Vowel production in aphasia: preliminary acoustic findings from arabic

* Significant difference between means at P <.05 or less utilizing the MannWhitney U-test (z -store correction for tied ranks).

Vowel duration. Figure 1 presents the average durations (ms) of the three long vowels (/a:,u:,i:/) for speakers with Broca's aphasia and healthy speakers. Although Broca's aphasics produced longer durations for all three vowels than the healthy speakers, the ANOVA analysis showed no significant differences among the speaker groups for the vowels /a:/ and /u:/ except for the vowel /i:/ being significantly longer for speakers with Broca's aphasia relative to the healthy speakers.

Figure 1 Box Plot of Mean Vowel Duration (Ms) for the Vowels /A:, U:,I:/ As Produced by the Healthy Speakers and Speakers with Broca's Aphasia

Vowel space. Figure 2 illustrates the acoustic vowel space of Broca's aphasics and healthy speakers. When comparing Boca's aphasics' finings to healthy speakers, differences between the groups were observed. More specifically, the front vowel /i:/ is more backed in Broca's aphasics than healthy speakers. Vowel backness is associated with F2 lowering, indicating reduced tongue advancement. Interestingly, Boca's aphasics produced /u:/ more fronted compared to the healthy speakers. Also, the acoustic vowel space showed that Broca's aphasics had the vowel categories approximately in the same regions of the vowel space in healthy speakers. Furthermore, unlike the healthy speakers, /a:/ was lower in the acoustic space in Boca's aphasics speech. Furthermore, speakers with Boca's aphasia produced (/i:, u:/) lower in the acoustic space than the healthy speakers whose (/i:, u:/) were located much higher.

As shown in Fig. 2, Broca's aphasics demonstrated a relatively reduced acoustic area compared to healthy speakers. The vowel ellipses of all vowel understudies in Broca's aphasics were noticeably larger than the healthy speakers, as can be seen in Fig. 3. Among the three vowels, Broca's aphasics produced the vowel /u:/ with the largest ellipse areas relative to the healthy speakers, suggesting tongue positioning deviations.

Figure 2 The Acoustic Vowel Space of Broca's Aphasics and the Healthy Speakers

Figure 3 The Acoustic Vowel Space of Broca's Aphasics and the Healthy Speakers

Furthermore, individual formant spaces showed notable variations regarding vowel placement patterns with most vowel distribution variability and overlap for all speakers with Broca's aphasia except for th e participant APH6, as shown in Fig. 4. In contrast, the healthy speakers showed no overlap while producing the target vowel categories, as presented in Fig. 4. The findings generally revealed differences between Broca's aphasics and the healthy speakers concerning vowel height and vowel frontness/backness.

Figure 4 F1 and F2 Frequencies for Normal Speakers (NOR)

Colored diamonds and circles represent the target vowels. There is no distribution overlap for healthy speakers NOR1 and NOR2.

Figure 5 F1 and F2 Frequencies for Speakers Diagnosed with Aphasia (APH)

Colored diamonds and circles represent the target vowels. There is a distribution overlap for the speakers with Broca's aphasia APH1, APH4 and APH7 with the exception of APH6.

Discussion

The present study was aimed to examine vowel production in Arabic-speaking individuals with Broca's aphasics compared to healthy speakers, focusing on vowel duration and formant frequencies. The findings showed that speakers with Broca's aphasia had longer vowel durations compared to healthy speakers, but no statistically significant differences were observed, except /i:/ being longer by speakers with Broca's aphasia. These results are consistent with previous studies (Baum & Pell, 1997; Baum & Boyczuk, 1999; Schirmer, 2004; Seddoh, 2008).

Such excessive prolongations indicate temporal impairments and motor planning deficits. They may also suggest inadequate tactile feedback during vowel production compared to consonants. Therefore, this may make Broca's aphasics are uncertain about the articulatory configurations and tongue movements required for producing vowels accurately and intelligibly (Williams & Seaver, 1986). The fact that slow speech also characterized Broca's aphasia supports such abnormal temporal durations, indicating self-monitoring problems and motor speech planning deficits. It is also interesting to note that our speakers with Broca's aphasia successfully followed the same durational pattern demonstrated by the control group, namely by producing the high vowels with shorter durations than the low vowels. Similar findings were also reported by Gandour and Dardarananda (1984).

