Spectral Modulation Detection Thresholds Research Articles

Spectral processing deficits appear to underlie developmental language disorders

More than 50 years of research has suggested something is awry in the auditory functioning of children with language disorders, yet there remains no consensus on exactly what the problem is. To address this gap in knowledge, we have proposed a model termed the Auditory Bases of Phonological Acquisition. This model is based on the language feature known as dual patterning, which describes two levels of structure: one consisting of meaningful words and how they can be combined and the other consisting of word-internal, meaningless units and how they can be combined. We hypothesized that acquisition of this second level—phonological structure—is most affected by deficits in auditory functions, especially those involving the spectral domain. To test this hypothesis, we assessed children with and without language disorders between 7 and 10 years old on their abilities to detect spectral and temporal modulation and to perform relevant language tasks. The largest group differences were found at younger ages for spectral modulation detection (SMD). SMD thresholds were also found to correlate most strongly with language measures, especially those based on phonological structure. Outcomes suggest that keen sensitivity to spectral structure in the speech signal is necessary for acquiring precise phonological representations.

The contribution of spectral processing to the acquisition of phonological sensitivity by adolescent cochlear implant users and normal-hearing controls.

This study tested the hypotheses that (1) adolescents with cochlear implants (CIs) experience impaired spectral processing abilities, and (2) those impaired spectral processing abilities constrain acquisition of skills based on sensitivity to phonological structure but not those based on lexical or syntactic (lexicosyntactic) knowledge. To test these hypotheses, spectral modulation detection (SMD) thresholds were measured for 14-year-olds with normal hearing (NH) or CIs. Three measures each of phonological and lexicosyntactic skills were obtained and used to generate latent scores of each kind of skill. Relationships between SMD thresholds and both latent scores were assessed. Mean SMD threshold was poorer for adolescents with CIs than for adolescents with NH. Both latent lexicosyntactic and phonological scores were poorer for the adolescents with CIs, but the latent phonological score was disproportionately so. SMD thresholds were significantly associated with phonological but not lexicosyntactic skill for both groups. The only audiologic factor that also correlated with phonological latent scores for adolescents with CIs was the aided threshold, but it did not explain the observed relationship between SMD thresholds and phonological latent scores. Continued research is required to find ways of enhancing spectral processing for children with CIs to support their acquisition of phonological sensitivity.

The effects of modulator shape and methods for expressing modulation depth on spectral modulation detection thresholds

The detection of sinusoidal modulation is commonly used for assessing the auditory perception of temporal, spectral, and spectro-temporal acoustic features. For temporal (amplitude) modulation, the sinusoidal modulator usually is expressed on a linear amplitude scale. For spectral modulation, the sinusoidal modulator has been specified on a linear amplitude scale, consistent with temporal modulation, or on a logarithmic amplitude scale, with the notion of approximating a sinusoidal excitation pattern. The definition of modulation depth depends on the measurement points (i.e., midpoint to peak or peak to peak) and order of operations when expressing depth in dB. Such differences can make it difficult to compare results for similar tasks among investigations. Here we quantify differences among methods and provide a complete matrix for translating among methods. Spectral modulation detection was measured for 9 normal-hearing listeners in ten conditions (linear vs. logarithmic shape at 0.5, 1, 2, 4, and 8 cycles/octave). Peak-to-peak values of the modulation envelope were equalized, thus only modulation waveform shape differed. Noise carriers had a passband from 400 to 3200 Hz. Thresholds revealed statistically significant effects of both spectral shape and spectral modulation frequency. Several methods for expressing threshold modulation depth were compared to highlight differences among the methods.

Spectral modulation detection in adolescents with normal hearing or cochlear implants predicts some language skills, but not others

Previous outcomes of an ongoing longitudinal study involving children with cochlear implants (CIs) have revealed that these children demonstrate especially large deficits in two areas: (1) language skills dependent upon phonological structure; and (2) speech-in-noise recognition. Morphosyntactic skills appear closer to typical. We have hypothesized that both the phonological deficit and the speech-in-noise problems arise from poor spectral resolution. To test this hypothesis we measured children’s spectral modulation detection (SMD). Participants were14-year-old children with normal hearing (49) or with CIs (42). A 3-interval, forced-choice procedure was used in which two stimuli were unmodulated noises and one was a noise modulated at a rate of 0.5 ripples per octave (rpo). The 70.7% threshold for SMD was estimated adaptively. Other measures included tests of phonological awareness and processing, vocabulary and syntactic abilities, and speech recognition in single- or multi-talker babble, as well as in speech-shaped noise. Children with CIs had slightly higher SMD thresholds, and poorer performance on all other measures. For both groups, SMD thresholds were related to phonological skills and speech-in-noise recognition, but not to vocabulary or syntactic abilities. Outcomes were interpreted as suggesting that phonological sensitivity and speech-in-noise recognition depend upon spectral resolution, but morphosyntactic skills do not.

