Jack Ryalls

Brain damage and cortical compensation in foreign accent syndrome

Foreign Accent Syndrome (FAS) is a well-known neurological deficit whose underlying cause has remained obscure despite almost a century of study. Combining structural and functional imaging, our studies suggest that FAS represents a compensatory response to impaired motor regulation of speech. We describe a patient who acquired FAS as a result of an ischemic stroke in the left basal ganglia. In addition to this case being exceptionally clean, we were able to confirm a specific lesion location as well as provide strong evidence that impaired motor speech regulation resulted in compensation by other areas of the cortical motor speech network.

An Atypical Case of Foreign Accent Syndrome.

A new case of Foreign Accent Syndrome is described. This American woman presented with a British‐ or Australian‐ sounding accent after stroke, which resulted in a lacunar infarct in the left internal capsule. The atypical etiology and apparent changes in lexical use are described. It is hypothesized that an abnormally tense vocal tract posture may account for phonetic changes in vowel quality and a higher average fundamental frequency.

A Comparison of the Prosodic Characteristics of the Speech of People with Parkinson’s Disease and Friedreich’s Ataxia with Neurologically Normal Speakers

The realization of prosody (speech rate, fundamental frequency, intonation) was investigated in a group of 10 individuals with Parkinson’s disease and a group of 10 individuals with Friedreich’s ataxia. Data from these two neurologically disordered groups were compared to individuals without neurological impairment. Both neurologically impaired groups retained some aspects of normal speech prosody, while other aspects were affected to a significant degree. The prosodic characteristics of speakers with Parkinson’s disease were distinct from those of speakers with Friedreich’s ataxia. These results were interpreted in terms of prosodic competence and prosodic performance.

Voice Onset Time production in older Caucasian- and African-Americans

In an earlier study (Ryalls, Zipprer and Baldauff, 1997) significant differences in Voice Onset Time (VOT) production were found between younger males and younger females, as well as between African-American and Caucasian-American speakers. In this study we attempted to replicate these significant effects for gender and ethnic background in a group of older speakers. Participants in this study were healthy normal speakers between 50 and 70 years of age and included 10 African-Americans and 10 Caucasian-Americans with 10 males and 10 females. Speakers in this study produced real words with each of the 6 stop consonants of English (/p/, /t/, /k/; /b/, /d/, /g/) combined with the three extreme vowels (/i/, /a/, /u/) in monosyllabic words ending with a voiceless stop. Each of these 18 stimuli were then read at least five times in random order within a repetition and recorded onto Digital Audio Tape (DAT). The Brown University Laboratory Interactive Speech System (BLISS, Mertus, 1999) was used to display waveforms and measure VOT for three of the repetitions for a total of 54 measures per speaker. An Analysis of Variance indicated no significant effects for either ethnic background or gender. However, a comparison with the previous VOT data from younger speakers revealed a significant difference with the present data from older speakers. The results of this study are discussed in the context of past findings in the published literature.

Voice Onset Time Production in Older and Younger Female Monozygotic Twins

A comparison of voice onset time (VOT) production in younger and older female monozygotic twins revealed greater similarity for the younger than the older female twin pairs. Based on findings from this study and a review of previous twin literature, it is hypothesized that ‘source’ characteristics of speech (voice properties) are under more constraints by genetics, while ‘filter’ characteristics of speech (VOT, formant frequencies) are freer to diverge due to environmental factors such as different linguistic environments. Since the older twins separated to two distinct linguistic areas of the United States at the age of 25 years, while the 2 younger twins continue to share domicile at 21 years, the findings of this study appear to support this preliminary hypothesis.

Preliminary investigation of voice onset time production in persons with dysphagia.

Abstract. The purpose of this study was to determine whether voice onset time (VOT) values of persons with dysphagia differed from those of a person with normal swallow function. Five male subjects with dysphagia (average age = 80.6 years) and a control subject (age = 79 years) read 18 consonant–vowel–consonant words in quasi-random order. These syllables began with the voiced and voiceless cognates from the three stop places of articulation (i.e., bilabial, alveolar, and velar). These consonants were followed by the vowels /i/, /a/, and /u/. Digital audio tape recordings were performed and speech was digitized onto disk. Measurements were completed using BLISS software (Mertus J: BLISS Users Manual. Providence: Department of Cognitive and Linguistic Sciences, Brown University, 1989) implemented on a 486 microcomputer. Averages and standard deviations of the VOT measures for the six stop consonants were compared between the two experimental groups. For the dysphagic speakers, average VOT values for voiceless stops were shorter, and there were larger negative VOT values for voiced stops. Standard deviations for the VOT productions pf the dysphagic subjects were smaller. Statistical comparisons showed significant differences between individual dysphagic speakers and the normal control for three of the five subjects. These preliminary data suggest that dysphagia affects the fine motor control required for accurate VOT production in speech.

Identification of code-ntixed words by fluent Taiwanese–English bilinguals

Abstract In speech perception, there are three predominant models (i.e. top-down, bottom-up or a combination processing model). Recent discussion has focused on use of a combination model by bilingual speakers. Normative information is essential when applied to bilingual participants, in order to clarify what type of language input will best facilitate learning or recovery of specific language abilities. The purpose of this investigation was to study the identification of code-mixed words among Taiwanese–English speaking bilingual individuals. This study posed the following two questions: Is there a difference in the identification of an English or Taiwanese code-mixed word in fluent bilingual speakers (i.e. Taiwanese–English)? Is there a difference in the listeners perception of the code-mixed stimuli according to length of residence (LOR), or to the amount of time a speaker has lived in the United States? Thirty-two Taiwanese–English fluent bilinguals with no reported speech or hearing deficits participated in this experiment. The participants were divided into three subgroups according to LOR (i.e. short LOR, 2–6 years; middle LOR, 7–10 years; and late LOR, 11–20 years). The participants heard sentences with the last word broken up into segments increasing in length (gates). Each targeted word was contained in a sentence presented under four different language conditions. The results of this study indicated that bilingual listeners were able to distinguish differences according to language but there were no significant differences by subgroup or differences in language by group interactions. The listeners were not able to differentiate items with regard to the specific language conditions being heard. However, a profile plot analysis indicated that differences did occur with regard to the language conditions. In addition, a profile of different English language acquisition was noted by the short, middle or long length of residence groups.