Distortion product otoacoustic emissions in an active nonlinear model of the cochlea

Relationship between distortion product otoacoustic emissions and frequency modulation difference limens

OBJECTIVE

To explore the relationship between distortion product otoacoustic emissions (DPOAEs) and frequency modulation difference limens (FMDLs) in normal-hearing subjects.

DESIGN

Fifteen normal-hearing adult subjects (age range = 20 to 29 yr; mean age = 21 yr, 6 mo) participated in the study. Each subject met the following criteria: 1) hearing thresholds <20 dB HL bilaterally at frequencies 0.25 to 8 kHz; 2) type A tympanograms bilaterally; and 3) no history of otologic disease. To reduce gender-related differences, only female subjects were recruited as subjects. DPOAE amplitudes and FMDLs were measured in both ears of 15 female adults at 1, 2, and 4 kHz. Ears were classified on the basis of DPOAE amplitude into large DPOAEs and small DPOAEs.

RESULTS

Results indicated that ears with larger DPOAEs showed better frequency discrimination (smaller FMDLs) than did ears with smaller DPOAEs at 2 kHz.

CONCLUSIONS

Results of the current study support frequency discrimination models that predict a relationship between frequency selectivity and frequency discrimination.

A model of cochlear micromechanics

A cochlear model is presented which has two degrees of freedom, the motion of the basilar membrane and that of the tectorial membrane (TM), in every cross section of the organ of Corti. It assumes that the reticular lamina is virtually rigid, so that the apical portion of the outer hair cells is firmly anchored to it, and also that the TM is directly driven by the BM through the marginal net of the TM, which anchor to the phalanges of the outermost row of Deiters' cells or Hensen's cells, or both. It is shown that the model can produce more than 40 dB of cochlear gain in the frequency-domain simulation and 30 dB in time-domain simulation. Transiently evoked otoacoustic emissions and cochlear microphonics are simulated in the time domain.

Spontaneous otoacoustic emissions in an active feed-forward model of the cochlea

Numerical simulation of spontaneous otoacoustic emissions (SOAE) was done using a 1-dimensional macromechanical cochlear model with an active 'feed-forward' force in every section of the basilar membrane (BM). When the activity of the force was increased the model showed more stability than a 'feed-back' model and could have excitation curves with larger tips without divergence of the solution. The stability broke up when either (1) the damping of the BM was made slightly irregular throughout the BM or (2) the feed-forward force was switched off at a certain part of the BM, and limit cycle oscillations (LCO) emerged within the cochlea. Critical feed-forward value for the emergence of LCO in the first setting of BM (1), which was intended to simulate physiological variations in the distribution of outer hair cells, was searched utilizing the 'ringing' of delayed evoked otoacoustic emissions (DEOAE). In the course of the search, smooth transition from a DEOAE to a set of SOAEs was found to occur keeping the same spectral fine structure of the emissions when the feed-forward force surpassed a certain value. It was, as a result suggested that the two kinds of emissions may have the same origin. In the second setting of BM (2), which was intended to simulate pathological cases. LCOs tended to be stronger than in the first one and they had similarity not only to real SOAEs but also to tinnitus in the way they showed up very close to or at the "edge' of the switched-off part of the BM.

Further studies on the mechanics of the cochlear partition in the mustached bat. II. A second cochlear frequency map derived from acoustic distortion products

It has been proposed that acoustic 2f1-f2 distortions reflect the frequency characteristics of a secondary cochlear filter mechanism (Brown et al., 1992; Allen and Fahey, 1993). This concept was used to construct a second cochlear frequency map that may represent aspects of tectorial membrane (TM) tuning. Within the frequency range of 15-105 kHz, for a given f2 frequency, f1 was varied and the frequency ratio f2/f1 determined that produced maximum levels of the 2f1-f2 distortion (best ratio). The second cochlear frequency map was derived by plotting the distortion frequency that corresponded to the best ratio f2/f1 against the cochlear place of f2 which was obtained from the HRP-frequency map of Pteronotus (Kössl and Vater, 1985b). Minimum best ratios of 1.0005 and hence practically identical characteristic frequencies of the putative tuning of basilar membrane (HRP) and TM (2f1-f2) were found at about 45% distance from the base, a point at which 62 kHz are represented on the BM. This frequency is associated with strong cochlear resonance and large evoked and spontaneous otoacoustic emissions. Between 45% and 20% distance from the base, the basilar membrane (BM) tuning progressively increases to about 70 kHz whereas the calculated TM tuning remains constant at a frequency close to 62 kHz. The range of constant TM tuning coincides with the sparsely innervated cochlear region of Pteronotus where BM thickness is maximal and TM mass and limbal attachment are reduced (Vater and Kössl, 1996). We suggest that here the TM oscillates strongly at 62 kHz and may carry most of the energy of cochlear resonance which is transferred into movement of the organ of Corti at and apical to the 45% location where the BM is tuned to 62 kHz.