Object There is increasing evidence that simulation provides high-quality, time-effective training in an era of resident duty-hour restrictions. Simulation may also permit trainees to acquire key skills in a safe environment, important in a specialty such as neurosurgery, where technical error can result in devastating consequences. The authors systematically reviewed the application of simulation within neurosurgical training and explored the state of the art in simulation within this specialty. To their knowledge this is the first systematic review published on this topic to date. Methods The authors searched the Ovid MEDLINE, Embase, and PsycINFO databases and identified 4101 articles; 195 abstracts were screened by 2 authors for inclusion. The authors reviewed data on study population, study design and setting, outcome measures, key findings, and limitations. Results Twenty-eight articles formed the basis of this systematic review. Several different simulators are at the neurosurgeon's disposal, including those for ventriculostomy, neuroendoscopic procedures, and spinal surgery, with evidence for improved performance in a range of procedures. Feedback from participants has generally been favorable. However, study quality was found to be poor overall, with many studies hampered by nonrandomized design, presenting normal rather than abnormal anatomy, lack of control groups and long-term follow-up, poor study reporting, lack of evidence of improved simulator performance translating into clinical benefit, and poor reliability and validity evidence. The mean Medical Education Research Study Quality Instrument score of included studies was 9.21 ± 1.95 (± SD) out of a possible score of 18. Conclusions The authors demonstrate qualitative and quantitative benefits of a range of neurosurgical simulators but find significant shortfalls in methodology and design. Future studies should seek to improve study design and reporting, and provide long-term follow-up data on simulated and ideally patient outcomes.