Independent Professional Judgment in AI-Assisted Psychology: A Conceptual Framework for Assessing and Preserving Cognitive Autonomy in Future Psychologists
The growing integration of generative artificial intelligence (AI) into psychological education introduces a fundamental challenge: how can future psychologists benefit from intelligent assistance while maintaining the capacity for independent professional judgment? Although AI systems can support case formulation, literature interpretation, intervention planning, and reflective learning, their recommendations may also influence students' reasoning before those students have developed sufficiently robust professional competencies. This article examines independent professional judgment as a multidimensional capacity rather than the simple ability to accept or reject AI-generated recommendations. Drawing on research concerning automation bias, cognitive offloading, metacognition, human–AI decision-making, and professional ethics, the article introduces the Independent Professional Judgment Preservation (IPJP) framework. The proposed framework distinguishes five interconnected dimensions: independent formulation, evidence discrimination, metacognitive calibration, accountable revision, and unaided competence retention. It further outlines mechanisms through which AI assistance may strengthen or weaken professional reasoning, specifies testable hypotheses, and proposes an experimental research design involving psychology students and simulated professional cases. The central argument is that independent professional judgment does not require intellectual isolation from AI. Rather, it requires the ability to generate, evaluate, revise, and defend professional conclusions using relevant evidence, contextual understanding, and ethical responsibility. The article proposes that the educational value of AI should be evaluated not only by immediate task performance but also by the extent to which students retain transferable reasoning competencies when AI assistance is unavailable
Authors
- Roman V. Shabalin (ORCID: https://orcid.org/0009-0002-6818-6297)
Institutions
- Novosibirsk State Technical University (RU)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-14
- DOI
- https://doi.org/10.5281/zenodo.24216660
- Primary Topic
- Artificial Intelligence in Healthcare and Education
- Type
- preprint