Masking Period Patterns Research Articles

Masking Period Patterns and Forward Masking for Speech-Shaped Noise: Age-Related Effects.

The purpose of this study was to assess age-related changes in temporal resolution in listeners with relatively normal audiograms. The hypothesis was that increased susceptibility to nonsimultaneous masking contributes to the hearing difficulties experienced by older listeners in complex fluctuating backgrounds. Participants included younger (n = 11), middle-age (n = 12), and older (n = 11) listeners with relatively normal audiograms. The first phase of the study measured masking period patterns for speech-shaped noise maskers and signals. From these data, temporal window shapes were derived. The second phase measured forward-masking functions and assessed how well the temporal window fits accounted for these data. The masking period patterns demonstrated increased susceptibility to backward masking in the older listeners, compatible with a more symmetric temporal window in this group. The forward-masking functions exhibited an age-related decline in recovery to baseline thresholds, and there was also an increase in the variability of the temporal window fits to these data. This study demonstrated an age-related increase in susceptibility to nonsimultaneous masking, supporting the hypothesis that exacerbated nonsimultaneous masking contributes to age-related difficulties understanding speech in fluctuating noise. Further support for this hypothesis comes from limited speech-in-noise data, suggesting an association between susceptibility to forward masking and speech understanding in modulated noise.

Masking Period Patterns and Forward Masking for Speech-Shaped Noise: Age-Related Effects

Masking Period Patterns and Forward Masking for Speech-Shaped Noise: Age-Related Effects

Testing a nonlinear computational channel model for masker phase effects

The masked threshold differences produced by Schroeder-phase maskers are most often attributed to the non-linear response of the normal cochlea (Summers et al., 2003). The nonlinear properties of the basilar membrane (BM) cause different response to the positive and negative Schroeder-phase maskers (i.e., + SCHR and -SCHR maskers, respectively) based on signal level, temporal synchrony, and on- and off-frequency harmonic components (Kohlrausch and Sander, 1995, Carlyon and Datta, 1997a,b, and Summers, 2000). In this study, manipulation of harmonic components of the maskers was used to explore nonlinear aspects of BM motion produced by the two different maskers. Specifically, masking period patterns (MPPs) for the + SCHR and -SCHR maskers were measured to show the influence of the phase relationships, and thus, the spectrotemporal characteristics of harmonic complexes on masking effectiveness. An enhanced channel model (Oh, 2013) provides a quantitative explanation for masking differences between + SCHR and -SCHR maskers by introducing nonlinear channel correlation across both frequency and time and a nonlinear decision criterion. [Research supported by a grant from the Office of Naval Research #N000140911017.]

Investigating temporal asymmetry using masking period patterns and models of peripheral auditory processing

Two experiments were conducted in conjunction with modeling to evaluate the role of peripheral nonlinearity and neural adaptation in the perception of temporally asymmetric sounds. In both experiments, maskers were broadband noises amplitude modulated with ramped and damped exponential modulators that repeated at 40 Hz. Masking period patterns (MPPs) were constructed by measuring detection threshold of a 5-ms, 1000-Hz tone burst as function of the signal's onset delay. Experiment I showed that varying modulator half-life from 1 to 16 ms led to differences in the damped and the ramped MPPs that were largest at the short half-lives and diminished at the longer half-lives. When masker level was varied (experiment II), the largest difference between ramped and damped MPPs occurred at moderate stimulus levels. Two peripheral auditory models were evaluated, one a simple auditory filter followed by a power-law nonlinearity and another, a model of auditory nerve processing [J. Acoust. Soc. Am. 126, 2390-2412 (2009)] that includes neural adaptation. Neither models predicted differences between the ramped and damped MPPs, providing indirect support that the central auditory system has a role in perceptual temporal asymmetry.

Relationship between basilar membrane compression estimates and masking release.

