Abstract: Foreign accent syndrome is a rare phenomenon

Abstract: Foreign accent syndrome is a rare phenomenon where a person assumes an accent that is atypical of their primary language following neurological damage. The purpose of this case study was to identify acoustic features related to the perception of foreign accent and provide plausible explanation of the underlying pathophysiology contributing to the abnormal features. The acoustic properties of utterances produced by an American English speaking stroke patient with "foreign accent syndrome" were evaluated using spectral and time analysis. Pitch contours, formant frequencies, and time-length measures will be presented with proposed interpretation on their relationship with the perception of foreign accent. Introduction: Foreign accent syndrome (FAS) is a rare phenomenon where a person assumes an accent that is atypical of their primary language following neurological damage (Whitaker, 1982). The utterances of a patient with FAS are often perceived as having different articulatory and prosodic patterns found in a language that is not spoken by the patient. Few cases have been reported in the literature that suggest the underlying cause of the perceived accent may be attributed to articulatory irregularities such as increased vowel durations and prosodic deviations (e.g., Blumstein et al., 1987) and higher F0 (e.g., Ryalls & Whiteside, 2006). However, a consistent pattern of articulatory and prosodic features of foreign accent syndrome has not been described in the reported cases. Cohelo & Robb (2001) examined three explanatory models of FAS in relation to a case that they followed. One model suggests that FAS results from limited tongue mobility where patients are unable to make the appropriate articulatory contacts (Moen, 1990). Another model states that FAS constitutes motor impairments characteristic of apraxia of speech (Aronson, 1980). A third model suggests that FAS results from a failure in cognitive planning of speech production (Varley & Whiteside, 1998). Cohelo and Robb (2001) found irregular articulatory patterns in VOT, manner of consonant articulation, and vowel formant transitions and durations in the speech productions of their patient, which supported all three models of FAS. They did not address the prosodic dimensions of the patient's utterances. The purpose of this case study was to examine the acoustic characteristics of a patient who presented with foreign accent syndrome following a stroke. Method: Subjects: The subject (MC) was an 81 year-old right handed female who sustained a large left Middle Cerebral Artery (MCA) Cerebrovascular Accident (CVA) with residual aphasia and a right hemiparesis. Her medical history was unremarkable. She was a native speaker of American English who had been born and raised in the U.S. She completed two years of college and then joined the Redcross during WWII. After the war, MC was employed as a radio/TV talk show host but quit this job to raise a family. She travelled extensively with her husband related to his employment. At one time, MC spoke some Spanish but not fluently. After raising her family, she moved to a retirement community in New England. MC was seen for a speech-language evaluation 5 years post onset the diagnosis of the left CVA. The results of the diagnostic revealed that she had a global aphasia. Her auditory comprehension was moderately impaired at word/sentence level. Repetition was severely impaired at syllable/word level. Her verbal expression was severely impaired at the syllable/word level. MC demonstrated an oral, verbal and to a lesser degree limb apraxia (with the L upper extremities). Furthermore, a mild unilat eral upper motor neuron (UUMN) dysarthria was noted. Interestingly, although MC exhibited motor speech deficits, they did not appear to impede her speech production, which was judged to be fluent. Her utterances consisted of strings of nonsense syllables that were judged to be unintelligible 98-99% of the time. She periodically produced intelligible automatic utterances such as, "I don't know", "yes" and "no". However, when she spoke to other American English speakers following her stroke, the perception of her speech was that she was from a Scandinavian country. A verification of her history revealed that she was not of Scandinavian decent, had never lived in a Scandinavian country and had never spoken any Scandinavian languages. Task & Procedure: Subject was seated in a quiet room and asked to answer some questions during a diagnostic session where the Western Aphasia Battery (Kertesz, 1982) verbal expression section was employed. A speech therapist asked questions and administered assessment tests while a digital video camera was recording. The recorded files were stored onto a digital video disk (DVD). Data Retrieval and Analysis: A laptop (NEC Versa E2000) equipped with a DVD player was used to play back the recorded DVD files (44KHz, 16 bit) via the Window Media Player 9.0 softward. The experimenter listened to the auditory playback and activated the Blaze Media Pro 7.0 (Mystik Media) signal conversion software to export as wav files the part of acoustic recordings containing the subjects verbal responses. The TF32 time-frequency analysis software (Paul Milenkolvic) was used to play back the acoustic signals for further signal segmentation and analysis. Speech Samples: Overall, 23 multiple-word utterances during spontaneous speech were obtained (see Table 1). These responses were facilitated by questions about: the onset of stroke (What happened to you?): Utterance No. 1 the effect of the stroke (What kind of problems?): No. 2-5 recent problems (Any other problems?): No. 6, 7 the subjects past job (Where were you based?): No. 8 the subjects past working experience: No. 9 how many children the subject has: No. 10 the subjects own expectation: No. 11 the subjects name and address: No. 12-15 (The subject was asked to describe a picture): No. 16-23 Table 1. Phonetic Transcription of 23 utterances Results: The subjects speech was characterized by frequent misplaced pauses and prevocalic noises, abnormal vowel prolongation (e.g., see Figures 1 & 2) or shortening, abnormally high number of pitch turns within a syllable (Figures 1 & 2), glottal fry or voice break (Figure 2), frequent sound or syllable repetitions, concatenation of an abnormally high number of short syllables, recurrent utterance (i.e., non-meaningful variety of concatenated CV syllables), and speech automatism (i.e., repeated and unchanging utterances made up of recognizable words). Pitch Contour: Normal pitch contour was percieved to be maintained in speech automatism, e.g., phrases such as Well, I know (n = 1), Well, yes (n = 1), Well, no (n = 2), Well, lets see (n = 1), I know (n =1), bye (n = 2), I dont know (n = 1), all the way (n = 1) and some recognizable single words, e.g., one (n = 5), well (n = 4), players (n = 1). Pitch within a single syllable could involve several unusual upward and downward turns (Figures 2 & 3). Single words and automatic phrases were lodged in separate breath groups with no continuous intonation variation at the sentence level and the timing of breath groups or phrasing was irregular (see Figure 4). Figure 4. Time waveform (top graph), pitch trace (middle), & spectrogram (bottom) for Utterance No. 1. Figure 1. Time waveform & pitch trace for [thu] in Utterance No. 3 Figure 2. Time waveform & pitch trace for [ri] in Utterance No. 3

