OK readers, later in this article, I'm going to use an example that will involve either a garden, a sailboat, a running man or a train. Can you accurately guess which one? In a forthcoming issue of the Journal of Personality and Social Psychology (JPSP), Cornell psychology professor Daryl Bem has published an article that suggests you can, possibly more often than the 25 percent of the time on average you might expect just by chance.
Entitled "Feeling the Future: Experimental Evidence for Anomalous Retroactive Influences on Cognition and Affect," the paper presents evidence from nine experiments involving over 1,000 subjects suggesting that events in the future may influence events in the past -- a concept known as "retrocausation." In some of the experiments, students were able to guess at future events at levels of accuracy beyond what would be expected by chance. In others, events that took place in the future appeared to influence those in the past, such as one in which rehearsing a list of words enhanced recall of those words, with the twist that the rehearsal took place after the test of recall.
As Director of Research at the Institute of Noetic Sciences, where, among other things, we study experiences that seem to transcend the usual boundaries of time or space (generically called "psi" experiences), I've already received a slew of comments and queries regarding the pre-print of the article that is making the rounds.
The comments range from, "Wow, that's amazing!" to, "That's not possible -- there must be some mistake." But most responses are along the lines of "Hello?? This isn't news. Hundreds of articles reporting significant results on psi experiments have already been published in dozens of academic journals. What's the big deal?"
So what is notable about the current publication? To begin, Bem is not just any psychologist; he is one of the most prominent psychologists in the world (he was probably mentioned in your Psych 101 textbook, and may have even co-authored it). And JPSP is not just any journal but sits atop the psychology journal heap; the article, especially given its premise, was subjected to a rigorous peer-review (where scientific colleagues critique the article and decide whether it is worthy of publication). Also, Bem intentionally adopted well-accepted research protocols in the studies, albeit with a few key twists, that are simple and replicable (they don't require lots of special equipment, and the analyses are straightforward). Even so, whether the larger scientific community will pay attention to this study remains to be seen.
Which begs the question: Why is the existing literature on psi phenomena routinely dismissed by the scientific community and virtually ignored within the broader academic community? As science journalist Jonah Lehrer says about research findings on psi phenomena, "They've been demonstrated dozens of times, often by reputable scientists ... Why, then, do serious scientists dismiss the possibility of psi? Why do rational people assume that parapsychology is bullshit? Because these exciting results have consistently failed the test of replication."
Such assertions drive some of my colleagues crazy, who point to a large body of literature in which psi experiments have been replicated numerous times over many decades, involving dozens of independent scientists and thousands of subjects, and published in peer-reviewed journals. Still, the majority of the scientific community has largely dismissed the concept of psi -- no matter how reputable the investigator or prestigious his or her affiliation -- as frivolous, artifactual, not replicable, or having effect sizes that are so small as to be meaningless regardless of statistical significance. Worse, skeptics accuse psi researchers of being outright fraudulent, or well-meaning but delusional. Young scientists are regularly advised to stay far away from studying psi and warned about the ATF (the anti-tenure factor) that is associated with such interests. Senior scientists, including Nobel Laureates, have been known to be disinvited from giving talks if their interest in psi is discovered. Even religious scholars, who make it their business to examine the spiritual aspects of human experience, have trouble with psi.
With respect to effect sizes, yes, if you look at the results of lots of studies combined, psi effects are statistically significant, though small. However, a double standard is applied to the potential importance of small effects. The effect sizes reported in Bem's and many previous psi studies were frequently much larger than the effect sizes associated with many well-accepted scientific facts, like taking aspirin to prevent heart attacks, for example, or the risks of blood clots from taking Tamoxifen.
More importantly, though, even if we were to agree that "size does matter" and that these effects are generally small, let's remember that it shouldn't be possible to peer into the future at all, even a little, given what we generally understand about how the world works. Time is only supposed to go one way. Perception is supposed to be limited to the past or the present and only to those phenomena immediately and locally accessible by our five senses. When exceptions to these rules are observed, particularly under controlled laboratory conditions, they deserve a closer look.
