Psychological Determinants of Academic Integrity in the Use of Generative AI in Higher Education

Abstract

This paper examines the psychological determinants that shape academically honest and dishonest uses of generative artificial intelligence (GenAI) in higher education. Rather than treating academic misconduct as a purely technological problem, the study conceptualizes academic integrity as a psychologically mediated decision process influenced by moral reasoning, perceived social norms, policy clarity, academic self-efficacy, AI literacy, performance pressure, and beliefs about authorship. Methodologically, the paper adopts a focused narrative review and conceptual synthesis design. A purposive corpus of 16 core publications, including peer-reviewed studies and policy-oriented texts published between 2022 and March 2026, was assembled through targeted searches using combinations of the keywords generative AI, academic integrity, academic misconduct, moral disengagement, AI literacy, and higher education. The reviewed literature suggests that students do not interpret all forms of AI assistance as cheating. Integrity risk increases when institutional guidance is vague, peer use appears normalized, academic pressure is high, and AI tools are perceived as legitimate substitutes for difficult cognitive labor. By contrast, assignment-level guidance, explicit disclosure norms, ethics-oriented instruction, and authentic assessment design appear to reduce integrity risk more effectively than detection-centered responses alone. Based on these findings, the paper proposes an integrative conceptual model in which institutional context shapes psychological appraisal, and psychological appraisal in turn influences disclosed, borderline, or dishonest GenAI use. The paper concludes that effective responses to GenAI-related integrity problems should combine policy clarity, pedagogy, AI literacy, and student support rather than relying only on prohibition or software-based surveillance.