Improvements of sound localization abilities by the facial ruff of the barn owl (Tyto alba) as demonstrated by virtual ruff removal

Abstract

Background: When sound arrives at the eardrum it has already been filtered by the body, head, and outer ear. This process is mathematically described by the head-related transfer functions (HRTFs), which are characteristic for the spatial position of a sound source and for the individual ear. HRTFs in the barn owl (Tyto alba) are also shaped by the facial ruff, a specialization that alters interaural time differences (ITD), interaural intensity differences (ILD), and the frequency spectrum of the incoming sound to improve sound localization. Here we created novel stimuli to simulate the removal of the barn owl's ruff in a virtual acoustic environment, thus creating a situation similar to passive listening in other animals, and used these stimuli in behavioral tests.

Methodology/principal findings: HRTFs were recorded from an owl before and after removal of the ruff feathers. Normal and ruff-removed conditions were created by filtering broadband noise with the HRTFs. Under normal virtual conditions, no differences in azimuthal head-turning behavior between individualized and non-individualized HRTFs were observed. The owls were able to respond differently to stimuli from the back than to stimuli from the front having the same ITD. By contrast, such a discrimination was not possible after the virtual removal of the ruff. Elevational head-turn angles were (slightly) smaller with non-individualized than with individualized HRTFs. The removal of the ruff resulted in a large decrease in elevational head-turning amplitudes.

Conclusions/significance: The facial ruff a) improves azimuthal sound localization by increasing the ITD range and b) improves elevational sound localization in the frontal field by introducing a shift of iso-ILD lines out of the midsagittal plane, which causes ILDs to increase with increasing stimulus elevation. The changes at the behavioral level could be related to the changes in the binaural physical parameters that occurred after the virtual removal of the ruff. These data provide new insights into the function of external hearing structures and open up the possibility to apply the results on autonomous agents, creation of virtual auditory environments for humans, or in hearing aids.

Figure 1. Differences between HRTFs of owl H and owl 39.

The differences between the ITDs (upper panels) are shown for owl H and owl 39 with (A) intact and (B) removed ruff, respectively. Equivalent plots for ILD differences (lower panels) between owl H and owl 39 with (C) intact and (D) removed ruff feathers, respectively. Coloration is explained in the bar plots on the side. Blue areas mark positions where the ITDs or ILDs of owl 39 are smaller (nearer to 0 µs or dB) than those of owl H, red areas mean that the ITDs or ILDs of owl 39 are larger (>0 µs or dB). The differences in both, ITDs and ILDs, are larger when the facial ruff was removed than if it was intact (compare A to B and C to D).

Figure 2. Latencies.

Pooled latencies for the owls and stimulus positions are similarly distributed for trials using individualized HRTFs (dark gray, median at 125 ms marked by dark gray line), owl 39 normal (medium gray, median at 125 ms) and owl 39 ruffcut (light gray, median at 127 ms). Trials with latencies larger than 500 ms were excluded from the analysis, because they indicated low motivation or other distracters. No systematic effect of ruff removal on response latency was observed.

Figure 3. Azimuthal head-turn angles.

(A) Head-turn angles in degree (mean±SD) are plotted against the azimuthal-stimulus position in degree, exemplary for owl H with individualized HRTFs at 0° elevation (black) and for stimulation with HRTFs of owl 39 with intact ruff (blue). Localization of the exact stimulus position would result in a line with a slope of 1 (black straight line). The curved black and blue lines are Boltzman fits (see Data analysis) to the azimuthal head-turn angles. Head-turn angles differed only at -100° azimuth (blue asterisk, Mann-Whitney test, p<0.05) between the two stimulus conditions (for the other owls, see Figure S4). The ranges of the number of trials (n) per day point are indicated. (B) The azimuthal-localization error (difference between stimulus angle and head-turn angle) is plotted as a function of the azimuthal-stimulus angle. Responses to individualized HRTFs are shown in black, those to owl 39 normal in blue. Localization errors were smaller for small stimulus angles than for large stimulus angles, which are reflected by an increasing localization error with increasingly peripheral stimulus angles (see also Figure S4).

