The lab that asked the wrong questions

Lucy Odling-Smee, Princeton
 

A medley of random-event machines, including a kaleidoscopic crystal ball on a pendulum, a pipe spurting water and a motorized box straddled by a toy frog, came to the end of their working lives yesterday at the Princeton Engineering Anomalies Research (PEAR) laboratory in New Jersey.

Only romantics — and some parapsychologists — are likely to lament the loss of this unique institution, which investigated whether people can alter the behaviour of machines using their thoughts. Many scientists think the lab's work was pointless at best. But the closure highlights a long-running question: how permissive should science be of research that doesn't fit a standard theoretical framework, if the methods used are scientific?

The PEAR lab was founded in 1979 by Robert Jahn, former dean of Princeton's school of engineering and applied sciences, and an expert on electric propulsion. Start-up funds came from aerospace pioneer James McDonnell, who believed that aircraft machinery was influenced by the mental states of pilots. The lab has relied on private funds ever since.

Over 28 years, PEAR researchers collected data from tens of millions of trials using random-event machines. When all the data are considered together, they show that human intention has a very slight effect, the researchers say. Whether the machine is a screen that flashes numbers or a fountain of water droplets, they say that, on average, people can shift 2–3 events out of 10,000 from chance expectations.

It was Jahn's decision to close the lab. He set out to prove the existence of the effect and, at 76, believes the work is done. But such tiny deviations from chance have not convinced mainstream scientists, and the lab's results have been studiously ignored by the wider community. Apart from a couple of early reviews (R. G. Jahn Proc. IEEE 70, 136–170; 1982 and R. G. Jahn and B. J. Dunne Found. Phys. 16, 721–772; 1986), Jahn's papers were rejected from mainstream journals. Jahn believes he was unfairly judged because of the questions he asked, not because of methodological flaws.

Even in other areas of parapsychology, opinion is divided on the lab's results. The difficulty is that it's virtually impossible to prove that such subtle effects aren't caused by some flaw in the methods or equipment. A recent meta-analysis (H. Bösch et al. Psychol. Bull. 132, 497–523; 2006) combined 380 studies on the phenomenon, often termed psychokinesis, including data from the PEAR lab. It concluded that although there is a statistically significant overall effect, it is not consistent and relatively few negative studies would cancel it out, so biased publication of positive results could be the cause.

Robert Park, a physicist at the University of Maryland, adds that if you run any test often enough, it's easy to get the "tiny statistical edges" the PEAR team seems to have picked up. If a coin is flipped enough times, for example, even a slight imperfection can produce more than 50% heads.

In the end, the decision whether to pursue a tiny apparent effect or put it down to statistical flaws is a subjective one. "It raises the issue of where you draw the line," says sceptic Chris French, an 'anomalistic psychologist' at Goldsmiths, University of London, who tries to explain what seem to be paranormal experiences in straightforward psychological terms. French thinks that even though the chances of a real effect being discovered are low, the implications of a positive result would be so interesting that work such as Jahn's is worth pursuing.

Many scientists disagree. Besides being a waste of time, such work is unscientific, they argue, because no attempt is ever made to offer a physical explanation for the effect. Park says the PEAR lab "threatened the reputation" of both Princeton and the wider community. He sees the persistence of such labs as an unfortunate side effect of science's openness to new questions. "The surprising thing is that it doesn't happen more often," he says.

William Happer, a prominent physicist at Princeton, takes the middle ground. He believes the scientific community should be open to research that asks any question, however unlikely, but that if experiments don't produce conclusive results after a reasonable time, researchers should move on. "I don't know why this took up a whole lifetime," he says.

Parapsychologists have a hard time trying to publish in mainstream journals.
The status of paranormal research in the United States is now at an all-time low, after a relative surge of interest in the 1970s. Money continues to pour from philanthropic sources to private institutions, but any chance of credibility depends on ties with universities, and only a trickle of research now persists in university labs.

Elsewhere the field is livelier. Britain is a lead player, with privately funded labs at the universities of Edinburgh, Northampton and Liverpool Hope, among others. Parapsychologist Deborah Delanoy at the University of Northampton suspects that the field is stronger in Britain because researchers tend to work in conventional psychology departments, and also do studies in 'straight' psychology to boost their credibility and show that their methods are sound. "We're seen to be in the same business as other psychologists," she says.

But parapsychologists are still limited to publishing in a small number of niche journals. French thinks the field is treated unfairly. "I'm convinced that parapsychologists have a hard time trying to publish in mainstream journals," he says, adding that he even has difficulty publishing his 'straight' papers on why people believe in paranormal events: "Simply because the paper mentions the word telepathy or psychokinesis, it isn't sent out to referees. People think the whole thing is a waste of time."