Elaine Ballard

Parent-rated measures of bilingual children’s speech accuracy: Implications for a universal speech screen

Abstract Purpose: This study investigated whether parents were reliable in judging their bilingual children’s speech and whether parent-rated measure of children’s speech could be used as a tool for a universal speech screen to identify children with speech sound disorder (SSD). Method: Single word samples from 33 Korean–English bilingual (KEB) pre-school children were analysed for percentage of consonants correct (PCC). Their parents completed the Intelligibility in Context Scale and a similarly constructed scale devised by Stertzbach and Gildersleeve-Neumann. Spearman rank correlations were used to examine the association between the PCC scores and the parent-rated measures. Result: A number of factors influenced parents’ reliability in judging their bilingual children’s speech, including language dominance and the nature of the questions that were asked. The perception of strangers on a child’s speech problems, as judged by the parents, could be significant in identifying children with SSD. When the cut-off ICS mean score derived from a previous monolingual research was applied to KEB children, over 40% of the sample was identified as requiring a comprehensive clinical assessment. Conclusion: Implementation of a universal speech screen utilising parent-rated measures of children’s speech cannot be introduced without further research.

Prosody perception and musical pitch discrimination in adults using cochlear implants

Abstract Objectives: This study investigated prosodic perception and musical pitch discrimination in adults using cochlear implants (CI), and examined the relationship between prosody perception scores and non-linguistic auditory measures, demographic variables, and speech recognition scores. Design: Participants were given four subtests of the PEPS-C (profiling elements of prosody in speech-communication), the adult paralanguage subtest of the DANVA 2 (diagnostic analysis of non verbal accuracy 2), and the contour and interval subtests of the MBEA (Montreal battery of evaluation of amusia). Study sample: Twelve CI users aged 25;5 to 78;0 years participated. Results: CI participants performed significantly more poorly than normative values for New Zealand adults for PEPS-C turn-end, affect, and contrastive stress reception subtests, but were not different from the norm for the chunking reception subtest. Performance on the DANVA 2 adult paralanguage subtest was lower than the normative mean reported by Saindon (2010). Most of the CI participants performed at chance level on both MBEA subtests. Conclusion: CI users have difficulty perceiving prosodic information accurately. Difficulty in understanding different aspects of prosody and music may be associated with reduced pitch perception ability.

The phonological skills of Samoan speaking 4-year-olds

To date there has been little research on the typical developmental patterns for children speaking Samoan. The research outlined in this paper serves to fill this gap by reporting on the phonological development of Samoan speaking children growing up in the English dominant language environment of Auckland New Zealand. In this study 20 children aged between 4;0 and 4;11 were assessed using a picture naming task that probed their knowledge of the Samoan phonemic inventory. The findings presented here give an indication of what children in their fifth year have as speech sounds in their phonetic inventory and the types of errors that they may still produce. These results demonstrate the similarities and differences that Samoan phonological development shares with other languages. In particular the differences that seem “atypical” when compared with studies on English are discussed. It is tentatively concluded that these differences may well be typical for Samoan speech acquisition given language specific factors and the bilingual context in which these children are growing up.

Estimating language skills in Samoan- and Tongan-speaking children growing up in New Zealand

The MacArthur–Bates Communicative Development Inventory (short form) was adapted for Samoan and Tongan speakers in New Zealand. The adaptation process drew upon language samples from Samoan and Tongan parent–child dyads with 20- and 26-month-old children and adult informants. The resulting 100-word language inventories in Samoan and Tongan, plus a single question about word combinations, were then administered to over 600 mothers of 2-year-olds in the Growing Up in New Zealand pre-birth longitudinal cohort study who identified their children as understanding Samoan or Tongan. Most mothers were able to complete the inventories without the help of an interpreter or interviewer. Important demographic correlates of children’s vocabulary and grammar were mothers’ country of birth, education, and deprivation level, and children’s birth order. Mothers’ birthplace was the single best predictor of children’s vocabulary development in Samoan and Tongan, with children of mothers who were born outside New Zealand having higher Samoan and Tongan vocabularies. Clinical implications are discussed, along with future analyses of the language development of these children from the Growing Up cohort.

Woods Cree /ð/: An Unusual Type of Sonorant1

/ð/ appears in only a handful of the world’s languages (Maddieson 1984 and Ruhlen 1975). In most of these, /ð/ is classified as an obstruent; however, in Woods Cree this segment exhibits a range of characteristics typical of a sonorant. This paper provides a detailed account of the evidence for /ð/’s status in Woods Cree based on its phonetic realizations, the structure of the phonological inventory, voicing patterns, register variation, and phonetic variability in loans from English. While these factors provide strong support that this segment is classified as a sonorant, internal evidence based on obstruent realizations of /ð/—the conflicting patterning of consonant clusters and voicing patterns in word‐final position—point to the ambiguous nature of this segment. Given this internal ambiguity and external factors, such as prolonged contact with English and language shift in the community, we raise the possibility that this Woods Cree segment is being reclassified phonologically from one structural category to another.

