Basilar membrane responses to two-tone and broadband stimuli

Abstract

The responses to sound of mammalian cochlear neurons exhibit many nonlinearities, some of which (such as two-tone rate suppression and intermodulation distortion) are highly frequency specific, being strongly tuned to the characteristic frequency (CF) of the neuron. With the goal of establishing the cochlear origin of these auditory-nerve nonlinearities, mechanical responses to clicks and to pairs of tones were studied in relatively healthy chinchilla cochleae at a basal site of the basilar membrane with CF of 8-10 kHz. Responses were also obtained in cochleae in which hair cell receptor potentials were reduced by systemic furosemide injection. Vibrations were recorded using either the Mössbauer technique or laser Doppler-shift velocimetry. Responses to tone pairs contained intermodulation distortion products whose magnitudes as a function of stimulus frequency and intensity were comparable to those of distortion products in cochlear afferent responses. Responses to CF tones could be selectively suppressed by tones with frequency either higher or lower than CF; in most respects, mechanical two-tone suppression resembled rate suppression in cochlear afferents. Responses to clicks displayed a CF-specific compressive nonlinearity, similar to that present in responses to single tones, which could be profoundly and selectively reduced by furosemide. The present findings firmly support the hypothesis that all CF-specific nonlinearities present in the auditory nerve originate in analogous phenomena of basilar membrane vibration. However, because of their lability, it is almost certain that the mechanical nonlinearities themselves originate in outer hair cells.

Figure 1

Effect of suppressor tones on basilar membrane intensity functions for a near-cf tone. Intensity functions are shown for an 8 kHz probe tone alone (open circles, dashed line) and in the presence of a 300 Hz suppressor at several intensities (filled symbols, solid lines): diamonds, 70 dB; triangles, 80 dB; circles, 90 dB; squares, 100 dB. Cochlea: L29.

Figure 2

Changes in basilar-membrane frequency tuning produced by a 12 kHz suppressor tone presented at 63 dB spl. The isovelocity tuning curves (scale in left ordinate) were derived from intensity functions for probe tones presented alone (open circles, dashed line) and in the presence of the suppressor tone (filled circles, solid line). The thin solid line indicates suppression magnitude (scale in right ordinate) as a function of probe frequency. The 12 kHz suppressor tone evoked a response 7.4 dB smaller than the 50 μm s⁻¹ isovelocity criterion. Cochlea: L14.

Figure 3

Frequency spectra of basilar membrane responses to a two-tone stimulus. The frequencies of equal-intensity tones (f₁, 13.33 kHz;f₂, 16.66 kHz) were chosen so that 2f₁–f₂ coincided with the cf of the basilar membrane site (10 kHz). Stimulus intensity (in dB relative to 20 μPa) is indicated in each panel. Cochlea: L47.

Figure 4

Magnitudes of 2f₁–f₂ distortion products as a function of primary-tone intensity and frequency ratio (solid lines). The frequencies of the two primary tones were chosen so that 2f₁-f₂ coincided with cf (10 kHz), (a) Open circles and dash line represent the velocity input–output function for a single tone at the distortion-product frequency. Closed symbols and solid lines indicate input–output functions for pairs of tones with f₂:f₁ ratios of 1.1, 1.2 and 1.3. (b) Distortion-product magnitudes are plotted as effective level, i.e. as intensity of a single tone at the distortion-product frequency required to produce a response of the same magnitude as the distortion product produced by the two-tone stimulus. Effective levels are expressed as decibels relative to primary-tone intensities. The dash line indicates that artifactual 2f₁-f₂ distortion products in the acoustic-stimulus system were about 70 dB less than the intensity of the primary tones. Cochlea: L47.

Figure 5

Velocity responses to rarefaction clicks of a basilar membrane site with cf of 9 kHz. Click intensity, expressed as peak pressure (in decibels relative to 20 μPa), is indicated for each tracing. Modified from Ruggero & Rich (1991a). Cochlea: L13.

Figure 6

Gain-versus-frequency functions for responses to (a) clicks and (b) tones of a single basilar membrane site. Gains were computed by dividing, frequency by frequency, spectral response magnitude by stimulus pressure. The spectral magnitudes for click responses were obtained by Fourier transformation of the data of figure 5. Click peak pressures are indicated next to each gain spectrum. Cochlea: L13.

Figure 7

Frequency-specific effects of furosemide upon the magnitude of basilar membrane responses to clicks. The velocity-magnitude frequency spectra were computed by Fourier transformation of responses to clicks with peak pressures of (a) 48, (b) 68 and (c) 88 dB (relative to 20 μPa). For each stimulus level, three curves are displayed, representing responses immediately preceding (solid line, − 13 min) and following (long-dash line, +9 min) a furosemide injection and when full recovery had occurred (short-dash line, +100 min). Redrawn from Ruggero & Rich (1991b). Cochleae: L14.