Effects of modelling examples in complex procedural skills training: a randomised study

CONTEXT: Learning complex procedural skills, such as bronchoscopy, through simulation training, imposes a high cognitive load on novices. Example-based learning has been shown to be an effective way to reduce cognitive load and enhance learning outcomes. Prior research has shown that modelling examples, in which a human model demonstrates the skill to a learner, were effective for learning basic surgical skills. However, principles derived from simple skills training do not necessarily generalise to more complex skills. Therefore, the present study examined the effectiveness of integrating modelling examples into simulation training for a more complex procedural skill - bronchoscopy. Moreover, this study extended previous simulation studies by using a physical demonstration rather than video-based modelling examples. METHODS: Forty-eight medical students were randomised into a modelling group and a control group. They all practised on eight bronchoscopy simulation cases individually, followed by standardised feedback from an instructor. Additionally, the modelling group watched three modelling examples of the simulated bronchoscopy, performed by the instructor. These modelling examples were interspersed between cases. Assessments were carried out at pre-, post- and 3-week retention tests with simulator-measured performance metrics. The primary outcome measure was the percentage of segments entered/minute. Other measures were wall collisions, red-out, the percentage of segments entered and the time to completion. Group differences were examined using repeated measures analysis of variance (anova). RESULTS: A clear learning curve was observed for both groups, but as hypothesised, the modelling group outperformed the control group on all parameters except the percentage of segments entered on the post-test and retained this superiority at the retention test. For the primary outcome measure, the percentage of segments entered/minute, the modelling group achieved a 46% higher score at the post-test and a 43% higher score at the retention test. CONCLUSIONS: The present study shows, that integrating modelling examples into the curriculum of bronchoscopy simulation training optimises the role of the instructor and enhances novices' learning outcomes, presumably by optimising cognitive load during training.