Modifications of a single saturating non-linearity account for post-onsetchanges in 2f1-f2 distortion productotoacoustic emission.
Paper Summarised from :
Lukashkin and Russell(2002), J. Acoust. Soc. Am. 112, 1561-1568.
Andrei N.Lukashkin and Ian J. Russell,
School of Biological Sciences, University of Sussex,
Falmer, Brighton, BN1 9QG,
UK
e-mail:A.Lukashkin@sussex.ac.uk
1. Introduction
Liberman et al.1 reported changes in the 2f1-f2component of the distortion product otoacoustic emission (DPOAE) during a fewhundred milliseconds following the onset of stimulation. These changes in theDPOAE are considerably reduced following section of the olivocochlear nervebundle (OCB). An intriguing feature of the post-onset changes is theirbi-directionality. The DPOAE amplitude can either decrease or increase afterthe onset depending on both the frequencies and intensities of the primarytones. Kim et al.2 has also reportedbiphasic post-onset changes, when a rapid reduction in DPOAE amplitude isfollowed by DPOAE growth. Kujawa and Liberman3 reported that these changes are closely associated withamplitude minima (notches) routinely observed at the DPOAE I/O functions.However, local notches in the growth functions of the DPOAEs can arises due tonon-monotonic behaviour of distortion components at the output of a singlesaturating non-linearity4-7. In this case, a single phenomenon, ashift of the DPOAE amplitude notch, can explain multidirectional changes of theDPOAE observed for different parameters of stimulation.
2. Methods
DPOAE was recorded from pigmented guinea pigs in response totone bursts of 512-ms with 3-s interval between them. Responses were averagedin time domain. During data analysis a temporal window of 2048 point in length(i.e. 20.48 ms of duration sampled at 100 kHz) was slid along the 512-msresponse in increments of 5-ms. Amplitudes and phase angles of the spectralpeaks were obtained by performing an FFT of this temporal window. The moment oftime that corresponded to the starting point of the temporal window wasconsidered as the time when the amplitude and phase angle of the DPOAE werecalculated. All procedures involving animals were performed in accordance withUK Home Office regulations.
3. Results
Amplitude alterations of the DPOAE at the 2f1-f2 frequency (where f1and f2 are frequencies ofthe primary tones and L1and L2 are theircorresponding levels) were associated with an amplitude minimum (notch) in allguinea pigs studied. Systematic changes in the amplitude of the DPOAE were notobserved during the onset of prolonged tone bursts for DPOAE levels that werenot effected by the notch. For example, when the DPOAE was generated bylow-level (as low as L1/L2 = 20/10 dB SPL) orhigh-level (up to L1/L2 = 75/65 dB SPL) primaries(Fig. 1).
Figure 1. Time dependence of the position ofthe amplitude notch for the 2f1-f2 frequency component ofthe DPOAE before and after intravenous injection of strychnine (1 mg/kg b.w.).Solid line shows DPOAE response at the beginning of the prolonged acousticalstimulation. This response does not change after the injection of strychnine.Dashed and dotted curves represent the DPOAE responses in 480-ms after theonset of the stimulation before and after the injection, respectively. f2and f2/f1 are indicated within the figure. L1step is 1 dB. L2 is 10 dB below L1.
Sliding windowanalysis revealed that the notch gradually shifts upwards with time in thathigher levels of the primaries elicit it at successive periods in timefollowing the onset of the tone bursts (Fig. 1). This displacement ofthe notch position is usually completed within the period of the 512-ms toneburst, but may continue beyond it (Fig. 2). The shift becomessignificantly smaller (Fig. 1, dotted lines) after intravenous injection ofstrychnine, which is known to block the action of the OCB8.Therefore the observed shift of the notch is at least partially mediated by theactivity of the MOC efferents.
This single shift of the amplitude notchalso explains all three types of the 2f1-f2 changes (increase,decrease and biphasic variations), which have been documented at the onset ofprolonged stimulation1-3. Figure 2 shows a fragment of thecurves from Fig. 1 in the vicinity of the notch. An increase in theDPOAE is observed for levels of primaries generating emission on the left slopeof theinitial notch (solid line). The left upward-pointing vertical arrow (Fig.2, top panel) specifies this increase of the amplitude for a particularlevel of primaries (Fig. 2, bottom left panel). In contrast, the DPOAE decreases for levels of primariesgenerating emission on the right slope of the steady state notch (dashedline). The right downward pointing vertical arrow (Fig. 2, top panel) indicatesthe DPOAE decrease for a specific level of primaries (Fig. 2, bottom rightpanel). A biphasic DPOAE response (i. e. a decrease followed by an increase)takes place for any primary levels generating DPOAE between initial and steadystate notch positions (Fig. 2, bottom middle panel). This range ofprimaries is indicated in Fig. 2, top panel, by a horizontaldouble-headed arrow. The biphasic response may be absent if the stimulusparameters are not optimal so that the notch is shallow, or hard to resolve,when the time-dependent shift of the notch is small. Similar conclusions thatthe DPOAE post-onset changes occur due to shift of the amplitude notch has beenindependently drawn by Kim et al.9.
