Why do reports of positive results for psi experiments, even by respected researchers in high-profile journals, tend to recede from view without consequence?  And is there any way that psi research could be re-accepted into mainstream science?

I’ve been away from this issue for a long while, but after the recent media flurry over Cornell psychologist Daryl Bem’s results, I thought I’d spend some time re-organizing my thoughts on it all.

First, as a sort of starting guidepost:  Any branch of science, even mainstream modern “Science” broadly defined, has its own culture with its own relatively inelastic ways of seeing – and not seeing.  Psi happens to be one of the many things that Science does not see; and in a sense, that’s not even remarkable, because for us humans, with our young civilization and small brains, probably most of the universe lies in our blind spots.  We should be grateful for what we do see!

On the other hand, psi isn’t an advanced, yet-to-be-discovered phenomenon.  It belongs to a rejected past, not to an inaccessible future.  Parapsychology was once a very active branch of Science, with a literature that was full of positive and seemingly repeatable results.  So the important question here is not, Why doesn’t Science see psi? but rather, How did Science come to unsee psi?

The Unseeing

During the heyday of parapsychology in the late 1800s, when it enjoyed the enthusiastic support of prominent philosophers and scientists (William Crookes, William James, Henry Sidgwick, Lord Rayleigh et al), and even a bit later when Freud and Jung spoke well of it, Science was still a bit “loose.”

I mean this in a theoretical sense and a social sense.  In the theoretical sense, Science’s conception of the universe and its laws and relationships was still coming together, and amid that tumult there was still some conceptual room for psi.  That conceptual framework has since tightened considerably, and psi never was able to connect strongly to anything within it – psi can’t be explained by reference to any other accepted phenomenon – and thus it has been excluded.  As two skeptical researchers wrote recently, “the positive evidence that has been reported [for psi] is merely ‘anomalous’” — in other words, it cannot be taken very seriously because it fits no accepted theory.

Science also has tightened up in a social sense:  Its institutions were relatively underdeveloped a century ago.  Sources of research funding were less nationalized and centralized; there were still even some gentleman-scientists who self-funded and could do pretty much what they liked.  Nowadays, what we might call the “elites” in any given discipline tend to determine whose research gets funding (usually government funding).  My guess is that parapsychological research nowadays is also more expensive, in real terms, than it was a century ago.  The point is that getting the elite support needed to do such research, at a level where it can make an impact on psychology or neuroscience, is probably harder now than ever.  There would have to be not one or two Daryl Bems on board, but perhaps dozens.

Then there is the character of psi itself.  Science since the late 1800s has become extremely elaborate and powerful.  Yet psi seems to mock its power.  If psi exists and – as its proponents claim – can place a person’s perceptual faculties anywhere in time or space, or move objects around from a distance, what’s the point of all the other humdrum details of life and the universe?  Psi as promoted by its enthusiasts is like a magical backdoor to reality, through which one can suspend the normal (and rather more painstakingly discovered) physical laws.  It’s a deus ex machina that spoils the narrative of history by rendering conventional human effort almost worthless.  Discovering that psi is real would be like learning, a la the Hitchhiker’s Guide to the Galaxy, that the key to the universe is “42.”

Psi’s historical and cultural baggage also offends.  In premodern societies and presumably prehistoric ones, psi beliefs appear to have been universal.  Yet when ethnologists have studied the tribal shamans and mediums who promote these beliefs in old-culture societies, they commonly have observed sleight-of-hand and other forms of trickery.  The same sort of trickery emerged in the early days of parapsychology research, inspiring public skepticism from the likes of Houdini. And of course nowadays in Western societies, psi is strongly associated with New Age faith-healers, spirit-mediums, tarot card readers, witches, toll-call psychics, Sci-Fi Channel ghost whisperers, and other characters of dubious abilities and questionable motives.  I suspect that to some of the more vehement skeptics out there, psi-related beliefs and behaviors are not even meant to be studied scientifically, but instead are seen as atavisms of an ancient, pre-scientific mindset:  Psi, in other words, is a hairy foot in the door of modernity, and one which mustn’t be allowed to enter.

A Noisy Signal

If all the factors that I have just listed were the only disadvantages facing modern parapsychologists, psi might by now have triumphed anyway.  But it has had one additional disadvantage, namely that it is, on the whole, a weak and fluky phenomenon.  It seems that its most reproducible results over the years have been with rare, gifted individuals, and even they have been “off” much of the time – sometimes for no evident reason, and sometimes (it is said) because they simply get bored with the protocols.

In principle, even that shouldn’t be a problem.  Even a slight long-term edge over random chance should offer intrepid researchers the opportunity to develop world-shaking applications, from espionage to trading.  Somehow no one has been able to develop such applications, at least not in a way that we all can see.