Figure 3. Time waveform & pitch trace for [Bru] in Utterance No. 16 Formant Frequency: The F1 and F2 frequencies averaged from 10 samples for each of the three corner vowels /i/, /a/, and /u/ respectively were shown in Figure 5, with bidirectional bars representing the standard deviations. The three corner vowels were percieved to be relatively distinguishable from one another, although the vowel space (shown as a red triangle in Figure 4) was relatively small mainly due to elevation of F1 frequency especially for the two high vowels /i/ and /u/ (suggesting lower tongue positioning or pharyngeal constriction) as compared with normal data (/i/: F1= 300 Hz, F2 = 2300 Hz; /a/: 750 Hz, F2 = 1750 Hz; /u/: F1 = 380 Hz, F2 = 950 Hz, Fry, p. 79). Time Length: Number of syllables in words were shown in Table 2. 2800 Time-length measurements include speech rate and articulation rate 2600 /i/ 2400 (Table 3) and length of syllables and between-syllable pauses in 2200 F2 (in Hz) 2000 three-syllable words (Figure 6). As the subjects speech was hardly 1800 /a/ 1600 intelligible, speech rate was defined as the syllable production rate in /u/ 1400 Table 2. Tabulation of syllable number a sentence and calculated as the ratio between the number of syllables 1200 1000 and the time length of a sequence of continuous speech sounds 200 300 400 500 600 700 800 900 F1 bounded by long pauses and perceived by the experimenter as a (in Hz) sentence. Articulation rate was Figure 5. defined as the syllable production F1-F2 Plot for /i/, /a/, & /u/ rate in a breath group. Two methods of estimating the Table 3. Speech and Articulation Rate articulation rate was used. Based (in no. of syllable per second) on a 250 ms method, the articulation rate was calculated as the ratio of the number of syllables to the time length of a sentence excluding only pauses longer than 250 ms. Based on a between-word method, the articulation rate was calculated as the ratio of the number of syllables to the time length of a sentence excluding all perceptible between-word pauses. All of these three Figure 6. Length of syllables & between-syllable pauses rates Discussion & Conclusion : The acoustic characteristics of the subject that have been identified in this case study as being potentially related to the perception of a foreign accent included: word-based (instead of sentence-based) pitch change, excessive within-syllable pitch variation were (similar toexpressed in numbers of syllables per second. tone language or Scandinavian languages), pre-vocalic noises, vowel prolongation/shortening, inappropriate within and between-word pauses, excessive number of syllables in one word, and idiosyncratic pitch and length pattern in multiple-syllable words. These findings suggest that the sign of foreign accent is mostly attributable to the lack of oral-laryngeal coordination in maintaining a normal sentence-based pitch contour and rhythm. Since a normal pitch pattern was found to be maintained in automatic speech, which is mostly word or phrase-based, this lack of oral-laryngeal coordination could be considered to be related to a higher-order deficiency in speech planning instead of a low-order motor problem as far as the overall pitch control is concerned. However, the excessive pitch variation within a syllable, vowel prolongation, and the occurrence of prevocalic noises (even in automatic speech), consonant/vowel repetitions, and excessive articulatoin rate in some unusually concatenated multiple-syllabe words are suggestive of a loworder motor coordination problem and its associated compensatory signs. In addition, although the extent of vowel distortion could not be evaluated in this case study because the speech was mostly unintelligible, observations on the formant frequencies of identifiable corner vowels revealed a tendency for lower than normal tongue positioning for high vowels. In conclusion, these preliminary findings suggest that low-order speech motor difficulties were at least partially responsible for the perception of a foreign accent. References: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Abercrombie, D. (1967). Elements of general phonetics. Edinburgh: Edinburgh University Press. Aronson, A. (1980). Clinical Voice disorders: An interdisciplinary approach. New York: Theime Publishing. Blumstein, S., Alexander, M., Ryalls, J., Katz, W., & Dworetzky, B. (1987). On the nature of foreign accent syndrome: A case study. Brain and Language, 31, 215-244. Cohelo, C., & Robb, M. (2001). Acoustic analysis of Foreign Accent Syndrome. An examination of three explanatory models. Journal of Medical Speech-Language Pathology, 9, 227-242. Fry, D. B. (1979). The Physics of Speech. London: Cambridge University Press. Kent, R. D. (1982). Prosodic Disturbance and Neurologic Lesions. Brain and Language, 15, 259-291. Kertesz, A. (1982). Western Aphasia Battery. New York: Grune & Stratton. Moen, I. (2000). Foreign accent syndrome: A review of contemporary explanations. Aphasiology, 14, 5-15. Ryalls, J. & Whiteside, J. (2006). An atypical case of foreign accent syndrome. Clinical Linguistics and Phonetics, 20 (2/3), 157-162. Varley, R., & Whiteside, S. (1998). Voicing in severe apraxia of speech: Perceptual and acoustic analysis of a single case. Journal of Neurolinguistics, 11, 259-273. Whitaker, H. (1982). Levels of impairment in disorders of speech. In R. Malatesha and L. Hartlage (Eds.), Neuropsychology and Cognition, Vol. 1. The Hague: Nijhoff. 2.