A guide to designing a naming test for an under-researched bilingual population: adapting the Boston Naming Test to Tongan

ABSTRACT The purpose of this paper is to provide guidelines on how to develop a naming test for an under-researched language, through adaptation of an English test. An additional consideration is how to make an assessment suitable for bilinguals. While test development involves a number of phases, the focus in this research is on test construction. To illustrate the steps required for this, a Tongan adaptation of the Boston Naming Test (BNT) for Tongan–English bilinguals is discussed. In the adaptation, language structure and cultural appropriateness are considered in item selection. This version retains just 7 of the original BNT words and includes 15 words of particular importance in Tongan culture. As the items in the BNT are ordered according to naming difficulty, the authors also consider how to determine graded difficulty for Tongan words, in the absence of a Tongan corpus or established information on word frequency/difficulty. This issue is resolved by asking Tongan–English bilinguals (n = 20) to rate a list of 175 words for frequency and familiarity. From this set, 120 words were then selected and presented as a picture naming assessment to an additional group of bilinguals (n = 42). Their responses were analysed according to accuracy and speed. A final list of 70 words with pictures was ordered on the basis of these responses for the next step of testing.

Age-related changes in segmental accuracy and error production in Korean-English bilingual children: implications for clinical speech sound assessment procedures

ABSTRACT A lack of information about typical phonological development in bilingual children presents as a challenge to many speech-language pathologists assessing bilingual children with suspected speech sound disorder. This study investigated age-related changes in speech accuracy and error production in 16 preschool Korean-English bilingual children, aged between 3;1 and 5;11, in New Zealand. We found considerable individual variations in the longitudinal data. In particular, we found some error patterns emerged during the course of development, instead of being progressively resolved with age. Our findings suggest that speech-language pathologists should take a considered approach to identifying bilingual children with speech sound disorder solely based on the information provided in cross-sectional studies. We suggest that a follow-up session may provide valuable information facilitating the clinical assessment procedure to identify bilingual children with speech sound disorder.

Prosody Perception and Production in Children with Hearing Loss and Age- and Gender-Matched Controls.

Background: Auditory development in children with hearing loss, including the perception of prosody, depends on having adequate input from cochlear implants and/or hearing aids. Lack of adequate auditory stimulation can lead to delayed speech and language development. Nevertheless, prosody perception and production in people with hearing loss have received less attention than other aspects of language. The perception of auditory information conveyed through prosody using variations in the pitch, amplitude, and duration of speech is not usually evaluated clinically. Purpose: This study (1) compared prosody perception and production abilities in children with hearing loss and children with normal hearing; and (2) investigated the effect of age, hearing level, and musicality on prosody perception. Research Design: Participants were 16 children with hearing loss and 16 typically developing controls matched for age and gender. Fifteen of the children with hearing loss were tested while using amplification (n = 9 hearing aids, n = 6 cochlear implants). Six receptive subtests of the Profiling Elements of Prosody in Speech‐Communication (PEPS‐C), the Child Paralanguage subtest of Diagnostic Analysis of Nonverbal Accuracy 2 (DANVA 2), and Contour and Interval subtests of the Montreal Battery of Evaluation of Amusia (MBEA) were used. Audio recordings of the childrens reading samples were rated using a perceptual prosody rating scale by nine experienced listeners who were blinded to the childrens hearing status. Study Sample: Thirty two children, 16 with hearing loss (mean age = 8.71 yr) and 16 age‐ and gender‐matched typically developing children with normal hearing (mean age = 8.87 yr). Data Collection and Analysis: Assessments were completed in one session lasting 1‐2 hours in a quiet room. Test items were presented using a laptop computer through loudspeaker at a comfortable listening level. For children with hearing loss using hearing instruments, all tests were completed with hearing devices set at their everyday listening setting. Results: All PEPS‐C subtests and total scores were significantly lower for children with hearing loss compared to controls (p < 0.05). The hearing loss group performed more poorly than the control group in recognizing happy, sad, and fearful emotions in the DANVA 2 subtest. Musicality (composite MBEA scores and musical experience) was significantly correlated with prosody perception scores, but this link was not evident in the regression analyses. Regression modeling showed that age and hearing level (better ear pure‐tone average) accounted for 55.4% and 56.7% of the variance in PEPS‐C and DANVA 2 total scores, respectively. There was greater variability for the ratings of pitch, pitch variation, and overall impression of prosody in the hearing loss group compared to control group. Prosody perception (PEPS‐C and DANVA 2 total scores) and ratings of prosody production were not correlated. Conclusions: Children with hearing loss aged 7‐12 yr had significant difficulties in understanding different aspects of prosody and were rated as having more atypical prosody overall than controls. These findings suggest that clinical assessment and speech‐language therapy services for children with hearing loss should be expanded to target prosodic difficulties. Future studies should investigate whether musical training is beneficial for improving receptive prosody skills.