An Exploratory Comparative Analysis of Mixed Reality Simulation Versus 3D‐Printed Models for Undergraduate Training in Third Molar Removal

ABSTRACT Purpose/Objectives To compare the effectiveness of MR simulation and 3D‐printed models in training undergraduate students in wisdom tooth removal, and to assess students’ perceptions of the MR simulator. Methods Students trained first on either the MR simulator or 3D‐printed plastic models and then repeated the training using the other method on the same day (“MR‐first” and “plastic‐first” groups). Post‐training perceptions were collected using a 7‐point Likert scale covering realism, usability, and effectiveness. Group differences in performance metrics were assessed using linear mixed models and generalized estimating equations. Likert scale scores were summarized descriptively. Results Twenty undergraduate dental students participated in a randomized crossover study. Task completion time did not differ significantly between simulation modalities, although participants tended to complete the second simulation faster, indicating a learning effect. Bone removal volume did not differ significantly between simulation modalities despite a trend toward less bone removal in the MR simulator ( β = −20.09 mm 3 , 95% CI −40.59 to 0.42, p = 0.055). Within‐modality analysis showed a significant reduction in bone removal between Period 1 and Period 2 for the MR simulator ( β = −33.29 mm 3 , 95% CI −53.92 to −12.66, p = 0.0016), while no such effect was observed for the plastic model. Procedural error rates differed: Lingual bone perforation was significantly more frequent in the MR simulator compared with the plastic model (OR = 5.72, 95% CI 1.60–20.43, p = 0.007), whereas second molar damage and incomplete root removal did not differ significantly between modalities. Students rated the MR experience highly (realism/usability: mean = 6.40, SD = 0.68; perceived learning effectiveness: mean = 6.23, SD = 0.87). Conclusions MR offers high‐perceived realism and allows repeated, standardized training, while 3D‐printed models provide tactile feedback. Together, these findings suggest that MR can serve as a valuable complement to existing teaching methods in third molar surgery training.