Joyce McDonough

An acoustic analysis of prosody in high-functioning autism

This paper examined the fundamental frequency variation in the narratives of individuals with high-functioning autism (HFA) and typical controls matched on age, IQ, and verbal abilities. Study 1 found increased fundamental frequency variation in the speech of 21 children and adolescents with HFA when compared to 21 typical controls. Study 2 replicated the findings from Study 1 with a younger sample of 17 children with HFA and 17 typical controls. In addition, Study 1 found evidence that acoustic measurements of prosody were related to clinical judgments of autism-specific communication impairments, although this was not replicated in Study 2. Taken together, these studies provide evidence for differences in expressive prosody in individuals with HFA that can be measured objectively.

Resolving ambiguity: A psycholinguistic approach to understanding prosody processing in high-functioning autism

Individuals with autism exhibit significant impairments in prosody production, yet there is a paucity of research on prosody comprehension in this population. The current study adapted a psycholinguistic paradigm to examine whether individuals with autism are able to use prosody to resolve syntactically ambiguous sentences. Participants were 21 adolescents with high-functioning autism (HFA), and 22 typically developing controls matched on age, IQ, receptive language, and gender. The HFA group was significantly less likely to use prosody to disambiguate syntax, but scored comparably to controls when syntax alone or both prosody and syntax indicated the correct response. These findings indicate that adolescents with HFA have difficulty using prosody to disambiguate syntax in comparison to typically developing controls, even when matched on chronological age, IQ, and receptive language. The implications of these findings for how individuals with autism process language are discussed.

Tamil liquids : An investigation into the basis of the contrast among five liquids in a dialect of Tamil

The Brahmin dialect of Tamil (Dravidian) has an unsual inventory of five distinctive liquid sounds : plain and retroflex rhotics, and plain and retroflex laterals, and a fifth liquid which has been variously described as a rhotic, a lateral, a glide and/or a fricative. This paper investigates the articulatory, acoustic, and perceptual properties of these liquids, and, in particular, the fifth liquid. Electropalatography (EPG) and static palatography were used to examine the articulatory properties of the liquids, the acoustic properties of the liquids were examined, and we tested the intelligibility of the EPG recordings in a perception experiment. Our intent is to propose a classification for the fifth liquid based on these studies. The fifth liquid is classified as a retroflex central approximant, with characteristics that make it distinct from the other liquids along three dimensions of contrast : static ∼ dynamic, central ∼ lateral and retroflex ∼ non-retroflex

Speech Coding in the Brain: Representation of Vowel Formants by Midbrain Neurons Tuned to Sound Fluctuations

Abstract Current models for neural coding of vowels are typically based on linear descriptions of the auditory periphery, and fail at high sound levels and in background noise. These models rely on either auditory nerve discharge rates or phase locking to temporal fine structure. However, both discharge rates and phase locking saturate at moderate to high sound levels, and phase locking is degraded in the CNS at middle to high frequencies. The fact that speech intelligibility is robust over a wide range of sound levels is problematic for codes that deteriorate as the sound level increases. Additionally, a successful neural code must function for speech in background noise at levels that are tolerated by listeners. The model presented here resolves these problems, and incorporates several key response properties of the nonlinear auditory periphery, including saturation, synchrony capture, and phase locking to both fine structure and envelope temporal features. The model also includes the properties of the auditory midbrain, where discharge rates are tuned to amplitude fluctuation rates. The nonlinear peripheral response features create contrasts in the amplitudes of low-frequency neural rate fluctuations across the population. These patterns of fluctuations result in a response profile in the midbrain that encodes vowel formants over a wide range of levels and in background noise. The hypothesized code is supported by electrophysiological recordings from the inferior colliculus of awake rabbits. This model provides information for understanding the structure of cross-linguistic vowel spaces, and suggests strategies for automatic formant detection and speech enhancement for listeners with hearing loss.

