Computer-Assisted Performance-Based Assessment for Mental Health: A Scoping Review.

Abstract

Adolescent mental health is foundational to personal development, yet it faces escalating challenges globally. While traditional assessment methods lack objectivity and ecological validity, integrating computer-assisted technology (CAT) into performance-based assessments (PBAs) offers a promising pathway. This review, following the PRISMA-ScR reporting standard, analyzed 89 articles (2015-2025) to map the assessed components, CAT applications, and scenario diversity in mental health PBAs. Analysis revealed a research emphasis on mental disorders, with critical domains for adolescent development remaining significantly understudied. CATs significantly enhanced PBAs through data analysis, data acquisition, scenario creation, and tool digitization. PBA scenarios are diverse, demonstrating the adaptability of PBAs for multidimensional mental health assessment. Prioritizing the design of PBAs for social-emotional and adaptive assessment is critical for the early identification of adolescent mental health issues. Furthermore, advancing predictive analytics and leveraging large language models for feedback generation are promising ways to unlock CAT's potential in enhancing PBAs. Importantly, integrating and adapting scenarios from validated scales by CATs into PBAs could further enhance assessment typicality and reliability.