Interaural correlation and the binaural summation of loudness

Sensitivity to a Break in Interaural Correlation in Frequency-Gliding Noises

This study was to investigate whether human listeners are able to detect a binaurally uncorrelated arbitrary-noise fragment embedded in binaurally identical arbitrary-noise markers [a break in correlation, break in interaural correlation (BIAC)] in either frequency-constant (frequency-steady) or frequency-varied (unidirectionally frequency gliding) noise. Ten participants with normal hearing were tested in Experiment 1 for up-gliding, down-gliding, and frequency-steady noises. Twenty-one participants with normal hearing were tested in Experiment 2a for both up-gliding and frequency-steady noises. Another nineteen participants with normal hearing were tested in Experiment 2b for both down-gliding and frequency-steady noises. Listeners were able to detect a BIAC in the frequency-steady noise (center frequency = 400 Hz) and two types of frequency-gliding noises (center frequency: between 100 and 1,600 Hz). The duration threshold for detecting the BIAC in frequency-gliding noises was significantly longer than that in the frequency-steady noise (Experiment 1), and the longest interaural delay at which a duration-fixed BIAC (200 ms) in frequency-gliding noises could be detected was significantly shorter than that in the frequency-steady noise (Experiment 2). Although human listeners can detect a BIAC in frequency-gliding noises, their sensitivity to a BIAC in frequency-gliding noises is much lower than that in frequency-steady noise.

The "Missing 6 dB" Revisited: Influence of Room Acoustics and Binaural Parameters on the Loudness Mismatch Between Headphones and Loudspeakers

Generations of researchers observed a mismatch between headphone and loudspeaker presentation: the sound pressure level at the eardrum generated by a headphone has to be about 6 dB higher compared to the level created by a loudspeaker that elicits the same loudness. While it has been shown that this effect vanishes if the same waveforms are generated at the eardrum in a blind comparison, the origin of the mismatch is still unclear. We present new data on the issue that systematically characterize this mismatch under variation of the stimulus frequency, presentation room, and binaural parameters of the headphone presentation. Subjects adjusted the playback level of a headphone presentation to equal loudness as loudspeaker presentation, and the levels at the eardrum were determined through appropriate transfer function measurements. Identical experiments were conducted at Oldenburg and Aachen with 40 normal-hearing subjects including 14 that passed through both sites. Our data verify a mismatch between loudspeaker and binaural headphone presentation, especially at low frequencies. This mismatch depends on the room acoustics, and on the interaural coherence in both presentation modes. It vanishes for high frequencies and broadband signals if individual differences in the sound transfer to the eardrums are accounted for. Moreover, small acoustic and non-acoustic differences in an anechoic reference environment (Oldenburg vs. Aachen) exert a large effect on the recorded loudness mismatch, whereas not such a large effect of the respective room is observed across moderately reverberant rooms at both sites. Hence, the non-conclusive findings from the literature appear to be related to the experienced disparity between headphone and loudspeaker presentation, where even small differences in (anechoic) room acoustics significantly change the response behavior of the subjects. Moreover, individual factors like loudness summation appear to be only loosely connected to the observed mismatch, i.e., no direct prediction is possible from individual binaural loudness summation to the observed mismatch. These findings – even though not completely explainable by the yet limited amount of parameter variations performed in this study – have consequences for the comparability of experiments using loudspeakers with conditions employing headphones or other ear-level hearing devices.

Human scalp evoked potentials related to the fusion between a sound source and its simulated reflection

The auditory system needs to fuse the direct wave (lead) from a sound source and its time-delayed reflections (lag) to achieve a single sound image perception. This lead-lag fusion plays crucial roles in auditory processing in reverberant environments. Here, we investigated neural correlates of the lead-lag fusion by tracking human cortical potentials evoked by a break in the correlation (BIC) between the lead and lag when the time delay between the two was 0, 2, or 4 ms. The BIC evoked a scalp potential consisting of an N1 and a P2 component. Both components were modulated by the delay. The effects of the delay on the amplitude of the two components were similar, an increase of the delay resulting in a decrease of the amplitude. In contrast, the delay differently modulated the latency of the two components, an increase of the delay resulting in an increase of the P2 latency but not an increase of the N1 latency. Similar to the P2 latency, the reaction time for subjective detection of the BIC also increased with the delay. These findings suggest that both the N1 and the P2 evoked by the BIC are neural correlates of the lead-lag fusion and that, relative to the N1, the P2 may be more closely related to listeners' perception of the fusion. Our study thus provides a neurophysiological and objective approach for investigating the fusion between the direct sound wave from a sound source and its reflections.

