Objective:Virtual reality (VR) simulation training can improve temporal bone (TB) cadaver dissection skills and distributed, self-regulated practice is optimal for skills consolidation. Decentralized training (DT) at the trainees' own department or home offers more convenient access compared with centralized VR simulation training where the simulators are localized at one facility. The effect of DT in TB surgical training is unknown. We investigated the effect of decentralized VR simulation training of TB surgery on subsequent cadaver dissection performance.Study Design:Prospective, controlled cohort study.Setting:Otorhinolaryngology (ORL) teaching hospitals and the Danish national TB course.Participants:Thirty-eight ORL residents: 20 in the intervention cohort (decentralized training) and 18 in the control cohort (standard training during course).Intervention:Three months of access to decentralized VR simulation training at the local ORL department or the trainee's home. A freeware VR simulator (the visible ear simulator [VES]) was used, supplemented by a range of learning supports for directed, self-regulated learning.Main Outcome Measure:Mastoidectomy final-product scores from the VR simulations and cadaver dissection were rated using a modified Welling Scale by blinded expert raters.Results:Participants in the intervention cohort trained decentrally a median of 3.5 hours and performed significantly better than the control cohort during VR simulation (p < 0.01), which importantly also transferred to a 76% higher performance score during subsequent cadaver training (mean scores: 8.8 versus 5.0 points; p < 0.001).Conclusions:Decentralized VR simulation training of mastoidectomy improves subsequent cadaver dissection performance and can potentially improve implementation of VR simulation surgical training.