Dysarthria in Mandarin-speaking children with cerebral palsy: Speech subsystem profiles

Li Mei Chen, Katherine C. Hustad, Ray D. Kent, Yu Ching Lin

Research output: Contribution to journalArticlepeer-review

14 Citations (Scopus)


Purpose: This study explored the speech characteristics of Mandarin-speaking children with cerebral palsy (CP) and typically developing (TD) children to determine (a) how children in the 2 groups may differ in their speech patterns and (b) the variables correlated with speech intelligibility for words and sentences. Method: Data from 6 children with CP and a clinical diagnosis of moderate dysarthria were compared with data from 9 TD children using a multiple speech subsystems approach. Acoustic and perceptual variables reflecting 3 speech subsystems (articulatory-phonetic, phonatory, and prosodic), and speech intelligibility, were measured based on speech samples obtained from the Test of Children’s Speech Intelligibility in Mandarin (developed in the lab for the purpose of this research). Results: The CP and TD children differed in several aspects of speech subsystem function. Speech intelligibility scores in children with CP were influenced by all 3 speech subsystems, but articulatory-phonetic variables had the highest correlation with word intelligibility. All 3 subsystems influenced sentence intelligibility. Conclusion: Children with CP demonstrated deficits in speech intelligibility and articulation compared with TD children. Better speech sound articulation influenced higher word intelligibility, but did not benefit sentence intelligibility.

Original languageEnglish
Pages (from-to)525-548
Number of pages24
JournalJournal of Speech, Language, and Hearing Research
Issue number3
Publication statusPublished - 2018 Mar

All Science Journal Classification (ASJC) codes

  • Language and Linguistics
  • Linguistics and Language
  • Speech and Hearing


Dive into the research topics of 'Dysarthria in Mandarin-speaking children with cerebral palsy: Speech subsystem profiles'. Together they form a unique fingerprint.

Cite this