Listeners with cochlear hearing loss often obtain less benefit, or masking release (MR), than do normal-hearing listeners when a steady-state masker is replaced by a temporally fluctuating one, even when overall audibility differences between the two listener groups are taken into account. One possible explanation for the reduced MR seen in these listeners is loss of the active cochlear process that results in less compression and hence less amplification for speech segments that coincide with temporal gaps in the masker. This study tests whether varying degrees of cochlear compression can predict the degree of loss of MR. Behavioral estimates of compression, using temporal masking curves, were compared with the magnitude of MR for speech and for pure tones in a group of hearing-impaired listeners and a control group of age-matched noise-masked normal-hearing listeners. Compression estimates were made at 500, 1500, and 4000 Hz. Masking period patterns in a 10-Hz square-wave-gated speech-shaped noise were measured, as was MR for band-limited (500–4000-Hz) IEEE sentences, using the same temporally gated noise. The results will be used to test for a relationship between masking release and residual cochlear compression. [Work supported by NIH grants R01DC03909 and R01DC008306.]

Masking period patterns of exponentially ramped and damped noises in hearing‐impaired listeners.

Noises that are modulated with either an exponentially rising (ramped), repeating modulator or a falling (damped), repeating modulator may evoke strong perceptual differences. To explore a possible cochlear contribution to these perceptual differences, masking period patterns (MPPs) of ramped and damped noises were measured for normal‐hearing listeners at different stimulus levels and for hearing‐impaired listeners. By time‐reversing the ramped MPP, comparisons can be made between thresholds for which the masker power was the same between ramped and damped noises. Previous results showed that for normal‐hearing listeners, the MPP of a ramped noise is considerably shallower than the MPP of a damped noise. The present results indicated that the differences between the damped and the time‐reversed ramped MPPs were the largest at moderate stimulus levels and the smallest at low and high stimulus levels. Listeners with hearing loss exhibited no difference in the damped versus time‐reversed ramped MPPs. Results...

Prediction of initial-phase effects of harmonic complex maskers on masking period patterns using auditory model

Prediction of initial-phase effects of harmonic complex maskers on masking period patterns using auditory model

The effect of basilar-membrane nonlinearity on the shapes of masking period patterns in normal and impaired hearing.

Masking period patterns (MPPs) were measured in listeners with normal and impaired hearing using amplitude-modulated tonal maskers and short tonal probes. The frequency of the masker was either the same as the frequency of the probe (on-frequency masking) or was one octave below the frequency of the probe (off-frequency masking). In experiment 1, MPPs were measured for listeners with normal hearing using different masker levels. Carrier frequencies of 3 and 6 kHz were used for the masker. The probe had a frequency of 6 kHz. For all masker levels, the off-frequency MPPs exhibited deeper and longer valleys compared with the on-frequency MPPs. Hearing-impaired listeners were tested in experiment 2. For some hearing-impaired subjects, masker frequencies of 1.5 kHz and 3 kHz were paired with a probe frequency of 3 kHz. MPPs measured for listeners with hearing loss had similar shapes for on- and off-frequency maskers. It was hypothesized that the shapes of MPPs reflect nonlinear processing at the level of the basilar membrane in normal hearing and more linear processing in impaired hearing. A model assuming different cochlear gains for normal versus impaired hearing and similar parameters of the temporal integrator for both groups of listeners successfully predicted the MPPs.

Measurement of auditory temporal processing using modified masking period patterns.

A common metric of auditory temporal processing is the difference in the threshold for a pure-tone signal masked by either unmodulated or amplitude-modulated noise. This technique may be viewed as a modification of the masking period pattern technique. Such measurements have been proposed as an efficient means of estimating auditory temporal resolution in a clinical setting, although in many cases threshold differences may reflect additional spectro-temporal processes. The primary purpose of the present experiment was to examine interactions among signal frequency and masker bandwidth and the effects of modulation frequency on modified masking period patterns. The results revealed unmodulated-modulated threshold differences that increased with increasing masker bandwidth and decreased with increasing modulation frequency. There was little effect of signal frequency for narrow-band noise maskers that were equal in absolute bandwidth across frequency. However, unmodulated-modulated threshold differences increased substantially with increasing signal frequency for bandwidths proportional to the signal frequency and for wideband maskers. Although the results are interpreted in terms of a combination of both within-channel and across-channel cues, the specific contributions of these cues in particular conditions are difficult to ascertain. Because modified masking period patterns depend strongly upon a number of specific stimulus parameters, and because it is difficult to determine with any precision the underlying perceptual processes, this technique is not recommended for use as a clinical measure of auditory temporal processing.