The stop contrasts of the Athabaskan languages

Abstract The Athabaskan languages have a particularly rich series of stop contrasts, plain stops and affricates, each exhibiting a three-way laryngeal contrast, unaspirated, aspirated and ejective. Several aspects of this inventory are interesting to phonetic and phonological studies, among them the length and ‘heaviness’ of the releases on the aspirated plain stops, the temporal properties of the ejectives, and the richness of the stop contrast set, approximately 21 distinct segments. This paper is an investigation of the phonetic realization of the stops in five Athabaskan languages: Dene Sųline (CL), Dene Sųline (FC), Dogrib, North Slavey, and Tsilhqut’in, compared with data from Navajo from McDonough [(2003). The Navajo sound system . Kluwer: Dordrecht]. Based on the phonetic patterns in the data, we argue that, among the consonants, the primary organizing feature of the contrasts is a temporal distinction, which we model as simplex–complex contrast, based on Laver [(1994). Principles of phonetics . Cambridge: Cambridge University Press], in which the temporal properties of medial phase of the segments articulation play a significant role in the contrast. This classification represents a major divide in the inventory between the unaspirated plain stops (orthographically ( b ) , d, g, ( q )) and the rest of the stops in the inventory. These data suggest first that aspirated plain stops (orthographically t, k ) are mislabeled. Instead, in keeping with much of the early literature on the Athabaskan languages, the t and k phonemes are the affricates /tx, kx/. Second, related to this, long release periods are a characteristic feature of all but the unaspirated stops in the inventory. As such they represent a feature of a larger grouping in the Athabaskan inventory, realized in a temporal domain and persistent in the family, in a pattern likely inherited from the parent language. We model this temporal distinction as a simplex–complex distinction, which separates out the ‘unaspirated’ plain stops, as simplex segments with short offsets, from the rest of the stops, including the plain ‘aspirated’ stops and ejectives, which have complex medial phases with long release periods. Furthermore, the proposal suggests that the languages have exploited the sets of simplex stops and fricatives to build their rich inventories of complex segments, as several linguists have observed. The analysis serves as a basis for understanding sound change and alternation patterns in the family.

A causal test of the motor theory of speech perception: a case of impaired speech production and spared speech perception

The debate about the causal role of the motor system in speech perception has been reignited by demonstrations that motor processes are engaged during the processing of speech sounds. Here, we evaluate which aspects of auditory speech processing are affected, and which are not, in a stroke patient with dysfunction of the speech motor system. We found that the patient showed a normal phonemic categorical boundary when discriminating two non-words that differ by a minimal pair (e.g., ADA–AGA). However, using the same stimuli, the patient was unable to identify or label the non-word stimuli (using a button-press response). A control task showed that he could identify speech sounds by speaker gender, ruling out a general labelling impairment. These data suggest that while the motor system is not causally involved in perception of the speech signal, it may be used when other cues (e.g., meaning, context) are not available.

Navajo vowels and universal phonetic tendencies

Navajo is one of the comparatively few languages in the world in which there are said to be four vowel qualities, [i,e,ɑ ,o]. Data from nine speakers of Navajo show that when the vowels are represented in terms of their formant frequencies, they are not approximately equidistant as predicted by Lindblom’s dispersion hypothesis, in that [i] and [e] are too close together. What is also not predicted by this hypothesis is that the vowel [o] has a greater variance than the other vowels. There are two possible reasons for the greater variance of back vowels. On the one hand, considering the vowel space as a triangle, the high back corner could be less sharp than the high front or low back corners, allowing high back vowels to be in various locations that are all well dispersed from [ɑ ] and [i]. On the other hand, high back vowels could be less well defined in articulatory‐acoustic terms; [i] and [ɑ ] are more in accord with Stevens’ quantal hypothesis in that they have vocal tract shapes that can be somewhat ...

Neural representations of vowels and vowel-like sounds

Neural representations of speech at every level of the neuraxis have nonlinear features that are not described by spectrograms or linear filter-banks. In this study, recent computational models for populations of cells in the auditory periphery, brainstem, and midbrain were used to explore the implications of vowel features for neural responses. Peripheral neural responses are characterized by strong periodicities, dominant frequencies, that depend upon the distribution of energy across the harmonics and vary across vowels [Fant, 1970]. Strong periodicities related to the fundamental and low-frequency harmonics are observed for peripheral neurons tuned to a wide range of frequencies [Delgutte & Kiang, 1984]. The computational model captures this feature of the physiological responses. These periodicities are interesting because many midbrain neurons are tuned to fluctuations in this frequency range. Single-formant vowels allowed systematic manipulations of the relationship between formant and harmonic fre...

