Fig. 1: Example experiment, annotated with common approaches for incorporating analytical variability.
From: Why experimental variation in neuroimaging should be embraced
In this illustration, we break down experiments into three coarse stages: sampling and data collection, processing and curation, and analysis. By looking at the elements of experiments one can perturb at each of the analytic stages, you can see how various forms of common experimental designs are performed. Importantly, any of these elements can be perturbed either jointly or independently, and doing such explorations is likely to provide insight into the generalizability or robustness of results.
