Sound Localization Mechanisms in the Auditory Cortex

Unlike the visual and somatosensory systems, spatial information is not directly represented at the sensory receptor epithelium in the auditory system. Spatial locations are computed by integrating neural binaural properties and frequency-dependent pinna filtering, providing a useful model to study how neural properties and peripheral structures are adapted for sensory encoding.

Although auditory cortex is necessary for sound localization, our understanding of how the cortex represents space remains rudimentary. The auditory cortex is necessary for sound localization.

The mechanisms that shape bicoordinate spatial representation in the auditory cortex remain unclear. Here we show that two functionally distinct regions of the pallid bat auditory cortex represent 2D space using different mechanisms.

Here, we addressed this issue by quantifying spatial receptive fields (SRFs) in two functionally distinct cortical regions in the pallid bat. The pallid bat uses echolocation for obstacle avoidance and listens to prey-generated noise to localize prey. Its cortex contains two segregated regions of response selectivity that serve echolocation and localization of prey-generated noise.

Pallid Bat

The main aim of this study was to compare 2D SRFs between neurons in the noise-selective region (NSR) and the echolocation region [frequency-modulated sweep-selective region (FMSR)].

Sound Localization in Bats: Echolocation and Neural Mechanisms

Key Differences Between NSR and FMSR

The data reveal the following major differences between these two regions:

Compared with NSR neurons, SRF properties of FMSR neurons were more strongly dependent on sound level.
As a population, NSR neurons represent a broad region of contralateral space, while FMSR selectivity was focused near the midline at sound levels near threshold and expanded considerably with increasing sound levels.
The SRF size and centroid elevation were correlated with the characteristic frequency in the NSR, but not the FMSR.

These data suggest different mechanisms of sound localization for two different behaviors.

Previously, we reported that azimuth is represented by predictable changes in the extent of activated cortex. The present data indicate how elevation constrains this activity pattern.

Spatial Receptive Field (SRF) Properties

The SRF of a NSR neuron recorded at MT + 10 dB illustrates the various properties quantified in this and subsequent figures. The snout of the bat is at 0° azimuth/elevation (white ellipse). The x-axis (azimuth) ranges from 75° IL (negative angles) to 75° CL (positive angles) in 15° increments (the 30° and 60° axis markers are omitted for clarity). The y-axis (elevation) ranges from 60° below to 60° above the snout of the bat in 30° increments. The color scale to the right indicates the response magnitude (total number of spikes to 20 stimulus repetitions) within the SRF. The thin dashed line marks the gyradius (40° for this neuron). The thicker dashed line delineates the 50% SRF defined as the contour within which response was >50% of maximum response. The “plus” sign marks the centroid azimuth/elevation of the SRF.

SRF of NSR Neuron

Stability of NSR Neuron SRF Properties

The SRF properties of NSR neurons were stable with sound level. Each panel shows the 2D SRF of a NSR neuron. Left column, Responses were calculated for noise presented at MT + 10 dB. Right column, Responses from the same neuron shown to the left, but measured at MT + 20 dB. The response magnitude color chart is shown to the right of each neuron and corresponds to both the MT + 10 dB and MT + 20 dB plots.

A, B, Neuron #41B05: CF = 19 kHz; the centroid azimuth at MT + 10 dB and MT + 20 dB was 54° and 53°, respectively. The centroid elevation at the two sound levels was −5° and −9°. The gyradii at the two different sound levels were 44° and 46°.
C, D, Neuron #70C02: CF = 30 kHz; centroid azimuth = 28°; 35°; centroid elevation = 42°, 37°; gyradius = 28°, 36°.
E, F, Neuron #83A01: CF = 21 kHz; centroid azimuth = 36°, 36°; centroid elevation = 3°, 2°; gyradius = 51°, 50°.

SRF Properties of NSR Neurons

Sound Level Dependence in FMSR Neurons

The SRF properties of FMSR neurons were strongly sound level dependent. The left and right columns correspond to SRFs recorded from the same FMSR neurons, but at MT + 10 dB and MT + 20 dB, respectively. The 50% SRF contours are not shown for neurons in E-H because of the fragmenting of the SRF into multiple peaks at the higher SPL.

