It's obviously not ideal, but it's not like you can recruit subject, do some baseline tests, and then prospectively assign some of them to a foreign embassy whe

It's obviously not ideal, but it's not like you can recruit subject, do some baseline tests, and then prospectively assign some of them to a foreign embassy where they /might/ be caught in the middle of something. Not everything needs to be an RCT, and we've learned a lot from case studies, especially where the "case" is due to some effect that's difficult or unethical to replicate. Some examples: - https://en.wikipedia.org/wiki/Henry_Molaison https://en.wikipedia.org/wiki/Henry_Molaison and his bilateral hippocampal lobectomy taught us a lot about memory. - https://en.wikipedia.org/wiki/Patient_DF https://en.wikipedia.org/wiki/Patient_DF has yielded some valuable information about visual processing streams. - https://en.wikipedia.org/wiki/S.M._(patient) https://en.wikipedia.org/wiki/S.M._(patient) and the neural basis of fear (and, more generally, emotions). - Barry Kidston accidentally invented--and validated on himself--a method for studying Parkinson's: (https://en.wikipedia.org/wiki/MPTP https://en.wikipedia.org/wiki/MPTP)