Improving reproducibility in neuroscience

Methods which have become standard in the genetics literature – use of large samples and replication of findings – are not always adopted in neuroscience, leading to a risk of false positive findings. Furthermore, flexible analytic pipelines make it possible to analyse any dataset to give a significant result. I will give some examples of good and poor methods from a review I am conducting of studies in the field of neurogenetics. Replication should become a standard practice; this, together with the need for larger sample sizes will entail greater emphasis on collaboration between research groups. Increased complexity in methods is also a barrier to reproducibility, as studies are often hard to understand, even for experts in the field. Moves towards standardized reporting may help overcome this problem. Finally, working with simulated datasets is an excellent way to make researchers aware of how easy it is to generate spurious results.