Harnessing artificial intelligence for automatic feedback in a virtual anatomy study tool: A Q-methodology study

Objective structured practical examinations (OSPEs) are standard methods of evaluation in anatomy education. Despite their widespread use, student preparedness for this unique exam structure can be severely limited due to the lack of access to independent study opportunities with bodies and mock OSPEs. Additionally, significant resources are required to administer and grade OSPEs. Resource constraints and lack of in-person access to the anatomy laboratory, exacerbated by the COVID-19 pandemic, prompted the development of an accessible online OSPE study tool powered and graded by artificial intelligence (AI). The AI OSPE is an innovative web-based educational resource that presents users with the opportunity to practice OSPE stations while utilizing AI to grade responses and return immediate personalized feedback. The tool was piloted as a study resource with a cohort of first-year Occupational Therapy students, and Q-methodology was used to assess their opinions and perceptions related to the AI OSPE. Following a by-person factor analysis, two groups emerged: the Spatial Learners and Practical Thinkers. The Spatial Learners emphasized their preference for 3D anatomical structures (versus 2D images), which they felt facilitated easier orientation and preferable view. In contrast, the Practical Thinkers appreciated the utility of the tool and its representativeness of the course material. The results of this study uncover students' preferences regarding a combination of virtual and AI-based anatomy study tools and inform priority areas for quality improvement of those tools.