Is Precognition Real? Cornell University Lab Releases Powerful New Evidence that the Human Mind can Perceive the Future

According to today’s conventional scientific wisdom, time flows strictly forward — from the past to the future through the present.  We can remember the past, and we can predict the future based on the past (albeit imperfectly) — but we can’t perceive the future.

But if the recent data from the lab of Prof. Daryl Bem at Cornell University is correct, conventional scientific wisdom may need some corrections on this particular point.

In a research paper titled Feeling the Future, recently accepted for publication in the prestigious Journal of Personality and Social Psychology, Bem presents some rather compelling empirical evidence
that in some cases — and with weak but highly statistically significant accuracy – many human beings can directly perceive the future.  Not just predict it based on the past.

A pre-publication copy of Bem’s paper is available on his website, and it should appear on the
journal’s website shortly. The article is already attracting considerable attention, including a
piece in Psychology Today
. Also, Bem reports that he has already received hundreds of requests
for “replication packages” — documentation and software allowing others to repeat the experiments he did.   If you want to try to replicate the work yourself, replication packages for some of the
experiments are already available at .

If Bem’s results are indeed replicated, this will shock some scientists, but many others will say “I told you so.”  A 2002 survey by the US National Science Foundation shows that 60% of adult Americans agree that some individuals possess psychic powers.  The percentage of scientists holding such opinions is much lower — but there is a small community of scientists, such as Dr. Bem, working to reconcile popular intuitions about paranormal phenomena with the scientific method and world-view.

[graphs are from ]

What we’re talking about here is precognition (consciously perceiving the future) and premonition (unconsciously sensing the future) — aspects of the general class of phenomena Bem calls “psi,” and others have called “paranormal” or “psychic.”  US government contractors SAIC used the phrase “anomalous cognition and perturbation,” in the context of their top secret work for the government investigating related effects.

I grew up very skeptical of claims of psychic power, jaded by stupid newspaper astrology columns and phony fortune-tellers claiming to read my future in their crystal balls for $20.  Clearly there are many frauds and self-deluded people out there, falsely claiming to perceive the future and carry out other paranormal feats.  But this is no reason to ignore solid laboratory evidence pointing toward the conclusion that, in some cases, precognition really does exist.

As Bem puts it in the abstract of his paper, a little more formally:

The term psi denotes anomalous processes of information or energy transfer that are currently unexplained in terms of known physical or biological mechanisms.  Two variants of psi are precognition (conscious cognitive awareness) and premonition (affective apprehension) of a future event that could not otherwise be anticipated through any known inferential process.  Precognition and premonition are themselves special cases of a more general phenomenon: the anomalous retroactive influence of some future event on an individual’s current responses, whether those responses are conscious or nonconscious, cognitive or affective.

His new paper explores precognition and premonition empirically, via reporting the results of 9 experiments, involving more than 1,000 Cornell University students.

Bem’s recent experiments share a common methodology.  Start with a well-established psychological effect involving certain stimuli leading to certain human responses.  Then, modify the standard experimental set-up so that the human subject’s responses are obtained before the “stimulus” events occur.  The question is whether the experiments still work this way.  Can the Cornell students, sometimes, directly feel the future?

The short answer is: yes

From Psi Skeptic to Psi Enthusiast

Daryl Bem didn’t start out as a psi researcher, and that’s not what he focused on for most of his scientific career.  His greatest prominence is in personality psychology, as the creator of the self-perception theory of attitude change.  He actually began his career as a physicist, with a BA in physics from Reed college and some physics graduate work at MIT; then he switched course to psychology in the midst of his graduate work, due to the civil rights movement of the early 1960s.  Now formally retired (though still active in research), he had a very successful research career with a host of publications in top journals, and spent two decades as a full professor at Cornell, after holding earlier positions at Harvard, Stanford, and Carnegie Mellon.

Bem got into psi research indirectly via one of his hobbies: he is a stage magician, and in the early 1990s he was invited to evaluate some ESP research being done at the Psychophysical Research Laboratories (using the “ganzfeld” procedure that I’ll discuss a little later).  At first he was skeptical of the lab’s ESP claims, and thought to use his knowledge of stage magic trickery to figure out what was really going on behind the scenes.  But after much careful study he became more and more convinced there might be some genuinely anomalous phenomena going on.  Ultimately his evaluation led to one of the classic publications in parapsychology, a review of ganzfeld ESP experiments in the Psychological Bulletin, a leading peer-reviewed academic psychology journal, coauthored by Bem and Honorton.  As time went on, Bem’s interest in psi led him to design and conduct his own experiments in the field including, most recently, the work reported in the Journal of Personality and Social Psychology paper that is our focus here.

Perceiving Erotic Stimuli from the Future

The first experiment described in Bem’s new paper involves perceiving erotic stimuli from the future — specifically, perceiving whether an erotic picture is going to appear in a certain location or not.  As usual in empirical psychology, the experimental setup is a bit involved — but if you want to really appreciate the evidence for precognition that Bem has obtained, there’s no substitute for actually understanding some of the experiments he did.  So I’m going to quote Bem’s paper at some length here, regarding his first experiment.

The setup was, in Bem’s words, as follows:

One hundred Cornell undergraduates, 50 women and 50 men, were recruited for this experiment using the Psychology Department’s automated online sign-up system.  They either received one point of experimental credit in a psychology course offering that option or were paid $5 for their participation.  Both the recruiting announcement and the introductory explanation given to participants upon entering the laboratory informed them that

[t]his is an experiment that tests for ESP.  It takes about 20 minutes and is run completely by computer.  First you will answer a couple of brief questions.  Then, on each trial of the experiment, pictures of two curtains will appear on the screen side by side.  One of them has a picture behind it; the other has a blank wall behind it.  Your task is to click on the curtain that you feel has the picture behind it.  The curtain will then open, permitting you to see if you selected the correct curtain.  There will be 36 trials in all.

Several of the pictures contain explicit erotic images (e.g., couples engaged in nonviolent but explicit consensual sexual acts).  If you object to seeing such images, you should not participate in this experiment.

The participant then signed a consent form and was seated in front of the computer.  After responding to two individual-difference items (discussed below), the participant was given a 3 minute relaxation period during which the screen displayed a slowly moving Hubble photograph of the starry sky while peaceful New Age music played through stereo speakers.  The 36 trials began immediately after the relaxation period.

The stimuli presented were as follows:

Most of the pictures used in this experiment were selected from the International Affective Picture System (IAPS; Lang & Greenwald, 1993), a set of 820 digitized photographs that have been rated on 9-point scales for valence and arousal by both male and female raters.  This is the same source of pictures used in most presentiment studies.  Each session of the experiment included both erotic and non-erotic pictures randomly intermixed.  […]

40 of the sessions comprised 12 trials using erotic pictures, 12 trials using negative pictures, and 12 trials using neutral pictures.  The sequencing of the pictures and their left/right positions were randomly determined by the programming language’s internal random function.  The remaining 60 sessions comprised 18 trials using erotic pictures and 18 trials using non-erotic positive pictures with both high and low arousal ratings.  These included 8 pictures featuring couples in romantic but non-erotic situations (e.g., a romantic kiss, a bride and groom at their wedding).  The sequencing of the pictures on these trials was randomly determined by a randomizing algorithm … and their left/right target positions were determined by an Araneus Alea I hardware-based random number generator.

The main hypotheses Bem made about the outcome of this experiment, before running it, were that

1.    “participants will be able to identify the position of the hidden erotic picture significantly more often than chance (50%).”
2.    “we should also find the hit rates on the erotic trials to be significantly higher than the hit rates on some types of non-erotic trials.”

And what were the results?

1.    “Across all 100 sessions, participants correctly identified the future position of the erotic pictures significantly more frequently than the 50% hit rate expected by chance: 53.1%.” (which is highly statistically significant given the number of trials involved, according to the calculations shown in the paper)
2.    “In contrast, their hit rate on the non-erotic pictures did not differ significantly from chance: 49.8.  This was true across all types of non-erotic pictures: neutral pictures, 49.6%; negative pictures, 51.3%; positive pictures, 49.4%; and romantic but non-erotic pictures, 50.2%.”

In other words the hypotheses made in advance of the experiment were solidly confirmed.  The experiment yielded highly statistically significant evidence for psychic precognition.  Much more than would be expected at random, given the number of subjects involved, the Cornell students were able to perceive the erotic stimuli from the future  —  but not, in this context, the non-erotic ones.

I won’t run through the other experiments described in the paper in similar detail.  You can read the paper itself for that, which is unusually clearly written for an academic article.  But you get the idea.  These are rigorously conducted psychology experiments, professionally automated and computerized — as is often done these days — with the only unusual aspect being that the response comes before the stimulus. The results are statistically compelling.  They don’t show any particular individuals who can foresee the future with total accuracy, every time.  But they show that on average, over a bunch of Cornell university students, the future can be foreseen more often than chance — if the future involves something of sufficient psychological valence (e.g. an erotic picture).  Some students appeared to have greater premonitory power than others, though nobody had anywhere near perfect accuracy on any of the experiments.

If Psi Exists, Why Aren’t the Observed Effects Stronger?