A Loudness Model for Time-Varying Sounds Incorporating Binaural Inhibition

This article describes a model of loudness for time-varying sounds that incorporates the concept of binaural inhibition, namely, that the signal applied to one ear can reduce the internal response to a signal at the other ear. For each ear, the model includes the following: a filter to allow for the effects of transfer of sound through the outer and middle ear; a short-term spectral analysis with greater frequency resolution at low than at high frequencies; calculation of an excitation pattern, representing the magnitudes of the outputs of the auditory filters as a function of center frequency; application of a compressive nonlinearity to the output of each auditory filter; and smoothing over time of the resulting instantaneous specific loudness pattern using an averaging process resembling an automatic gain control. The resulting short-term specific loudness patterns are used to calculate broadly tuned binaural inhibition functions, the amount of inhibition depending on the relative short-term specific loudness at the two ears. The inhibited specific loudness patterns are summed across frequency to give an estimate of the short-term loudness for each ear. The overall short-term loudness is calculated as the sum of the short-term loudness values for the two ears. The long-term loudness for each ear is calculated by smoothing the short-term loudness for that ear, again by a process resembling automatic gain control, and the overall loudness impression is obtained by summing the long-term loudness across ears. The predictions of the model are more accurate than those of an earlier model that did not incorporate binaural inhibition.

Testing and refining a loudness model for time-varying sounds incorporating binaural inhibition

This paper describes some experimental tests and modifications to a model of loudness for time-varying sounds incorporating the concept of binaural inhibition. Experiment 1 examined the loudness of a 100% sinusoidally amplitude-modulated 1000-Hz sinusoidal carrier as a function of the interaural modulation phase difference (IMPD). The IMPD of the test sound was 90° or 180° and that of the comparison sound was 0°. The level difference between the test and the comparison sounds at the point of equal loudness (the LDEL) was estimated for baseline levels of 30 and 70 dB sound pressure level and modulation rates of 1, 2, 4, 8, 16, and 32 Hz. The LDELs were negative (mean = -1.1 and -1.5 dB for IMPDs of 90° and 180°), indicating that non-zero IMPDs led to increased loudness. The original version of the model predicted the general form of the results, but there were some systematic errors. Modifications to the time constants of the model gave a better fit to the data. Experiment 2 assessed the loudness of unintelligible speech-like signals, generated using a noise vocoder, whose spectra and time pattern differed at the two ears. Both the original and modified models gave good fits to the data.

The detailed shapes of equal-loudness-level contours at low frequencies

High-resolution equal-loudness-level contours (ELCs) were measured over the frequency range 10-250 Hz using 19 normal-hearing subjects. Three levels of the 50-Hz reference sound were used, corresponding to the levels at 50 Hz of the 30-, 50-, and 70-phon standardized ELCs given in ISO-226:2003. The dynamic range of the contours generally decreased with increasing reference level, and the slope was shallow between 10 and 20 Hz, consistent with previous studies. For the lowest level, the ELCs were sometimes but not always smooth and on average followed the standardized 30-phon contour for frequencies above 40 Hz. For the two higher levels, the individual ELCs showed a distinct non-monotonic feature in a "transition region" between about 40 and 100 Hz, where the slope could reach near-zero or even positive values. The pattern of the non-monotonic feature was similar across levels for the subjects for whom it was observed, but the pattern varied across subjects. Below 40 Hz, the slopes of the ELCs increased markedly for all loudness levels, and the levels exceeded those of the standardized ELCs. Systematic deviations from the standardized ELCs were largest for frequencies below 40 Hz for all levels and within the transition region for the two higher levels.