The results also showed that speakers with Broca's aphasia had significantly higher F1 values than healthy speakers. The present study results are consistent with previous studies reporting deviated formant patterns in speakers with Broca's aphasics compared to healthy speakers (Keller, 1975; Ziegler & Hoole, 1989; Haley at al., 2001). Additionally, our findings generally support the common consensus that formant frequencies can predict severity and variability among speakers with Broca's aphasia (Ryalls, 1986; Liu et al, 2005; Tsao et al., 2006).

Some studies have found a positive correlation between F1 values and gender, with F1 values being higher in female speakers than male speakers (Kim et al., 2011; Lee et al., 2016). However, gender distribution between our Broca's aphasics was not equal, where only two females participated in the study (2/8 females, see table 2). So, it is unlikely that this asymmetric gender distribution drives high F1 values. Therefore, although we view this gender imbalance as a limitation of the current study, there is no substantial evidence that this may have affected the results introduced here. However, higher F1 values generally suggest an articulatory setting with more oral cavity opening and lowered mandible (Lee et al., 2016).

The vowel acoustic spaces and ellipses areas (Figures 2 and 3) showed that speakers with Broca's aphasia displayed more dispersions in the back vowels than the front vowels (Figures 2 and 3). As mentioned earlier, the acoustic features of vowel formant frequencies indicate tongue placement. The results suggest that speakers with Broca's aphasia showed imprecise articulatory positioning of tongue height and advancement, indicating vowel fronting and backing.

The inspection of the acoustic vowel space in our speakers with Broca's aphasics showed a reduced vowel acoustic area and more centralized vowels than the healthy speakers, indicating tongue positioning constraints, confirming previous findings (Ryalls, 1986). The acoustic analysis also suggests that Broca's aphasics in the present study are less intelligible due to their relatively rescued size of vowel space as compared to the healthy speakers. Such results provide strong evidence to previous findings indicating that the acoustic vowel space can serve as an index for vowel accuracy and speech intelligibility (Kim et al., 2011; Karlsson & van Doorn, 2012).

Speakers with aphasia in the present study also reported variability of formant frequencies. The results also revealed acoustic overlaps between the vowels categories in Broca's aphasics, as shown in figure 5. These findings are consistent with those reported by Haley et al. (2010). Acoustic overlap may suggest a relative lack of formant differentiation, imprecise articulatory configurations, and more vowel centralization (Mefferd & Green, 2010; Karlsson & van Doorn, 2012). A closer inspection of the vowel ellipses areas also showed that our participants with Broca's aphasia demonstrate greater dispersions in the high vowels /u:/ and /i:/ than in the low vowel /a:/ (Figures 2 and 5).

Taken together, the results of the present study would give more insight into the vowel acoustic characteristics of individuals with Broca's aphasia and expand our understanding of the underlying nature of Broca's aphasia. The results also suggest that vowels should become an integral part of speech therapy approaches and design.

For example, a speech therapist should include those vowels with more remarkable acoustic overlaps in their training and therapy activities to help patients to identify the differences between vowel pair affected by this acoustic overlap.

Limitations of the Study

The present study is limited by several aspects, including the number of participants, stimuli, and analysis scope. First, the sample size, which included eight individuals (six males and two females) with Broca's aphasia, was relatively small. Including more female speakers can also help determine the gender effect on formant frequencies and vowel durations, allowing for more comparisons between speaker groups.

Secondly, the speech stimuli are relatively small as it includes three vowels (/a:,u:,i). This number of vowels may not allow for general conclusions regarding vowel production in Arabic-speaking individuals with Boca's aphasics. However, the present study's findings are preliminary, and further research is needed to gain better understanding of vowel deficits in Broca's aphasia. Finally, the analysis did not address the formant trajectory and how formants change in time, which was found to be crucial for vowel duration (Liss & Weismer, 1992).

Conclusion

The present paper examined the acoustic features of vowel production in Palestinian Arabic-speakers with Broca's aphasics compared to healthy speakers. The findings showed speakers with Broca's aphasia produced all vowel categories with more variability compared to the healthy speakers. The results also revealed significantly higher F1 values for speakers with Broca's aphasics than the healthy speakers. Reduced vowel acoustic spaces and overlaps were noted in the speakers with Broca's aphasia. Such acoustic variability might reflect articulatory inconsistency and tongue positioning constraints.

Furthermore, although speakers with Broca's aphasia produced longer vowel durations than the healthy speakers, they were not significantly different between speaker groups. The present findings support previous studies concerning abnormal temporal and spectral patterns in Broca's aphasia. Acoustic measures of vowels can help speech therapists develop speech training programs to enhance speakers' speech intelligibility with Broca's aphasia.