You may wonder why the results of Bem’s experiments weren’t stronger.  Why only 53%, why not 95%?  OK, so he didn’t find any experimental subjects who were so powerfully psychic they could predict the erotic pictures almost all the time — but then couldn’t he have set up a different sort of experiment, yielding a stronger effect?

Of course, outside the lab, people have reported many apparent cases of extremely dramatic psi effects.  But the long history of parapsychology lab research, going back far before Bem to Rhine’s ESP work in the 1930s, shows that when you bring psi into the lab, it tends to become more of a systematic statistical biasing factor than a source of individual mind-blowing “miracle events.”

A reasonable analogy might be the study of “falling in love.”  In the wild, in real life, falling in love can be a dramatic and overwhelming phenomenon.  But imagine studying it in the lab by pairing men and women with each other in various combinations and contexts, and measuring the “in love” tendencies that arise.  “Falling in love in the lab,” empirically measured in contrived laboratory situations, would very likely present itself as a weak and mercurial effect, a mere systematic biasing of the behavior of certain men and women toward each other — statistically meaningful, perhaps; but qualitatively different from its everyday-life manifestation, which is unpredictable by nature, and, partly for this reason, so overwhelming.

I don’t believe I have any particularly strong psi abilities myself, but in my everyday life, I’ve witnessed some rather striking examples of psi phenomena involving others.  For instance, one day a few years ago, a friend and I were walking in the forest with her beloved dog, and the dog (as was common) ran far away from us, exploring the woods and chasing animals.  Then, all of the sudden, my friend said, “She [the dog] is looking at a turtle.  I can see it right now as if it were in front of me.”

I was understandably skeptical: “Yeah right.  How could you know?”  Turtles were not that commonly seen in those woods.

I was going to call the dog, but my friend asked me not to.  Instead we quietly looked for the dog, and she was about 100 feet away staring intently at a turtle, which was sitting there peacefully by a stream.

Strange and striking — and like so many other real-life anecdotes of psi phenomena, damnably hard to replicate in a lab.

I’m reminded of another dog story.  My Japanese Chin, Crunchkin, once surprised us by showing how well he recognized himself in the mirror.  He looked at himself in the mirror curiously.  Then he walked across the room and picked up a sock, and stared at himself in the mirror with the sock in his mouth.  Then he put the sock down and picked it up again, all the while observing his mirror image do the same.  Finally, he apparently concluded the dog in the mirror was just him, somehow, and lost interest.  It was brilliant – but I know if we tried to replicate it in a laboratory setting (or in the house for that matter) it wouldn’t work — the dog would run away or act silly or something.  One of the general challenges of laboratory psychology is to bring out, in weak but systematic form, phenomena that occur much more strongly, but much more capriciously, in everyday life situations.

Evidence for ESP

Perhaps the most frequently replicated psi experiment is the “ganzfeld” ESP experiment, which was the reason Daryl Bem became involved in psi research in the first place.

In 1983, due to his combined expertise as a stage magician and research psychologist, Bem was asked to perform a careful evaluation of Charles Honorton’s ganzfeld work at the Psychophysical Research Laboratory (PRL) in Princeton, New Jersey.  PRL was funded by the McDonnell Foundation, founded by John McDonnell, one of the founders of the aerospace firm of McDonnell-Douglas.

“Ganzfeld” is German for “whole field,” and it refers to the procedure of placing an experimental subject in a setting of mild sensory deprivation, thus enabling (so it seems) greater sensitivity to psi phenomena.  The desired mind state is generally achieved by isolating the subject in a comfortable chair in a darkened room and having them listen to “white noise” through headphones.  In Honorton’s version of the experiment, halved ping-pong balls were taped over the subject’s eyes, and these were flooded with red light, creating a homogenous visual field.

Under these conditions, Honorton’s experimental subjects were then asked to try to perceive a video “target” that was being played in another isolated room, and being watched by a “sender.”  The perceiver was asked to discourse during the session, commenting on the visual images they see in their mind’s eye.  Afterwards they were supposed to rate either four still pictures or four moving videos, in order to judge which one they felt the “sender” was watching during the session.  The process was completely automated, removing any role for subjective bias on the part of the experimenter.

When Bem arrived, Honorton had just started a new series of ESP experiments that used the ganzfeld procedure — and after studying this work, Bem was convinced that results needed to be published in a mainstream science journal.  As Bem said, “I looked over the protocol, and was quite impressed….  I had read Honorton’s debate with [psi skeptic] Ray Hyman, and thought that the one talent I have is that I am able to reach the mainstream journals.”

Bem was unable to find any fatal flaw in Honorton’s work.  He became more and more interested in extending his research focus from personality and social psychology to psi research.  In 1994, Bem and Honorton co-authored a landmark article on psi in the mainstream psychology journal Psychological Bulletin. The article described the results of a thorough statistical meta-analysis of eleven ganzfeld studies.  (A meta-analysis involves combining data from a series of similar experiments conducted over a period of time, to come to an overall conclusion.)   The result of the meta-analysis was striking: subjects obtained overall target “hit” rates of approximately 35 percent, far above the 25 percent that chance performance would predict.

Bem and Honorton also carefully evaluated the possibility of a “file drawer effect.” This is the name used to illustrate a tendency to publish studies with positive results, while studies with negative results don’t get reported.  Their analysis shows that, to explain the obtained results using a file drawer effect, would require around 50 unpublished negative studies for each published positive one.  Given the large amount of cost and effort required to run a serious psi experiment, this really doesn’t seem plausible.  (It should be noted that many results in other areas of science would have to be thrown out, if one were to adopt significance criteria so strict as to rule out the ganzfeld data due to the possibility of an extreme file drawer effect like this.)

In the years since, there have been further meta-analyses of the ganzfeld ESP database, and there has been some back-and-forth with psi skeptic Ray Hyman about the results (which can be found on Bem’s website.  But ultimately, no skeptic has managed to explain away the results of these meta-analyses in a remotely convincing way.

Why Is Replication of Psi Experiments So Difficult Sometimes?

The chief bugaboo of scientific psi research has been replicability.  It has proved frustratingly difficult to precisely replicate the results of many psi experiments.  While the results of meta-analyses like the one Bem and Honorton did are compelling, it’s nevertheless frustrating that one often needs to proceed by analyzing the results of multiple experiments in a statistical way, rather than by simply doing a precise replication of “exactly” the same experiment over and over again, and getting exactly the same results each time.

As Ray Hyman, the most serious critic of the ganzfeld meta-analyses, has said: “Every field has ‘paradigm experiments’ where you can get results.  There are thousands of experiments in psychology that can be replicated, but in parapsychology there isn’t one where you can get that.  In no other field is there something similar.”

Perhaps the experiments described in Bem’s new paper, Feeling the Future, will finally resolve this problem and provide robustly replicable experiments demonstrating psi phenomena.  I certainly hope so.  But it’s worth briefly reflecting on possible explanations for the relative difficulty of replicating psi experiments.  The nonexistence of psi is one possible explanation, of course, but certainly not the only one.

Consider the experiment of dropping objects from the air, and observing that they fall to the ground.  This can be replicated perfectly easily.  But what if there’s a high wind?  Then the objects may blow away.  They may even “fall” up instead.

The point is that, in order to “replicate” an experiment, one must know the relevant factors to be held constant between the original experiment and the replication.  There is, after all, no such thing as an exact replication.  In the case of falling objects, we know that wind conditions are one thing that must be held roughly constant in order to have a fair replication.  But what things must be held roughly constant in order to have a fair replication of a psi experiment?

Many early psi researchers noted the possibility of “experimenter effects,” wherein the mindset of the experimenter might affect the results.  This is not uncommon in psychology experiments of various kinds.  Modern psi research protocols minimize this possibility via using automated set-ups, enabled by modern computer technology.

A more perplexing phenomenon is the apparent dependence of some psi phenomena on solar activity and geomagnetic fields.

If you’ve studied any astronomy, you may recall that a sidereal day is approximately 23 hours, 56 minutes, 4.091 seconds – shorter than the regular 24-hour solar day.  A sidereal day is defined as the time it takes for the Earth to complete one rotation relative to the vernal equinox (the point the Sun passes in March on its way from south to north, when the Earth’s equator is in the same plane as the center of the sun).  Astronomers use sidereal time to keep track of the direction to point their telescopes to view a given star in the night sky.

Why is this relevant to psi?  Because a meta-analysis has shown that the outcomes of psi experiments are related to sidereal time!  Specifically, in an analysis by Ed May and James Spottiswoode, the correlation between psi effect size and the Earth’s magnetic field was found to be greater for experimental trials occurring around 13 hours Local Sidereal Time.

May and Spottiswoode’s conclusion?  “AC [anomalous cognition] performance is modulated by a parameter which varies with solar activity.”

This sort of dependency is exactly the sort of thing that could help make psi experiments difficult to replicate.  We know that the arc of a falling ball depends on the wind, but we don’t know whether psi depends on solar activity and the earth’s magnetic field or not. If it does have such dependencies, we don’t know how they work.

It’s the sort of consideration that highlights the subtle interdependency between theory and experiment in science.  We create theories to explain experiments.  But then, to validate an experimental observation requires replicability — and the articulation of the conditions required to constitute “replication” in a given context requires some theoretical understanding of the phenomena in question.  Tricky? Yes.  But science has not progressed as far as it has by hiding its head in the sand and ignoring subtle, tricky aspects of the world.

