Yes, Graham, that effect has been found so often it has a name in the parapsychological literature, the 'decline effect', other researchers termed it 'shyness effect'. Some of the once successful subjects even scored below average later, which is called 'psi-missing'. Words, words, signifying nothing.

The decline effect has not only been seen within experiments, but also between experiments. A once successful paradigm of research with promising data over two decades usually loses the ability to bring forth significant results, both in new laboratories studying the effect and in the formerly successful laboratories.

Critics like to think that this has to do with improved experimental controls. But there is an alternative interpretation (just to show that parapsychologists can also create theories and not only invent words as 'decline effect'): Over the years, when more critics read the experimental reports, they do not want it to be true and use their mental powers to make the experiments come out with a nil result. The more people with a negative mindset read an article the less likely this line of research will be sucessfull. So the very result of a decline is a proof of the power of mind.

About once every twenty years parapsychologists turn to a new experimental design and are still wondering why mainstream scientists are skeptical. Critics just don't want there to be proof of what they don't want to be true and don't want to listen to beautiful explanations as e.g. by Walter von Lucadou (1984) who said ....psi correlations are an emergent property of macroscopic self-referential systems which are phenomenologically equivalent to non-local correlations in Quantum Physics......

But all this has nothing to do with the line of research that has started this thread. I didn't like the particular article, but I'm willing to defend this line of experiments.

Wolfgang