Recently Viewed Presentations

  • AmeriHealth Caritas Pennsylvania Community HealthChoices Presented by: Christopher

    AmeriHealth Caritas Pennsylvania Community HealthChoices Presented by: Christopher

    Provider Support and Partnership. AmeriHealth Caritas Pennsylvania Community HealthChoices. Work with participating HealthChoices providers to obtain contracts to help ensure that all Participants continue to receive their services, without interruption, as we implement Phase 3 CHC.
  • How to make a simple tempeh incubator. - Shop

    How to make a simple tempeh incubator. - Shop

    How to make a simple tempeh incubator. For all information regarding , how to make tempeh or purchasing of materials for tempeh/tofu making please visit . www.mrtempeh.com.au. Completed simple incubator.
  • Perspective Lecture - WordPress.com

    Perspective Lecture - WordPress.com

    Rembrandt perfected this manner of Chiaroscuro. Tenebrism made value an instrument of strong contrast that lends itself to a dramatic and even theatrical style of using light and dark contrast. The tenebrists were interested in peculiar lighting causing mood or...
  • Cerebrovascular Scanning Techniques

    Cerebrovascular Scanning Techniques

    Tel: 0117 32 88412 | Fax: 0117 344 8499 | E-mail: [email protected] or [email protected] Anglia Ruskin Universityv - Cambridge PdC/PgD/MSc Medical Ultrasound- Distance learning/part-time.For details: contact Vicki Elliott, Medical Ultrasound Pathway Leader, Department of Radiography, Institute of Health and Social...
  • Helping Your Child to Read

    Helping Your Child to Read

    Fun Writing activities. Write in wet sand/mud/with chalk/paint. Write in the bath in steam. Use magnetic letters to make words. Make letters out of play dough
  • Graduation Transitions 12

    Graduation Transitions 12

    Graduation Transitions 12 (What you need to know to get it done!)
  • Assessment Without Levels…

    Assessment Without Levels…

    The Reading Test… The English reading test has a strong focus on fictional texts. There is also a emphasis on the comprehension elements of the curriculum. The test consists of a reading booklet and a separate answer booklet. Pupils will...
  • PowerPoint Presentation

    PowerPoint Presentation

    Today's Lecture How to find astronomical objects Constellations, coordinates, precession Motion of stars on the sky First homework is due at the start of class on Monday.