Distortion-product otoacoustic emission suppression tuning curves in humans

Distortion-product otoacoustic emission (DPOAE) suppression data as a function of suppressor level (L ₃) for f ₂ frequencies from 0.5 to 8 kHz and L ₂ levels from 10 to 60 dB sensation level were used to construct suppression tuning curves (STCs). DPOAE levels in the presence of suppressors were converted into decrement versus L ₃ functions, and the L ₃ levels resulting in 3 dB decrements were derived by transformed linear regression. These L ₃ levels were plotted as a function of f ₃ to construct STCs. When f ₃ is represented on an octave scale, STCs were similar in shape across f ₂ frequency. These STCs were analyzed to provide estimates of gain (tip-to-tail difference) and tuning (Q _ERB). Both gain and tuning decreased as L ₂ increased, regardless of f ₂, but the trend with f ₂ was not monotonic. A roughly linear relation was observed between gain and tuning at each frequency, such that gain increased by 4-16 dB (mean ≈ 5 dB) for every unit increase in Q _ERB, although the pattern varied with frequency. These findings suggest consistent nonlinear processing across a wide frequency range in humans, although the nonlinear operation range is frequency dependent.