But How Can It Be That Way?

Some people, who believe psi phenomena are real, think they lie intrinsically beyond the scope of science — perhaps in the same rationally inexplicable domain as the divine and the soul.  Most psi researchers disagree.  Daryl Bem declares confidently that “there will be a physical explanation.”

I tend to agree.  But what kind of explanation?

Nobody knows yet for sure, but the most likely direction seems to be quantum physics.  Regarding precognition, in particular, there is much reason for hope here.  Quantum physicists, with no thought at all to psi, are prone to discussing the possible ways in which the future may affect the present (see the recent Discover magazine article, for example.)   Strange though it may sound to the layperson, the foundational equations of quantum physics don’t provide much support for the common-sense notion that time only flows forwards.

In 2006, the American Association for the Advancement of Science (AAAS) organized an interdisciplinary conference of physicists and psi researchers specifically to discuss the physics of time and retrocausation.  The proceedings were published as a book by the American Institute of Physics.

Quantum theory itself is not understood perfectly, and its equations can’t (yet) be exactly solved except in very simple cases. So it’s conceivable that psi phenomena can be explained by modern physics as is, once we learn to solve and interpret the equations better.  But many researchers (myself included) suspect that some subtle tweaks to quantum physics may be needed to create a detailed explanation for psi phenomena.

Nobel Prize-winning physicist Brian Josephson and University of California statistician Jessica Utts have noted one promising research direction:

… theories that presuppose that quantum theory is not the ultimate theory of nature, but involves (in ways that in some versions of the idea can be made mathematically precise) the manifestations of a deeper “subquantum domain.”  In just the same way that a surf rider can make use of random waves to travel effortlessly along, a psychic may be able to direct random energy at the subquantum level for her own purposes.  Some accounts of the subquantum level involve action at a distance, which fits in well with some purported psychic abilities.

This feels right to me,but for the purposes of this article such details aren’t really to the point.  Rather, what seems worth emphasizing is that there may well be some variant of quantum theory that allows coherent behavior among molecules in the brain in a manner that could potentially support psi phenomena such as precognition (as reported in the Bem paper).  We don’t know for sure, but given that quantum physicists themselves are prone to talk about influences going backwards as well as forwards in time, this sort of notion is certainly not off the wall or pseudoscientific.

The Dangers of Pseudo-Skepticism

So there is significant laboratory evidence of psi and a plausible direction for finding a scientific explanation, even if there is no explanation yet.  What, then, should the scientific-minded individual’s attitude be?  Isn’t a great deal of skepticism still warranted?

Certainly, any individual’s claims of dramatic psi capabilities should be viewed with great skepticism.  There have been too many frauds and tricksters fooling themselves and/or others.

On the other hand, skepticism can be overdone. defines a skeptic as “One who instinctively or habitually doubts, questions, or disagrees with assertions or generally accepted conclusions.”  A certain amount of skepticism is important for science.  While science has sometimes upheld traditional beliefs, it has often upturned them.  Telescopes and spacecraft revealed airless space and hunks of rock in the sky, not gods or ancestors.  But there’s one thing skepticism is not about — it is not about rejecting observed data because it disagrees with current theoretical understanding.  One of the disturbing features of psi research in the last few decades has been the exceptional derision it’s faced from some individuals and groups purporting to advocate rationality and skepticism.

For instance we have CSICOP (the Committee for the Scientific Investigation of Claims of the Paranormal, pronounced “Psi Cop”), a well-funded organization devoted specifically to debunking claims of psi phenomena, with multimillion dollar headquarters in New York and LA.  In 2006, CSICOP changed its name to CSI (the Committee for Skeptical Inquiry), but continues to use the Internet domain.

On their website, CSICOP states that their brand of skepticism “does not reject claims on a priori grounds, antecedent to inquiry, but examines them objectively and carefully.”  And yet, listen to Lee Nisbet, former Executive Director of CSICOP: “[Belief in psi is] a very dangerous phenomenon. Dangerous to science, dangerous to the basic fabric of our society… We feel it is the duty of the scientific community to show that these beliefs are utterly screwball.”  To my mind, this isn’t skepticism — it’s dogma.  A real skeptic would want to investigate claims like Bem’s with an open mind, to understand the real truth, regardless of beliefs or biases one way or another.

Fortunately, some others associated with CSICOP have taken a more open and scientific attitude than Nisbet.  In 2000, CSICOP’s Skeptical Inquirer magazine contained an article on Bem’s earlier work, entitled The Best Case for ESP?, that didn’t perform any debunking, and certainly did not reveal Bem as an “utterly screwball.”  Psi skeptic Ray Hyman was quoted complaining about the lack of clear replications in the psi research literature, but then also noting that psi critics have rarely risen to the level of sophistication of serious psi researchers: “Most of the criticism of the field is of straw people.  The criticism has been very bad.”  The article concluded on the following ambiguous and reasonable note:

The history of parapsychology has not offered much hope that future generations will witness the scientific confirmation of psychic ability, but the prospect remains provocative and tantalizing to the imagination.  Daryl Bem’s research may stand as a major guidepost on the road to discovery, or go down in future decades as one of many promising findings never to be replicated nor confirmed.

(Ben Goertzel and psi “skeptic” James Randi, at the 2010 Singularity Summit)

And then there’s James Randi, the well-known stage magician, a founding member of CSICOP, who has made a career out of debunking televangelists and other fraudulent pseudo-psychics.  I met Randi at the 2010 Singularity Summit in San Francisco, where we both spoke.  My talk was on the applications of AI to genetics, and he expressed his appreciation for it.  His talk focused on some of his successful debunking efforts, and was compelling and hilarious.  However he never mentioned the existence of serious academic psi research, focusing instead on the bogosity of various evangelical faith healers and showman-style stage psychics.  In my brief chat with him, I chose not to broach the topic of psi, not wanting to disrupt the pleasant vibe.  But one of his helpers did express interest in having me speak at one of Randi’s “Amazing Meeting” conferences.  I wonder if they’ll still be interested after reading this article!

One of Randi’s claims to fame is his “One Million Dollar Paranormal Challenge” — a million-dollar prize he offers to “any person or persons who can demonstrate any psychic, supernatural or paranormal ability of any kind under satisfactory observing conditions.”  This sounds interesting at first, but gets less so when you dig into the fine print.  For instance, the introduction to the rules says, “All tests must be designed in such a way that the results are self-evident, and no judging process is required.”

As the more science-oriented psi skeptic Ray Hyman has noted, this doesn’t really mesh with the way science works: “Scientists don’t settle issues with a single test, so even if someone does win a big cash prize in a demonstration, this isn’t going to convince anyone.  Proof in science happens through replication, not through single experiments.”

After all, think about it: how would Daryl Bem win Randi’s prize for the sort of work described in his new paper?  Note that, in order to win, applicants must pay the expenses to run their demonstrations at Randi’s own site.  Further, they must sign a contract effectively giving Randi control over all the publicity derived from the demonstration.

So: Dr. Bem would have to set up a lab full of computer equipment in Randi’s facility in Florida at his own expense.  Rather than using Cornell University students, he would have to recruit local individuals to take part in the experiments and pay them for their participation.  Presumably, he would also have to bear the cost of enabling Randi and his staff to monitor the experiment on video to enable thorough fraud detection.

And after all that, would Randi judge the results as sufficiently “self-evident”?  If not, then Bem would have wasted a lot of time and money replicating the experiment according to Randi’s requirements.  It seems it would be much more worthwhile for other university psychology labs to replicate Bem’s methodology with their own students.

Overall, it seems clear that the community of “professional skeptics” is not really adequately equipped to evaluate the scientific evidence for psi.  They are well suited to evaluate supposed faith healers and stage psychics, but that’s a different matter.  The scientific evidence for psi is going to have to be evaluated by the scientific community.

It’s damn hard to get papers on psi accepted into mainstream scientific journals due to the overall anti-psi bias of the scientific community.  The Journal of Parapsychology is reasonably high-quality — and follows many admirable practices such as accepting papers without regard to whether their results are positive or negative regarding the presence of psi phenomena — but it isn’t taken as seriously as it should be outside the psi research world.  Bem’s recent paper shows it is possible to publish psi results in high-quality mainstream science journals, and I hope we’ll see more of this.

Further Reading

If this article and Bem’s new paper have stoked your curiosity about psi phenomena, where can you turn next for more (high-quality, scientifically-sound, BS-free) information?

If you want to read more scientific papers on psi, Bem’s website is one place to start.  There are also some interesting and relevant articles on Dr. Jessica Utts’ website. Utts is  a statistician at UC Irvine who has worked for the US government evaluating apparent psi phenomena.  Her article, An Assessment Of The Evidence For Psychic Functioning, is particularly compelling, containing descriptions of a variety of experimental results, and concludes:

It is clear to this author that anomalous cognition is possible and has been demonstrated. This conclusion is not based on belief, but rather on commonly accepted scientific criteria.  The phenomenon has been replicated in a number of forms across laboratories and cultures.  The various experiments in which it has been observed have been different enough that if some subtle methodological problems can explain the results, then there would have to be a different explanation for each type of experiment, yet the impact would have to be similar across experiments and laboratories.  If fraud were responsible, similarly, it would require an equivalent amount of fraud on the part of a large number of experimenters or an even larger number of subjects.

