# Human-in-the-loop in artificial intelligence in education: A review and entity-relationship (ER) analysis **Author:** Bahar Memarian and Tenzin Doleck **Published:** 2024 **Length:** 36 mins (9,342 words) ## In Summary Reviewing 28 studies, Memarian and Doleck examine how humans and AI interact in educational and related settings. Using an entity-relationship framework, they find that most systems involve two or three participants or components, with predominantly one-directional interactions between humans and AI. Research commonly positions AI as either an expert providing feedback or a system learning from human input, rather than enabling genuinely reciprocal relationships. The authors argue for more nuanced models that recognise diverse human roles, contexts and opportunities for shared development. **Source:** [https://www.sciencedirect.com/science/article/pii/S2949882124000136](https://www.sciencedirect.com/science/article/pii/S2949882124000136) ## What might this mean for us? How might Higher Education move beyond treating human involvement as simply checking or correcting AI outputs? Electric Dreams' dialogic assessment and agent explorations offer opportunities to examine relationships between students, educators and intelligent systems. Could mapping who initiates, interprets and validates decisions help reveal where agency resides, and how technologies might support rather than constrain collective academic judgement? ## Links in this reading - [Personalized education and artificial intelligence in the United States, China, and India, Bhutoria](https://doi.org/10.1016/j.caeai.2022.100068) - [BeParrot: Efficient Interface for Transcribing Unclear Speech via Respeaking](https://doi.org/10.1145/3490099.3511164) - [PrISM-Tracker: Multimodal Procedure Tracking with User-Driven Error Handling](https://doi.org/10.1145/3569504)