OBJECTIVE:: To investigate the impact of instructor feedback versus no instructor feedback when training a complex operational task on a laparoscopic virtual reality simulator. BACKGROUND:: Simulators are now widely accepted as a training tool, but there is insufficient knowledge about how much feedback is necessary, which is useful for sustainable implementation. METHODS:: A randomized trial complying with CONSORT Statement. All participants had to reach a predefined proficiency level for a complex operational task on a virtual reality simulator. The intervention group received standardized instructor feedback a maximum of 3 times. The control group did not receive instructor feedback. Participants were senior medical students without prior laparoscopic experience (n = 99). Outcome measures were time, repetitions, and performance score to reach a predefined proficiency level. Furthermore, influence of sex and perception of own surgical skills were examined. RESULTS:: Time (in minutes) and repetitions were reduced in the intervention group (162 vs 342 minutes; P <0.005) and (29 vs 65 repetitions; P <0.005). The control group achieved a higher performance score than the intervention group (57% vs 49%; P = 0.004). Men used less time (in minutes) than women (P = 0.037), but no sex difference was observed for repetitions (P = 0.20). Participants in the intervention group had higher self-perception regarding surgical skills after the trial (P = 0.011). CONCLUSIONS:: Instructor feedback increases the efficiency when training a complex operational task on a virtual reality simulator; time and repetitions used to achieve a predefined proficiency level were significantly reduced in the group that received instructor feedback compared with the control group. Trial registration number: NCT01497782.