Effects of an AI-supported inquiry model on AI literacy and authentic performance: A quasi-experimental study with preservice teachers

As generative AI becomes more common in teacher education, programs need instructional models that help preservice teachers use AI purposefully, critically, and productively in authentic academic work. However, evidence on how structured AI-supported inquiry improves both AI literacy and performance remains limited. Grounded in Deweyan Inquiry and the Practical Inquiry model, this quasi-experimental study examined the effects of QUEST + AI, an AI-supported inquiry model built around five phases: Question, Understand, Engage, Solve, and Teach. Ninety-five preservice teachers in an educational research methods course participated in a 10-week study using two intact classes (experimental n = 52; comparison n = 43). Both groups received the same in-class instruction, but the experimental group completed two QUEST + AI cycles with coached generative AI use, whereas the comparison group completed conventional homework. Outcomes included a multidimensional AI literacy measure and a capstone research proposal scored with a common rubric. After controlling for pretest, gender, and grade, the experimental group showed higher overall AI literacy, with small but meaningful gains concentrated in applying AI, AI-supported problem solving, and emotion regulation during AI use. No clear group differences were found for more concept-focused or evaluative dimensions. The experimental group also earned higher scores on the final research proposal, indicating a moderate advantage in authentic performance. These findings suggest that structured AI-supported inquiry can strengthen applied and self-regulatory aspects of AI literacy while improving discipline-relevant performance. Limitations include the modest single-site sample, intact-group design, brief intervention, and reliance on several self-report subscales. Future research should use larger and more diverse samples, longer interventions with follow-up measures, and more performance-based assessments triangulated with protocol-adherence data.