Generative artificial intelligence-supported self-regulated learning (GenAI-SRL) in L2 writing: scale development, validation, and short-form construction

Abstract The integration of generative artificial intelligence (GenAI) into foreign language writing instruction has brought about human-AI co-regulation, in which textual production arises from learner-algorithm interaction. Investigating how learners regulate their writing processes in GenAI-mediated environments becomes necessary because existing self-regulated learning (SRL) frameworks were developed for human-centered teaching methods. These frameworks may not fully capture the regulatory complexities when GenAI affects text production. We examined SRL within GenAI-supported writing contexts through developing a scale for gauging GenAI-supported SRL in writing (GenAI-SRL) by taking into consideration six regulatory dimensions: cognitive, metacognitive, motivational, affective, social-behavioral, and environmental processes. We recruited two independent samples of Chinese English as a foreign language (EFL) learners. Exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) were conducted with Sample one ( N = 305) and Sample two ( N = 342), respectively, for scale development and validation. EFA revealed a six-factor structure, which CFA further confirmed alongside satisfactory reliability, convergent validity, discriminant validity, criterion validity, and measurement invariance across gender. Ant Colony Optimization (ACO) yielded a 12-item abbreviated form that retained the psychometric properties of the full-length instrument. The study advances SRL theory within GenAI-mediated writing contexts and provides empirical tools for diagnostic and instructional applications.