Using a Visual Analogue Scale (VAS) to assess the effect of speech therapy in children with cleft lip and palate

Tools

Dokovova, Marie and Cleland, Joanne and Crampin, Lisa and Campbell, Linsay (2023) Using a Visual Analogue Scale (VAS) to assess the effect of speech therapy in children with cleft lip and palate. In: ICPLA 2023: 2023 Symposium of the international clinical phonetics and linguistics association, 2023-07-04 - 2023-07-07, Salzburg, Austria.

[thumbnail of Dokovova-ICPLA-Using-a-Visual-Analogue-Scale-(VAS)-to-assess-the-effect-of-speech-therapy-in-children-with-cleft-lip-and-palate]
Preview
 Text. Filename: Dokovova-ICPLA-Using-a-Visual-Analogue-Scale-_VAS_-to-assess-the-effect-of-speech-therapy-in-children-with-cleft-lip-and-palate.pdf
Final Published Version
License: Strathprints license 1.0
Download (425kB)| Preview

Abstract

Background: Most speech intervention studies use metrics such as percentage consonants correct to measure changes in speech production after intervention. This typically takes a binary view of correctness of targeted speech sounds. However, emerging evidence suggests that some children make phonetically gradient changes to their speech during, and after intervention. This study measures changes in speech production using a visual analogue scale (VAS) in a 7-year-old child with repaired unilateral cleft lip and palate receiving articulatory intervention for backing of consonants. Methods: Speech materials were 30 /t/-initial word tokens (12 unique words), which could be minimal pairs with /k/, and 18 /k/-initial word tokens (6 unique words), which could be minimal pairs with /t/. Each word was produced with a schwa at the start (e.g., /ə’kap/). The recordings of the /t/-initial words were derived across six speech therapy sessions. The recordings of /k/ were derived from assessment sessions before and after therapy. Listeners were undergraduate Speech and Language Therapy students in their final (4th) year of training. They used a digital visual analogue scale to rate how close each initial consonant was to /t/ or /k/. Responses were converted to continuous responses from 0 to 100, where 0 was most /t/-like and 100 was most /k/-like. A linear mixed effects model (LMEM) was fitted with VAS as an outcome variable, phoneme as a predictor, and accounting for the multiple ratings per listener and per word. The effect of therapy session on VAS was investigated using a LMEM on a subset of the data containing only /t/. Results and conclusion: Data collection and analysis is ongoing. Preliminary analysis suggests that listeners had significantly higher VAS ratings for /k/ than for /t/. VAS shows potential to be used to measure speech production changes in children with cleft lip and palate.

ORCID iDs

Dokovova, Marie ORCID logoORCID: https://orcid.org/0000-0002-4350-6082, Cleland, Joanne ORCID logoORCID: https://orcid.org/0000-0002-0660-1646, Crampin, Lisa and Campbell, Linsay;
CORE (COnnecting REpositories)