Nuclear physicist Dr. Ed May has a number of interesting articles on the site of the Laboratory for Fundamental Research, including information on the US government’s StarGate project to investigate “remote viewing” (the use of ESP to perceive far-off locations via the minds of others who are currently experiencing them).

I have not found any truly great textbooks about psi research, but a fairly thorough historical grounding in the area is provided by the edited volume Basic Research in Parapsychology put together by Dr. K. Ramakrishna Rao, an Indian researcher who is currently Head of the Department of Psychology and Parapsychology (as well as Dean of Faculty of Arts and Social Sciences, and Vice Chancellor) at Andhra University in Visakhapatnam, India.  This is an academic work, with some chapters that are as readable as a newspaper feature article and others that are more involved and technical.  It provides a high-quality overview of serious psi research from the 1960s through the 1990s.

On the less academic side, Damien Broderick’s 2005 book Outside the Gates of Science gives an extremely sensible and readable “popular science” level summary of psi research and the thinking that underlies it.  (Broderick is best known as a science fiction writer, but devoted transhumanists may recall that his excellent book The Spike covered essentially the same ground as Kurzweil’s The Singularity is Near, but a half-decade earlier.)

Finally, if you have a historical bent you may get a kick out of Upton Sinclair’s 1930 book Mental Radio.  Best known as a novelist and social advocate, Sinclair also had a deep interest in psi research, due partly to his wife Mary’s claim to have telepathic abilities.  Over a three year period, he invented and conducted a series of 300 tests probing these abilities, which are reported in his book.  For instance, in some of these experiments, Sinclair would sit in one room, make a drawing and place it into a sealed envelope; while at the same time, Mary would attempt to psychically detect the image and draw a copy of it.  Other experiments involved variations of the protocol, for example longer-distance transmissions.  Mary’s success rate, as recorded by Sinclair, was vastly greater than chance.

Sinclair’s Mental Radio contains the following wonderful quote, which I urge you to keep in mind if you choose to further explore the work of Bem and other modern psi researchers:

It is foolish to be convinced without evidence, but it is equally foolish to refuse to be convinced by real evidence.

See Also

How Sports Fans Can Help Save the World

You may also like...

81 Responses

  1. Bob says:

    I’ve probably had several precognition experiences, though two I can remember including one from this morning. Both of these precognitions happened while I was at the tail end of my sleep cycle. The first one happened while in college. There was a girl I liked, and one morning I dreamed of her asking me to go with her somewhere. She has never done this before, and I had never had any dreams of her before. Sure enough, she asked me, and it was even in the same place I saw in my dream. The second one occurred this morning. I was sort of waking up, but not yet conscious at the time. A random sound of a girl talking came into my mind. It was a strange occurrence because it made no sense to me whatsoever. Then literally two or three second later a girl walked by outside my window, which was opened, and she said the exact same thing I heard in the exact same tone of voice. It sounded like she was talking on the phone with someone.

    I’ve also had many experiences of sudden feelings of being touched and then thinking of someone. I’d log onto Facebook or check my email and sure enough they’d be logged on, just sent me a message or email.

    Also at work I seem to know when a person is about to laugh hysterically out loud before it happens. I don’t even have to know what they are talking about. I just get this sudden weird feeling in my chest before it happens.

    As someone trained in the sciences and statistical math, it’s hard to just explain these events away as chance.

  2. JERRY TELLE says:

    I have witnessed one extremely strong precognition. It occurred while sunbathing on a beach in Cabo San Lucas. I was listening to my tape player and soon discovered, after some scary disbelief, that whenever a score was about to play it played in my head for about 5 seconds before the actual tape score played. The precognitions ended at the end of the side of tape in question — no matter how hard I tried to duplicate the previous ambience.

    This may seem insignificant but I assure everyone it was a very “mystical” more than a little frightening experience.

    I am ADD BIPOLAR and have other, not psychic, but explicit experiences others can only guess at via fMRI. And more flashes not yet recognized.

    These include flashes of the behavior determining process, i.e., flashes of the cue, the selection of behaviors either habitual or decision, the accompanying affective experience associated with every near in time activity, flashes of the activity as it occurs, including expectations of secondary and primary affect outcomes.

    If the selected activity does not occur as predicted, the outcomes are weighed and discrepancies calculated via the RPE (Reward Prediction Error) process. And associations of the outcomes for future predictions are stored.

    Other explicit activities include the imitation learning process, 3 near amnesic experiences, and various other experiences others consider implicit — not withstanding ridicule from various professions?? just as I assumed they would. Small minds in big places. A lifelong situation.

    I have 3 exercise machine patents and other creations that occur all the time.

    Regardless I enjoyed your straightforward writing.

    All correspondence welcome.


  3. Tom says:

    In 1995 I saw an accident in exact detail 20 -25 minutes before it happened. My wife was in the car with me as I drove through the intersection (where the accident later took place) relayed what I saw to my wife. Dropped her and my son off for his nap at home. I then went to buy a cigar, drove back to that intersection and there was the accident, phoned my wife, relayed the info, insisted she wake my son, raced home, drove her to the scene, upon her seeing it, she looked at me jaw dropped and said “you”re an alien”.

    While I am not an alien, pre-cognition is real, and wish I had the scientific background to explain it.

  4. Jerry Telle says:

    Believe it or not I have had, very few, but apparently very real precognitions. I also have almost complete awareness of on going pre- activity cognitions.

    These “implicit?” cognitions are well known to fMRI research scientists. I flash! on the predicted activity, and flash on the physical but mostly affective Emotional/feelings/ Sensations outcomes — before activity.

    I also flash on what I call the LSA — that is the Last Specific Activity at the onset of said activity. Terribly interesting!

    Would love to join your Humanist movement, but paypal will not accept my one and only card. they are the only ones!

    Very frustrating — they of course have no “contact Us” icon.


    Jerry Telle

    Previous volunteer:
    suicide life line. hospice counselor, adult literacy and crime victims advocate.

  5. Anon says:

    I’ve look into the first experiment (the paper then the data). I can’t explain the result, but there is obviously something wrong. I’ve averaged each picture from the DPE 60 (18 erotic picture, 18 non-erotic). For each average, binomial(n=60, p=0.5) gives a standard deviation of around 6.5%, so the number should be 95% between 63 and 37. For erotic picture he got higher, but even for non-erotic pictures where he got out of 18 values 2 values above 63 and 2 below 37. It cancels each-other, but the probability of it happening randomly has a p of less than 0,3%. Said otherwise, it has very few probability that non-erotic pictures were answered randomly. That enough should be fatal to his study.

    I’m not an expert in statistic so a second opinion would be nice.

  6. Jeff says:

    I don’t know how you will be able to measure or capture this ability but I do know that it is genuine because I have experienced it myself. And I have not told many people about it as I do not want to labelled as a nutcase but I KNOW that it exists.

  7. leslie williams says:

    Im Pre Cog and I have a Thought your going to say “” Ya Right”

  8. Sceptic says:

    If its truly a real phenomenon then it would shake the already established laws of physics! for instance sake say: one criminal(who is to be executed within a few minutes) had a precognition of himself murdering a person in between that time! Now my question is how he ll be able to know that he was precognitive? untill unless he commits one within that time before he is executed, which has a void possibility! Now for sucess rate of the following psi experiments done above lets say: the experiment of pictures with skin conductance, the reason for which i believe that some of it gave a posetive considerable statistical output is: perhaps the types of pictures were shown to the test subjects in a certain pattern unconsciously by the experementer(which he was not aware of during the conducted experiment) which was being percieved by the test subjects also in an unconscious manner, therefore the test subjects were able to predict the type of pictures which were to be shown by the experementer beforehand in an unconscious manner based on that pattern which inturn created a response stimulus before the pictures were being shown! which made both of them believe that they have sucessfully experimented the phenomenon to be true.

  9. Sydney S. says:

    I have just this to say: there seems to be no study of how the power of predicting the future is transmitted from generation to generation as occurs in the traditional societies of all continents and among shamans. This is because psi has not yet been proved in an irrefutable way. However, I have been married for 27 years to a wife who belongs to a family belonging to that category, hailing from a very ancient priestly caste. Sometimes falling in trance or just looking a somebody’s face she, and now our three children, have visions of that person’s future and have regularly, almost weekly, sometimes same dreams of the future events that actually occur. The problem is that they cannot actually predict the date when they will occur and some of them actually occur almost the next day or so, sometimes after some months or even years. But they occur exactly as seen by her or our children. Nevertheless, this keeps guiding us safely through life. Believe it or not, I just don’t care! I think that researchers like Bern are doing a great job. There must be something somewhere that explains this sort of time reversal. And the fact is that time does not exist elsewhere than in the human mind.

  10. Jessica says:

    Precognitions are real, those who never experienced them can have their doubts. However, it gives the believe doubts too.

    I video tape mine, which include descriptions of sudden deaths, facts about unborn children and traumatic events.

    I am awake while I see a vision or receive answers in my mind. It appears just like a stored memory. One day, I wake up and all of a sudden I have that memory, except it is of the future. It can also come to me while I am around a loved one, and I can just feel it.

    • mike gallego says:

      I also have precog but in dreams. Its mostly in my close proximity and mostly tragic. I have them most of my life. So now I tell my wife what will happen and we wait for the outcome. It has become part of my normal routine in life.

