Aim To compare the perceptions of students when using five different tooth replicas and to detect common errors in the performance of students that might be attributed to a specific type of tooth replica. Methodology Five groups (n = 10 each) of artificial first maxillary molars (DEPT, DRSK, Nissin, DENTALIKE and TrueTooth) were used. All 50 teeth were mounted individually in opaque containers, distributed in 10 packages containing a sample from each with an assigned random order for students to perform root canal treatments. Ten postgraduate students each performed a root canal treatment on the five replicas, in the assigned order, and completed a satisfaction questionnaire. Three trained and calibrated endodontic educators, each with more than 15 years of experience, evaluated their performance using a grading rubric and completed a questionnaire to detect common errors attributed to a specific tooth replica. Inter‐rater reliability was calculated with the interclass correlation coefficient for both consistency and absolute agreement. A two‐way related measures anova was used to assess the interaction amongst evaluators and tooth groups on the average scores of students. Post hoc T3 Dunnet was used to compare groups. The perceptions of students amongst groups were compared with chi‐square and linear‐by‐linear association tests. Results Inter‐rater reliability was very high for both consistency (ICCC = 0.939; 95% confidence interval (CI) 0.902–0.964) and absolute agreement (ICCA = 0.940; 95% CI 0.904–0.965). No significant differences were found amongst the ratings of evaluators; however, students performed differently when using the various tooth replicas (P < 0.05). Overall, 60% of students preferred the DRSK replica for root canal treatment training purposes, followed by DENTALIKE (30%). The least preferred was TrueTooth (70% responses) due to its complex anatomy and poor resistance to instruments and heat pluggers. Evaluators detected several common errors in specific tooth replicas and preferred tooth replicas manufactured based on microCT scans of natural teeth. Conclusions Tooth replicas manufactured based on microCT scans of natural teeth (TrueTooth and DENTALIKE) had much better acceptance amongst evaluators, although students rated and performed less well in TrueTooth replicas due to their greater level of difficulty.

Results of the evaluation of an instructional system which take account of how past experiences and knowledge influence perception, comunication and learning are presented. Experimental activities, problem solving, and discussion period for contrasting previous and new learning have been emphasized in the instructional design. Instructional strategies such as conceptual maps, Gowin's heuristic V and problem solving have been used. The instructional system was evaluated with a sample of 76 secondary school. The selection was based on previous knowledge of optics and the determination of misconceptions in cognitive structure of the same population. Results show relevant differences in tests between controlled and experimental groups.