Effects of hearing impairment and presentation level on masking period patterns for Schroeder-phase harmonic complexes.

Masking period patterns (MPPs) for Schroeder-phase harmonic complexes containing equal-amplitude harmonics of a 100-Hz fundamental were determined for 5-ms tonal probes at 4,000 and 1,000 Hz. Maskers consisted of harmonics 2-50 (200-5,000 Hz bandwidth) for 4,000-Hz probes and harmonics 2-20 (200-2,000 Hz) for 1,000-Hz signals. Masked thresholds were determined for probe onsets 153, 155.5, 158, 160.5, and 163 ms following masker onset (masker duration=460 ms). Overall, results were similar for both probe frequencies. For listeners with normal hearing, MPPs for positive Schroeder-phase complexes masking 60 dB SPL probes were highly modulated and became flatter when probe level was increased to 80 dB SPL. MPPs were less modulated for listeners with sensorineural hearing loss than for normally hearing listeners at both 60 and 80 dB SPL probe levels. Thresholds in negative Schroeder-phase maskers were more similar across the two groups of listeners and across differences in probe position and probe level. The findings support an interpretation involving differences in the shape of the basilar-membrane waveform generated by each masker and possible influences of nonlinear cochlear processing on these internal responses. For normally hearing listeners, 60 dB SPL probes were most difficult to detect when temporally positioned so that probe frequency and masker instantaneous frequency were closely matched. For 80 dB SPL probes and for hearing-impaired listeners, probes presented at these same positions were often more easily detected than probes at other positions. The latter result appears to involve benefit associated with in-phase addition of the probe to a portion of the masker similar to the probe in both frequency and phase. This benefit was reduced or entirely eliminated when probe phase was altered so that this in-phase addition did not occur.

Masking period patterns in listeners with cochlear hearing loss

Masking period patterns (MPPs) were measured in listeners with hearing loss of cochlear origin, using a short 6-kHz probe presented at different times within the envelope of a 100% sinusoidally amplitude-modulated (AM) masker (4-Hz modulation rate). The carrier frequency of the AM masker was 3 kHz (off-frequency masking) or 6 kHz (on-frequency masking). MPPs obtained from hearing-impaired listeners were compared to those obtained from normal-hearing listeners using the same stimuli. A model used previously to simulate MPPs from normal-hearing listeners [Wojtczak et al., J. Acoust. Soc. Am. 106, 2147 (1999)] was used to predict data from the hearing-impaired listeners. In contrast to MPPs from normal-hearing ears, MPPs from ears with moderate to profound hearing loss did not exhibit large differences in MPP shapes for on- and off-frequency maskers, i.e., they did not exhibit longer MPP valleys for the off-frequency masking situation. Results of the modeling suggest that the similar shapes for on- and off-frequency MPPs reflect a lack of compression (or reduced compression) in the hearing-impaired listeners. The implications of this effect for speech understanding in the presence of competing talkers will be discussed. [Work suported by NIH NIDCD DC00149 and the Lions 5M Hearing Foundation.]

Effect of BM compression on masking period patterns (MPPs)

MPPs were measured in listeners with normal and impaired hearing using tonal 4-Hz SAM maskers and short tonal probes with frequencies that were either identical to or higher than the carrier frequency of the masker. The probe frequencies were 500, 1200, and 3000 Hz for on-frequency masking, and 1200, 2400, and 6000 Hz for off-frequency masking. In normal-hearing listeners MPPs measured with off-frequency probes had valleys that were much longer and deeper than valleys observed with on-frequency probes. A similar result was observed in hearing-impaired listeners in the frequency region of mild hearing losses, where significant residual compression was presumably operating. However, in the frequency region with substantial hearing loss where compression is substantially reduced or absent, MPPs measured with the on- and off-frequency probes were very similar. A model consisting of peripheral filtering, compressive nonlinearity, and a sliding temporal window was used in an attempt to predict the data and to estimate the compression index. The results suggest that the similarity of on- and off-frequency temporal resolution in hearing-impaired listeners may be due in part to the lack of the compressive nonlinearity that is evident at the level of the basilar membrane in normal-hearing listeners. [Work supported by NIH-NIDCD DC00149.]

