 4 ms, 0.5 Hz: 2 ms, 1-4 kHz: 1 ms, notched-noise-levels 0.25-0.5 Hz: +10 dB, 1-4 kHz: +/-0 dB). Deviations from behavioral thresholds did not exceed +/-5 dB in 93% of the measures. The results indicate that frequency specific auditory evoked responses provide reliable threshold estimations. Future experiments are required to confirm the threshold deviations for hard of hearing individuals.