Figure 4. ITDs and azimuthal head-turn angle.

(A) The azimuthal head-turn angles were pooled for stimulation with individualized HRTFs (dotted, owls H and S), responses to owl 39 normal (black, all three owls) and to owl 39 ruffcut (blue, all three owls). Stepsize was 20°. Arrows mark ±140° stimulus position, where the azimuthal head-turn angle decreased highly significantly (Mann-Whitney test, p<0.0001) in the ruffcut condition. Note that, in contrast to the ruffcut condition, the head-turn angles with intact ruff (individualized and owl 39 normal) approach a plateau at about ±60°. Significant differences between stimulus conditions are marked with asterisks depending on the significance level (**p<0.01, ***p<0.001) in black (individualized versus owl 39 normal) respectively in blue (owl 39 normal versus owl 39 ruffcut). Each data point includes at least 96 trials, unless indicated otherwise by the number of trials (n). (B) The ITD in µs contained in the HRTFs at 0° elevation is plotted against stimulus azimuth in degree for owl 39 normal (black) and owl 39 ruffcut (blue). Note the sinusoidal course of the ITD. The ITD decreased at peripheral azimuths for both intact as well as for removed ruff. The ITD range was smaller in the ruffcut condition.

Figure 5. Prediction of azimuthal head-turns from the ITD.

The ITD contained at 0° stimulus elevation in the HRTFs of (A) owl H, (B) owl S, and (C) owl P were plotted against the azimuthal head-turn angle in degree. A linear regression (dotted line) through all head-turn angles shows that the owls responded well to the ITD in the HRTFs within the frontal field (≤±60°). Linear equations and goodness of fit (R²) of the regression are stated in each panel. With individualized HRTFs at ±140° (black arrows), however, the head-turn angles significantly deviated from the regression line (Mann-Whitney test, p<0.05).

Figure 6. Elevational head-turn angles.

The elevational head-turn angles (mean±SD) are plotted against stimulus angles for (A–C) owls H, S, and P individually as well as (D) the pooled data from all owls. Since the owls reacted to stimulus elevation even with individualized HRTFs only in the frontal area, stimulus angles of <±60° were included. With individualized HRTFs (A,B, black line and circles), there was a significant increase in the elevational turning angle with stimulus elevation (Mann-Whitney test, p<0.001). The slope of this increase was lower, but still significant with non-individualized HRTFs of owl 39 normal (blue). In the ruffcut condition (red), the slope was significantly different from 0 only for owl H, but neither for owl S nor for owl P. Owl P reacted similar to non-individualized HRTFs (gray: owl H's HRTFs, blue: owl 39 normal) as the other two owls, but located stimuli at 40° elevation lower than those at 0° elevation. However, the general characteristics of the elevational head-turn behavior were preserved in that the increase of head-turn angle with stimulus elevation was strongly reduced in the ruffcut condition for all owls compared to HRTFs recorded with intact ruff (D). The linear equations are given for each stimulus condition. For each pair of stimulus angles (−40° and 0°, 0° and 40°, and −40° and 40°), head-turn angles were compared with a Mann-Whitney test; significant differences are marked with asterisks (***p<0.001, **p<0.01, *p<0.05). Each data point includes at least 18 trials.

Figure 7. Elevational localization related to ILDs.

For the frontal field (±60°), elevational head-turn angles were plotted against the ILDs in (A) individualized and (B) non-individualized HRTFs with intact ruff. Both factors were significantly correlated at the indicated level (p) within the frontal field, where ILDs strongly varied with elevation (see Figure S4). (C) At more peripheral positions, where ILDs did no longer vary with elevation, ILDs and elevational head-turn angles were not correlated. (D) The same held for the ruffcut condition, where ILDs were not correlated with the elevational head-turn angle neither in the frontal field (D), nor in the periphery (data not shown).