Repeatability of distortion product otoacoustic emissions in normally hearing humans.

Distortion product otoacoustic emissions (DPOAEs) at the frequency of 2f1-f2 were measured in one or both ears of 12 young adults during 4 test sessions over a 6-week period. The purpose was to determine the variability in DPOAE amplitudes and 'detection thresholds' over repeated measurements using a computer-based time-averaging system. DPOAEs were generated with f1 and f2 relative to (f1f2)1/2 in two basic paradigms: (a) fixed levels of L1 = L2 of 70 and 55 dB SPL over a stimulus range from 0.8 to 8 kHz in 0.2-octave intervals; (b) input-output functions in stimulus regions of 0.8, 1, 1.5, 2, 3, 4 and 6 kHz, L1 from 35 to 70 dB SPL changing in 5-dB steps and L2 at 6 dB below the amplitude of L1. The mean variability of DPOAE amplitudes with equilevel stimuli was 1.8 dB (SD = 1.8) for L1 = 70 dB SPL and 2.9 dB (SD = 2.7) for L1 = 55 dB SPL. It was 1.7 dB (SD = 1.7) and 2.4 dB (SD = 2.0) for comparable levels of L1 with L2 at 6 dB below L1. Variability in amplitude of the DPOAEs for the fixed-level condition was greatest overall above 6 kHz and below 1 kHz and in the 2-kHz region for one third of the subjects. Neither individual differences in emission amplitudes nor the presence of spontaneous otoacoustic emissions had a significant influence on the amount of amplitude variability within ears. Variability was not influenced by the length of time between measurements from 1 to 6 weeks.(ABSTRACT TRUNCATED AT 250 WORDS)