Peter Xinya Zhang

On the ability of human listeners to distinguish between front and back

In order to determine whether a sound source is in front or in back, listeners can use location-dependent spectral cues caused by diffraction from their anatomy. This capability was studied using a precise virtual reality technique (VRX) based on a transaural technology. Presented with a virtual baseline simulation accurate up to 16 kHz, listeners could not distinguish between the simulation and a real source. Experiments requiring listeners to discriminate between front and back locations were performed using controlled modifications of the baseline simulation to test hypotheses about the important spectral cues. The experiments concluded: (1) Front/back cues were not confined to any particular 1/3rd or 2/3rd octave frequency region. Often adequate cues were available in any of several disjoint frequency regions. (2) Spectral dips were more important than spectral peaks. (3) Neither monaural cues nor interaural spectral level difference cues were adequate. (4) Replacing baseline spectra by sharpened spectra had minimal effect on discrimination performance. (5) When presented with an interaural time difference less than 200 micros, which pulled the image to the side, listeners still successfully discriminated between front and back, suggesting that front/back discrimination is independent of azimuthal localization within certain limits.

Binaural models and the strength of dichotic pitches

Modern physiologically based models of the binaural system incorporate internal delay lines in the pathways from left and right peripheries to central processing nuclei. Different binaural models for the formation of dichotic pitch employ these delay lines in different ways. Consequently, the different models make different predictions for the relative strengths of dichotic pitches made with particular phase conditions. The differences are magnified for dichotic pitches at low frequencies where especially long delay lines may be required. Data from four low-frequency pitch strength experiments on pure-tone-like dichotic pitches (two on Huggins pitch and two on binaural coherence edge pitch) are consistent with models of the equalization-cancellation type and not consistent with the central activity pattern model.

Lateralization of sine tones–interaural time vs phase

Listeners estimated the lateral positions of 50 sine tones with interaural phase differences ranging from -150 degrees to +150 degrees and with different frequencies, all in the range where signal fine structure supports lateralization. The estimates indicated that listeners lateralize sine tones on the basis of interaural time differences and not interaural phase differences.

Transaural experiments and a revised duplex theory for the localization of low-frequency tones

The roles of interaural time difference (ITD) and interaural level difference (ILD) were studied in free-field source localization experiments for sine tones of low frequency (250-750 Hz). Experiments combined real-source trials with virtual trials created through transaural synthesis based on real-time ear canal measurements. Experiments showed the following: (1) The naturally occurring ILD is physically large enough to exert an influence on sound localization well below 1000 Hz. (2) An ILD having the same sign as the ITD modestly enhances the perceived azimuth of tones for all values of the ITD, and it eliminates left-right confusions that otherwise occur when the interaural phase difference (IPD) passes 180°. (3) Increasing the ILD to large, implausible values can decrease the perceived laterality while also increasing front-back confusions. (4) Tone localization is more directly related to the ITD than to the IPD. (5) An ILD having a sign opposite to the ITD promotes a slipped-cycle ITD, sometimes with dramatic effects on localization. Because the role of the ITD itself is altered by the ILD, the duplex processing of ITD and ILD reflects more than mere trading; the effect of the ITD can be reversed in sign.

Lateralization of the Huggins pitch

The lateralization of the Huggins pitch (HP) was measured using a direct estimation method. The background noise was initially N0 or Nπ, and then the laterality of the entire stimulus was varied with a frequency‐independent interaural delay, ranging from −1 to +1 ms. Two versions of the HP boundary region were used, stepped phase and linear phase. When presented in isolation, without the broadband background, the stepped boundary can be lateralized on its own but the linear boundary cannot. Nevertheless, the lateralizations of both forms of HP were found to be almost identical functions both of the interaural delay and of the boundary frequency over a two‐octave range. In a third experiment, the same listeners lateralized sine tones in quiet as a function of interaural delay. Good agreement was found between lateralizations of the HP and of the corresponding sine tones. The lateralization judgments depended on the boundary frequency according to the expected hyperbolic law except when the frequency‐independent delay was zero. For the latter case, the dependence on boundary frequency was much slower than hyperbolic. [Work supported by the NIDCD grant DC 00181.]