Simple reaction time for broadband sounds compared to pure tones

Although many studies have explored the relation between reaction time (RT) and loudness, including effects of intensity, frequency, and binaural summation, comparable work on spectral summation is rare. However, most real-world sounds are not pure tones and typically have bandwidths covering several critical bands. Since comparing to a 1-kHz pure tone, the reference tone, is important for loudness measurement and standardization, the present work focuses on comparing RTs for broadband noise to those for 1-kHz pure tones in three experiments using different spectral and binaural configurations. The results of Experiments 1 and 2 yield good quantitative agreement with spectral loudness summation models for moderate and high sound pressure levels, measured using both pink noise covering almost the entire hearing range and bandpass-filtered pink noise with different center frequencies. However, at lower levels, the RT measurements yield an interaction of level and bandwidth, which is not in line with loudness scaling data. In Experiment 3, which investigated the binaural summation of broadband sounds, the binaural gain for white noise was determined to be 9 dB, which is somewhat larger than what had been found in previous RT measurements using 1-kHz pure tones.

Binaural Loudness Constancy

In binaural loudness summation, diotic presentation of a sound usually produces greater loudness than monaural presentation. However, experiments using loudspeaker presentation with and without earplugs find that magnitude estimates of loudness are little altered by the earplug, suggesting a form of loudness constancy. We explored the significance of controlling stimulation of the second ear using meatal occlusion as opposed to the deactivation of one earphone. We measured the point of subjective loudness equality (PSLE) for monaural vs. binaural presentation using an adaptive technique for both speech and noise. These stimuli were presented in a reverberant room over a loudspeaker to the right of the listener, or over lightweight headphones. Using the headphones, stimuli were either presented dry, or matched to those of the loudspeaker by convolution with impulse responses measured from the loudspeaker to the listener position, using an acoustic manikin. The headphone response was also compensated. Using the loudspeaker, monaural presentation was achieved by instructing the listener to block the left ear with a finger. Near perfect binaural loudness constancy was observed using loudspeaker presentation, while there was a summation effect of 3-6 dB for both headphone conditions. However, only partial constancy was observed when meatal occlusion was simulated. These results suggest that there may be contributions to binaural loudness constancy from residual low frequencies at the occluded ear as well as a cognitive element, which is activated by the knowledge that one ear is occluded.

The effect of stimulus bandwidth on binaural loudness summation

Binaural loudness summation is an important property of the human auditory system. This paper presents an experimental investigation of how binaural loudness summation varies with stimulus bandwidth. Loudness matches were obtained between dichotic stimuli, with interaural level differences (ILDs) of 2-12 dB, and diotic stimuli. The stimuli were noise bands with seven center frequencies and four bandwidths. Results showed that the loudness of dichotic stimuli increased nonlinearly with ILD, the increase being slightly less with broader bandwidths. There was a bandwidth-dependent difference between the listening tests results and the predictions of Moore and Glasberg's [(2007) J. Acoust. Soc. Am. 121, 1604-1612] loudness model. The size of the difference was, however, small. A characteristic function was derived describing how overall loudness depends on stimulus bandwidth and ILD.

Untrained listeners experience difficulty detecting interaural correlation changes in narrowband noises

Interaural correlation change detection was measured in untrained normal-hearing listeners. Narrowband (10-Hz) noises were varied by center frequency (CF; 500 or 4000 Hz) and diotic level roving (absent or present). For the 500-Hz CF, 96% of listeners could achieve threshold (79.4% correct at the easiest testing level) if roving was absent, but only 36% of listeners could if level roving was present. No one could achieve threshold at the 4000-Hz CF, unlike trained listeners in the literature. The results raise questions about how individual differences affect learning and generalization of monaural and binaural cues related to interaural correlation detection.

Similar Impacts of the Interaural Delay and Interaural Correlation on Binaural Gap Detection

