KMAN Publication Inc. (KMANPUB) Journal of Assessment and Research in Applied Counseling (JARAC) 8 Serial Number 30 2026 07 01 Artificial Intelligence-Based Assessment and Intervention for Specific Learning Disabilities: A Systematic Review Artificial Intelligence-Based Assessment and Intervention for Specific Learning Disabilities: A Systematic Review 1 17 EN Journal Article 2026 02 14 <table> <tbody> <tr> <td> <p><strong>Objective:</strong>  This study aimed to systematically review the existing evidence on the application of artificial intelligence-based methods in the assessment, screening, prediction, diagnosis, intervention, and progress monitoring of specific learning disabilities.</p> <p><strong>Methods and Materials: </strong>This systematic review was conducted according to PRISMA-based principles. Scientific databases including Scopus, Web of Science, PubMed/MEDLINE, ERIC, PsycINFO, IEEE Xplore, and ScienceDirect were searched for studies published between January 2014 and May 2026. The search combined terms related to artificial intelligence, machine learning, deep learning, expert systems, adaptive learning, intelligent tutoring, dyslexia, dyscalculia, dysgraphia, and specific learning disabilities. From 1,286 initially identified records, 362 duplicates were removed, 924 titles and abstracts were screened, and 153 full-text articles were assessed for eligibility. Finally, 42 studies met the inclusion criteria and were analyzed through descriptive and narrative synthesis.</p> <p><strong>Findings: </strong>Inferential synthesis of the included studies indicated that artificial intelligence-based models showed the strongest evidence in assessment-related functions, particularly early screening, diagnostic classification, and academic-risk prediction. Classification and prediction models generally demonstrated acceptable to high performance, especially when multidimensional cognitive, linguistic, academic, behavioral, handwriting, or digital-learning data were used. Evidence was strongest for dyslexia and reading disorder, while dyscalculia and dysgraphia were less frequently investigated but showed promising results, particularly in handwriting analysis and mathematics-risk detection. Intervention-related findings indicated that AI-based adaptive learning systems, intelligent tutoring platforms, mobile applications, serious games, and personalized feedback tools were associated with improvement in short-term academic outcomes, especially reading accuracy, decoding, mathematical performance, writing accuracy, engagement, and progress monitoring. However, evidence for long-term intervention effectiveness, transfer, and sustained educational outcomes remained limited.</p> <p><strong>Conclusion: </strong>Artificial intelligence has considerable potential to support the early identification, individualized assessment, adaptive intervention, and continuous monitoring of learners with specific learning disabilities. Nevertheless, current evidence is stronger for screening and classification than for intervention effectiveness. Future studies should prioritize external validation, longitudinal designs, explainability, ethical implementation, cultural adaptation, and integration with professional educational and clinical judgment.</p> </td> </tr> </tbody> </table> https://www.journals.kmanpub.com/index.php/jarac/article/download/5689/10879