Vowel Discrimination Task Research Articles

Brazilian Portuguese vowel discrimination by Nungon and Tashelhit speakers

The vowel system of speakers’ native languages are known to affect perception of vowels in foreign languages (Elvin et al., 2021). In a field-based behavioral experiment, we tested the perception of Brazilian Portuguese vowels (seven-vowel system) by speakers of the Papuan language Nungon (six-vowel system) and the Berber language Tashelhit (three-vowel system). 60 native speakers of Nungon and 38 native speakers of Tashelhit participated. The acoustics of Nungon vowels were previously analyzed using a 30-point analysis technique (Sarvasy et al., 2020). For the present study, we also analyzed Tashelhit vowels using the same technique. We then used discriminant analysis to analyze the relationship between speakers’ performance on the Brazilian Portuguese vowel discrimination task and the acoustics of vowels in their native languages. We present preliminary results of these analyses, and address their possible contributions to four areas: (a) description of the acoustics of under-described languages, (b) the relationship between native vowel systems and perception of foreign vowels, (c) methods for undertaking rigorous acoustic and experimental research in noisy, field environments, and (d) theoretical frameworks for phonetic learning such as the L2LP model (Escudero, 2005; van Leussen and Escudero, 2015).

L2 Perception of Contrastive Vowel Nasality in Brazilian Portuguese

AbstractAlthough not undisputed, it is generally agreed that Brazilian Portuguese (BP) has lexically contrastive vowel nasality, for instance between [si] ‘if; oneself‘ and [sĩ] ‘yes.‘ It is known that second-language (L2) learners of BP struggle with oral-nasal vowel contrasts in perception, but less is known on how L2 learners perform in perception. This paper reports on a study that investigated the perception of BP contrastive vowel nasality by a group of English-native learners of BP and a native speaker control group to assess how non-native listeners perform in pre-lexical discrimination and lexical identification of contrastive vowel nasality. Although results from a vowel discrimination task revealed no differences between L2 and L1 listeners in terms of pre-lexical perception, a lexical identification task revealed that some oral-nasal vowel contrasts impeded lexical access in L2 listeners. These findings highlight how L2 listeners can perform comparably to L1 listeners in perception of non-native sound contrasts (here, vowel nasality) at the pre-lexical level, but may still struggle in encoding those contrasts at a lexical level.

Bilateral age-related atrophy in the planum temporale is associated with vowel discrimination difficulty in healthy older adults

In this study we investigated the association between age-related brain atrophy and behavioural as well as electrophysiological markers of vowel perception in a sample of healthy younger and older adults with normal pure-tone hearing. Twenty-three older adults and 27 younger controls discriminated a set of vowels with altered second formants embedded in consonant-vowel syllables. Additionally, mismatch negativity (MMN) responses were recorded in a separate oddball paradigm with the same set of stimuli. A structural magnet resonance scan was obtained for each participant to determine cortical architecture of the left and right planum temporale (PT). The PT was chosen for its function as a major processor of auditory cues and speech. Results suggested that older adults performed worse in vowel discrimination despite normal-for-age pure-tone hearing. In the older group, we found evidence that those with greater age-related cortical atrophy (i.e., lower cortical surface area and cortical volume) in the left and right PT also showed weaker vowel discrimination. In comparison, we found a lateralized correlation in the younger group suggesting that those with greater cortical thickness in only the left PT performed weaker in the vowel discrimination task. We did not find any associations between macroanatomical traits of the PT and MMN responses. We conclude that deficient vowel processing is not only caused by pure-tone hearing loss but is also influenced by atrophy-related changes in the ageing auditory-related cortices. Furthermore, our results suggest that auditory processing might become more bilateral across the lifespan.

