Bad Behavior: Improving Reproducibility in Behavior Testing

Systems neuroscience research is increasingly possible through the use of integrated molecular and circuit-level analyses. These studies depend on the use of animal models and, in many cases, molecular and circuit-level analyses. 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.