Interest in educational applications of Digital Twins (DTs) is growing, but translation of theoretical benefits to measurable student experiences has not been investigated in depth. This systematised review evaluates empirically measured learning effects caused by Digital Twins in engineering education, in comparison to traditional teaching methods. It aims to identify key learning outcomes and experiences caused by DTs across cognitive, behavioural, and affective dimensions, and to assess current strength of evidence supporting their use in education. A keyword search of Scopus, Web of Science, and IEEE Explore yielded twenty-six papers. Strength of evidence was found to be tentative for behavioural outcomes, but more promising for cognitive outcomes, especially when DTs are used as an enabling technology for new ways of teaching and learning. Affective outcomes were found to be under-researched. Other research gaps include formalised qualitative outcomes, long-term effects, and experiences of marginalised students. To prevent misconceptions about technological feasibility and benefit, greater transparency is required when defining and describing Digital Twin implementations.