The effects of carrier bandwidth and intensity on spectral ripple perception in listeners with hearing loss

Listeners with hearing loss often need to listen to speech signals at high intensities to ensure proper audibility. Spectral processing deteriorates at high intensities, but signal acoustics such as bandwidth and filtering can be used to mitigate this deterioration. This study evaluated the effect of carrier bandwidth on broadband spectral processing at high intensities for individuals with hearing loss. Spectral modulation detection thresholds were measured using a rippled noise carrier with varying bandwidths from 1 to 4 octaves, to assess whether individuals with hearing loss were more susceptible to the deleterious effects of high-intensity compared to their counterparts with normal hearing. Results show that spectral processing degrades at high intensities. Listeners with hearing loss benefited from the increase in signal bandwidth at high intensities much more than their counterparts with normal hearing. Our findings suggest that spectral processing for broadband signals involve within and across ch...

Speech Understanding With Various Maskers in Cochlear-Implant and Simulated Cochlear-Implant Hearing: Effects of Spectral Resolution and Implications for Masking Release.

The purpose of this study was to investigate the relationship between psychophysical spectral resolution and sentence reception in various types of interfering backgrounds for listeners with cochlear implants and normal-hearing subjects listening to vocoded speech. Spectral resolution was measured with a spectral modulation detection (SMD) task. For speech testing, maskers included stationary speech-shaped noise (SSN), four-talker babble, multitone noise, and a competing talker. To explore the possible trade-offs between spectral resolution and susceptibility to different types of maskers, the degree of simulated current spread was varied within the vocoder group, achieving a range of performance for SMD and speech tasks. Greater simulated current spread was detrimental to both spectral resolution and speech recognition, suggesting that interventions that decrease current spread may improve performance for both tasks. Better SMD sensitivity was significantly correlated with improved sentence reception. In addition, differences in sentence reception across the four maskers were significantly associated with SMD across the combined group of cochlear-implant and vocoder subjects. Masking release (MR) was quantified as the signal-to-noise ratio difference in speech reception threshold between the SSN and competing talker. Several individual cochlear-implant subjects demonstrated substantial MR, in contrast to previous studies, and the degree of MR increased with better SMD thresholds across subjects. The results of this study suggest that alternative masker types, particularly competing talkers, are more sensitive than stationary SSN to differences in spectral resolution in the cochlear-implant population.

Relationship Between Auditory Function of Nonimplanted Ears and Bimodal Benefit

Patients with a cochlear implant (CI) in one ear and a hearing aid in the other ear commonly achieve the highest speech-understanding scores when they have access to both electrically and acoustically stimulated information. At issue in this study was whether a measure of auditory function in the hearing aided ear would predict the benefit to speech understanding when the information from the aided ear was added to the information from the CI. The subjects were 22 bimodal listeners with a CI in one ear and low-frequency acoustic hearing in the nonimplanted ear. The subjects were divided into two groups-one with mild-to-moderate low-frequency loss and one with severe-to-profound loss. Measures of auditory function included (1) audiometric thresholds at 750 Hz or lower, (2) speech-understanding scores (words in quiet and sentences in noise), and (3) spectral-modulation detection (SMD) thresholds. In the SMD task, one stimulus was a flat spectrum noise and the other was a noise with sinusoidal modulations at 1.0 peak/octave. Significant correlations were found among all three measures of auditory function and the benefit to speech understanding when the acoustic and electric stimulation were combined. Benefit was significantly correlated with audiometric thresholds (r = -0.814), acoustic speech understanding (r = 0.635), and SMD thresholds (r = -0.895) in the hearing aided ear. However, only the SMD threshold was significantly correlated with benefit within the group with mild-to-moderate loss (r = -0.828) and within the group with severe-to-profound loss (r = -0.896). The SMD threshold at 1 cycle/octave has the potential to provide clinicians with information relevant to the question of whether an ear with low-frequency hearing is likely to add to the intelligibility of speech provided by a CI.

