Discuss as:

Can science prove we're psychic?

Twentieth-Century Fox / DreamWorks

In a scene from the 2002 film "Minority Report," Tom Cruise's character helps out a woman (Samantha Morton) who has precognitive powers. Do such powers actually exist, even to a small extent?

Scientists are buzzing over a peer-reviewed study that suggests humans have predictive powers, but it’s too early to predict whether or not the research will hold up.

The 61-page paper, titled "Feeling the Future," was written by Cornell psychology professor emeritus Daryl Bem and is due for publication in the Journal of Personality and Social Psychology. Bem says his experiments support the idea that there really is something to human precognition of events that haven't yet occurred.

You could argue that this is a case of science imitating sci-fi -- particularly considering that precognition provided a key element of the plot for "The Minority Report," a Philip K. Dick short story that was made into a movie starring Tom Cruise in 2002. You might be forgiven if you think this is the latest trick from a professor who used to be a stage magician. But Bem is dead serious about the experiments, and his submission to the journal is no work of fiction.

"My very first publication was 50 years ago in that journal, which would make a nice capstone," Bem told me today.


Bem said each of the experiments described in the paper simply takes a well-known method for testing how sensory input affects the brain's output "and turns it around backwards" in time sequence.

Here are three examples:

  • Retroactive recall: One experiment was based on the truism that practice makes perfect. Suppose you set up a vocabulary study drill on a list of 24 words. Then you show the students a 48-word list, including those 24. When you ask the students later to recall as many of the words as they can, it's more likely that they'll remember the 24 from their vocabulary drill. "In my version, we reverse it," Bem said. "We show you the 48 words, then we give you the test, then we give you practice. The prediction is that you will have done better on the test with the words that you were going to practice than the ones you won't practice. And it works."
  • Retroactive priming: Suppose you show 100 people a series of pleasant images (like a Hubble space photo) mixed in with unpleasant images (like a terrible car crash). Lots of experiments have been done about the effect of seeing a "primer" word before the image. If the primer word is something like "peace" or "lovely," the subjects will register a pleasant image more quickly than they would an unpleasant image. But in Bem's experiment, the image was shown before the primer word. The experimental subjects registered the images correctly as pleasant or unpleasant roughly 15 to 25 milliseconds more quickly if the images were followed by a primer word that matched the picture's content.
  • Precognitive selection: A hundred subjects were asked to predict which of two computer screens will flash up a picture rather than an empty space. They're told in advance that some of the images will be erotic in nature. The computer didn't make its random selection of which images would appear where until after the human subjects made their choice. The subjects correctly identified the future position of the arousing images 53.1 percent of the time -- while the success rate for the non-arousing images was merely the expected 50-50. A separate experiment, involving 150 subjects, came up with a 51.7 percent "hit rate" for selecting preferred images over negative images.

These experiments led Bem to conclude that there's a slight but statistically significant precognition effect, particularly if the person making the prediction is a stimulus-seeking extrovert. Such "stimulus-seekers" recorded a slightly higher success rate on Bem's tests.

Bem acknowledged that a 51.7 to 53 percent success rate might seem very close to what would be expected by chance. "But that's about the edge that a casino has over you, and guess what? They're still in business," he said. "We don't ignore things just because they're 53 percent."

What's the explanation?
Lots of people believe in psychic phenomena, but Bem's findings have attracted more than usual attention because he's claiming a quantifiable psychic (or "psi") effect, and because such claims are being published in a reputable scientific journal.

Outside researchers generally acknowledge that the findings seem sound, but they just can't bring themselves to believe Bem's conclusion. Joachim Krueger, a psychology professor at Brown University, said he remains unconvinced that the psi effect is real.

"I am not only bothered by the lack of a positive theory, but also by the contradictions between psi and ordinary scientific assumptions," he wrote in a Psychology Today blog posting. Precognition appears to run counter to the conventional view that causes precede their effects, he said. (Which reminds me to check in on the status of experiments in quantum retrocausality.)

The way Bem sees it, the weird implications of his results are the most fascinating aspects of the research. "What does this say about the real world, because this stuff violates our notion of cause and effect," he said. "That's the part that totally intrigues me."

Bem told me it's "absurd" that he should be expected to come up with a theory to explain his data. He just knows that the data indicate there's a slight, subtle but statistically solid phenomenon worthy of further investigation.

"The odds against its just being chance are actually 7 billion to 1," Bem said. "In any experiment like this, critics are permitted to come up with alternatives to explain the results of the experiment, and some of those explanations are non-psi. That's fair game. But one thing we're sure of is that these results aren't due to chance."

Are the results reproducible?
Bem acknowledged that the experiments would have to be replicated in order to confirm that precognition is a real effect. Two other researchers, Jeff Galak of Carnegie Mellon University and Leif Nelson of the University of California at Berkeley, have already tried to replicate one of Bem's experiments (the one with the word recall test) and failed to get any significant results.

"Why do we not see any evidence of precognition?" Galak and Nelson wrote in their research paper. "There are obviously a multitude of possibilities for why we failed to obtain a result similar to Bem, ranging from the mundane (e.g., our sample was more heterogeneous than Bem's) to the exotic (e.g., the quantum mechanics that allow for the detection of future events are also contingent on the specific physical features of the original experiment rooms.)

"For the purposes of this paper we really only care about one possibility: Do we fail to detect precognition because precognition does not exist? In answer to this question we emphatically say, 'We don't know. On the one hand, we fail to replicate the effect, but on the other hand, our single failure to replicate is hardly sufficient to seriously undermine an entire paper.'"

Bem said Galak and Nelson went ahead with their trial without having full information about how his own experiments were conducted. They also had their experimental subjects take the test over the Internet rather than in person. When the test is conducted online, "you lose total control over it," Bem said. (Galak and Nelson admit that it's "unclear" how closely their subjects attended to the key elements of the experiment.)

But never fear: Since word about Bem's results got out, he's received dozens of requests for information about how the experiment was done. The experiments are sure to be repeated, and the results from those experiments should confirm whether precognition is for real. Who knows? Some of you may already have a sixth sense about this. Feel free to weigh in with your own premonitions -- or even your plain old opinions -- in the comment section below.


For more about Bem's findings, check out this report from New Scientist and this one from H Plus magazine. But if you want to try replicating his experiments, don't just download the replication packages. Let Bem know so that he can keep track.

Additional unsuccessful attempts to replicate Bem's experiments have come to light, including "Retro-priming et re-test" by Thomas Rabeyron and Caroline Watt; and "Precognitive Habituation: An Attempt to Replicate Previous Results," by Gergo Hadlaczky.

Connect with the Cosmic Log community by "liking" the log's Facebook page or following @b0yle on Twitter. You can also check out "The Case for Pluto," my book about the controversial dwarf planet and the search for new worlds.