Auditory processing of spectral cues for sound localization in the inferior colliculus

Abstract

The head-related transfer function (HRTF) of the cat adds directionally dependent energy minima to the amplitude spectrum of complex sounds. These spectral notches are a principal cue for the localization of sound source elevation. Physiological evidence suggests that the dorsal cochlear nucleus (DCN) plays a critical role in the brainstem processing of this directional feature. Type O units in the central nucleus of the inferior colliculus (ICC) are a primary target of ascending DCN projections and, therefore, may represent midbrain specializations for the auditory processing of spectral cues for sound localization. Behavioral studies confirm a loss of sound orientation accuracy when DCN projections to the inferior colliculus are surgically lesioned. This study used simple analogs of HRTF notches to characterize single-unit response patterns in the ICC of decerebrate cats that may contribute to the directional sensitivity of the brain's spectral processing pathways. Manipulations of notch frequency and bandwidth demonstrated frequency-specific excitatory responses that have the capacity to encode HRTF-based cues for sound source location. These response patterns were limited to type O units in the ICC and have not been observed for the projection neurons of the DCN. The unique spectral integration properties of type O units suggest that DCN influences are transformed into a more selective representation of sound source location by a local convergence of wideband excitatory and frequency-tuned inhibitory inputs.

Figure 1

Response maps for pure tones (top) and notched noise (bottom). Data were obtained from a DCN type IV unit (3.02 exp. 4/10/00, BF = 7 kHz) and an ICC type O unit (1.04 exp. 3/13/00, BF = 14 kHz). The bandwidth of the notch was 3.2 kHz. The horizontal lines represent average spontaneous rates. Black (gray) regions indicate stimulus conditions that elicited excitatory (inhibitory) discharge rates. The magnitude of the rate response is indicated by the scale bar at the bottom left of each map. Stimulus levels are noted in dB attenuation by numerical labels along the right margin of each map. 0 dB attenuation is near 100 dB SPL re 20 µPa for tones and 40 dB re 20 µPa/√Hz for noise spectrum level.

Figure 2

Frequency distribution of notch excitatory responses. A–C. Rate profiles of type O units with an excitatory peak below BF, surrounding BF, or above BF. The gray-filled regions indicate the range of spontaneous rates (SR); vertical bars mark BF. D. Scatterplot of peak frequency versus BF for all type O units in the present study. Notch bandwidths were 1.6 (squares) or 3.2 kHz (circles). The unity line indicates peaks that were centered on BF. Units above the ordinate did not show excitatory peaks (Xs).

Figure 3

Interactions of notch frequency and bandwidth on the excitatory responses of type O units. A, B. For comparison, data are shown for a representative type IV unit(4.01 exp. 9/5/00, BF = 19 kHz) and type O unit (1.06 exp. 6/24/99, BF = 21 kHz). Insets show the location of spectral edges for notches that elicited maximum excitatory responses from the type O unit. C. Scatterplot of the rising edge of excitatory notches versus BF for type O units with below-BF notch sensitivity. Symbols identify different notch bandwidths. The equation describes a statistically significant linear fit to the data. A unity line is superimposed to emphasize the slight negative displacement of edge frequency relative to BF. D. Responses of a type O unit (2.05 exp. 3/13/00, BF = 11 kHz) to spectral edges versus spectral notches. As illustrated by the inset, stimulus conditions were low-pass noise (LP, dotted line), high-pass noise (HP, dashed line), and notched noise that represented the sum of the bandpass stimuli (NN, solid line).

Figure 4

Response patterns of high-rate type O units. A, B. Pure-tone response map and notched noise rate profile for a representative high-rate type O unit (4.01 exp. 9/12/00). C. Scatterplot of peak frequency versus BF for all sampled high-rate type O units. Different symbols indicate different notch bandwidths: diamonds, 0.4 kHz; stars,0.8 kHz; squares, 1.6 kHz; and circles, 3.2 kHz. The diagonal line indicates peaks that were centered on BF. D. Responses of the type O unit to spectral edges versus spectral notches. The inset shows the complementary frequency arrangement of the three stimulus conditions.

Figure 5

Sensitivity of type V and I units to notched noise. Response maps for pure tones (top) and notched noise (bottom) for a type V unit (2.01 exp. 3/13/00) and type I unit (1.02 exp. 4/20/00). The bandwidth of the notch was 0.4 kHz for the type V unit, and 1.6 kHz for the type I unit.

Figure 6

Effects of notch bandwidth on the responses of type IV and type O units. A, B. Rate-level functions for a type O unit (1.12 exp. 4/30/97, BF = 23 kHz) and type IV unit (1.04 exp. 3/18/93, BF = 15 kHz) at four notch bandwidths (see legend in A). C. Effects of bandwidth on the average driven rates of type O units (filled circles) and type O units (squares) at a noise level 10 dB re threshold. D. Average driven rates at a 40-dB noise level. Data for type IV units are taken from Nelken and Young (1994).

Figure 7

Effects of bicuculline on the responses of a type O unit (2.18 exp. 7/12/99) to notch frequency sweeps. Discharge rates were recorded before (solid lines) and during (dashed lines) bicuculline administration at noise levels 10 and 40 dB above threshold. Ranges of spontaneous rates before and during drug application are indicated by the width of gray and stippled bars, respectively.

Figure 8

Circuit diagram summarizing hypothetical sources of input to a prototypic O unit. The tonotopic organization of excitatory (filled symbols) and inhibitory inputs (open symbols) are indicated at the bottom of the figure. IV: On-BF excitatory input from a DCN type IV unit. INH: Inhibitory input with below-BF frequency tuning. WBE: Excitatory input with wideband spectral integration properties.

Figure 9

Sensitivity of a type O unit to directional properties of the cat's head-related transfer function. A. HRTFs for three elevations (ELs) at 0° azimuth (AZ). The frequency of the large spectral notch between 8 and 15 kHz is related to the direction of the sound source in the median plane. B. HRTFs for three locations thatproduce the same 11-kHz notch frequency. C. Spatial receptive field for a type O unit (2.04 exp. 9/9/99, BF = 12.6 kHz). HRTF filter shapes were used to synthesize binaural “virtual space” stimuli at azimuths and elevations where gridlines intersect. The unit achieved its highest discharge rates along the spatial contour where midfrequency notches fell near 11 kHz (bold line).