  11. Sven says:

    Don’t discount the JREF Million Dollar Challenge so readily.

    “Dr. Bem would have to set up a lab full of computer equipment in Randi’s facility”

    What a couple of laptops?

    ” Rather than using Cornell University students, he would have to recruit local individuals”

    Probably could get volunteers without paying them. I’d look at porn for free.

    “And after all that, would Randi judge the results as sufficiently “self-evident”?”

    The standards for success would be established before the test. In other words, the specific percentage of successful hits would be determined ahead of time and agreed to by both sides.

    If it truly is statistically significant, and replicable, I’m sure the University could foot the bill for being tested as the Million dollars payoff would be a sure thing.

  12. Walter van Groningen says:

    Hi there from the Netherlands,
    sry for my bad englisch, i’m a little drunk and onely (a bit) higher level high-school educated.

    I could tell you a long time before, that the future lies in your dreams. I’ve had, and still have, dreams that come true in unbelevibly precise details.
    beleve it or not. (and no, i don’t really care but i like to share this with you as a second confirmation about futureseen-theories)
    i’ve also seen that the future can be changed, by just don’t let certain things happen. this was just happend becouse i couldn’t handle things in my life.
    wich excactly means: origin -> effect.
    this is the scientificly theorie if im correct.

    my opinion: everyone’s the same, and everyone could dream his/her future.

  13. Mike Puckett says:

    Perhaps this slight leakage of information from future to past, assuming it is real, is the universes way of subtly directing us and it to a desired end-state

  14. jesse m says:

    When looking for data patterning so subtle, ANY unseen variences in the test, equipment, calculations, collection procedure, subjectivity, etc….can cause the ‘magical’ effect in question.

    Also, even if we assume something “real” is happening, why can’t we assume the causality is the other way around….that these people are bringing their imagined reality to physical reality.

  15. Hi again,
    the metaanalysis you refer to (abstract)

    is 17 years old. If there are so many interesting result in this field, one should think there would be a more recent metaanalysis?

  16. Hi,

    Most of the links in your article are dead.

  17. anna says:

    As I child, I could find my lost pets by walking to the end of the sidewalk in front of my home and sense the direction in which they were. Don’t ask me how. I don’t know. But, I would find them.

    Also as a child of about 8 or so I was able to direct my parents on how to find my sister and her husband who had gotten separated from us in their car. I directed them to where they had stopped to fill up for gas, and then at a dead end in the hwy, which direction to go to find them.

    As an adult, I’ve also had experiences, but not as vivid. I had precognition in a vivid dream that I was about to be fired the next day (dreaming that I and each of my coworkers were tied up to our office chairs and put on an assembly line like cattle hanging from a slaughter hook, gagged and bound, traveling down the assembly line…….more details too long to write). I wrote my resignation letter when I got to work, which I handed to my supervisor as he was about to fire me. I’ve sensed when someone was about to call me and they would call me, or would enter into my home knowing that a certain person had called me and would wait until they called again in a few minutes……no answering machines at the time. And, no regular schedule for them to call. I could just hear the phone ringing in my head when I walked into the room.

    My mother and my sister often picked up the phone to call the other one, only to have the other person was already on the line, and had already called but the phone had not rung. I’ve twice been able to see myself passing in front of cars in a parking lot through the eyes of someone else, or walking down an alley across from my college, and turned my head to see the person that was looking at me. It was the same person each time………..a teacher with whom I was very fond of.

    More, but these are an example. So, yes, I believe in both precognition and premonition. My mother had the same powers as you can read above, and often told my father that he was going to be offered a certain job and to accept it.

  18. Anonymous says:

    I wonder if Bem does his own programming for his experiments. If he doesn’t, then I ask the question…

    Which is more likely, his results are real or his programmer is screwing with him for some personal reason?

  19. matt_k says:

    Thanks for the info about Daryl Bem’s work; it’s an interesting set of experiments.
    It’s very curious the amount of emotionality any mention of anything to do with ‘anomalous perception’ can elicit in some people, both favorable and unfavorable emotional responses. The negative responses tend to be the most vociferous and least rational (add an exponential amplitude to both if they happen to be a scientist).

    For a fairly comprehensive and painstakingly meticulous meta-analysis of over 100 years of psi research, you might want to take a look at Dean Radin’s ‘Entangled minds: extrasensory experiences in a quantum reality’. A meta-analysis is primarily what this book is, with some discussion of current trends in thinking on just what exactly is going on with the observed phenomena in physics terms. Lab experiments tend to yield positive but low significance (with the exception of the early remote viewing experiments at SRI, in which the researchers took the novel approach of intentionally selecting a core group of experimental subjects with already established above average track records both in psi function, not self selected, but as identified by others, and in being successful in the business of regular everyday life instead of a bunch of college student volunteers- duh!). However, after some hundred or so years of accumulated data, this frequency of low level psi function in the general population acquires an astronomical odds against chance factor. The book’s hardly a Dan Brown thriller, maybe good bedtime reading for insomniacs, but it’s pretty solid info on the subject.

    It’s decidedly no longer ‘if’, but ‘how’.

    ‘Precognition’ is particularly interesting for its potential implications for very fundamental physics models. All that is known of what is called precognition involves ‘feedback’- that is, the person experiencing the precognition actually personally experiencing in some way the predicted event when it finally rolls around (hearing about it, seeing pictures, reading about it or actually being there for the event). As one researcher described it, ‘It’s like remembering the future’. It is tempting to think in terms of using precognition to alter the course of events leading to an undesirable future occurrence, but, if you change the circumstances to prevent the anticipated event, there’s nothing in the future to ‘remember’ in the past (although there may be a potential to mitigate to some extent…).

    And there is the small matter of not being able to recognize as ‘precognitive’ what might actually be an accurate impression but one never becomes aware of the fulfillment of the precognition through more customary sensory input.

    It would have very significant implications for physics if it could be shown that accurate precognition could occur without feedback. No possibility of feedback occurring for the experimental subject. Since were dealing with seeming temporal anomalies here, that’s EVER. That would be tricky to set up.
    There is at least one relatively famous instance where one person described in great detail a past situation (some 50 years past) which was entirely unknown to the researchers, who thought an error had been made, and unknown to the test subject. But feedback finally arrived confirming the accuracy of the description after the subject who presented the description had died. One of the researchers eventually stumbled upon a picture of the site as described in a history book. There was a complete break in any potential direct personal access to the information for the test subject. (See R. Targ and K. Harary’s book, ‘The Mind Race’) To the best of my knowledge there has been no similar event for a precognitive perceptual disjunction ever recorded in any of the research. Such a separation would be some significant evidence in favor of the notion that the future already exists in some way and, as for someone traveling from New York to San Francisco, San Francisco can be said to already exist, but just not yet for the traveler.
    If precognition is not possible to achieve without ‘feedback’, the nature of the future may actually be even a much more unsettled matter than it already seems for many of us. It would also instruct a physics model for how psi function could work.

    Love the picture of you and Randy, very fun. :-)

    Thanks for the article.
    matt k.

  20. Paul Tiffany says:

    At another science/tech outlet, PhysOrg,

  21. Anonymous says:

    Comic strip poking fun at Bem’s recently accepted studies on premonition (psi).

  22. Anonymous says:

    I have stopped trying to run round after so-called scientists. They are some of the most arrogant, audacious, ignore-ant, angry, foul-mouthed fools I have ever had the misfortune to meet, though have learned many lessons from such encounters. I am speaking about the ‘science advocates’ that frequent–seemingly eternally–the online forums supposedly dedicated to scientific exploration, and I include Richard Dawkin’s forums too, and especially.

    NONE of these clowns seem to have heard of scientism…..! Theuy believe that term has no meaning, YET, YET, they are members of the cult of scientism.

    None of them know what is going ON. Either with the culture, or…*anything*. Blind as bleedin bats are they. In fact I am sure fundamentalist Christians were be better company and that’s saying something!

    So why do I include in the tile ‘corporate facism’? because this is the sign of how dumb and stupid these people who pretend a scientific superiority over us, and all past ancestors…are. They have no idea about the world. They will support it, and any mention of anything not right, like 9/11 being an obvious fals flag they will call you ‘woo woo’ about questioning that like they will call you woo woo about wanting to ask questions about UFOs, and other unexplainable phenomena. They are fools–the lot of em. VERY rarely do you meet a true scientist. I would regard Dr Mack one. But these foul people do not even have the dency not to speak ill of the dead. At the Richard Dawkins forums when I mentioned Dr Mack their abuse of him was foul.

    So I just say to people–encouragingly–do NOT waste your time with these people. An old Chinese saying: you cannot wake a man up who is pretending to be asleep.

    • Cat Detective says:

      You should really seek professional help of some kind. You are having paranoid delusions and you’d be much less angry/miserable/terrified all of the time if you got help.

      Also, accusing scientists of “scientism” when they are addressing scientific questions is like accusing mechanics of being “mechanists” when they are fixing your car.

      If a scientist tells you that there’s an empirical way that he can prove Elton John is a better musician than David Bowie, that’s scientism. He’s trying to use science to answer a question that is beyond the scope of scientific inquiry. If your mechanic tries to offer you a mechanical explanation of how the human brain works, then he’s being a “mechanist”. He’s misapplying his knowledge to topics that are beyond his purview.