The phonetics of native North American languages

At the time of European contact, about 300 languages from around 16 or more language families were spoken on the North American continent north of the Rio Grande (Mithun, 1999). Despite the significant variation found among them, the languages in North America represent a distinct typology. Mithun points out that none of these languages are truly analytic. These languages tend to be richly inflectional or agglutinative; their words are complex entities and are likely to encode considerable structure. For this reason the documentation and investigation of North American languages has proven to be an important source in the development of syntactic, morphological and typological theories of human language. Intrinsic to morphological and phonological structure is the phonetic structure of the language. But, until recently, not much attention has been paid to the phonetic structure of Native American languages. This has not always been the case. Instrumental fieldwork on North American indigenous languages began in the early 20th century. Pliny Earl Goddard collected kymographic and palatographic records in Northern Alberta in the early 1900s (Goddard, 1907, 1912, 1929). The Danish phonetician Uldall worked on Maidu and Achumawi, languages spoken on the northern coast of California in the 1930s and 1950s. Though there are no published reports, he refers to kymographic records in his footnotes (1935, 1954). Hogan (1976) published a study of Dene Suline ejectives, and Mona Lindau (1984) included Navajo data in her instrumental cross linguistic study of ejectives. There are likely other studies, but the numbers are small, the studies are limited, or the data is unpublished or inaccessible. A revival of phonetic fieldwork has taken place, reflected in this volume, in large part relaunched by Peter Ladefoged and Ian Maddieson at the UCLA phonetics lab in the 1990s, through their National Science Foundation grant Sounds of the World’s Languages. The Ladefoged–Maddieson team called attention to the importance of the documentation of phonetic phenomena in the field, developing methodologies for the collection of instrumental data, together with experimental techniques, and the training of field phoneticians (including one of the co-editors and several of the authors of papers in this volume). This revival has been bolstered by the development of new technology and equipment which allow linguists to expand the type of data they can collect in field situations. Through these developing technologies, linguists are introducing new types of data for use in language documentation and for analysis and theory building, shifting the focus away from an exclusively text-based approach to the spoken word. Integral to this shift is the availability of good phonetic data on these languages. Additionally, since the structures found in less studied languages have proven important to other linguistic domains, it stands to reason that they likely to be equally important to phonetic typology and theory. Without this kind of data, we are limited in our understanding of the type and variety of phonetic phenomena that may occur in human languages. Orthographies, when they do exist, are often not standardized and have a limited range of use. Many language communities have had the experience of having an outsider (a linguist or a missionary, for instance) propose an orthography, which is then used to produce a limited amount of materials like grammars and dictionaries for language preservation. Given the fact that the best (i.e., most practical) orthographies gloss over phonetic detail, these records are usually of limited value for phonetic science. Even the best IPA transcriptions require a segmentation and linearization that is often at odds with the phonetic phenomena at hand. Unless we make consistent and durable recordings of the languages as they are still spoken, we are at

Duration and Timing

In describing of the profile of the speech patterns of a community, segmental duration is a revealing aspect of the sound system (Lehiste 1970, Klatt 1976). Quantitative duration differences serve a function at every level in the grammar, from syllable (vowels are longer before voiced consonants in English) to utterance level (final lengthening). The implementation of local differences in duration mark a wide range of phenomena: in the realization of metric structure and stress, of prosodic position-in-word and -utterance, focus and intonation, markedness, and extra-linguistic phenomena such as speech rate. Duration is a quantitative measure explicitly associated with different phonological, prosodic and psychological phenomena, making it one of the more problematic aspects of an interface between phonetics and phonology. In the same vein, duration effects are also to a large degree language specific. In many languages with metric stress systems, unstressed vowels are shorter than stressed ones, but the implmentation of that reduction vary cross-linguistically (Lehiste 1970, Crosswhite 2001). That is to say, duration, while quantitative, is not determined by purely physiological phenomena. This means that specific information about duration and lengthening must be encoded in the implementation of lexical representations. In particular, because of the close association of lengthening with the expression of metric stress, pitch accent and prosodic features like intonation and focus, the prosodic organization of a language will be explicitly reflected to one extent or another in its duration profile, in a similar way that a pitch contour reflects tonal structure.