Take running the four-minute mile. If we as scientists had studied even thousands of people in the 1950s, we might have concluded that running a four-minute mile was not humanly possible. Over time, however, it was found that a few people could actually do it -- an extremely small effect to be sure, but these anomalies proved that it was, in fact, possible. Not only do we now know that running a four-minute mile is possible, it is the standard for professional middle-distance runners (for those of you paying attention, that was the example with the running man).
Perhaps the oft-quoted maxim "extraordinary claims require extraordinary evidence" should be accompanied by a counter-maxim: "extraordinary anomalies deserve special attention." For example, a new drug to treat depression that resulted in some relief in one out of 100 people might not be worth a second glance, but if a new drug was claimed to cure AIDS in one out of 100 patients, it would justify further examination. When evidence runs contrary to prior probabilities, it calls for special consideration, not a knee-jerk out-of-hand dismissal.
As for replication, as noted earlier, psi proponents argue that there have been numerous replications -- often far more than many other scientifically supported "facts" that are taken for granted. Indeed, scientists familiar with this area of research view Bem's studies as clever conceptual replications that rest upon a large body of previous work. These scientists are now going beyond the idea of mere existence of these effects and forging ahead into studying what conditions may enhance them -- inherent individual traits, training, genetics? In small, underfunded labs around the world, scientists are working to improve research designs, measures and methods to better study psi.
There is also a growing recognition that it might not be quite so simple as developing one good experiment and then replicating it to death. An article published in the Dec. 13, 2010 issue of The New Yorker highlights a phenomenon that is well known to scientists, not only in the field of psi but across many disciplines: Initial experiments can show very strong results, but when the experiments are repeated again and again, the effects can decline. Gamblers may recognize this phenomenon as "beginner's luck." Of course this isn't true for all natural phenomena. When you drop a rock it will head toward the ground pretty much every time. But for more complex phenomena, we may need to contend with the "decline effect," along with observer effects and other design and measurement complexities.
Does this mean that the effects aren't real and that these topics are inherently "unscientific" and shouldn't be studied? Of course not. Recall that in the early 19th century, it took many years for Faraday to demonstrate the existence of electromagnetism to his colleagues, and still, he did not live to see his theory that electromagnetic forces extended out into empty space around a conductor validated. Many research topics are extremely complex, requiring decades of research, and all kinds of new measures, methods, controls and technologies to adequately explore them. Cancer remains a profound mystery despite the efforts of tens of thousands of scientists and billions of dollars spent looking for a cure. Sequencing the human genome was a vast and complicated undertaking. Even "evidence-based" drugs for treating depression, on which a multi-billion dollar industry is based, are being called into question as being not much better than a placebo after all. Unless the object of study is extremely simple, science is mostly a long, winding, painstaking, incremental and challenging pursuit.
Problems with fluctuating effect sizes, experimenter effects, finding adequate controls and so on, are inherent in studying phenomena with complex interactions and poorly understood mechanisms. So I don't think we can attribute resistance to evidence for psi to these, nor can we blame complexities of measurement, difficulties with replication or even the challenge of pinning down an underlying theory. I think it's fear that some of our most cherished beliefs about how the world works and about who and what we are may be wrong. On a deeper level, there may be a collective, protracted, post-traumatic stress disorder resulting from that period in human history when reliance on blind faith in supernatural explanations of reality led to a very dark time when priests determined what was true and rational thought and systematic observation were prohibited.
Bem's article and its supporting body of literature, combined with serious discussions of retrocausation in physics, suggest that retrocausation in human experience may indeed be possible. But the real significance of the article lies in the fact that the dialogue about psi has been brought once again into the arena of intelligent debate in a public forum, where it deserves to be. While a long period of cautiousness regarding the commingling of science and anything considered supernatural -- like perceiving the future or the impact of consciousness on physical systems -- has been an understandable and adaptive response, surely we can trust ourselves in the 21st century to examine these issues intelligently without losing our heads. Such examination may lead to radical revisions in our understanding of how the world works and our human potentials.
Update: Thanks for the big response and thoughtful debate. I've written a follow-up to this post at http://www.noetic.org/blog/responding-to-the-debate-psi-research-at-a-tipping/.