      Similarly, you don’t want to use political pundits and conspiracy blogs as sources of evidence about things that are better off being scientifically investigated. Also, again, get some help. You’ll feel a lot better.

    • Anonymous says:

      Have you ever heard of ‘ad hominem’?

  23. Ananta says:

    To explore the phenomenon of precognition, one should understand the nature and its laws. Collection of statistical values has nothing to do with this phenomenon unless it is related with the natural laws. Based on the experience of nature, I can say that sun will definitely rise tomorrow. I can definitely say that early in the morning once I wake up I have to go to bathroom. These are only few things which any human being can foretell. To go deep into the phenomenon of precognition, one has to understand the law of Karmas (action). There is a saying “as you sow so shall you reap”. This is not happening automatically or God is ruling on that. But there is a mechanism how our own actions are encoded in our consciousness (it happens in present), how they are collected into layer by layer (past are within us) and ultimately how they flows out (past is stepping out into the future) and give the present shape to us. Whenever some important things is going to happen to one’s life, certain feeling comes first. That point, we have to understand that the karma is flowing outward. Law of karma can give us answer for everything. It can give us what is our origin and where we are destine to. Psychology will be only a part under the laws of karma. There fore its time for scientist to think about how to understand the mechanism on how the three stages of Karma are interacting with our life. I would rather advice to open a new branch of science call “Karmalogy”

    • Anonymous says:

      I am going to assume this is a joke.

      • Ananta says:

        Hope you have a good laught at my comment.
        What if i say, we are the product of our own past and nothing else.We have only present in our hand and future is the product of our past and present. Actually, though we are in our body, we are a lump of informations. The informations which has been encoded in our subtle body in previous births has activated at present and our body is just a mean to interact with those informations. These informations are formed in our subtle body as per our own ‘karma ‘. There are 5 doors of sences through which the informations pass into our subtle body.As i already explained in my comment above that the saying ‘ As you sow, so shall you reap ‘ is not controlled by any devine power. But there is mechanism how it works.

  24. Mike Austin says:

    A researcher friend (Stephan A. Schwartz) turned me on to this article, and I found its tone especially interesting. I have a well-developed respect for thoughtful scientific research and enjoyed most of it.

    Something worth considering for y’all are these things: (1.) it could likely turn out that providing scientific validation for precognition – or other things common in psi – could be similar to proving the existence of God. It could depend on individual perceptions and nothing else. (2.) While there are many excellent researchers now trying to confirm or deny the existence of psi phenomena like precognition, there are many of us who simply use psi abilities on a daily basis because we’ve found them to be useful.

    It’s fortunate that no one need feel obligated to anyone else to prove or disprove their individual experience; if I spent my time justifying my experiences instead of doing work which I find more productive for myself, I’d feel caught in an endless, positive feedback loop where I’d never accomplish anything.

    • Anonymous says:

      Regarding your first point, what if even just one person correctly identified the position of the picture close to 95% of the time? How is this similar to proving the existence of God or dependent on ones perception if it is clearly demonstrated and documented? If you’d said collective rather than individual perception it may hold a bit more weight. The fact that someone with the world view of a religious fundamentalist might (and do) react by attributing the phenomena to evil forces, but does not question the phenomena itself shows you can only go so far in that denial.

      Frankly your second point kind of reeks of ego, however subtle it may be which adds nothing to the discussion itself and rather contradicts the first point. If it is only dependent on individual perception then either everyone is literally in their own universe (even though something outside it obviously has an impact) or it’s simply delusion.

      And this leads to your final point regarding work *you* find more productive for *yourself*. Psi phenomena by its very nature involves interaction with something else, which makes this kind of attitude rather illogical even though it is rather standard and understandable. Instead of viewing it in an open ended way in terms of better understanding of the phenomena you are framing it in terms of a question of proving or disproving. If the perception is mistaken in the first place (asking the wrong question) then the answer to that approach is NEITHER, but that does not seem to concern people who are only interested in what they can get out of it. So it starts to look like interest in the truth of the matter vs. self interest, and the latter may very well be another feedback loop that actually gets nothing done. In other words, it is only productive in terms of the way you identify and therefore references self interest, rather than any discovery of truth.

  25. Sanjeev says:

    ….need extraordinary evidence.

    53% is not above the noise floor. I speak from a common sense PoV, would you trust a person who speaks truth only 53% time and lies the rest? Will any reputed scientist invest his time and funds in replicating the experiments for such a figure? I suspect that with a big enough sample size it will come down to 50.

    Now, this does not mean that the psi is impossible. There can be some truth in it, but this experiment falls short of the mark. I agree with the author (of this article) that Psi is difficult to test in a lab and the traditional methods are mostly not very conclusive.

    My question to Ben Goertzel , who is experimenting with AGI at present, is — what are the implications of Psi being true for AGI (and superintelligent machines) ?

    If Psi is true, there is no hope of ever achieving an AGI with current hardware and software, because it means that the consciousness is not merely a material phenomenon.

  26. David Pearce says:

    Precognition would be good news for Maxwell’s demon
    ('s_demon )
    - and very bad news for the Second Law of Thermodynamics.
    I think we should be extremely cautious in interpreting this study.

  27. Paul Barrett says:

    The title on my post put up just now should have been “doing science” vs “doing statistics”: Bem seems unaware …

    sorry …

  28. Paul Barrett says:

    Well, Gigerenzer’s (2004) paper entitled “Mindless Statistics” just about sums up the level of scientific thinking and ‘thoughtful’ data-analysis shown in Daryl Bem’s article.

    The key issue which seems to completely evade his consciousness is that of ‘aggregation’, and the specific form of hypotheses that can be tested using inferential data-model statistics.

    The effect size parameter is a ratio between aggregate-data variances, or the scaled difference between two aggregate-based parameters (the means).

    But, what is the actual form of hypothesis which can be tested this way?

    H1: If the hypothesis is that aggregate statistical effects can be shown to be larger than zero, then Bem did a good job. But the hypothesis says nothing about “people show psi ability” – because “people” defined as constituting many single-unit effect-producing entities were never examined. All that was examined were aggregates of all the “single-unit” scores/outcomes.

    H2: On the other hand, IF the hypothesis was to be: all humans show evidence of psi ability, then every individual must show that ability (however tiny) in order for such a hypothesis to be supported.

    A careful reading of Bem’s paper shows that he continually wants to argue for H2 (as an evolutionary advantageous property of being human, like having a prefrontal cortex etc., we all have it in varying degrees), but implements a hypothesis testing procedure which can only address H1.

    We know from the pitiful effect sizes that many people did not show psi (assuming normally distributed data implied by the methods he used). Hence, H2 is already disproven by Bem, although he is so committed to a statistical view of phenomena that he doesn’t recognize what his own results imply.

    Consider the fact that some people in his sample may have shown truly outstanding evidence of psi, some (fewer) will have actually produced behaviors less than chance would have expected, some will be 50/50 … you average them and what do you get? Exactly what Bem found, slightly above chance effect sizes.

    But what have you got in the real world (not in his statistician’s view of the world)

    1. Some people seem to possess psi.
    2. Some people don’t possess any psi; they respond at chance expected levels.
    3. Some people seem to have responded ever worse than chance.

    As a scientist, #1 is the really important finding as psi need not be a property of every human; what’s important is being able to demonstrate conclusively that some individuals really do show precognition/psi.

    Who cares about an effect size of 0.15 for an entire sample when perhaps 5 individuals in that sample show a psi effect that is consistently and replicably 80% accurate above chance-expected levels? Instead of this silly paper, and it is silly, we’d be reading the work in Science or Nature – with our collective scientific jaws dropping around the world.

    And that’s the bigger issue – do you test specific hypotheses which are framed as causal for each human being in whatever population you have in mind, or hypotheses about “aggregates” of individuals?

    In essence, does your hypothesis say that an effect (caused by some property/combination of attribute properties) should be observable in every individual in your sample – or – does it state that in a sample of individuals, only some will show the “causal” effect.

    To state that “in a sample of individuals, only some will show the “causal” effect” begs the question “why”. That statement cannot support claims that causal theory X explains the outcome Z. Because it doesn’t. It does for some, not for others.

    Statistics allows you to avoid basic principles of causation observed and practised in the physical sciences because it treats individuals not showing the effect as “error”. It doesn’t seek to explain what error means; it simply defines it as an effect not consistent with the hypothesized effect. Furthermore, its explanation of cause is now only observable in a statistical aggregate, not in any individual.

    That’s not science, it’s insurance-salesman actuary. Fine if all you wish aspire to is counting behaviors attributable to the “the average man or woman”, as an insurance salesman might, but useless if you want to try and understand phenomena as a scientist might.

    What price the early physics of black-body radiation if a Kirchoff-specified blackbody device produced the following results from 100 such blackbodies…

    1. Some bodies of equal substance and shape composition, given constant applied heat of x minutes duration seem to radiate in accordance with theory.
    2. Some bodies of equal substance and shape composition, given constant applied heat of x minutes duration produce unexpected radiation.
    3. Some bodies of equal substance and shape composition, given constant applied heat of x minutes duration produce no radiation.

    What exactly would averaging such results have contributed to a causal theory of such radiation? Nothing.

