Speech sound disorder (SSD) is a high-incidence developmental disability that can result in long-term negative impacts on academic and career achievement. Although the impacts of SSD can be mitigated through evidence-based interventions, the challenge of diagnosing and managing SSD is greatly exacerbated when working with clients from multilingual backgrounds. A clinician evaluating a bilingual child must avoid over- diagnosing disorder in cases better characterized as differences while also guarding against under-diagnosis of SSD in the bilingual population. This need to reduce misdiagnosis is particularly pressing in the context of under-studied language combinations: languages interact differently depending on their typological properties, and by focusing on a small set of relatively well-documented cases (e.g., Spanish-English), we risk arriving at an overly narrow model of bilingual speech development and disorders. By studying children acquiring Jamaican Creole (JC) and English, this proposal not only benefits an underserved population (Washington, 2012) but will also broaden the theoretical and empirical knowledge base for guiding clinical management of speech development and disorders in a bilingual context. Although recent psycholinguistic models have made strides toward understanding bilingual development in the context of speech perception (e.g., PRIMIR, Curtin et al., 2011), an equivalent theoretical model is lacking for bilingual production. We propose to model bilingual speech development using the Articulatory Map ([A-map]; McAllister Byun et al., 2016). Although initially proposed in a monolingual context, the A-map model indirectly posits a possible mechanism to explain interaction between the phonetic production inventories of two languages developing in parallel (Stoehr et al., 2018). The long-term goal of this research is to arrive at a theoretically coherent understanding of how speech production development differs in bilingual versus monolingual learners, with the goal of developing improved diagnostic markers for bilingual learners. Using transcription-based measures of accuracy and variability across repeated productions, we will test the predictions of the A-map model in a sample of preschool-aged bilingual children acquiring JC and English, as well as a matched monolingual English-speaking sample. Statistical modeling of the relative contributions of acoustic- and transcription-based measures of variability and accuracy will be used to identify the optimal diagnostic criteria for bilingual JC-English speakers. Insights from JC-English bilinguals could generalize to other cases of acquisition of a creole and its lexifier language (e.g., Haitian Creole and French), two languages with extensive cognates (e.g., Spanish and Catalan), or even divergent dialects of one language (e.g., Standard English and African American English). The data we generate in this exploratory study will (1) fill a need to document speech production accuracy and variability in an under-studied bilingual population, and (2) support a continuum of research to develop screening and diagnostic assessment protocols that will reduce misdiagnosis of SSD in bilingual children.
Bilingual children are routinely misdiagnosed for speech sound disorder, with under-diagnosis having lifelong negative impacts on academic and career achievement and over-diagnosis putting the child at risk for social stigma while posing an unnecessary cost to the school system. By targeting an under-studied language combination (Jamaican Creole-English bilinguals), this research will broaden the theoretical and empirical knowledge base that guides clinical management of speech development and disorders in bilingual children. Using a combination of transcription- and acoustic-based measures of variability and accuracy, we will compare typical development across monolingual and bilingual speakers and identify the factors that most accurately separate typical and disordered speech development in the bilingual population.