Benefits of nonlinear active gain for detection of brief probes in Schroeder-phase complex maskers

Masking period patterns for positive and negative Schroeder-phase complexes were determined for tonal probes at 1 and 4 kHz. Maskers included components from 200–5000 Hz of a 100-Hz fundamental. The 5-ms probes had onsets occurring 153, 155.5, 158, 160.5, or 163 ms following masker onset. For listeners with normal hearing, thresholds in the positive Schroeder-phase masker varied by as much as 25 dB depending on the position of the probe within the masker period. Thresholds in the negative Schroeder-phase maskers were nearly constant as probe onset varied. The maskers were more nearly equal in effectiveness for listeners with sensorineural hearing loss and in testing at high presentation levels. The findings support an interpretation involving differences in the shape of the basilar-membrane waveform generated by each masker, and influences of nonlinear active gain on these internal responses. The positive Schroeder-phase masker appears to produce a highly modulated basilar membrane response. Nonlinear active gain enhances the signal-to-masker ratio for signals presented during low-amplitude ‘‘troughs’’ of these modulated internal waveforms. Consistent with this account of the present findings, active gain is reduced or eliminated at high presentation levels and with cochlear damage.

Masking period patterns of Schroeder-phase complexes: effects of level, number of components, and phase of flanking components.

Masking period patterns (MPPs) were obtained for maskers consisting of harmonics 2-20 of a 100-Hz fundamental. The signal was always a 5-ms 1100-Hz sinusoid presented 152, 154, 156, 158, or 160 ms after the start of a 400-ms masker. Experiment 1 replicated the finding that, for a masker level of 69 dB component, the shape of the MPP depended strongly on the phases of the components: Summing them in positive Schroeder phase led to a threshold variation of about 18 dB across the MPP, but summing them in negative Schroeder phase produced a flat MPP [A. Kohlrausch and A. Sander, J. Acoust. Soc. Am. 97, 1817-1829 (1995)]. Reducing the level of the positive-phase masker resulted in a systematic flattening of the MPP, whereas the negative-phase MPPs were flat both at high and at low levels. Experiment 2 showed that removing all components of a positive-phase masker except those close to the signal raised thresholds at the minimum of the MPP. In contrast, a similar manipulation applied to the negative-phase masker produced a uniform elevation of the MPP. Experiment 3 showed that an analogous effect could be obtained by manipulating the phases of masker components remote from the signal. It is shown that several features of the data can be simulated using a nonlinear model of the auditory periphery [C. Giguère and P.C. Woodland, J. Acoust. Soc. Am. 95, 331-342 (1994)].

Modified masking period patterns as a function of frequency region and masker bandwidth

The auditory system must encode dynamic aspects of acoustic stimuli at frequencies throughout the audible spectrum. This coding is limited by the temporal resolution of the ear. Recent measures of modulation detection and temporal gap detection thresholds with narrow-band noises of fixed bandwidth indicate that temporal resolution does not vary with noise carrier frequency from 500 to 4400 Hz. The present study investigated temporal processing using a modified masking period pattern technique. Masked thresholds were measured for pure tones of 500, 2000, and 4000 Hz that were 400 ms in duration. Maskers consisted of narrow-band noises centered on the signal frequency that were either unmodulated or sinusoidally amplitude modulated (modulation depth=1.0) at frequencies from 4 to 256 Hz. The noise bandwidth ranged from 200 to 1600 Hz at each center frequency. By comparing masked thresholds for the unmodulated and amplitude modulated noises, one may assess both within-channel (narrowband noises) and across-channel (wideband noises) temporal processing. Preliminary results indicate that thresholds for a given masker bandwidth depend strongly on modulation frequency but vary little with signal frequency over the range tested. The data will be discussed relative to measures of temporal resolution and comodulation masking release.

Masking period patterns in normal-hearing and hearing-impaired listeners.

