Development, Content Validation, and Preliminary Effectiveness of the NEXUS Multicomponent Educational Protocol Integrating AI-Mediated Socratic Inquiry, Identity Meaning-Making and Impactful Expression in Adolescents: A Small-Cluster Randomized Trial with Three-Month Follow-Up

Background: Generative artificial intelligence (GenAI) can support explanation and feedback, but unrestricted answers may encourage cognitive offloading, weak source evaluation, and diminished learner authorship. This study developed and content-validated NEXUS, which restricts GenAI to Socratic coaching, and examined preliminary effectiveness for learning agency, intrinsic motivation, and critical-thinking disposition among male upper-secondary students in Qazvin, Iran. Methods: Phase I combined needs mapping in ۴۶۳ students, co-design, cognitive interviews, and two-round Delphi validation by ۱۵ experts. Phase II randomized ۱۲ classes (۱۰۶ students; six clusters per arm) to ten weekly NEXUS sessions or a dose- and technology-matched active control. Three-month follow-up was primary. Mixed-model estimates were paired with CR۲ inference, CR۳ sensitivity analysis, restricted wild-cluster bootstrap-t intervals, and exact within-pair randomization tests. Model-based Hedges g used total class-plus-student variance. Results: The ۵۵-element manual achieved S-CVI/Ave = ۰.۹۱; universal agreement was ۰.۳۱, and Fleiss kappa was ۰.۶۶ for relevance and ۰.۶۱ for necessity. Follow-up completion was ۸۹.۶%. Adjusted follow-up differences were ۰.۴۴ for agentic engagement (۹۵% CI ۰.۱۰ to ۰.۷۸; Holm-adjusted p = ۰.۰۴۲; g = ۰.۴۳), ۰.۳۱ for intrinsic motivation (۹۵% CI -۰.۰۴ to ۰.۶۶; adjusted p = ۰.۰۸۶; g = ۰.۳۰), and ۶.۱ for critical-thinking disposition (۹۵% CI ۰.۶ to ۱۱.۶; adjusted p = ۰.۰۷۸; g = ۰.۳۸). The exact global multivariate randomization test was p = ۰.۰۳۱. Fidelity averaged ۸۶.۸%; minor AI-output and participant-burden events occurred, but no serious related event was observed. Conclusion: NEXUS showed credible content validity and a promising joint signal, strongest for agentic engagement. With only ۱۲ clusters, incomplete blinding, and no prospective registration, independent replication is required.

کلیدواژه ها:

adolescent learning ، Artificial intelligence ، critical thinking ، intrinsic motivation ، learning agency ، Socratic questioning