The Impact of Unscaffolded GenAI Use on Pre-Service Teachers' AI Readiness, Self-Regulated Learning, Critical...

Although GenAI has been increasingly applied in pre-service teacher education, limited evidence is available on how permitted but unscaffolded GenAI use affects pre-service teachers' learning and professional development in authentic course contexts. Grounded in cognitive load theory and the zone of proximal development, this quasi-experimental study examined the effects of unscaffolded GenAI use in an 11-week instructional design course. Two intact sophomore classes at a normal university participated, with one class permitted to use GenAI without prompt templates or instructional guidance and the other not permitted to use GenAI. Data were analyzed using paired-samples t-tests and a one-way analysis of covariance (ANCOVA). After controlling for pretest scores, no significant group differences were found in AI readiness, self-regulated learning, or critical thinking, whereas the control group showed stronger instructional design performance. Within-group comparisons showed that both groups improved in AI readiness and instructional design performance, but not in self-regulated learning or critical thinking. These findings suggest that, in this course context, unscaffolded GenAI access alone may be insufficient to support pre-service teachers' professional learning and may be less favorable for their instructional design performance.