Intelligibility of English Vowels Produced by Nigerian and Malaysian Speakers

The widespread use of the English language raises concerns about maintaining the mutual intelligibility across different nonnative English varieties. Some have viewed nonnative English varieties as distorted forms of English that cannot stand on their own that would cause the language to disintegrate into mutually unintelligible varieties. This study is an attempt to ascertain if there is mutual intelligibility between Nigerian and Malaysian English speakers, as there is little exploration of the intelligibility of African English varieties to Malaysians and vice versa. Forty Nigerians and 80 Malaysian undergraduate ESL students took part in a vowel discrimination task. The Nigerians listened to the words recorded by Malaysian speakers while the Malaysians listened to the words recorded by Nigerian speakers. Seven pairs of vowels were chosen as target vowel contrasts tested. Results showed that the Nigerians and Malaysians performed well in the discrimination task with performance above the guessing threshold for most of the selected pairs of vowel contrasts. Findings suggest that mutual intelligibility exists between the 2 recognized nonnative English varieties.

Uncertainty-based informational masking in a vowel discrimination task for young and old Mongolian gerbils

Informational masking emerges with processing of complex sounds in the central auditory system and can be affected by uncertainty emerging from trial-to-trial variation of stimulus features. Uncertainty can be non-informative but confusing and thus mask otherwise salient stimulus changes resulting in increased discrimination thresholds. With increasing age, the ability for processing of such complex sound scenes degrades. Here, 6 young and 4 old gerbils were tested behaviorally in a vowel discrimination task. Animals were trained to discriminate between sequentially presented target and reference vowels of the vowel pair/I/-/i/. Reference and target vowels were generated shifting the three formants of the reference vowel in steps towards the formants of the target vowels. Non-informative but distracting uncertainty was introduced by random changes in location, level, fundamental frequency or all three features combined. Young gerbils tested with uncertainty for the target or target and reference vowels showed similar informational masking effects for both conditions. Young and old gerbils were tested with uncertainty for the target vowels only. Old gerbils showed no threshold increase discriminating vowels without uncertainty in comparison with young gerbils. Introducing uncertainty, vowel discrimination thresholds increased for young and old gerbils and vowel discrimination thresholds increased most when presenting all three uncertainty features combined. Old gerbils were more susceptible to non-informative uncertainty and their thresholds increased more than thresholds of young gerbils. Gerbils’ vowel discrimination thresholds are compared to human performance in the same task (Eipert et al., 2019).

Perceiving isn't believing: Divergence in levels of sociolinguistic awareness

AbstractThe influence of social knowledge on speech perception is a question of interest to a range of disciplines of language research. This study combines experimental and qualitative approaches to investigate whether the various methodological and disciplinary threads of research on this topic are truly investigating the same phenomenon to provide converging evidence in our understanding of social listening. This study investigates listeners’ perceptions of Spanish and Quechua speakers speaking Spanish in the context of a contact zone between these two languages and their speakers in central Bolivia. The results of a pair of matched-guise vowel discrimination tasks and subsequent interviews demonstrate that what peopleperceive, as measured by experimental tasks, is not necessarily what theybelievethey hear, as reported in narrative responses to interview prompts. Multiple methodological approaches must be employed in order to fully understand the way that we perceive language at diverging levels of sociolinguistic awareness. (Perception, sociophonetics, sociolinguistics, awareness, Andean Spanish)

Uncertainty in location, level and fundamental frequency results in informational masking in a vowel discrimination task for young and elderly subjects

Informational masking (IM) is defined as the compromised ability to perceive and analyze signals from a single sound source in a cacophony of sounds from other sources even if the excitation patterns produced by these signals in the auditory periphery are well separated from those produced by the sounds from the other sources. IM that causes an elevation of discrimination thresholds is affected by the similarity between target and masker and by stimulus uncertainty. Here, six young and six elderly subjects were asked to discriminate between sequentially presented reference and target vowels of the vowel pairs /I/-/i/, /æ/-/ε/, and /α/-/Λ/. Psychometric functions were collected characterizing the discrimination of target vowels from reference vowels. Target vowels differed from the reference by one of seven steps shifting the three formants of a reference vowel towards the formants of the corresponding target vowel. Stimulus statistics were varied, generating uncertainty by non-informative but potentially distracting location, level, and fundamental frequency changes or all three combined. Young subjects tested with distracting changes applied to the target vowels only, the reference vowels only, or the target and reference vowels showed similar amounts of IM for all three conditions. Elderly subjects were tested with distracting changes applied to target vowels only. Applying uncertainty only to the target vowels led to worse vowel discrimination thresholds for young and elderly subjects and thresholds increased most for the three distracting changes combined. Elderly subjects showed higher vowel discrimination thresholds than young subjects, but the increase in vowel discrimination thresholds due to IM did not differ between young and elderly subjects. The temporal fine structure processing of elderly subjects was degraded in comparison to young subjects, but it was only correlated with the discrimination threshold for vowel pair /I/-/i/.

