Acoustical and electrical biasing of the cochlea partition. Effects on the acoustic two tone distortions f₂−f₁ and 2f₁−f₂

Abstract

Low frequency acoustical biasing of the cochlear partition with 5 Hz tones produces phase correlated changes of the acoustic two-tone distortions 2f₁−f₂ and f₂−f₁. Pronounced changes of f₂−f₁ and only small changes of 2f₁−f₂ for lower bias tone levels indicate that there is a close relation between changes in the difference tone f₂−f₁ and changes in the operating point of the cochlear amplifier (Frank and Kössl, 1996). To further investigate this relationship, the cochlear partition was additionally biased by current injection into the scala media of the gerbil. The injection of low frequency (5 Hz) AC currents (max. 1.3 μA) has a similar effect to that caused by low frequency tones in that both produce phase correlated changes of the two distortions (so-called biasing patterns), with stronger effects on f₂−f₁. For bias tone levels of about 105 dB SPL and current values of 1.3 μA, the effects are approximately of the same size. A change in the f₂−f₁ biasing pattern that can be found for increasing bias tone levels can also be seen for increasing primary levels. Changing the setpoint of the cochlear amplifier through the injection of DC current into the scala media during acoustical biasing of the cochlear partition produces the same changes of f₂−f₁ biasing patterns as increasing the primary levels. This indicates that the operating point of the outer hair cells that respond to the primary tones is not only influenced by low frequency biasing stimuli but also by shifts with increasing primary levels.

Introduction

High frequency resolution and sensitivity of the mammalian cochlea is most probably achieved through mechanical amplification of the incoming sound. One main feature of this amplification process is its nonlinear response characteristic, with the outer hair cells (OHCs) as the most likely candidates for the putative cochlear amplifier (see Dallos, 1992for a review). Voltage dependent fast movements of the OHCs found in vitro (Brownell et al., 1985; Ashmore, 1987; Mammano and Ashmore, 1993) may feed energy back into the transduction process on a cycle by cycle basis (reverse transduction). That the reverse transduction is present in vivo is indicated by measurements of electrically evoked otoacoustic emissions (EEOE; Hubbard and Mountain, 1983; Mountain and Hubbard, 1989; Xue et al., 1993b; Murata et al., 1991) and current induced travelling waves (Nuttall and Dolan, 1993a, Nuttall and Dolan, 1993b). OHCs from the apical part of the guinea pig cochlea show an asymmetrical operating range with DC receptor potentials measurable already for low stimulus levels (Dallos, 1985). In contrast basal turn OHCs show DC receptor potentials only for high stimulus levels (Russell et al., 1986; Cody and Russell, 1987) and their operating range is more symmetrical (Russell et al., 1986). This led to the proposal that there must be a tight regulation of the setpoint (=resting position of the operating point) of the OHC transfer function for low stimulus levels (Russell et al., 1986, Russell et al., 1989).

The generation of the acoustic two tone distortions (DPOAE), measurable at the tympanum, is a consequence of nonlinear mechanical amplification by the OHCs. The 2f₁−f₂ distortion is acoustically most prominent and depends on asymmetric components of the non-linear transfer characteristics. A decrease of the endocochlear potential (EP) (e.g. Mills et al., 1993) and the destruction of OHCs with antibiotics (e.g. Brown et al., 1989) reduce the 2f₁−f₂ amplitude to a great extent. Symmetrical components of the mechanic transfer characteristic determine the quadratic difference tone f₂−f₁ and therefore it may reflect changes in the operating point, i.e. the symmetry, of the cochlear amplifier. F₂−f₁ is susceptible to anesthesia, prolonged sound stimulation (Brown, 1988) and changes in the EP (Mountain, 1980; Mills et al., 1993). The amplitude changes of f₂−f₁ due to a furosemide induced EP decrease are more complex compared to 2f₁−f₂ (Mills et al., 1993).

In an earlier paper (Frank and Kössl, 1996) we described how 2f₁−f₂ and f₂−f₁ could be changed through a shift of the operating point along a nonlinear transfer function. Fig. 1 illustrates how 2f₁−f₂ and f₂−f₁ change in dependence on the operating point of a presumed OHC transfer function. The Boltzmann function involved (Fig. 1A) was taken from 3 state models of mechano-electrical transduction (Corey and Hudspeth, 1983; Crawford et al., 1989; Kros et al., 1992). Its initial setpoint was at a symmetrical position:

$$\begin{array}{l}\text{y}=[1+exp({\text{a}}_2({\text{x}}_2-\text{x}))*(1+exp({\text{a}}_1-\text{x}))){]}^{-1}\\ \text{with}{\text{x}}_1={\text{x}}_2=-0.06\text{and}{\text{a}}_1=3{\text{a}}_2=12.8\end{array}$$

The primary frequencies, f₁ and f₂, were processed by this function and the resulting levels of 2f₁−f₂ (solid curve) and f₂−f₁ (dashed curve) are displayed against a shift of the operating point from the zero position (Fig. 1B). The gray highlighted area of Fig. 1B is to emphasize that small shifts around the zero position lead to slight changes of 2f₁−f₂ and steep variations of f₂−f₁. When the gerbil cochlea was biased with a low frequency tone, the acoustic distortions changed as predicted by the above simulation, in agreement with a shift of the operating point (Frank and Kössl, 1996). Depending on the phase of the biasing tone the amplitude changes of 2f₁−f₂ and f₂−f₁ generally were reversed as suggested in Fig. 1B and f₂−f₁ was most sensitive to small shifts of the operating point around the zero position.

The purpose of the present study was to further investigate the relationship between the acoustical two tone distortions f₂−f₁ and 2f₁−f₂ and a change in the operating point of the putative cochlear amplifier. We extended the acoustical biasing experiments in that we varied not only the bias tone level systematically but also the primary levels. To additionally bias the cochlear partition, low frequency AC current was injected into the scala media (SM). Previous studies have shown that under current stimulation the OHCs exert force and are able to induce basilar membrane (BM) motion (Xue et al., 1995; Nuttall et al., 1995).

It is likely that the injection of AC current induces an alternating shift of the operating point in hyper- and depolarizing direction. When DC current is used, the operating point should be shifted in only one direction, depending on the polarity of the current stimuli. A comparison of the acoustical biasing pattern with and without DC current injection should reveal how the biasing pattern is changed through a setpoint shift.