Abstract
Speech and language impairments are common pediatric conditions, with as many as 10% of children experiencing one or both at some point during development. Expressive language disorders in particular often go undiagnosed, underscoring the immediate need for assessments of expressive language that can be administered and scored reliably and objectively. In this paper, we present a set of highly accurate computational models for automatically scoring several common expressive language tasks. In our assessment framework, instructions and stimuli are presented to the child on a tablet computer, which records the child's responses in real time, while a clinician controls the pace and presentation of the tasks using a second tablet. The recorded responses for four distinct expressive language tasks (expressive vocabulary, word structure, recalling sentences, and formulated sentences) are then scored using traditional paper-and-pencil scoring and using machine learning methods relying on a deep neural network-based language representation model. All four tasks can be scored automatically from both clean and verbatim speech transcripts with very high accuracy at the item level (83−99%). In addition, these automated scores correlate strongly and significantly (ρ = 0.76–0.99, p < 0.001) with manual item-level, raw, and scaled scores. These results point to the utility and potential of automated computationally-driven methods of both administering and scoring expressive language tasks for pediatric developmental language evaluation.
Highlights
Untreated and undiagnosed developmental language disorder (DLD) is prevalent in young children (Tomblin et al, 1997; Conti-Ramsden et al, 2006; Grimm and Schulz, 2014; Rosenbaum and Simon, 2016) and can have serious behavioral and educational consequences (Clegg et al, 2005)
Using child language data collected both with this computerized instrument and with standard paper-and-pencil administration, we demonstrate the accuracy and feasibility of an automated scoring system for four expressive language tasks
These results demonstrate the promise of computerized approaches to scoring expressive language tasks, which in turn can support clinicians tasked with diagnosis and extend the reach of services for children with developmental language disorder
Summary
Untreated and undiagnosed developmental language disorder (DLD) is prevalent in young children (Tomblin et al, 1997; Conti-Ramsden et al, 2006; Grimm and Schulz, 2014; Rosenbaum and Simon, 2016) and can have serious behavioral and educational consequences (Clegg et al, 2005). Wide-reaching language assessment is urgently needed for early identification of DLD and for planning interventions and tracking the efficacy of these interventions. Such efforts, add strain to scarce and overtaxed clinical resources. Add strain to scarce and overtaxed clinical resources To address this challenge, clinicians, educators, and researchers have begun to explore alternatives to standard assessment paradigms that can be more and more reliably administered and scored. Conventional language test administration and scoring is a labor-intensive and time-consuming task relying on significant clinical expertise. Assessment is typically conducted during a clinical
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have