Articulatory peripherality modulates relative attention to the mouth during visual vowel discrimination

Masapollo, Polka, and Ménard (2016) have recently reported that adults from different language backgrounds show robust directional asymmetries in unimodal visual-only vowel discrimination: a change in mouth-shape from one associated with a relatively less peripheral vowel to one associated with a relatively more peripheral vowel (in F1-F2 articulatory/acoustic vowel space) results in significantly better performance than a change in the reverse direction. In the present study, we used eye-tracking methodology to examine the gaze behavior of English-speaking subjects while they performed Masapollo et al.'s visual vowel discrimination task. We successfully replicated this directional effect using Masapollo et al.'s visual stimulus materials, and found that subjects deployed selective attention to the oral region compared to the ocular region of the model speaker's face. In addition, gaze fixations to the mouth were found to increase while subjects viewed the more peripheral vocalic articulations compared to the less peripheral articulations, perhaps due to their larger, more extreme oral-facial kinematic patterns. This bias in subjects’ pattern of gaze behavior may contribute to asymmetries in visual vowel perception.

Suprathreshold compound action potential amplitude as a measure of auditory function in cochlear implant users

Electrically evoked compound action potential (eCAP) amplitudes elicited at suprathreshold levels were assessed as a measure of the effectiveness of cochlear implant (CI) stimulation. Twenty-one individuals participated; one was excluded due to facial stimulation during eCAP testing. For each participant, eCAPs were elicited with stimulation from seven electrodes near the upper limit of the individual's electrical dynamic range. A reduced-channel CI program was created using those same seven electrodes, and participants performed a vowel discrimination task. Consistent with previous reports, eCAP amplitudes varied across tested electrodes; the profiles were unique to each individual. In 6 subjects (30%), eCAP amplitude variability was partially explained by the impedance of the recording electrode. The remaining amplitude variability within subjects, and the variability observed across subjects could not be explained by recording electrode impedance. This implies that other underlying factors, such as variations in neural status across the array, are responsible. Across-site mean eCAP amplitude was significantly correlated with vowel discrimination scores (r2 = 0.56). A single eCAP amplitude measured from the middle of the array was also significantly correlated with vowel discrimination, but the correlation was weaker (r2 = 0.37), though not statistically different from the across-site mean. Normalizing each eCAP amplitude by its associated recording electrode impedance did not improve the correlation with vowel discrimination (r2 = 0.52). Further work is needed to assess whether combining eCAP amplitude with other measures of the electrode-neural interface and/or with more central measures of auditory function provides a more complete picture of auditory function in CI recipients.

Degraded neural and behavioral processing of speech sounds in a rat model of Rett syndrome