The subjective representation of the sounds delivered to the two ears of a human listener is closely associated with the interaural delay and correlation of these two-ear sounds. When the two-ear sounds, e.g., arbitrary noises, arrive simultaneously, the single auditory image of the binaurally identical noises becomes increasingly diffuse, and eventually separates into two auditory images as the interaural correlation decreases. When the interaural delay increases from zero to several milliseconds, the auditory image of the binaurally identical noises also changes from a single image to two distinct images. However, measuring the effect of these two factors on an identical group of participants has not been investigated. This study examined the impacts of interaural correlation and delay on detecting a binaurally uncorrelated fragment (interaural correlation = 0) embedded in the binaurally correlated noises (i.e., binaural gap or break in interaural correlation). We found that the minimum duration of the binaural gap for its detection (i.e., duration threshold) increased exponentially as the interaural delay between the binaurally identical noises increased linearly from 0 to 8 ms. When no interaural delay was introduced, the duration threshold also increased exponentially as the interaural correlation of the binaurally correlated noises decreased linearly from 1 to 0.4. A linear relationship between the effect of interaural delay and that of interaural correlation was described for listeners participating in this study: a 1 ms increase in interaural delay appeared to correspond to a 0.07 decrease in interaural correlation specific to raising the duration threshold. Our results imply that a tradeoff may exist between the impacts of interaural correlation and interaural delay on the subjective representation of sounds delivered to two human ears.

Loudness of low-frequency pure tones lateralized by interaural time differences

Directional loudness is that phenomenon by which the loudness of a sound may vary according to the localization of its source. This phenomenon has been mainly observed for high-frequency sounds, for sources located in the horizontal plane. Because of the acoustic shadow of the head, the left and right ear pressures are modified depending on the source azimuth and the global loudness resulting from a summation process may vary accordingly. Directional loudness has also been reported to occur at 400 Hz, where shadowing effects are usually rather small. It might therefore be suspected that directional loudness effects could be influenced by other parameters involved in the localization process. This study assessed the loudness of low-frequency pure tones (200 and 400 Hz) lateralized with headphones by applying an interaural time difference (ITD) but no interaural level difference. It showed small but significant variations of loudness with respect to ITD at a low loudness level (40 phon): ITD values associated with virtual azimuths of ±60° and ±90° led to a small but significant increase in loudness (up to 1.25 dB). However, there was no such effect at a moderate loudness level (70 phon).

Development and current status of the "Cambridge" loudness models

This article reviews the evolution of a series of models of loudness developed in Cambridge, UK. The first model, applicable to stationary sounds, was based on modifications of the model developed by Zwicker, including the introduction of a filter to allow for the effects of transfer of sound through the outer and middle ear prior to the calculation of an excitation pattern, and changes in the way that the excitation pattern was calculated. Later, modifications were introduced to the assumed middle-ear transfer function and to the way that specific loudness was calculated from excitation level. These modifications led to a finite calculated loudness at absolute threshold, which made it possible to predict accurately the absolute thresholds of broadband and narrowband sounds, based on the assumption that the absolute threshold corresponds to a fixed small loudness. The model was also modified to give predictions of partial loudness—the loudness of one sound in the presence of another. This allowed predictions of masked thresholds based on the assumption that the masked threshold corresponds to a fixed small partial loudness. Versions of the model for time-varying sounds were developed, which allowed prediction of the masked threshold of any sound in a background of any other sound. More recent extensions incorporate binaural processing to account for the summation of loudness across ears. In parallel, versions of the model for predicting loudness for hearing-impaired ears have been developed and have been applied to the development of methods for fitting multichannel compression hearing aids.

Measurement and modeling of binaural loudness summation for hearing-impaired listeners

The summation of loudness across ears is often studied by measuring the level difference required for equal loudness (LDEL) of monaural and diotic sounds. Typically, the LDEL is ∼5-6 dB, consistent with the idea that a diotic sound is ∼1.5 times as loud as the same sound presented monaurally at the same level, as predicted by the loudness model of Moore and Glasberg [J. Acoust. Soc. Am. 121, 1604-1612 (2007)]. One might expect that the LDEL would be <5-6 dB for hearing-impaired listeners, because loudness recruitment leads to a more rapid change of loudness for a given change in level. However, previous data sometimes showed similar LDEL values for normal-hearing and hearing-impaired listeners. Here, the LDEL was measured for hearing-impaired listeners using narrowband and broadband noises centered at 500 Hz, where audiometric thresholds were near-normal, and at 3000 or 4000 Hz, where audiometric thresholds were elevated. The mean LDEL was 5.6 dB at 500 Hz and 4.2 dB at the higher center frequencies. The results were predicted reasonably well by an extension of the loudness model of Moore and Glasberg.