References

1. Al-Ani, S. (1970). Arabic phonology; an acoustical and physiological investigation. The Hague: Mouton.

2. Abou Haidar, L. (1994). Norme linguistique et variabilitd dialectale: analyse formantique du systeme vocalique de la langue arabe. Revue de honetique Appliquee, 110, 1-15.

3. Alotaibi, Y. & Hussain, A. (2009). Formant based analysis of spoken Arabic vowels. Lecture Notes Computer Science, 5707, 162-169.

4. Aziz, M. A., Hassan, M., Razak, R., & Garraffa, M. (2020). Syntactic abilities in Malay adult speakers with aphasia: A study on passive sentences and argument structures. Aphasiology, 34(7), 886-904. https://doi.org/10.1080/02687038.2020.1742283

5. Baum, S., & Boyczuk, J. (1999). Speech timing subsequent to brain damage: Effects of utterance length and complexity. Brain and Language, 67(1), 30 45. https://doi.org/10.1006/brln.1999.2047

6. Baum, S., & Pell, M. (1997). Production of affective and linguistic prosody by brain-damaged patients. Aphasiology, 11(2), 177-198. https://doi.org/10.1080/02687039708248463

7. Belkaid, Y. (1984). Les voyelles de 1'arabe littdraire moderne. Analyse spectrographique. Travaux de 1'Institut de Phonetique de Strasbourg, 16, 217-240.

8. Boersma, P. & Weenink, D. (2020). Praat: Doing phonetics by computer (version 1.43) [computer program]. http://www.praat.org.

9. Boye, K., & Bastiaanse, R. (2018). Grammatical versus lexical words in theory and aphasia: Integrating linguistics and neurolinguistics. Glossa: A Journal of General Linguistics, 3, 1-18.

10. Caramazza, A. (1997). How many levels of processing are there in lexical access? Cognitive Neuropsychology, 14(1), 177-208. https://psycnet.apa.org/doi/10.1080/026432997381664

11. Chodroff, E., & Wilson, C. (2020). Acoustic-phonetic and auditory mechanisms of adaptation in the perception of sibilant fricatives. Attention, Perception, & Psychophysics, 82(4), 20272048. https://doi.org/10.3758/s13414-019-01894-2

12. Ferguson, S., & Kewley-Port, D. (2007). Talker differences in clear and conversational speech: Acoustic characteristics of vowels. Journal of Speech, Language, and Hearing Research, 50(5), 1241-1255. https://doi.org/10.1044/1092-4388(2007/087)

13. Fyndanis, V., Arcara, G., Christidou, P., & Caplan, D. (2018). Morphosyntactic production and verbal working memory: Evidence from Greek aphasia and healthy aging. Journal of Speech, Language, and Hearing Research, 61(5), 1171-1187. https://doi.org/10.1044/2018 jslhr-l-170103

14. Gandour, J., & Dardarananda, R. (1984). Prosodic Disturbance in Aphasia: Vowel Length in Thai. Brain and Language, 23(2), 206-224. https://doi.org/10.1016/0093-934x(84)90064-6

15. Garraffa, M., & and Fyndanis, V. (2020). Linguistic theory and aphasia: an overview. Aphasiology, 34(2), 905-926. https://doi.org/10.1080/02687038.2020.1770196

16. Haley, k., Bays, G. & Ohde, R. (2001). Phonetic properties of aphasic-apraxic speech: A modified narrow transcription analysis. Aphasiology, 15(12), 1125-1142. https://doi.org/10.1080/02687040143000519

17. Haley, K., Ohde, R. Wertz, T. (2011). Vowel quality in aphasia and apraxia of speech: Phonetic transcription and formant analyses. Aphasiology, 15, 1107-1123.

18. Johansson, M. (2012). Aphasia and Communication in Everyday Life: Experiences of persons with aphasia, significant others, and speech and language pathologists. Doctoral thesis of Uppsala University.