Individuals with Rett syndrome have greatly impaired speech and language abilities. Auditory brainstem responses to sounds are normal, but cortical responses are highly abnormal. In this study, we used the novel rat Mecp2 knockout model of Rett syndrome to document the neural and behavioral processing of speech sounds. We hypothesized that both speech discrimination ability and the neural response to speech sounds would be impaired in Mecp2 rats. We expected that extensive speech training would improve speech discrimination ability and the cortical response to speech sounds. Our results reveal that speech responses across all four auditory cortex fields of Mecp2 rats were hyperexcitable, responded slower, and were less able to follow rapidly presented sounds. While Mecp2 rats could accurately perform consonant and vowel discrimination tasks in quiet, they were significantly impaired at speech sound discrimination in background noise. Extensive speech training improved discrimination ability. Training shifted cortical responses in both Mecp2 and control rats to favor the onset of speech sounds. While training increased the response to low frequency sounds in control rats, the opposite occurred in Mecp2 rats. Although neural coding and plasticity are abnormal in the rat model of Rett syndrome, extensive therapy appears to be effective. These findings may help to explain some aspects of communication deficits in Rett syndrome and suggest that extensive rehabilitation therapy might prove beneficial.

Asymmetries in vowel perception: Effects of formant convergence and category “goodness”

The mechanisms underlying directional asymmetries in vowel perception have been the subject of considerable debate. One account—the Natural Referent Vowel (NRV) framework—suggests that asymmetries reflect a language-universal perceptual bias, such that listeners are predisposed to attend to vowels with greater formant convergence (Polka & Bohn, 2011). A second (but not mutually exclusive) account—the Native Language Magnet (NLM) theory—suggests that asymmetries reflect an experience-dependent language-specific bias favoring “good” exemplars of native language vowel categories (Kuhl, 1993). We tested the above hypotheses by investigating whether listeners, from different language backgrounds, display asymmetries influenced by formant proximity and/or language experience. Specifically, we examined monolingual English and French listeners’ performance in a within-category AX vowel discrimination task, using variants of /u/ that systematically differed in both their degree of formant proximity (between F1 and F2) and category “goodness” judgments. Results revealed asymmetries that pattern as predicted by NRV when pairs of /u/ tokens exhibited a relatively larger difference in their F1-F2 convergence patterns, and as predicted by NLM when the pairs of /u/ tokens exhibited a relatively smaller difference in their F1-F2 convergence patterns. These findings suggest that language-universal perceptual biases and specific language experience interact to shape vowel perception.

Acoustical and categorical tasks differently modulate activations of human auditory cortex to vowels

The present study compared activations to prototype, nonprototype, nonphonemic, and cross-category vowel pairs during vowel discrimination, category discrimination, 2-back, and visual tasks. Our results support previous findings that areas of the superior temporal gyrus (STG) are sensitive to the speech-level difference between prototype vs. nonprototype and phonemic vs. nonphonemic vowels. Further, consistent with previous studies, we found enhanced activations in anterior-posterior STG and inferior parietal lobule (IPL) during the vowel discrimination and 2-back tasks, respectively. Unlike the vowel discrimination task, the category discrimination task was associated with strong IPL activations. Our results provide evidence that activations in STG and IPL strongly depend on whether the task requires analysis of detailed acoustical information or operations on categorical representations. Based on previous studies investigating activations during categorical pitch and spatial tasks, we argue that this distinction is probably not specific to speech.

Activations of human auditory cortex to phonemic and nonphonemic vowels during discrimination and memory tasks

We used fMRI to investigate activations within human auditory cortex (AC) to vowels during vowel discrimination, vowel (categorical n-back) memory, and visual tasks. Based on our previous studies, we hypothesized that the vowel discrimination task would be associated with increased activations in the anterior superior temporal gyrus (STG), while the vowel memory task would enhance activations in the posterior STG and inferior parietal lobule (IPL). In particular, we tested the hypothesis that activations in the IPL during vowel memory tasks are associated with categorical processing. Namely, activations due to categorical processing should be higher during tasks performed on nonphonemic (hard to categorize) than on phonemic (easy to categorize) vowels. As expected, we found distinct activation patterns during vowel discrimination and vowel memory tasks. Further, these task-dependent activations were different during tasks performed on phonemic or nonphonemic vowels. However, activations in the IPL associated with the vowel memory task were not stronger during nonphonemic than phonemic vowel blocks. Together these results demonstrate that activations in human AC to vowels depend on both the requirements of the behavioral task and the phonemic status of the vowels.

Prenatal exposure to antidepressants and depressed maternal mood alter trajectory of infant speech perception.