Auditory Mismatch Negativity: Detecting Spectral Contrasts in a Modulated Noise

The ability to detect sinusoidal spectral envelope from an unmodulated (flat) spectrum has shown a strong correlation with speech perception in quiet and noisy listening conditions in normal hearing, hearing-impaired, and cochlear implant users. In this study, the mismatch negativity (MMN) was collected from 10 normal-hearing adult listeners and used to determine if modulation detection thresholds derived electrophysiologically correlate with the psychophysically measured spectral modulation detection thresholds. MMNs were present for most participants with spectral contrasts of 10 and 20 dB, suggesting that the MMN may be a promising neurophysiologic measure of spectral envelope perception when nonbehavioral measures are desired.

Spectral modulation transfer function and vowel identification.

Psychophysical and physiological studies have demonstrated selectivity for spectral modulation frequency in the auditory system, suggesting that auditory perception of complex spectra might be based on spectral modulation channels. However, an early study [Liu and Eddins (2008)] reported that vowel identification was significantly reduced by filtering in spectral modulation frequency domain, although there was no significant correlation between spectral modulation detection thresholds and vowel identification. The latter result may have been due to the small number of listeners (n = 5) in the previous study. Using a larger number of listeners, the present study measured spectral modulation transfer functions and vowel identification performance with and without spectral modulation filtering in 11 normal‐hearing listeners. Vowel identification performance was significantly correlated with spectral modulation detection thresholds at 0.5 cyc/oct (i.e., higher spectral modulation detection thresholds associated with poorer vowel identification scores). In addition, average spectral modulation detection thresholds were significantly correlated with the change in vowel identification associated with filtering in the spectral modulation domain. Both of these results highlight the relation between the perception of sinusoidal spectral modulation and vowel identification.

Spectral modulation masking patterns reveal tuning to spectral envelope frequency

Auditory processing appears to include a series of domain-specific filtering operations that include tuning in the audio-frequency domain, followed by tuning in the temporal modulation domain, and perhaps tuning in the spectral modulation domain. To explore the possibility of tuning in the spectral modulation domain, a masking experiment was designed to measure masking patterns in the spectral modulation domain. Spectral modulation transfer functions (SMTFs) were measured for modulation frequencies from 0.25 to 14 cycles/octave superimposed on noise carriers either one octave (800-1600 Hz, 6400-12,800 Hz) or six octaves wide (200-12,800 Hz). The resulting SMTFs showed maximum sensitivity to modulation between 1 and 3 cycles/octave with reduced sensitivity above and below this region. Masked spectral modulation detection thresholds were measured for masker modulation frequencies of 1, 3, and 5 cycles/octave with a fixed modulation depth of 15 dB. The masking patterns obtained for each masker frequency and carrier band revealed tuning (maximum masking) near the masker frequency, which is consistent with the theory that spectral envelope perception is governed by a series of spectral modulation channels tuned to different spectral modulation frequencies.

Spectral modulation detection as a function of modulation frequency, carrier bandwidth, and carrier frequency region

The present study investigates the nature of spectral envelope perception using a spectral modulation detection task in which sinusoidal spectral modulation is superimposed upon a noise carrier. The principal goal of this study is to characterize spectral envelope perception in terms of the influence of modulation frequency (cycles/octave), carrier bandwidth (octaves), and carrier frequency region (defined by lower and upper cutoff frequencies in Hz). Spectral modulation detection thresholds measured as a function of spectral modulation frequency result in a spectral modulation transfer function (SMTF). The general form of the SMTF is bandpass in nature, with a minimum modulation detection threshold in the region between 2 to 4 cycles/octave. SMTFs are not strongly dependent on carrier bandwidth (ranging from 1 to 6 octaves) or carrier frequency region (ranging from 200 to 12 800 Hz), with the exception of carrier bands restricted to very low audio frequencies (e.g., 200-400 Hz). Spectral modulation detection thresholds do not depend on the presence of random level variations or random modulation phase across intervals. The SMTFs reported here and associated excitation pattern computations are considered in terms of a linear systems approach to spectral envelope perception and potential underlying mechanisms for the perception of spectral features.