19. Karlsson F, & van Doorn, J. (2012). Vowel formant dispersion as a measure of articulation proficiency. Journal of the Acoustical Society of America, 132(4), 26332641. https:ZZdoi.org/10.1121Z1.4746025

20. Keller, E. (1978). Parameters for vowel substitutions in Broca's aphasia. Brain and Language, 5(3), 265-285. https://doi.org/10.1016/0093-934x(78)90026-3

21. Kempler, D. & Van Lancker, D. (2002). Effect of Speech Task on Intelligibility in Dysarthria: A Case Study of Parkinson's Disease. Brain and Language, 80(3), 449464. https://doi.org/10.1006/brln.2001.2602

22. Kurowski, K., Blumstein, S., Palumbo, C.,Waldstein, R., &Burton, M. (2007). Nasal Consonant Production in Broca's and Wernicke's Aphasics: Speech Deficits and Neuroanatomical

23. Correlates. Brain and Language, 100(3'), 262-275. https://doi.org/10.1016/j.bandl.2006.10.002

24. Lee, J., Shaiman, S., & Weismer, G. (2016). Relationship between tongue positions and formant frequencies in female speakers. Journal of the Acoustical Society of America, 139(1), 426440. https:ZZdoi.org/10.1121Z1.4939894

25. Liu, H., Tsao, F., & Kuhl, P. (2005). The effect of reduced vowel working space on speech intelligibility in Mandarin-speaking young adults with cerebral palsy. Journal of the Acoustical Society of America, 117(6), 3879-3899. https://doi.org/10.112V1.1898623

26. Martinez-Ferreiro, S., Bastiaanse, R., & Boye, K. (2020). Functional and usage-based approaches to aphasia: The grammatical-lexical distinction and the role of frequency. Aphasiology, 34(8), 927-942. https://doi.org/10.1080/02687038.2019.1615335

27. Matchin, W. (2018). A neuronal retuning hypothesis of sentence-specificity in Broca's area.

28. Psychonomic Bulletin & Review, 25(5), 1682-1694. https://doi.org/10.3758/s13423-0171377-6

29. Mefferd, A., & Green, J. (2010). The articulatory-to-acoustic relationship in response to speaking rate and loudness manipulations. Journal of Speech, Language, and Hearing Research, 53(5), 1206-1219. https://doi.org/10.1044/1092-4388(2010/09-0083)

30. Menn, L., & Obler, L. (1990). Agrammatic Aphasia: A Cross-Language Narrative Sourcebook. Amsterdam und Philadelphia: John Benjamins.

31. Metoui, M. (1995). Phono Lab: Computerprogramm zur Artikulatorisch-Akustischen Datenanalyse. Arbeitsberichte des Instituts fur Allgemeine und Vergleichende Sprachwissenschaft der Universitdt Mainz, 1, 1-100.

32. Newman, D., & Verhoeven, J. (2002). Frequency analysis of Arabic in connected speech, Antwerp Papers in Linguistics, 100, 77-86.

33. Paradis, M. (1987). The assessment of bilingual aphasia. Hillsdale, NJ: Lawrence Erlbaum Associates.

34. Ouden, D., Malyutina, S., Basilakos, A., Bonilha, L., Gleichgerrcht, E., Yourganov, G., Hillis, A.

35. E., Hickok, G., Rorden, C., & Fridriksson, J. (2019). Cortical and structural-connectivity damage correlated with impaired syntactic processing in aphasia. Human Brain Mapping, 40(7), 2153-2173. https://doi.org/10.1002/hbm.24514

36. Ryalls, J. (1986). An acoustic study of vowel production in aphasia. Brain and Language, 29(1), 48-67. https://doi.org/10.1016/0093-934x(86)90033-7

37. Schimer, A. (2004). Timing speech: a review of lesion and neuroimaging findings. Cognitive Brain Research, 21(2), 269-287. https://doi.org/10.1016/j.cogbrainres.2004.04.003

38. Seddoh, S. (2008). Conceptualisation of deviations in intonation production in aphasia.

39. Aphasiology, 22(12), 1294-1312. https://doi.org/10.1080/02687030701640289

40. Tsao, Y., Weismer, G., & Iqubal, K. (2006). The effect of intertalker speech rate on acoustic vowel space. Journal of the Acoustical Society of America, 119(2), 10741082. https://doi.org/10.112V1.2149774

41. Verhaegen, C., Delvaux, V., Fagniart, S., Huet, K., Piccaluga, M. (2020). Phonological and phonetic impairment in aphasic speech: an acoustic study of the voice onset time of six French-speaking aphasic patients. Clinical Linguistics & Phonetics, 34(3), 201221. https://doi.org/10.1080/02699206.2019.1619095

42. Williams, S., & Seaver, J. (1986). A comparison of speech sound durations in three syndromes of aphasia. Brain and Language, 29(1), 171-182. https://doi.org/10.1016/0093-934X(86)90041-6

43. Ziegler, W., & Hoole, P. (1989). A combined acoustic and perceptual analysis of the tense-lax opposition in aphasic vowel production. Aphasiology, 3, 449-463.

Размещено на Allbest.ru