The ability to predict the future is a skill that could greatly improve everyone’s life. Just think of how you could bolster your financial security if you could tell what the stock market would be doing tomorrow or what the winning lottery ticket number will be. If you’re a student, your grades in school could be boosted immeasurably by seeing into the future and knowing what will be on your next test. Knowing what lies ahead in the future could also save your life. If you knew that a large truck would be veering through an intersection at the moment you would be crossing the street, you would cross another street to get to your destination.
According to research by Cornell psychologist Daryl Bem, we actually can “feel the future.” Bem has devoted a good deal of his career to documenting the existence of “psi” in the laboratory. Psi is a form of telepathic communication, or what Bem calls “anomalous processes of information or energy transfer that are currently unexplained in terms of known physical or biological mechanisms.” In other words, psi is the ability to perceive events that have not actually occurred. Bem’s experiments tested a version of psi called “retroactive influence,” or time-reversing. He devised a method to find out whether participants would respond at Time 1 as if they already knew what would happen at Time 2. Through a series of 9 experiments, Bem claimed to have proven that we behave in the present as if we knew what would happen in the future; so-called "pre-cognition".
Here’s the basic framework of one of Bem’s experiments, the so-called “White Queen” test. If you remember your basic Lewis Carroll, the White Queen in Through the Looking Glass claims that the citizens of her country can remember things that happened in the future ("it's a poor sort of memory that only works backwards"). Bem’s test involved the following scenario: Participants saw a set of 48 common nouns, drawn from 4 categories (foods, clothes, occupations, and animals). Each word appeared on a computer screen for 3 seconds. While each word was presented, the participants were instructed to visualize it (e.g. try to see a shirt when given the word “shirt”). Then, as is typical for this type of experiment, the participants were told to write down as many words as they could from this list. The participants didn’t actually know ahead of time that they’d be asked to try to remember the words.
Now comes the crucial part of the experiment. Would participants have better memory for the words they saw after the recall test? Here’s what Bem did to test this out. After the participants completed the recall of the 48 words, he showed them 24 out of the original 48, drawn from each of the 4 categories (called the “practice” words). He told the participants to click on the 6 food words they saw on the screen, and then retype them into blank spaces on the screen. After this, they had to do the same thing for each of the other 3 categories of words. Participants did not see the other 24 words (the “non-practice” or control words). Of course, the participants weren't really practicing the words because the test had already been given. All this exercise did was to focus their attention on those words, an act that would bolster the claim that they "knew" they'd see them later during the memory test.
To measure psi, Bem calculated the difference between the practice and non-practice words. If psi existed, participants should have better recall for the practice words than the non-practice words. The findings as reported by Bem showed that, in fact, participants did recall the words practice words at a slightly, but significantly, better rate in the memory test. The precise advantage they showed was about 2% (out of a range of minus to plus 100%). To push the test results further, Bem then tried an additional manipulation in which he showed the 24 “practice” words within each of their respective categories. This time, the effect was even stronger, averaging 4% (significant statistically, though small in an absolute sense, we should note).
Based on the finding of these and 7 other experiments on over 1,000 college undergraduates, Bem concluded his paper with the wish that psychologists would be more open to considering “impossible things” like the ability of the future to influence the present. He was convinced he's done everything needed to prove things from his end, and the fact that the paper was published in a highly reputable journal would have assured that this was in fact a correctly done experiment.
The scientific world of academic psychology was, to say the least, horrified that such a paper could have been published. Not only did the premise of the study fail to make sense (you should apparently study for a test after taking it), but the small so-called "effect sizes" seem to have been magnified by a lack of statistical controls.
The study's publications spawned several attempts at replication. Each of these has shown, beyond a shadow of a doubt, that the findings were spurious at best and bogus at worst. One of the most extensive attempt to replicate the findings not only failed to reproduce them in several labs around the world, but also examined all the available evidence for the effect in a large number of other published studies. Carnegie Mellon psychologist Jeff Galak and collaborators (2012, p. 11) concluded that “it is unclear how Bem (2011) could find significant support for a hypothesis that appears to be untrue.”
