Large Language Model–assisted curriculum audit in psychology: A prompt-based method

Resumen

Curricula in psychology and other disciplines face increasing pressure to align intended learning outcomes with assessment evidence while responding to changing professional demands and the growing presence of generative artificial intelligence (GenAI) in higher education. This methodological paper introduces the Prompt Playbook, a prompt-supported workflow for programme-level curriculum audit and competency mapping, with an optional later stage of the process for redesign, designed to structure faculty deliberation and improve documentation for quality assurance. The Playbook specifies three steps: (1) Programme Audit to surface misalignments among programme aims, course learning outcomes, and assessments; (2) Illustrative Competency Mapping to prioritise knowledge, skills, and abilities using external signals triangulated with disciplinary standards; and (3) Curriculum Integration to revise learning outcomes and design aligned learning activities and assessments, including paired AI-supported and AI-restricted tasks. A governance layer addressing ethics, data protection, equity, and validation is embedded through logging and verification protocols. Examples in the main text are primarily illustrative and are intended to demonstrate the structure of outputs rather than report new empirical findings; in addition, a supplementary worked example presents one actual captured LLM run to show how the validation protocol handles imperfect responses. The paper concludes with implementation pathways, limitations, and a pilot-study design for future evaluation.