Psi arguably had its best shot when the US intelligence community funded the well-known “remote viewing” program.  But despite a quarter century of research and development and operational practice, which included a prominent unclassified (albeit controversial) publication in Nature in 1974, the program never found a permanent home among the armed services and intelligence agencies.  High ranking officers and espiocrats and even legislators supported the program, and there is evidence that it was useful on various occasions, but it became more, not less controversial over the years, until it was finally shut down in 1995.

The epitaph on the RV program was “needs more study,” but with very rare exceptions such as Bem, academic psychologists and neuroscientists haven’t wanted even to study it.  They’ve preferred to just ignore it – and they’ve been able to get away with that because of psi’s lack of sustained, obvious and real-world-practical efficacy.  No one has been getting in Science’s face, so to speak, with successful, psi-derived earthquake predictions, or astronomically unlikely trading gains, or solutions to great crime mysteries, or the location of Bin Laden or other fugitives.

Perhaps some skeptics have even rationalized that if psi were real, parapsychologists should all be rich from precognitive trading or betting successes, and therefore able to fund their own research, without taking scarce resources from conventional disciplines.  Would such an expectation be unfair?  If so, the proponents of psi haven’t really articulated why.  I’m surprised that they haven’t – to my knowledge – invoked the principle put forward last year to explain why the LHC kept breaking down:

… the hypothesized Higgs boson, which physicists hope to produce with the collider, might be so abhorrent to nature that its creation would ripple backward through time and stop the collider before it could make one, like a time traveler who goes back in time to kill his grandfather.

Disconfirmation Bias

Finally there is the whole issue of disconfirmation.  The skeptics’ conceit here seems to be that, yes, the psi researchers publish positive results (as they must, if they are not to discredit their profession and perhaps also their religious beliefs) but when more skeptical and sober-minded scientists try their hand, using more rigorous experimental protocols, they find no evidence of psi.  Researchers already are trying to make disconfirmation the epilogue to the Bem story.

I think the most interesting and illustrative recent case of disconfirmation — of psi generally, not of Bem’s experiments — comes from the laboratory of Stephen Kosslyn, a Harvard psychologist who is well known for his use of functional MRI (fMRI) and other techniques to explore the correlates of mental imagery in the brain.  Kosslyn is very accomplished and influential in his field, and also, according to a bio I found on the web, “has served on several National Research Council committees to advise the government on new technologies.”  If there is still an active debate within the US intelligence community over the utility of remote viewing or psi generally, I guess that Kosslyn is in the loop.

In a paper two years ago in the Journal of Cognitive Neuroscience, Kosslyn and one of his grad students, Samuel Moulton, reported on a psi experiment which must be one of the first in the open literature, maybe even the first ever, to use fMRI.  They claimed that the results were “the strongest yet obtained against the existence of paranormal mental phenomena.”

They recruited twins and had them serve as sender/receiver pairs.  One sat in an isolated room and was shown emotion-evoking images; the other lay in an fMRI machine and tried to guess which of two presented images was the one being sent by the sender.  In a way it was like the old ganzfeld experiments, but the researchers’ idea was that the fMRI data might contain evidence of psi (that is, a telltale change when the image being “sent” by the sender was shown) even if the choices made by the receivers scored only at chance levels.  To make a long story short, they reported finding no evidence of psi either in the receiver choices or in the fMRI data.

I think it’s fair to say that one is likely to find this sort of result convincing only if one already disbelieved in psi.  If one were a believer, one could ask:  Were the researchers skeptical to begin with?  Did their skepticism have the subconscious effect (a negative suggestion, as it were) of suppressing psi in their subjects?  Were the fMRI parts of the protocol truly sensitive enough to distinguish subtle influences of psi on brain activity, had such influences been present?  Wouldn’t an outbound remote viewing protocol, in which the congruence between viewer data and target data is scored independently, have been likelier to provide positive psi results?  And finally, wouldn’t subjects who had already seemed to demonstrate positive psi results in other experiments – psychics who routinely help police, or former government RVer Joe McMoneagle, for example – have been better for this experiment than a group of relatively psi-naïve individuals?  I mean, conceivably Moulton and Kosslyn are right, and the entire anecdotal and experimental evidence for psi can be explained away by reference to fraud and various cognitive biases; but I can see how a psi proponent would view their experiment as having been designed to look good but to fail.

