Contents
Professor Stian Reimers (Director for Educational Enhancement and Digital Innovation) School of Health and Psychological Sciences, City, University of London.
[Paper 1]
This session will cover the way GenAI can allow students to have conversations with a simulated client or service user and then get summary feedback on that interaction, to build communication and clinical skills. It’s necessarily formative but is a novel, scalable way of providing feedback.
Artificial intelligence is increasingly used in medical and health sciences, notably in training students for professional roles (for a review see Stamer, Steinhäuser & Flägel, 2023). We looked at using Generative AI (GenAI) like ChatGPT and other Large Language Models in two ways during interprofessional learning days (Reimers & Myers, 2024), specifically in roleplay and other simulation activities which can be challenging to get students to engage successfully with (Stevenson & Sander, 2002).
1. GenAI played the role of a patient in the roleplay of a complex case. Students discussed among themselves how to tackle the case and what to say to the virtual patient. The GenAI patient had been designed to throw out challenges to the students, who would have to learn about each other’s professional roles and work out which agency should handle parts of the case.
2. Students conducted a conventional roleplay, with one of them playing the part of the patient. In this case, other students asked the patient a series of questions from their professional perspective, and wrote up their notes which were then submitted to a well-briefed GenAI interface. The GenAI evaluated the notes for clarity and completeness, and pointed out any information that the students had failed to obtain from the patient. This session will give an overview of these tasks, how they were implemented and run, issues encountered, and feedback from students involved. Although much of the focus will be on GenAI in skills building, the session will be linked to the event’s themes through discussion of its potential for providing scalable, personalised formative feedback to students. There will also be discussion of the wider potential for GenAI in simulation work like roleplays, and of the progress of extending this work to the Law School and Bayes.
References
Reimers, S., & Myers, L. (in press). Using Generative AI agents for scalable roleplay activities in the health sciences. In Beckingham, S. (Ed.) Using Generative AI effectively in HE: sustainable and ethical AI for the common good. Routledge.
Stamer, T., Steinhäuser, J. and Flägel, K., 2023. Artificial Intelligence Supporting the Training of Communication Skills in the Education of Health Care Professions: Scoping Review. Journal of Medical Internet Research, 25, p.e43311. Available at: https://doi.org/10.2196/43311
Stevenson, K. and Sander, P., 2002. Medical students are from Mars-business and psychology students are from Venus-University teachers are from Pluto? Medical Teacher, 24(1), pp.27-31.