Language acquisition reflects a complex interplay between biology and early experience. Psychotropic medication exposure has been shown to alter neural plasticity and shift sensitive periods in perceptual development. Notably, serotonin reuptake inhibitors (SRIs) are antidepressant agents increasingly prescribed to manage antenatal mood disorders, and depressed maternal mood per se during pregnancy impacts infant behavior, also raising concerns about long-term consequences following such developmental exposure. We studied whether infants' language development is altered by prenatal exposure to SRIs and whether such effects differ from exposure to maternal mood disturbances. Infants from non-SRI-treated mothers with little or no depression (control), depressed but non-SRI-treated (depressed-only), and depressed and treated with an SRI (SRI-exposed) were studied at 36 wk gestation (while still in utero) on a consonant and vowel discrimination task and at 6 and 10 mo of age on a nonnative speech and visual language discrimination task. Whereas the control infants responded as expected (success at 6 mo and failure at 10 mo) the SRI-exposed infants failed to discriminate the language differences at either age and the depressed-only infants succeeded at 10 mo instead of 6 mo. Fetuses at 36 wk gestation in the control condition performed as expected, with a response on vowel but not consonant discrimination, whereas the SRI-exposed fetuses showed accelerated perceptual development by discriminating both vowels and consonants. Thus, prenatal depressed maternal mood and SRI exposure were found to shift developmental milestones bidirectionally on infant speech perception tasks.

Different timescales for the neural coding of consonant and vowel sounds.

Psychophysical, clinical, and imaging evidence suggests that consonant and vowel sounds have distinct neural representations. This study tests the hypothesis that consonant and vowel sounds are represented on different timescales within the same population of neurons by comparing behavioral discrimination with neural discrimination based on activity recorded in rat inferior colliculus and primary auditory cortex. Performance on 9 vowel discrimination tasks was highly correlated with neural discrimination based on spike count and was not correlated when spike timing was preserved. In contrast, performance on 11 consonant discrimination tasks was highly correlated with neural discrimination when spike timing was preserved and not when spike timing was eliminated. These results suggest that in the early stages of auditory processing, spike count encodes vowel sounds and spike timing encodes consonant sounds. These distinct coding strategies likely contribute to the robust nature of speech sound representations and may help explain some aspects of developmental and acquired speech processing disorders.

Preference patterns in infant vowel perception

Infants show directional asymmetries in vowel discrimination tasks that reveal an underlying perceptual bias favoring more peripheral vowels. Polka and Bohn (2003) propose that this bias is language independent and plays an important role in the development of vowel perception. In the present study we measured infant listening preferences for vowels to assess whether a perceptual bias favoring peripheral vowels can be measured more directly. Monolingual (French and English) and bilingual infants completed a listening preference task using multiple natural tokens of German /dut/ and /dyt/ produced by a male talker. In previous work, discrimination of this vowel pair by German-learning and by English-learning infants revealed a robust directional asymmetry in which /u/ acts as a perceptual anchor; specifically, infants had difficulty detecting a change from /u/ to /y/, whereas a change from /y/ to /u/ was readily detected. Preliminary results from preference tests with these stimuli show that most infants between 3 and 5 months of age also listen longer to /u/ than to /y/. Preference data obtained from older infants and with other vowel pairs will also be reported to further test the claim that peripheral vowels have a privileged perceptual status in infant perception.

The time course of learning during a vowel discrimination task by hearing-impaired and masked normal-hearing listeners

Vowel discrimination was compared between a group of young, well-trained listeners with mild-to-moderate sensorineural hearing impairment (YHI), and a matched group of normal hearing, noise-masked listeners (YNH). Unexpectedly, discrimination of F1 and F2 in the YHI listeners was equal to or better than that observed in YNH listeners in three conditions of similar audibility [Davis et al., J. Acoust. Soc. Am. 109, 2501 (2001)]. However, in the same time interval, the YHI subjects completed an average of 55% more blocks of testing than the YNH group. New analyses were undertaken to examine the time course of learning during the vowel discrimination task, to determine whether performance was affected by number of trials. Learning curves for a set of vowels in the F1 and F2 regions showed no significant differences between the YHI and YNH listeners. Thus while the YHI subjects completed more trials overall, they achieved a level of discrimination similar to that of their normal-hearing peers within the same number of blocks. Implications of discrimination performance in relation to hearing status and listening strategies will be discussed. [Work supported by NIHDCD-02229.]

