I recently contributed to The BPS Research Digest for a series of articles on the best psychology experiment that's never been done. My suggestion is the 'Truman Show Experiment'.
While the greatest psychology experiment imaginable has never been done, it has been filmed. The film is The Truman Show in which the main character Truman Burbank, played by Jim Carrey, lives in an entirely manufactured world, and has done since birth. The island on which he lives is a stage, his wife is an actress along with all his friends, neighbours and acquaintances - indeed everyone on the island is playing a part.
In the film, Truman's every move is broadcast to an adoring global audience of millions - a run-of-the-mill Hollywood dream. But in the psychologist's dream, Truman's every move would be broadcast back to a waiting team of analysts.
All breeds of psychologists would be in on the act. Developmental psychologists examining how Truman changes over his life-span, social psychologists testing his obedience, conformity and social identity and cognitive psychologists checking his memory and attention. All in a controlled environment, and over a controlled lifespan.
The problem is that a sample size of one doesn't play well in the academic journals. So more 'participants' would be required. Truman is soon joined by an ever-growing cast of participants. He would need a 'real' wife, 'real' children and 'real' in-laws. But as more participants are added, the environment becomes less controlled, more chaotic, open to the vagaries of human behaviour. All those nicely controlled experiments would start to break down as real participants interacted with each other in unexpected ways.
No, we need many Trumans, each in separate artificial environments - each subjected to slightly different environments and having slightly different genetic make-ups...
Right, I'm off to write my grant application. Do you think a few billion pounds ought to cover it?