Earedness: Left‐eared and right‐eared listeners

The Huggins pitch (HP) stimulus known as HP‐ is created with a broadband background noise having an interaural phase difference of zero, together with a narrow boundary region wherein the interaural phase varies with frequency. At the spectral center of the boundary region the interaural phase is 180 deg. Therefore, HP‐ is symmetrical with respect to the two ears. Despite the symmetry, most listeners hear the HP image strongly lateralized to one side of the head. Some hear it on the right; others hear it on the left. Two surveys, involving 51 listeners, found that these perceptions do not change when the headphones are reversed. Extensive experiments with five listeners found that the lateralization directions were usually insensitive to variations in the frequency of the boundary region (more than two octaves). The left or right preference was strong enough that listeners chose alias locations (differing from a more central location by 360 deg) on the preferred side when various frequency‐independent interaural delays and phase shifts were added to the HP stimulus. The experiments suggest that given ambiguous stimuli, listeners exhibit earedness—a preference similar to, but not as strong as, handedness. [Work supported by the NIDCD Grant DC 00181.]

Salient phase density model for the lateralization of binaural pitches

Binaural pitches are sensations of pitch generated from a broadband noise having a narrowband boundary region where the interaural phase difference varies rapidly as a function of frequency. These pitch sensations are perceived to be lateralized in the head. A binaural model, the Salient Phase Density model, has been derived to predict the lateralizations. The model begins by separating the boundary from the background. Because the boundary is narrow, the boundary components can be represented statistically by a density. Components of the boundary with phases similar to the background are discounted in the density as not salient. The density is then used to compute the cross‐correlation function of the boundary region. The predicted lateralization is based on the interaural time difference of a template that best matches the cross‐correlation of the boundary. The model was tested against psychoacoustical experiments using four types of Huggins pitch boundary‐linear phase, stepped phase, uncorrelated phase...

Cancelled harmonics—How high does the effect go?

Demonstration Number 1 in the IPO‐NIU‐ASA collection of auditory demonstrations (compact disk) periodically cancels and reinserts a harmonic of a complex tone having a 200‐Hz fundamental and 20 equal‐amplitude harmonics. This procedure causes a listener to hear out the manipulated harmonic as a separate tone. In this way, the demonstration exposes harmonics 1 through 10. The following question arises: What is the highest harmonic that can be made audible, and what is responsible for the limitation? Listening experiments, using random harmonic phases, fundamental frequencies (f0) from 50 to 2000 Hz, and a maximum harmonic frequency of 20 kHz, show that for high fundamental frequencies (f0>200 Hz) the highest audible harmonic frequency is insensitive to f0 and is only about 10 percent less than the highest audible sine frequency in quiet. For lower fundamental frequencies, the highest audible harmonic number tends to be insensitive to f0 and is 50–70 for normal hearing listeners. In this region the highest ...

On the lateralization of the Huggins pitch

The central activity pattern (CAP) model of Raatgever and Bilsen [J. Acoust. Soc. Am. 80, 429–441 (1986)] correctly predicts that Huggins pitch (HP+) is lateralized in the center whereas HP− is lateralized to the left or the right. Experiments show that some listeners (left‐eared listeners) always hear the pitch sensation on the left and others always hear it on the right. Still others can hear it on one side or the other. The CAP model also predicts that the laterality of HP− should follow a hyperbolic function of the boundary frequency. To test this prediction, laterality was measured in careful laterality estimation experiments, wherein HP− was combined with a set of interaural time differences (ITDs). Although laterality estimates followed predictions for finite (ITDs), on those trials where the ITD was zero the hyperbolic law was violated for five out of five listeners. Instead, the laterality of HP− was very insensitive to the boundary frequency over the range tested, 200 to 1000 Hz. A search for a ...

Lateralization of sine tones—Interaural time versus interaural phase

Five listeners estimated the lateral positions of 50 sine tones in a headphone experiment designed to determine whether the human sense of sound location correlates better with the interaural phase difference (IPD) or the interaural time difference (ITD). In any experimental block the IPD values ranged from −150 to +150 degrees, and the frequencies were chosen such that the ITDs ranged from −1000 to +1000 microseconds. The frequencies were all in the range where human listeners are known to be able to lateralize tones based on the ITD in the waveform fine structure. It was found that the lateralization responses correlated with the ITD much better than with the IPD. The average variance was five times smaller for the ITD hypothesis compared to the IPD hypothesis, and only the ITD hypothesis led to a well‐fitting (compressive) function. For the ITD function, individual compressive exponents varied considerably and averaged 0.75. For the IPD function the exponents were too small to be meaningful. Comparison...