Systems neuroscience research is increasingly possible through the use of integrated molecular, genetic, and circuit analyses. These studies depend on the use of animal models, and in many cases, behavioral analyses to investigate changes associated with genetic, pharmacologic, epigenetic, and other types of environmental manipulations. We illustrate typical pitfalls resulting from poor validation of behavior tests. We describe experimental designs and enumerate controls needed to improve reproducibility in investigating and reporting of behavioral phenotypes.