    Statistics describes aggregated data. That’s really helpful when you want to talk about trends and “likelihoods”, and “inferences”. But for hypotheses which are required to test theories of “causality” … no.

    The really important scientific data remain sight-unseen within the raw datasets from Bem’s studies; what he has produced is just more “atheoretical response counting, the province of clerks” (Kline, 1998).

    I have no bias one way or another about the existence of psi; what we have with Bem’s work is a series of experimental studies which completely miss the target question at hand: do individuals possess psi ability?

    But, he can answer that question -IF- he digs deep into his data – and isolates then retests those individuals who appear to show substantive ‘psi’ effects (and just as important) those who show really substantive below chance response rates.

    Will he? No, because to do so requires the thinking processes and “causal outlook” of a scientist, and not that of a ‘mindless’ statistician working with numbers.

    Paul Barrett

  29. Tim Tyler says:

    The paper cites Lewis Caroll: “Near the end of her encounter with the White Queen, Alice protests that “one can’t believe impossible things,” a sentiment with which the 34% of academic psychologists who believe psi to be impossible would surely agree. The White Queen famously retorted, “I daresay you haven’t had much practice. When I was your age, I always did it for half-an-hour a day. Why, sometimes I’ve believed as many as six impossible things before breakfast””.

    If this study is a joke, it seems like a fairly elaborate and expensive one.

  30. Anonymous says:

    that’s what i was going to say

    how did you know

  31. David Pearce says:

    Even an extremely weak ability to foretell the future would allow one to clean up at the casino and on the stock market. This hasn’t happened to date. I think the Standard Model is safe for now.

    • Anonymous says:

      So whatever hasn’t been discovered yet could not possibly exist because it has not be utilized already? What? The fact that nobody has made use of it does not necessarily imply that psi does not exist: rather, it implies that it either does not exist *or* that if it exists no good way to control it in order to make its effects more apparent has been found *yet*.

    • Perhaps esp abilities is above a measurable threshold only for a very few people, and these people don’t gamble and don’t play the stock market.

      I am agnostic about esp. It would be cool but I have never seen evidence for it myself. At the same time I can imagine many possible mechanisms for esp, all compatible with our scientific knowledge and outlook.

    • Stefan R says:

      If there are psychic powers, i think, the people who would tend to have these are more likely to be spiritual and perceive the current world more as a “layer of many existences” concerning space and time.
      Because of their view on the world as just one of many realities, they don’t focus that much on material wealth and money rather than on developing their spirit which should be kept clean from getting poisoned by greed and material possessions.

      Thus they are not likely to become addicted to gambling or spend a whole lot of time and money in education which would bring them into trading positions in the financial markets.

      On the other hand, traders and gamblers concentrate pretty much of their effort on gaining material wealth and money.
      So they don’t think of more than this reality thus do not develop spiritual thinking which may be necessary for any kind of psychic powers.

      Of course, it’s my personal assumption that being spiritual is needed for having psychic powers, if there are any.

      But it would explain, why there aren’t succesful psychic stock market gurus meditating in front of a crystal sphere containing a dollar bill.

      Nevertheless, i don’t believe kinds of psychic power can be really well trained enough to really predict future events on demand and so not in real professional and accurate fortune tellers.

      I think of psychic events more as of “sparks” trough space and time, which are just being spontanously received by more spiritually people who may have a better “connection” for sub-reality events.

      But due to our primitve minds we can not recognize more than just a little glimbse

      … not yet, of course.. ;)
      let’s see how following forms of intelligences will do.

    • Theo Ibrahim says:

      How do you know people haven’t cleaned up at the casino or stock market using such abilities? :-) I think looking at patterns of deviation from the null hypothesis in those institutions would be very interesting.

    • Brian says:

      best to see

      Remarkable book causes weird stuff to happen.

    • Anonymous says:

      The “Oracle of Omaha” seems to have cleaned up quite nicely on the stock market. Maybe we should be testing HIM for psi :)

    • matt k. says:

      why is it that the first thing people think of is making money with this? if they’ve got a psi tallent, it’s probably busy working overtime in the background just trying to keep ‘em out of trouble.

      well, this has happened on occasion, intentionally using precognition to predict a money producing future event and then acting on that prediction. i knew someone who was a member of a group of amateur remote viewers. they tasked themselves with various upcoming sporting events and placed small bets on line based on the results of the pooled predictions. the person i knew had parlayed a $10 initial investment into a little over $2,000 over the course of about a year. if i did the math right, that’s roughly 200% return on investment. not too shabby.
      there’s also Russel Targ’s famous silver futures experiment where some serious money was made [and lost, but that's anther story, more about what influences psi performance- never let your client and rv team come face to face. ever.] also using a form of remote viewing.
      Pat Price, one of the stars of the early remote viewing experiments at SRI was a retired sheriff? in law enforcement. he had used his remarkable talents on several occasions to solve crimes in his jurisdiction. this sort of practical day to day application is somewhat typical of how natural talents who are aware of their abilities employ them.
      yes, and the ‘Oracle of Omaha’, remarkably ‘lucky’. some aren’t necessarily conscious of their psi abilities but respond subconsciously.
      it would be interesting to subject J K Rowling to some tests. i strongly suspect she plucks ideas out of thin air [she seems to have stolen my trademark jagged scar on the forehead and seriously blinged it up a bit].
      there’s more applications of the broad spectrum of phenomena lumped under what is called ‘psi’ than merely making money.
      the CIA’s remote viewing program lasted some 20 years. it wouldn’t have lasted that long if it weren’t productive. the program wouldn’t have even gotten off the ground had they not been able to clearly demonstrate a usable product. they provided intel for a number of government agencies, drug enforcement, coastguard, pentagon.
      India has picked up remote viewing; i understand they’ve come up with some interesting new innovations in working with it.
      then there’s that jolt- a surprised sense of recognition of a complete stranger- upon meeting that special someone who you will spend some years with- remembering future happiness.
      remembering the future…
      science? ach, i try to keep up with what’s happening in this area but i like my mysteries. they’re much more fun and you don’t have to struggle with trying to prove anything to pseudoskeptics. ;-)

    • DreamSayer says:

      Not True. What is being perceived are future events. This anomaly does not strengthen numerical memorization of random numbers(much less in a half asleep state), as it is mostly based in Synesthesia. Those who believe this means potential lottery winners are not assessing the concept. Precognition is merely observing the future, so if you werent already going to win, you would never have a vision of it. I would love to see you remember numbers & dates right when you wake up, as they begin to vanish from the conscious mind very quickly. That concept is based off of myth created by the Skeptics who bolster precognition into the realm of superpowers. It’s time to actually try and understand the anomaly instead of wildly denying it with ideas of condescending magical rhetoric.

    • EdG says:

      Your assumption that you would know if someone “cleaned up” at the casino or stock market is fallacious. You state with certainty that this hasn’t happened to date.

      How would you, personally, know? Why would such a person advertise themselves to you or to anybody? How can you be certain that people who are successful in the stock market aren’t using a psychic ability? Wouldn’t you just assume they’re extremely lucky or that they have some secret stock picking formula?

      Nobody ran a sub-4 minute mile until somebody ran a sub-4 minute mile. Nobody cured polio until somebody cured polio. These is so much we don’t know about so many things that making absolutist pronouncements is just plain foolish.

  32. Anonymous says:

    “…the programming language’s internal random function”.

    Unless the experimenters are using a large enough sample of people, with a sufficiently random seeding of the random number generator, and that random number generator is producing ‘luxury’ random numbers with maximum decorrelation, then there will be a bias detected.

    Even the GNU GSL random number generators with luxury random numbers require an adequately large sample size so that Law of Large Numbers statistical convergence is genuinely obtained.

    • Anonymous says:

      True that there may be a some correlation in the pseudorandom numbers depending on the method used to generate them, but this does not necessarily imply that this will results in a bias in the test subject’s inference of them.

      I completely agree with you w.r.t. to obtaining a large enough sample size – I strongly suspect that these results will not be consistently replicated in the future [but what do I know, I am not a psychic ;)].

  33. Brian Josephson is one of the geniuses whom I pay attention to, and I know from face-to-face conversations that he is open to the possibility suggested by Prof. Daryl Bem. One of my greatest influences is the late Robert Anson Heinlein, Dean of American Science Fiction, who explicitly included Psi in some of his fiction, even though he and his wife were experienced and pragmatic Engineers. My late co-author Sir Arthur C. Clarke was likewise extremely rational, an engineer (worked on Radar in England during World War II) equally a pioneer in the Space Program (inventor of the synchronous communications satellite), and yet had Psi in some of his fiction. No single experiment will convince me one way or the other. But I am not such a fool as to ignore Brian Josephson, Robert Heinlein, and Arthur C. Clarke. Hats off to Ben Goertzel and h+ magazine for this fine article. I shall read the scientific paper, and stay tuned to see what happens as other labs replicate it, and perhaps the paradigm shifts — in a scientific revolution.

    • Anonymous says:

      Clarke and Heinlein both used faster than light travel as storytelling devices in their fiction (emphases on “storytelling devices” and “fiction”). That does nothing to change the fact that there is no evidence that such devices can be physically realized. Their literary achievements have nothing whatsoever to do with the topic at hand, which neither of them studied scientifically, as far as I’m aware. Fiction isn’t considered an acceptable way to espouse a viewpoint on a scientific phenomenon.

