AI for Students with Learning Disabilities: A Systematic Review

Abstract

This review study aims to uncover how artificial intelligence (AI) has been employed to support students with learning disabilities (SWLDs). Of the 16 reviewed studies, 10 were focused on dyslexia, with only one focused on dyscalculia and the remaining fo- cused on learning disabilities in general. The study suggests that only 50% of studies focused on school-age children. Seven types of AI applications were used to support SWLDs, including adaptive learning, facial expression, chat robots, communication as- sistants, mastery learning, intelligent tutors, and interactive robots. Adaptive learning was the most widely used. Employing the SAMR-LD (i.e., substitute, augment, modify, and redefine—learning disability) model, we found that AI had been utilized in various ways to support SWLDs (4 substitution, 6 augmentation, 2 modification, and 4 redefinition lev- els). Findings revealed the potential of AI in supporting SWLDs, but the small number of empirical studies also implies significant gaps and the need for more research on how AI can support SWLDs beyond just identifying and diagnosing a learning disability.

Keywords: learning disabilities, artificial intelligence (AI), dyslexia, dyscalculia, adaptive learning, SAMR-LD