Part of the reason that Bem reported findings that others haven’t may lie in a bit of statistical legerdemain. Without directly calling Bem a statistical magician, Galak and others point to several possible sources of bias. One is the infamous “file drawer problem.” Did Bem report only the results that supported his hypothesis? (though to be fair, one of the 9 studies did not). A second has to do with probabilities, called the “fallacy of the transposed conditional” (Wagenmakers et al., 2012). There’s no easy way to translate that phrase, but the upshot is that if Bem was right about the existence of psi, one person with this paranormal ability could “bankrupt all casinos on the planet” (p. 428).
Other critics argue that the Bem article reflects a failure of the scientific review process- the gold standard for all of published articles. As noted in The Guardian, scientific journals are less likely to public replications than they are the original articles especially, according to one view, when the results show no effects (the file drawer problem, again). Fortunately, however, the null finding replication was eventually published in PlosOne, which you can read in its entirety. And the original journal that published the Bem study, the Journal of Personality and Social Psychology, is about to publish the set of critical papers including Galak's. Another criticism of the original Bem paper was that some reviewers may have felt they should tread lightly on critiquing the paper given Bem’s prominence in the field and his otherwise distinguished career.
Making the evaluation of scientific data even more problematic for you, the reader, is the fact that studies showing cool or sexy findings quickly hit the virtual “front page” of the Internet, sweeping through site after site of science-ish websites. Those that retract those findings- well, not so much. For example, the title “Startling New Study ‘Proves’ We Can See the Future.’ remains posted without the corrections from the subsequent refutations.
Ironically, Daryl Bem himself offered advice that would have averted these messy situations in a classic episode produced in the 1980s from Zimbardo’s Discovering Psychology PBS series (go to 13:59 on the video). If you’re one of the many millions of high school and college students who saw this series, you might remember the scene in the Research Methods episode in which Bem "The Magician" correctly “guesses” the card that a subject picks from a small number on a table. Bem pulls up his shirt and dramatically reveals a T-shirt underneath it with the exact same card on its front. Seems like ESP, right? However, as Bem “The Psychologist” reveals, he had a matching item in the room for each of the cards the subject could have picked. He explains: “There was never any possibility that chance was operating. I don’t leave things to chance, but a psychologist with a well-designed experiment, would have ruled out all of these things.”
Did this famous psychologist correctly peer into his own future and see that someday he would be reporting on a study that would eventually fail his own test of scientific rigor? That would indeed be ironic. However, like many psychologists and non-psychologists who are fascinated by the paranormal, it’s just too easy to be misled into falling into the trap of letting our desire to see the future influence our ability to see the data smack dab in front of us.
You might ask yourself why you should care about this silly study and the tempest in the scientific teapot it has caused. The answer is that this time the finding is perhaps not all that important (if not believable). The next time it might not be quite that innocuous. What if a study claims to show the benefits of a method of therapy that in fact doesn’t work or even causes harm? You would want to be protected by the scientific establishment to ensure that the best minds go into prescribing the remedies to help you and your loved ones when you need help.
The moral of the story is that whether a research producer or consumer, the scientific method should trump all. The more you learn about how good science is conducted, the better you can use the insights it can produce to improve your own everyday life, both now- and in the future.
Follow me on Twitter @swhitbo for daily updates on psychology, health, and aging. Feel free to join my Facebook group, "Fulfillment at Any Age," to discuss today's blog, or to ask further questions about this posting.
Copyright Susan Krauss Whitbourne, Ph.D. 2012
Galak, J., LeBoeuf, R. A., Nelson, L. D., & Simmons, J. P. (2012). Correcting the Past: Failures to Replicate Psi. Journal Of Personality And Social Psychology, doi:10.1037/a0029709
Wagenmakers, E., Wetzels, R., Borsboom, D., & van der Maas, H. J. (2011). Why psychologists must change the way they analyze their data: The case of psi: Comment on Bem (2011). Journal Of Personality And Social Psychology, 100(3), 426-432. doi:10.1037/a0022790