Identification of vowel speech sounds by skilled and less skilled readers and the relation with vowel spelling

Reading and spelling errors of vowels are reported in many studies (Bryson and Werker 1989; Fowler, Liberman, and Shankweiler 1977; Fowler, Shankweiler, and Liberman 1979; Goswami 1993; Landerl, Wimmer, and Frith 1997; Shankweiler and Liberman 1972). The present study tested the hypothesis that spelling errors involving vowels are linked to difficulties in vowel perception. Second to fourth graders (total n=155) were divided into five groups according to reading skill and were tested on a variety of measures involving vowel identification, vowel discrimination, and vowel spelling. Despite little difficulty on the vowel discrimination tasks, participants made many errors on the vowel identification measures. Vowel identification errors were linearly associated with reading skill with least skilled readers having significantly more difficulty with stressed “short” vowels as in dip than with stressed “long” vowels as in deep, presented in identical contexts. Vowel identification errors were also associated with vowel spelling errors. It is hypothesized that errors in vowel spelling may relate to weak access to the phoneme at the oral language level and may indicate a lack of constancy in the representation of vowels by less skilled readers. Weaknesses in vowel perception can be detected with a simple vowel identification test in which phonological similarity of test items is used as linguistic manipulation, and where phonemes must be identified based on presentation of a single test item in a forced choice format.

Adaptive Mixtures of Local Experts.

We present a new supervised learning procedure for systems composed of many separate networks, each of which learns to handle a subset of the complete set of training cases. The new procedure can be viewed either as a modular version of a multilayer supervised network, or as an associative version of competitive learning. It therefore provides a new link between these two apparently different approaches. We demonstrate that the learning procedure divides up a vowel discrimination task into appropriate subtasks, each of which can be solved by a very simple expert network.

Adults and infants show a “prototype effect” for speech sounds

Dennis Klatt provided valuable assistance to many speech perception researchers. His advice on the synthesis of speech signals was particularly helpful, and investigators relied on his expertise when problems arose in the development of stimuli designed to approach a problem in a new way. This was the case in the preparation of stimuli to test whether there are “prototypes” for vowel sounds. The experimental question was whether or not adults and infants responded differently in a within‐category vowel discrimination task when the “standard” stimulus was an exceptionally good instance of the vowel /i/—a prototype of the category—as opposed to a nonprototypic /i/ vowel. The results showed that when the prototype of the category served as the standard stimulus, it was more difficult to hear differences between it and novel /i/ vowels than it was to hear differences between the nonprototypic stimulus and novel /i/ vowels. In other words, the prototype was perceived to be more similar to new members of the category than was the nonprototypic stimulus. The effect was observed both in adults and in 6‐month‐old infants. Described first will be the hypothesis underlying the prototype test, the method used to construct the stimuli (the test stimuli for the prototype and the nonprototype were scaled psychophysically using the “mel” scale), and the results on English adults and infants. Described next will be the second phase of the research program, which entails cross‐language tests on Swedish adults and infants. The cross‐language tests are designed to assess two different explanations for the prototype effect: (1) that particular vowel stimuli (/a,i,u/) are inherently more resistent to the effects of articulatory/acoustic change (Stevens' quantal theory) and (2) that the effects observed in American adults and infants are attributable to experience in listening to English, even in the first 6 months of life. Careful preparation of the English and Swedish stimuli was critical in designing these experiments. Dennis Klatt was always ready to provide advice on such matters; his assistance is gratefully acknowledged and sorely missed. [Work supported by NIH.]