Figure 2. Dependence of the post-onset changesof the DPOAE on the levels of the primaries. Top panel shows the responsepresented in Fig. 1 in the vicinity of the amplitude notch. Vertical arrowsindicate changes of the DPOAE amplitude for particular pairs of the primarylevels; L1/L2 = 53/43 dB SPL (left arrow) and L1/L2= 57/47 dB SPL (right arrow). Temporal courses of these changes are shown inthe bottom left and right panels, respectively. Middle panel of the bottom rowshows biphasic time dependence of the DPOAE for the L1 rangeindicated by the horizontal arrow in the top panel. L1 is indicatedwithin each of the bottom panels. The other notations are the same as in Fig.1.
Dependence of the position of the amplitude notch on theparameters of stimulation (e.g. the ratio of the primary frequencies andchanges in the primary levels) has been extensively studied6,10.Therefore, it is possible to predict the DPOAE temporal behaviour in thetwo-dimensional space of the levels of the two primaries, and for theirdifferent frequency ratios. Detailed analysis of these predictions is given inthe original paper.
4. Conclusions
It is likely that all post-stimulus alterations of the 2f1-f2 component that have been described to date1-3,11originate in the upward shift of the amplitude notch that is reported here. Itis, therefore, difficult to make any quantitative comparisons between thesedata. Indeed, the current study shows that the magnitude of the DPOAEalterations depends considerably on the specific position of the DPOAE in thevicinity of its local minimum. Thus, variations in the magnitude of the effectin the different studies can be explained simply by different positioning ofthe DPOAE in the vicinity of the notch.
The magnitude of the DPOAE changes is not the most suitableparameter to characterise the post-onset changes in 2f1-f2 DPOAE.The shiftof the notch minimum provides a more objective and comparableparameter, and hence is a better characteristic of the process.
It is obvious fromthe current study that the properties of the post-onset process do not resemblethose normally associated with physiological adaptation and the usage of theterm “adaptation” seems to be inappropriate.
This work was supported by a grant from the Wellcome Trust.
5. References
1. Liberman,MC et al. (1996).“The ipsilaterally evoked olivocochlear reflex causes rapid adaptation of the 2f1-f2 distortion product otoacoustic emission,” J. Acoust.Soc. Am. 99, 3572-3584.
2. Kim,DO et al. (2001),“Adaptation of distortion product otoacoustic emission in humans,” JARO 2,31-40.
3. Kujawa,SG, and Liberman, MC (2001), “Effect of olivocochlear feedback on distortion productotoacoustic emission,” JARO 2, 268-278.
4. Weiss,TF, and Leong, R (1985),“A model for signal transmission in an ear having hair cells with free-standingstereocilia. IV. Mechanoelectric transduction stage,” Hear. Res. 20, 175-195.
5. Lukashkin,AN, and Russell, IJ (1998), “A descriptive model of the receptor potential nonlinearitiesgenerated by the hair cell mechanoelectrical transducer,” J. Acoust. Soc. Am.103, 973-980.
6. Lukashkin,AN, and Russell, IJ (1999), “Analysis of the f2-f1 and 2f1-f2 distortion componentsgenerated by the hair cell mechanoelectrical transducer: Dependence on theamplitudes of the primaries and feedback gain,” J. Acoust. Soc. Am. 106,2661-2668.
7. Bian, L et al. (2002). “Deriving a cochlear transducer function from low-frequencymodulation of distortion product otoacoustic emissions,” J. Acoust. Soc. Am.112, 198-210.
8. Desmedt,JE, and Monaco, P (1961), “Mode of action of the efferent olivocochlear bundle on theinner ear,” Nature 192, 1263-1265.
9. Kim,DO et al. (2002),“Effects of the medial olivocochlear reflex on cochlear mechanics: experimentaland modelling studies of DPOAE,” In Biophysics of the Cochlea: from Molecule toModel, organised by A. W. Gummer, Titisee, Germany, 2002.
10. Lukashkin,AN, and Russell, IJ (2001), “Origin of the bell-like dependence of the DPOAE amplitude onprimary frequency ratio,” J. Acoust. Soc. Am. 110, 3097-3106.
11. Sun,XM, and Kim, DO (1999),“Adaptation of distortion product otoacoustic emission in young-adult and oldCBA and C57 mice,” J. Acoust. Soc. Am. 105, 3399-3409.