The perception of apparent auditory source width in hearing-impaired adults

In a previous study [Whitmer, Seeber and Akeroyd, J. Acoust. Soc. Am. 132, 369-379 (2012)], it was demonstrated that older hearing-impaired (HI) listeners produced visual sketches of headphone-presented noises that were insensitive to changes in interaural coherence. The current study further explores this insensitivity by comparing (a) binaural temporal fine-structure (TFS) resolution and (b) sound localization precision to (c) auditory source width judgments. Thirty-five participants aged 26-81 years with normal to moderately impaired hearing (a) discriminated interaurally phase-shifted tones from diotic tones presented over headphones, (b) located 500-ms speech-spectrum filtered click trains presented over loudspeakers between ±30° in quiet, and (c) sketched the perceived width of low-pass, high-pass, and speech-spectrum noise stimuli presented over loudspeakers from 0° and simultaneously from ±45° at attenuations of 0-20 dB to generate partially coherent stimuli. The results showed a decreasing sensitivity to width with age and impairment which was related to binaural TFS threshold: the worse one's threshold-which was correlated with age-the less the perceived width increased with decreasing interaural coherence. These results suggest that senescent changes to the auditory system do not necessarily lead to perceptions of broader, more diffuse sound images based on interaural coherence.

Apparent auditory source width insensitivity in older hearing-impaired individuals

Previous studies have shown a loss in the precision of horizontal localization responses of older hearing-impaired (HI) individuals, along with potentially poorer neural representations of sound-source location. These deficits could be the result or corollary of greater difficulties in discriminating spatial images, and the insensitivity to punctate sound sources. This hypothesis was tested in three headphone-presentation experiments varying interaural coherence (IC), the cue most associated with apparent auditory source width. First, thresholds for differences in IC were measured for a broad sampling of participants. Older HI participants were significantly worse at discriminating IC across reference values than younger normal-hearing participants. These results are consistent with senescent increases in temporal jitter. Performance decreased with age, a finding corroborated in a second discrimination experiment using a separate group of participants matched for hearing loss. This group also completed a third, visual experiment, with both a cross-mapping task where they drew the size of the sound they heard and the identification task where they chose the image that best corresponded to what they heard. The results from the visual tasks indicate that older HI individuals do not hear punctate images and are relatively insensitive to changes in width based on IC.

Contributions of von Békésy to psychoacoustics

This paper reviews the contributions of von Békésy to psychoacoustics, comparing his findings and interpretations to those that have emerged since his work. The areas covered include the perception of pitch for pure tones and complex tones, the effect of frequency on the apparent location of pure tones, estimation of the velocity of the traveling wave on the basilar membrane using judgments of lateralization, and the relative loudness of monaural and diotic sounds. While subsequent research has failed to replicate some of his findings, other findings have stood the test of time. There is no doubt that von Békésy made very substantial contributions to psychoacoustic research.

Binaural loudness summation for speech presented via earphones and loudspeaker with and without visual cues

Preliminary data [M. Epstein and M. Florentine, Ear. Hear. 30, 234-237 (2009)] obtained using speech stimuli from a visually present talker heard via loudspeakers in a sound-attenuating chamber indicate little difference in loudness when listening with one or two ears (i.e., significantly reduced binaural loudness summation, BLS), which is known as "binaural loudness constancy." These data challenge current understanding drawn from laboratory measurements that indicate a tone presented binaurally is louder than the same tone presented monaurally. Twelve normal listeners were presented recorded spondees, monaurally and binaurally across a wide range of levels via earphones and a loudspeaker with and without visual cues. Statistical analyses of binaural-to-monaural ratios of magnitude estimates indicate that the amount of BLS is significantly less for speech presented via a loudspeaker with visual cues than for stimuli with any other combination of test parameters (i.e., speech presented via earphones or a loudspeaker without visual cues, and speech presented via earphones with visual cues). These results indicate that the loudness of a visually present talker in daily environments is little affected by switching between binaural and monaural listening. This supports the phenomenon of binaural loudness constancy and underscores the importance of ecological validity in loudness research.

