Optimizing Skull Base Anatomy Education: Qualitative Insights From Learner Interaction With 2D, Monoscopic 3D, and Virtual Reality Models.

Objective: Several models are currently employed to teach anatomy to medical students and residents, including cadaveric, two-dimensional (2D), monoscopic three-dimensional (3D), and virtual reality (VR) models. While many studies have focused on measuring how much and how quickly each model can help residents and medical students attain and remember anatomical facts, few studies have compared how models might help learners achieve other aspects of learning (such as synthesis of facts). We sought to investigate how residents and faculty interact with 2D, 3D, and VR models as learning tools for complex head and neck anatomy to improve and inform the design of learning modalities of head and neck anatomy for otolaryngology residents. Methods: A convenience sample of faculty and residents in the otolaryngology and neurosurgical departments was invited to complete semi-structured interviews regarding their perceptions of learning models in different dimensions. 2D, 3D, and VR models of infratemporal fossa (ITF) anatomy were created from the same cadaveric head using photogrammetry. Semi-structured interviews were performed as participants moved through three stations (2D, 3D, and VR), exploring participant perceptions and how they would use each model to assess affordances and barriers to improve and inform design of these learning models. Three investigators analyzed the transcripts using thematic content analysis. Results: Eleven interviews were conducted (four attendings and seven residents). Four themes arose related to perceptions and barriers of the learning models: 2D is preferred for learning head and neck anatomy for the first time; 3D and VR are preferred for self-driven discovery for affording deeper understanding of and for reviewing anatomy; models that prepare participants for the operation room are more preferable; and current models do not adequately balance clarity versus comprehensiveness. Conclusions: These findings help inform design principles for anatomical models tailored to resident level of training. Level of Evidence: N/A.