A, B, Neuron #96A01: CF = 45 kHz; centroid azimuth = 11°, 9°; centroid elevation = 4°, −3°; gyradius = 26°, 40°.
C, D, Neuron #77A02: CF = 48 kHz; centroid azimuth = 9°, 14°; centroid elevation = −1°, −1°; gyradius = 41°, 54°.
E, F, Neuron #36A04: CF = 35 kHz; centroid azimuth = 13°, 7°; centroid elevation = 10°, −3°; gyradius = 38°, 54°.

SRF Properties of FMSR Neurons

SRF Size Comparison

SRFs of NSR neurons were larger in elevation than azimuth. The 50% SRF is the width of the SRF, inside of which the response was within 50% of the maximum. The 50%SRF-AZ and 50%SRF-EL were calculated as the SRF width, with the centroid as reference.

Centroid Distribution

Across the population, NSR neuron centroids were more broadly distributed in 2D space compared with FMSR neurons.

A, B: Distribution of centroid azimuths in FMSRs and NSRs recorded at 10 dB (A) and 20 dB (B) above threshold.
C, D: The distribution of centroid elevation in FMSRs and NSRs recorded at 10 dB (C) and 20 dB (D) above threshold.

The dashed lines in A and B demarcate ipsilateral (IPSI) and contralateral (CONTRA) azimuth locations.

Centroid Distribution

Level Tolerance

FMSR neurons were less level tolerant than NSR neurons.

A: The gyradius instability index indicates the extent to which the gyradius increased per decibel increase in SPL.
B: Centroid azimuth instability index shows the degree to which the centroid azimuth changed position per decibel increase in SPL.
C: The centroid elevation instability index shows the degree to which the centroid elevation changed position per decibel increase in SPL.

Level Tolerance

Relationship Between CF and SRF Properties in NSR Neurons

SRFs of eight NSR neurons that exemplify the relationship between CF and centroid elevation wherein the high-CF neurons have centroids at higher elevation. The gyradius, indicative of SRF size, also decreased with CF in these neurons. All SRFs were obtained with broadband noise at 20 dB above threshold.

A, Neuron #87A01: CF = 12 kHz, centroid azimuth = 42°, centroid elevation = 3°, gyradius = 47°.
B, Neuron #68B04: CF = 18 kHz, centroid azimuth = 40°, centroid elevation = 3°, gyradius = 52°.
C, Neuron #93B03: CF = 18 kHz, centroid azimuth = 9°, centroid elevation = 0°, gyradius = 61°.
D, Neuron #41B03: CF = 20 kHz, centroid azimuth = 55°, centroid elevation = 8°, gyradius = 45°.
E, Neuron #40A01: CF = 24 kHz, centroid azimuth = 37°, centroid elevation = 19°, gyradius = 40°.
F, Neuron #34A01: CF = 24 kHz, centroid azimuth = 21°, centroid elevation = 17°, gyradius = 41°.
G, Neuron #40B01: CF = 29 kHz, centroid azimuth = 32°, centroid elevation = 27°, gyradius = 35°.

CF and SRF Properties in NSR Neurons

Relationship between CF and centroid azimuth or centroid elevation of NSR and FMSR neurons recorded at two different sound levels, MT + 10 dB (left column) and MT + 20 dB (right column).

A, B: In NSR neurons, the CF and centroid elevation were significantly correlated at both sound levels.
C, D: In NSR neurons, no significant relationship was observed between CF and centroid azimuth.

Relationship Between CF and SRF Properties

Activity Distribution in NSR

Schematic representation of how activity distribution in the NSR changes based on overlapping ILD and frequency maps, and the 2D location of a source generating broadband noise. Each panel shows the same cortical region but with activity patterns for specific 2D locations in space (indicated above each panel). Sound source moves from IL to CL space in the left-column-to-right-column direction and from low to high elevations in the bottom-row-to-top-row direction. Cortical orientation is provided in the bottom left panel. M, Medial; L, lateral; R, rostral; C, caudal. The color saturation scales with activity levels. The diagonal arrow indicates that the tonotopic direction in the NSR is in the caudolateral-to-rostromedial direction.

Activity Distribution in NSR