The effect of loudness on the reverberance of music: reverberance prediction using loudness models

This study examines the auditory attribute that describes the perceived amount of reverberation, known as "reverberance." Listening experiments were performed using two signals commonly heard in auditoria: excerpts of orchestral music and western classical singing. Listeners adjusted the decay rate of room impulse responses prior to convolution with these signals, so as to match the reverberance of each stimulus to that of a reference stimulus. The analysis examines the hypothesis that reverberance is related to the loudness decay rate of the underlying room impulse response. This hypothesis is tested using computational models of time varying or dynamic loudness, from which parameters analogous to conventional reverberation parameters (early decay time and reverberation time) are derived. The results show that listening level significantly affects reverberance, and that the loudness-based parameters outperform related conventional parameters. Results support the proposed relationship between reverberance and the computationally predicted loudness decay function of sound in rooms.

Generating partially correlated noise--a comparison of methods

There are three standard methods for generating two channels of partially correlated noise: the two-generator method, the three-generator method, and the symmetric-generator method. These methods allow an experimenter to specify a target cross correlation between the two channels, but actual generated noises show statistical variability around the target value. Numerical experiments were done to compare the variability for those methods as a function of the number of degrees of freedom. The results of the experiments quantify the stimulus uncertainty in diverse binaural psychoacoustical experiments: incoherence detection, perceived auditory source width, envelopment, noise localization/lateralization, and the masking level difference. The numerical experiments found that when the elemental generators have unequal powers, the different methods all have similar variability. When the powers are constrained to be equal, the symmetric-generator method has much smaller variability than the other two.

Trading of intensity and interaural coherence in dichotic pitch stimuli

When a signal is added to noise in the NoSπ binaural configuration, a reduction in interaural coherence, ρ, occurs at the signal frequency and increases in tone intensity decrease ρ. Corresponding manipulations of ρ result in the perception of a phantom signal which increases in loudness as ρ decreases [Culling et al. (2001). J. Acoust. Soc. Am. 110, 1020-1029]. In the present study, a narrow sub-band of noise (462-539 Hz) embedded within a broadband (0-3 kHz) diotic noise was manipulated in both intensity and ρ in a 3-interval, odd-one-out task. In the reference intervals, ρ was zero and the spectrum was flat. In the target interval, both ρ and the intensity of the target band were incremented giving opposing effects on loudness. Correct identification of the target interval followed a V-shape as a function of the size of intensity increment. The minimum of this function was often at chance performance, indicating that monaurally and binaurally evoked loudness were fully traded. These results show that reduction in ρ at a given frequency produces increased loudness at that frequency equivalent to up to 6 dB and consistent with an equalization-cancellation mechanism whose binaural output is strongly weighted compared to monaural excitation.

The loudness of sounds whose spectra differ at the two ears

Moore and Glasberg [(2007). J. Acoust. Soc. Am. 121, 1604-1612] developed a model for predicting the loudness of dichotic sounds. The model gave accurate predictions of data in the literature, except for an experiment of Zwicker and Zwicker [(1991). J. Acoust. Soc. Am. 89, 756-764], in which sounds with non-overlapping spectra were presented to the two ears. The input signal was noise with the same intensity in each critical band (bark). This noise was filtered into 24 bands each 1 bark wide. The bands were then grouped into wider composite bands (consisting of 1, 2, 4, or 12 successive sub-bands) and each composite band was presented either to one ear or the other. Loudness estimates obtained using a scaling procedure decreased somewhat as the number of composite bands increased (and their width decreased), but the predictions of the model showed the opposite pattern. This experiment was similar to that of Zwicker and Zwicker, except that the widths of the bands were based on the ERB(N)-number scale, and a loudness-matching procedure was used. The pattern of the results was consistent with the predictions of the model, showing an increase in loudness as the number of composite bands increased and their spacing decreased.

Interaural coherence for noise bands: waveforms and envelopes

This paper reports the results of experiments performed in an effort to find a formulaic relationship between the interaural waveform coherence of a band of noise gamma(W) and the interaural envelope coherence of the noise band gamma(E). An interdependence described by gamma(E)=pi/4+(1-pi/4)(gamma(W))(2.1) is found. This relationship holds true both in a computer experiment and for binaural measurements made in two rooms using a KEMAR manikin. Room measurements are used to derive a measure of reliability for the formula. Ultimately, a user who knows the waveform coherence can predict the envelope coherence with a small degree of uncertainty.