SureBuilt Series
From Answer-Getting to Sense-Making: A Semi-Systematic Review of Student Use of Large Language Models in Engineering and AEC Higher Education
- School of Applied Management, University of Westminster, London, The United Kingdom
Abstract
This semi-systematic review synthesises 30 learner-focused empirical studies published between 2022 and 2026 on higher-education learners’ use of large language models (LLMs), with particular relevance to Engineering and AEC education. Web of Science and ERIC were searched using core and process-focused query streams, followed by deduplication and staged screening. Four overlapping dimensions were identified: verification and evaluative judgement; metacognition and self-regulation; collaboration and feedback-oriented use; and interaction strategies involving prompting, iteration and dialogue. Across these dimensions, sense- making was associated with purposeful task framing, monitoring, checking, feedback-informed revision and justified decisions about generated outputs. Answer-getting was associated with transactional prompting, passive acceptance, limited checking and minimal revision. The paper contributes an evidence-informed concept-to-indicator framework to support AI-permissive task and assessment design and future process-based research in Engineering/AEC education. The University-Workplace Continuum:
Retrieved from https://surebuiltseries.com/papers/from-answer-getting-to-sense-making-a-semi-systematic-review-of-student-use-of-large-language-models-in-engineering-and-aec-higher-education/