Comparative Performance of Dental Students on Two Virtual Reality Simulators: A Crossover Study With Simodont and Virteasy

This study aimed to explore the convergent validity of two widely used virtual reality (VR) haptic dental simulators, Simodont and Virteasy, by comparing dental students' performance on comparable abstract psychomotor tasks. Secondary objectives were to examine short-term learning effects, simulator order, and the potential influence of additional clinical exposure.

A two-period crossover study was conducted with 74 fourth-year dental students corresponding to 222 paired exercise datasets. Each participant completed three progressively complex tasks (direct line, indirect line, indirect cross) on both simulators, 6 months apart. Primary outcome measures were the percentage of target removed on Simodont and the percentage of progression achieved on Virteasy. Secondary outcomes included simulator-specific scores, drill time, and total task time. Paired t-tests and within-subject comparisons were used for analysis.

Differences were observed between simulators for target-related metrics and drill time outcomes. Virteasy generated higher target progression values, whereas longer drill times were recorded on this platform. Given the substantial differences in scoring logic, ergonomics, and workflow implementation between systems, these metrics should not be interpreted as indicators of simulator efficiency or learner superiority. No significant period effect was observed (+0.6%, p = 0.36). Both simulators demonstrated short-term learning effects, with the greatest improvement between the first and second attempts. Simulator order had no significant influence on performance. Precision on Virteasy did not correlate with Simodont composite scores.

Both simulators supported effective psychomotor task execution but produced non-equivalent performance outcomes due to differences in feedback, ergonomics, and scoring logic. These findings support the view that Simodont and Virteasy assess partially overlapping constructs rather than interchangeable competencies. Future research should focus on mapping simulator-specific metrics to defined competency frameworks rather than harmonizing scores across platforms.