STARTLING evidence that the human mind can exert paranormal control over objects has been uncovered by researchers whose findings have confounded even hardened sceptics. Experiments conducted by a team at Princeton University are being hailed as the most convincing demonstration yet of so-called psychokinesis (PK), the supposed ability of thought to affect inanimate objects. Until now, most claims for the existence of PK have rested largely on anecdotes of poltergeists wrecking homes and demonstrations by stage performers such as Uri Geller, who claims to be able to bend forks by thought alone. 

Since the early Eighties Prof. Robert Jahn and colleagues of the Princeton Engineering Anomalies Research project have been perfecting a series of tightly controlled laboratory tests of PK, to discover once and for all whether the phenomenon exists. The experiments focus on electronic random number generators, which produce an utterly unpredictable sequence of ones and zeroes. Subjects are simply asked to concentrate on a display showing the output of the generators, and try to change the numbers it produces. Left to themselves, the devices will produce equal numbers of ones and zeroes in the long run. If PK exists, however, it should reveal itself in a bias away from chance expectation as subjects "will" the output upwards or downwards. 

Now, after 12 years of experiments involving more than 100 subjects in thousands of trials, Prof. Jahn and his team have uncovered astonishing evidence that the electronic devices can be controlled by thought. The human subjects proved capable of altering the output of the devices so much that the chances of getting such a bias by fluke alone is calculated to be less than one in 1,000 billion. "We believe that we now have pretty incontrovertible evidence for this phenomenon," Prof. Jahn said. "These effects seem to be broadly spread among human operators - it seems to be a common ability." 

Past research into PK based on electronic devices has been criticised for not carrying out thorough checks to ensure the devices are unbiased in the first place, and for relying too much on the success of a handful of subjects. The Princeton team insists that these criticisms are no longer valid: the effect appeared with different devices, all of which were thoroughly tested beforehand, and with many different subjects. Out of nine different sets of experiments, six showed statistically significant evidence for PK. In contrast, experiments using random number generators based on fixed mathematical formulas - which should be immune from psychic influence - did not produce any evidence for PK, exactly in line with prediction. "We would now lay claim to have the largest datasets and the most systematic experiments ever performed," Prof. Jahn said.

The Princeton evidence follows the discovery of equally impressive evidence for the existence of telepathy by researchers at Edinburgh University, highlighted by The Sunday Telegraph earlier this year. 
Experiments by Prof. Robert Morris and colleagues at the university's Koestler Parapsychology Unit suggest that people can mentally "transmit" images to others by thought alone. 

Until now, orthodox scientists have dismissed all such claims as the result of incompetence or fraud. However, even hardened sceptics now admit that these charges are becoming hard to sustain. "I have a lot more problems with these results as a sceptic," said Prof. Stephen Donnelly, a physicist at Salford University and deputy editor of UK Skeptic.
 
But Chris French, the head of psychology at Goldsmiths'College, London, and another long-standing critic of claims for the paranormal, said that he was concerned by the tiny size of the supposed psychic influence. "The effect sizes are so staggeringly small that some people would argue that any sensible person would prefer a non-PK explanation. There's also a worry that with the huge number of trials needed, conventional statistical theory starts to break down." 

Prof. Jahn insists the data is now so strong that the arguments over the paranormal must move towards explaining how it works. "We don't see much point in continuing the collection of yet more data," he said. "We're setting up experiments to get a better comprehension of these phenomena."