I-Corps: English language teaching program providing personalized instruction in sentence construction

Drexel University, Philadelphia PA

Investigators

Abstract

The broader impact/commercial potential of this I-Corps project is to fill a void in the products and services that teach language to children with autism and to non-native English speakers. Approximately one-third of individuals with autism exhibit a deficit in expressing themselves in sentences. The traditional language remediation for autism is in-person therapy. Computerized therapies offer more intensive instruction at lower cost, but existing programs do not fully address the language deficits seen in autism. In addition, the National Council for Education Statistics reports a population of English as a Second Language (ESL) students that grew from 8.1 percent in 2000 to 9.5 percent in 2015. The program developed here offers both ESL and autistic students personalized, automated instruction, along with a comprehensive curriculum, in core areas of English: grammar and syntax. Mastering these requires a level of extensive practice, personalized learning, and linguistic expertise that exceeds both the capacities of in-person therapies and teacher-directed instruction, and the capabilities of the language software programs that are currently on the market. This I-Corps project involves the development and commercialization of an English language teaching program providing personalized instruction in sentence construction. Informed by research showing that mastering a language requires active practice, the system uses pictures and verbal prompts (spoken and written) to prompt learners put words together into specific types of phrases and sentences. These targeted language structures gradually increase in length and complexity. Learners construct their responses from scratch through open-ended word-button clicking, keyboard typing, or speech. The system provides interactive, step-by-step feedback (both spoken and written) that is informed by principles of explicit and implicit learning. This feedback, enhanced through color-coded highlighting, diagnoses problems with word choice, word endings, and word order, guiding users through self-corrections to well-formed sentences. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.