Abstract :
[en] Currently, the diagnosis of developmental language disorder (DLD) relies on the use of static assessments that compare an individual’s performance to a normative reference sample. Such assessments are susceptible to biases that can affect the diagnosis accuracy, particularly for individuals from diverse linguistic, cultural, and educational backgrounds. Dynamic assessment has emerged as an assessment method that can reduce the influence of these biases, since it focuses not on an individual’s acquired skills, but on their abilities to learn. It is based on the assumption that children with DLD have lower learning potential, and its measure has proved relevant in differentiating between children with and without DLD. Despite this, dynamic assessment tasks are rarely used in practice. The aim of this rapid review is therefore to identify studies using dynamic assessment in the diagnosis of DLD in monolingual or multilingual children (aged between 0-18 years) to provide speech-language pathologists with an overview of current dynamic assessment tasks. A rapid review methodology was used to achieve this objective. An extensive literature search was conducted through four bibliographic databases: Eric, Medline ALL, PsycINFO, and Scopus. The search strategy was adapted for each database and included all identified keywords and index terms. It focused on two concepts – dynamic assessment and DLD. Furthermore, the reference lists of all selected studies were manually searched for additional relevant articles. The selection of included studies and data extraction was performed by one reviewer. A second reviewer verified 20% of a random sample for accuracy. This review identified 15 studies that extracted information on the characteristics of the participants assessed, the language used in DA task, the design of DA tasks (sandwich, cake or both), the linguistic domains targeted by the DA tasks, and the presence of psychometric proprieties related to the diagnostic of DA tasks (i.e., sensibility and specificity). If we focus on the sensibility and specificity, 10 studies evaluated the sensibility and specificity of DA tasks. The tasks showed better specificity than sensibility. In conclusion, the inclusion of DA in the diagnosis of DLD seems relevant, as it tends to reduce the influence of environmental factors (e.g., linguistic diversity). But if DA tasks is to find its place in assessment practices, it’s necessary to consider the measures (e.g., post test score, modifiability, number of aids needed) and design to be used in a DA task in order to achieve good specificity, and especially, good sensibility. It’s also important to develop DA protocols that are easy to access and adapted to clinical reality to increase their use in clinical practice.
