EXPLORING HUMAN–AI DISTRIBUTED CRITICAL THINKING (HADCT): PILOT VALIDATION OF A CRITICAL THINKING SCALE WITH VIETNAMESE EFL LEARNERS

Abstract

The increasing integration of Generative Artificial Intelligence (GenAI) in education has prompted renewed attention to how learners think and reason within AI-mediated environments. While prior studies have explored AI’s potential to enhance reasoning and argumentation, few validated instruments capture the distinctive cognitive, metacognitive, and interactive processes underlying such engagement. Addressing this gap, this study develops and validates the Critical Thinking in GenAI-Assisted Learning Scale (CrTAI), a self-report instrument encompassing three dimensions: analytical reasoning, source verification, and metacognitive selfregulation. Data from 100 English-major undergraduates at a Vietnamese university demonstrates strong internal consistency (α =.93) and satisfactory construct validity (EFA: 63.8% variance explained; CFA: CFI = 0.96, TLI = 0.95, RMSEA = 0.07 [90% CI.03–.11], SRMR = 0.04). Drawing on these findings, the study advances Human–AI Distributed Critical Thinking (HADCT) as a reconceptualization of critical thinking that takes shape within a third space — a shared cognitive territory where human analytical judgement and metacognitive regulation intersect with AI-generated reasoning to co-construct understanding. The CrTAI therefore functions as both an assessment instrument and a conceptual lens, enabling evaluation of learners’ critical engagement while also illuminating how such reasoning unfolds within the human–AI third space through analysis, verification, and reflective regulation.