Indications of different distortion product otoacoustic emission mechanisms from a detailed f(1),f(2) area study

The primary site of generation on the basilar membrane for the 2f(1)-f(2) d istortion product (DP) is generally considered to be near where the higher- frequency stimulus tone peaks. This site has also been shown to be a source of DP otoacoustic emission (DPOAE) in the ear canal, but a second source o f emission is known to exist in the region of the DP frequency place. The D POAE phase versus frequency gradient provides a means of investigating the emission mechanisms. ''Wave-fixed'' and ''place-fixed'' mechanisms have bee n proposed to account for the very different phase gradients found dependin g on whether the 2f(1)-f(2) DPOAE is evoked by a small or large stimulus-fr equency ratio. DPOAE phase versus frequency gradients can be investigated e ither by sweeping f(1),f(2) or by sweeping both frequencies maintaining a c onstant frequency ratio. Each manipulation gives only a partial description of DP behavior. In this study, the place-fixed/wave-fixed dichotomy is ana lyzed using extensive 2 f(1)-f(2) and 2f(2)-f(1) DP stimulus-frequency swee p data presented on matrices of f(1) vs f(2) and f(2)/f(1) ratio versus DP frequency: These show how the DPs are related and provide a more complete p icture of 2f(1)-f(2) and 2f(2)-f(1) DPOAE phase and amplitude versus freque ncy behavior. The phase data contain evidence for a systematic variation in the proportions of wave- and place-fixed emission. The results suggest tha t 2f(1)-f(2) DPOAEs with a wide stimulus frequency ratio are wave fixed, wh ile all other DPOAEs are place fixed. A transition occurs within the 2f(1)- f(2) DP data region at a frequency ratio of about f(2)/f(1) = 1.1. The 2f(1 )-f(2) DP and 2f(2)-f(1) Dp phase behavior is continuous across the f(2)/f( 1) = 1 boundary. As the 2f(2)-f(1) DP generation region must be strongly in fluenced by the DP frequency place, the results imply that the place-fixed component of the 2f(1)-f(2) DP is also linked to its frequency place. A sim ilar pattern was obtained with the 3f(1)-2f(2) and 3f(2)-2f(1) DPs. The res ults support the following model: For the limited set of stimulus condition s that gives rise to 2f(1)-f(2) wave-fixed emissions, DP energy is largely generated in the f(2) region and is emitted directly. All other DPOAEs are place-fixed emissions, and while nonlinearity within the f(2) stimulus enve lope remains the generator, the DP is not directly emitted but travels apic ally until it is re-emitted basally via a separate reflection mechanism in the region of the DP place. (C) 2000 Acoustical Society of America. [S0001- 4966(99)05212-1].