There is a popular idea about scientific discovery that it can easily be the product of an overeager researcher’s imagination, whereas the real test comes when more skeptically minded scientists try to disconfirm that discovery.  In other words, the disconfirmations are weighted more heavily than the initial report of discovery.  Certainly it is true that some “discoveries” haven’t panned out.  However, I think that there is a basic misconception here about how discoveries work.  To discover a new phenomenon one often has to be strongly motivated by the belief that the phenomenon is real, and discoverable, if one is going to optimize the experiment to pull the phenomenon out of the noise.  In cognitive terms, it’s a top-down process in which the firm expectation of something enhances your perceptual channels so that you can see that it is there.  Of course, scientists include control experiments to reduce the chance that they are deluding themselves.  But if they start by believing in and hoping to prove the null hypothesis, then their bias is more likely to lead to an experimental setup that produces a null result.

In this sense, believing is seeing, and unbelieving is unseeing — and that is why some scientific controversies can be so intractable.  The challenge for a psi researcher, or any would-be discoverer, is not only to find a way to see the new phenomenon but also to engineer an experimental protocol that works very robustly – so robustly that it will generate positive results even in others’ hands.  (This is one reason why academics often refer to the “construction of scientific knowledge.”)  But it’s an inherently difficult task, which is why a failure by a skeptic to confirm a positive result has little evidentiary value in these debates — just as a believer’s positive result typically fails to impress the skeptics.  Essentially, neither side trusts the other to conduct a rigorously fair experiment.  In the case of psi, a mass of anecdotal and early experimental evidence also supports the believers; so the skeptics, to end the debate, would have to generate not only a cascade of negative experimental results but also some convincing “non-paranormal” explanation for the existing body of positive evidence.

Does all this seem uncomfortably far from the conventional wisdom about scientific discoveries?  Well, perhaps it is.  But the conventional wisdom has to be wrong, because it cannot explain the persistence of controversies such as the one over psi.

Is There a Way Back?

Conceivably a believer such as Bem would develop a successful protocol, publish his findings, and then help other, relatively sympathetic colleagues to get it to work in their hands.  They in turn would propagate the technique to other labs, and with each iteration the blinders of skepticism would be discarded by more people within the field, until a “critical mass” of support was reached and the remaining skeptics threw in the towel and began repositioning themselves (“I was never that skeptical,” “I just thought it needed more study” etc.).  But even that strikes me as an optimistic scenario, given the widespread antipathy towards psi that I suspect exists  — though I hope I am wrong — among the elites of neuroscience and psychology.  In other words, I doubt that Bem will be able to find enough sympathetic colleagues to help him propagate his positive results, even if they are valid.

If there isn’t (as I suggested above) some cosmic ban on psi’s rejoining mainstream Science and deeply altering our view of reality, then how could it accomplish that feat?  One possibility, I think, would be a theoretical development, perhaps in physics or consciousness research, that firmly connects to psi, so that psi is longer “merely anomalous” but at last has a home in the framework of Science.  Mainstream researchers then would have a motive to see evidence for psi in their experiments.

Another way, perhaps easier in the sense that psi nowadays is mostly a popular phenomenon, would be for someone or some group to come up with a psi technique that has a robust real-world application – for example, a protocol that enables a team of remote viewers to systematically make huge sums in securities trading or sports betting, or to repeatably predict earthquakes or big sociopolitical events.  In this way, psi would “get in the face” of Science and force mainstream researchers to start studying it – non-skeptically.

I do think it would take a major event to put psi over the top, though.  Consider two other phenomena that have been relatively ignored despite what would seem to be strong evidence of their existence and significance.  One, consciousness, hits us right between the eyes every waking hour of every day, yet it is clearly, like psi, a paranormal phenomenon, for there is nothing in the rest of science that explains it.  As David Chalmers and others have suggested, it may be a fundamental property in the universe that emerges in certain circumstances — a property that we’re just going to have to accept, for now, even though we can’t yet relate it to quantum electrodynamics or gravity or string theory or whatever.  In any case, even though consciousness should be an undeniable phenomenon, the fact that it is paranormal has left consciousness research with a marginal status until recently — and even now the field (which includes the search for consciousness’s neural correlates as well as theories about how consciousness emerges) still gets very little funding, considering its apparently fundamental importance in our lives.

Hypnosis has a similar story.  It played a big role in the birth of parapsychology, and thus carries that unwanted historical baggage.  So despite having been used (in the pre-ether days) for surgical anaesthesia, and despite its continued popularity for various therapeutic applications, and despite not even being considered paranormal, it is routinely dismissed as mere “suggestion” (as if that explains everything) and probably gets even less psych/neuroscience funding than consciousness research.  Thanks to this neglect, hypnosis’s precise neural mechanisms remain mysterious.

Unfortunately, barring the development of some spectacular theoretical shift or real-world application, psi’s mechanisms are likely to remain just as neglected, and just as mysterious.