Abstract
This study examines the feasibility of automating protocol analysis in Japanese as a Foreign Language (JFL) research through artificial intelligence (AI) applications. Drawing on written recall protocols from ten JFL learners, we evaluated text comprehension using both AI-based assessments and human raters. The findings show comparable reliability between automated and manual evaluations, underscoring significant methodological implications for L2 reading research and the automation of protocol analysis. This innovation addresses the labor-intensive nature of qualitative second language acquisition (SLA) research while preserving analytical rigor in comprehension assessment.
https://ror.org/027tyzk91 
