# Effect of Training Sequence on Learning Outcomes Using a Haptic Virtual Simulator for Endodontic Access Cavities: A Controlled Experimental Study

## TL;DR

This study found that the order of using a haptic simulator versus natural teeth in dental training does not affect learning outcomes for endodontic cavity preparation.

## Contribution

The novel contribution is demonstrating that training sequence does not impact skill acquisition in endodontic access cavity preparation.

## Key findings

- No significant differences were found between training sequences in simulator metrics or natural-tooth performance.
- Students showed significant improvement in simulator performance between the first and second attempts.
- Learning trajectories improved overall across attempts, with no effect of training sequence.

## Abstract

Background: Haptic virtual simulators are increasingly incorporated into dental education, yet it remains unclear whether the sequence of simulation-based and natural-tooth training influences early endodontic skill acquisition. Objective: The objective of this study was to compare the effect of two training sequences—haptic simulation followed by natural-teeth practice, versus natural-teeth practice followed by haptic simulation—on performance in endodontic access cavity preparation among undergraduate dental students. Methods: Thirty-eight third-year dental students were randomly assigned to two groups. All participants completed three consecutive attempts on a haptic simulator (Simodont®) and one access cavity preparation on extracted mandibular incisors. Simulator metrics included progress, precision, target volume removed, and excess volume removed. Natural-tooth preparations were scored by two blinded endodontists (ICC range = 0.75–0.88). Data were analyzed using Mann–Whitney U tests with Holm correction, Wilcoxon signed-rank tests, and a linear mixed-effects model to characterize learning trajectories. Results: No significant between-group differences were found in any simulator metric (Holm-adjusted p = 0.47–0.62; effect sizes r = 0.12–0.20, 95% CI −0.14 to 0.43) or in natural-tooth performance (all Bonferroni-adjusted p = 1.00). Students demonstrated significant improvement between the first and second simulator attempts (p < 0.05), with a clear learning plateau thereafter. Mixed-effects modelling confirmed significant overall improvement across attempts (p < 0.001), with no effect of training sequence or attempt × group interaction. Conclusions: Training sequence did not influence learning outcomes or final clinical-quality access preparations. Early performance gains suggest a rapid familiarization effect, and both modalities provide complementary—but non-hierarchical—learning affordances. Haptic simulation can therefore be integrated flexibly within preclinical endodontic curricula without compromising educational effectiveness.