Masking period patterns (MPPs), developed by Zwicker [J. Acoust. Soc. Am. 59, 166–175 (1976)] as a psychophysical correlate of the physiological period histogram, were obtained from four normal-hearing and four cochlear hearing-impaired subjects. The just audible level of a brief, relatively high-frequency, test tone was measured as a function of its temporal position within the period of a 40-Hz masker. Zwicker’s findings were replicated in normal listeners. At low masker levels a single maximum (Max 1) was observed between 0 and 90 deg of the masker period. At higher masker levels a second maximum (Max 2) developed between 180 and 270 deg. Zwicker’s model [Biol. Cyb. 23, 49–60 (1976)] suggests that Max 1 reflects level-independent suppression effects and Max 2 reflects some combination of level-dependent excitation and suppression effects, which is consistent with physiological biasing data. For impaired listeners, in regions of hearing loss, the degree of modulation in the MPP was reduced relative to that seen in normal-hearing listeners. The pattern of modulation was usually altered as well. These results are discussed in the context of outer hair cell damage at the place in the cochlea corresponding to the test tone frequency. [Supported by NIDCD.]

Temporal resolution within the upper ‘‘accessory excitation’’ region for maskers with fluctuating envelopes.

During Zwicker’s numerous investigations of the time structure of masking, he observed that envelope masking period patterns were larger for frequencies above the masker (in the upper ‘‘accessory excitation’’ region) than for frequencies near the masker (in the ‘‘main excitation’’ region). For example, the peak-to-trough ratio of the masking period pattern for a 4-Hz 100% amplitude-modulated masker was only 12 dB when the probe tone was within the main excitation region, at 1000 Hz, while the ratio was as large as 36 dB when the probe tone was within the accessory excitation region, at 3200 Hz [E. Zwicker, Acustica 36, 113–120 (1976)]. In the present research, masking at the peak and within the trough of 500-Hz amplitude-modulated maskers was investigated as a function of test frequency and masker level for modulation frequencies from 4 to 128 Hz in normal-hearing and hearing-impaired ears. Peak and trough masked thresholds were measured using amplitude-modulated signals with phasic and antiphasic envelopes. The results from normal-hearing ears confirm Zwicker’s finding that temporal resolution can be up to three times better in the upper accessory excitation region than in the main excitation region. Masking during envelope peaks remains constant while masking during envelope troughs increases with modulation frequency. This finding, and the slopes of the growth of masking, support Zwicker’s interpretation that masking during the envelope peaks and troughs are associated with simultaneous and nonsimultaneous masking, respectively. Peak-to-trough ratios in the upper accessory region are considerably reduced in hearing-impaired ears, which has important implications for processing speech signals in the presence of maskers with fluctuating envelopes. [Work supported by NIDCD.]

The effect of a low-frequency masker on loudness

Intense low-frequency tones produce masking-period patterns in which the masking of a higher-frequency tone burst varies by up to 25 dB over the period of the masker. In this experiment, observers were asked to match the loudness of partially masked test-tone bursts in one ear by adjusting the level of unmasked bursts presented to the other ear. It was found that the variation in masked threshold over the period of the masker also affects loudness matches. This effect on loudness, although it decreases in size with increasing level above masked threshold, persits even 25 dB above masked threshold.

Frequency‐based and temporal‐based estimates of auditory filter bandwidth

Estimates of auditory filter bandwidths were examined from measurements of frequency selectivity and temporal acuity. Frequency selectivity was measured using psychophysical tuning curves and comb‐filtered noise masking patterns. Temporal acuity was measured using gap detection and temporal masking period patterns. Measurements for all four experimental paradigms were made at 500 and 3500 Hz using both normal‐hearing and hearing‐impaired listeners. The results are discussed in terms of the relation between frequency selectivity and temporal acuity as predicted by the auditory linear filter model. The results of the hearing‐impaired listeners are compared to the results of the normal‐hearing group and are discussed re: the predicted relation between frequency selectivity and temporal acuity.

Masking-period patterns: presentation of a model compatible with an enlarged set of psychophysical and physiological results.

Masking-period patterns (MPP) have been regarded by Zwicker [J. Acoust. Soc. Am. 59, 166-175 (1976)] as psychoacoustic equivalents of period histograms (PH) measured in auditory neurons. Various models have been proposed in the literature to account for his results. We present here a set of results on the MPP produced by a 40-Hz triangular masker that cannot be reproduced by any of these models. This leads to the elaboration of a new model for predicting MPP assuming the existence of nonlinearities in the basilar membrane, and based on neural fast adaptation and rectification, and a simple detection device. This model is shown to be able to account for the whole set of available results, and thus to provide a good basis for the use of MPP as a psychoacoustic tool for the study of PH.