      With regard to Josephson, he was undeniably a smart guy who made substantial contributions to physics but in his later years he seems to have succumbed to Nobelist Syndrome (the tendency of Nobel Laureates and other scientists of that caliber to believe in pseudoscience and other forms of bunkum and insanity). Simply put, these days anyone with a lick of sense regards him as a complete crank.

      See also: Transistor inventor William Shockley (enormous racist, believed in eugenics), DNA co-discoverer James Watson (racist, general nutcase, espouses mild form of eugenics), Phillipp Lenard (literal, honest-to-God Nazi), PCR inventor Kary Muller (doesn’t believe HIV is caused by AIDS, in addition to many many other insane things), Robert Millikan (believed in eugenics, thought cosmic rays were created by God), Guglielmo Marconi (distinguished member of Italian Fascist party serving under Benito Mussolini).

      The list is a mile long. The point is that someone can do brilliant and groundbreaking work in one field and be a complete dunce about almost everything else, and further, this seems to be true about a high percentage of the people who do brilliant and groundbreaking scientific work.

  34. termm says:

    I understand that the 53.1% perception is statistically significant, especially considering that there were 1000 students. But, I think many people will blow this off saying “it’s just 50%” which I could understand. Too bad it wasn’t 60% or something accuracy, that would be saying something.

  35. Anonymous says:

    I once (via TV, so this is highly anecdotal and nothing i myself would take as much proof of anything) heard about some study showing that human will and expectations can influence machines, the experiment was using some random algorithm in that case too.

    Unless i’ve misunderstood the experiment this would be the same thing, the computer picked erotic pictures randomly while the participant were guessing(and probaby expecting) certain results.

    This experiment should be repeated with randomly pre-genereated lists of the selections.

    And no, i don’t really belive in psychic powers, but i don’t mind exotic tests as long as they are well constructed, it keeps things interesting in science to not have any doors barred.

  36. Anonymous says:

    In other news, the CIA is enslaving people using technological mind control.

    I can’t believe fluff like the cornell article gets press while real abuses like this one get ignored or ridiculed.

  37. Michael Swan says:

    The erotic pictures may have been mostly comprised of skin colour.

    Reflected light could have easily escaped the curtain, slightly varying the colour of the wall. Our eyes are very, very good at picking up slight colour changes like this.

    As an animation lecturer, I taught students that, all objects reflect light.

    Imagine a generated Orange 3D sphere placed in real world environment with trees, sky etc.

    To make it more realistic, the reflected light from the environment, green for trees, blue for sky, must affect the orange ball, and the orange ball must now affect the environment

    A trick you can do is place a light on a tree with 3% sampled tree green strength, and point to to the ball, so it receives that lighting from the tree.

    People will tell that the orange ball in the scene with the 3% more light is “more real” than without it. They won’t tell it’s because it looks like it has 3% more tree green light! So sub-consciously, people figured out which curtain to choose without wholly being aware of what they were doing, because it is largely an automatic process, ie like the control processes of your heart.

    Retry the experiment where the monitors are behind a brick wall/ or the subject is blindfolded.

    Or, a good image recognition program is likely to predict the location, and could prob. be trained to be more “psi” ic than humans lol.

  38. DaveM says:

    “But, I think many people will blow this off saying “it’s just 50%” which I could understand. Too bad it wasn’t 60% or something accuracy, that would be saying something.”

    I don’t think so. A 53.1% number, if proven, does require explanation. I doubt scientists will say “too bad it wasn’t 60%”.

    I think the crux is “proven”, which means that many independent laboratories have to be able to replicate that 53.1% number. This is really what’s missing in psi-research. The small effect size is not so much of a problem as the lack of replicability.

  39. Knew you'd look says:


    Reality is just pre-buffering. To keep the show flowing uninterrupted, things need to be calculated a bit in advance.

    Occasionally some of this buffer leaks into the “now” stream, when there’s not enough time to calculate everything and reality “skips a frame” revealing the already calculated and “happened” future just a bit too early (typically no more than seconds or some fractions of) to the consciousness. It’s like when sound and video going out of sync for a few frames.


    The world (simple and easy stuff like stars and planets) and the consciousnesses (complex and slow) within it are computed in parallel on different processors and the processing threads fail to stay in the right execution order while they’re swapping data with each other, perhaps due to simple fluctuations in processor speeds or because the ‘mind’ processor wasn’t fast enough to process everything in time. The ‘world’ processor is feeding the next moment’s data to the ‘mind’ processor while it’s still processing ‘the now’.

  40. Mark Plus says:

    If humans really have precognitive abilities, why, then, do transhumanists suck so badly at predicting “the future”?

  41. Anonymous says:

    Two documents challenging the Bem paper. The former is conceptual and the latter is empirical.

  42. Anonymous says:

    pictures of curtains.

    not actual curtains.

  43. Anonymous says:

    It was pictures of curtains, not actual curtains, and the people clicked on them to reveal the images. If you had actually read the study, you would have seen that. But no, you had to express your uninformed “critique.”

  44. Anonymous says:

    I like the first one: “We don’t like his results and thus it’s time to reinvent statistics.”

  45. Anonymous says:

    comment on comment by Knew You’d Look…

    Dude that was awesome. Have you written anything else? I’d like to read it.

  46. Graham says:

    Even if it was 60% it still doesn’t mean anything. The effect is reported at a p<.01 level. I.e. there is only a 1 in a hundred chance that this result was obtained by accident. So it looks “significant”. However, if you consider how many thousands of psi experiments have been run in the past 50 years, we would expect that an effect that happens once in a hundred times by accident should have happened loads of times.

    Given the propensity for only ‘positive’ research to be published (failure to replicate is a yawn) and only controversial findings to make it into the press, we are clearly on the receiving end of a very strong bias effect. Good meta-analyses are needed to sort this out (where they look at a large sample of published results and consider the rate of “significant” results against the probablility of such a rate.) And many of these have been done. And they show that results like this are just what they look like – statistical flukes, overwhelmed by the massive amount of evidence showing no psi effect at all.

    Results like this can happen for all kinds of reasons – pure chance, experimenter bias, inappropriate statistical treatment, methodological flaws, and so on. The time to start believing in them is when many other labs have replicated the result and it is shown to be robust. Remember Jacques Benveniste’s evidence for ‘water memory’ that was supposed to be evidence for homeopathy? Don’t take a single result from a single lab as strong evidence. It is not.

  47. Anonymous says:

    This is a response to Graham.

    “Don’t take a single result from a single lab as strong evidence. It is not.”

    Very well said. Reality is not decided by single experiments. Nor by prizes. Which is why applying for the $1M prize – which is entertainment, not science – is not helpful.

    “Good” meta-analyses are needed? I suggest that competent statisticians have indeed reviewed the meta-analyses and have concluded that the psi effect is statistically significant. IOW we already have ‘existence proofs’.

    The real problem is finding a mechanism. I wish the researchers luck with that – but that sort of thing is way beyond my skill set and I leave it to physicists to work it out (if, indeed, the mechanism can be pinned down).

    P.S. Jacques Benveniste’s real error was desiring publication before independent confirmation. AFAIK he did not do much personal verification, either.

  48. Cat Detective says:

    This sort of nonsense has been pretty thoroughly debunked. Look up electrogaiagrams and the Global Consciousness Project for more info. They basically have a large number of gizmos constantly generating random numbers, and then wait for an event to occur which coincides with an apparently “non-random” sequence from one or more of the gizmos, which they can then point to and say “Look, our gizmo predicted this would happen! We must be measuring global human consciousness (or whatever the crackpot terminology du jour is).”

    It’s strange how they never know exactly what their gizmos will predict until after it happens, but I guess they’ve adopted a more convenient definition of the word “prediction.”

    It all boils down to confirmation bias and intellectual sleight of hand.

  49. Cat Detective says:

    When were these results confirmed by meta-analyses? As far as I’m aware, the meta-analyses that have been conducted of these types of studies have been overwhelmingly negative. There is no “existence proof” that I’m aware of.

  50. Anonymous says:

    But here is a successful replication of two effects reported by Bem, retrocausal habituation and induction of boredom by Alexander Batthyany:

    Retrocausal Habituation and Induction of Boredom: A Successful Replication of Bem (2010; Studies 5 and 7) on SSRN:

  51. Anonymous says:

    Very intelligent comment.


  1. May 4, 2012

    [...] are changing — and will change — human beings in fundamental ways.” Go ahead and read: Is Precognition Real? Cornell University Lab Releases Powerful New Evidence that the Human Mind can …, Ben Goertzel, November 4, [...]

  2. September 7, 2012

    [...] Is Precognition Real? Cornell University Lab Releases Powerful …Quick overviews and basic info on Precognition. The existence of precognition, as with other forms of extrasensory perception, …. namely a selective memory fo… Latest news and videos on Precogni… [...]

  3. September 27, 2012

    [...] Un article sur le site [...]

  4. March 2, 2013

    [...] evolutionary advantage to being able to perceive the future possibility of sex – essentially, precognitive detection of sexually available partners would aid in successful reproduction, and thus select for itself. In [...]

  5. December 5, 2013

    […] known as precognition.  More often an experience of highly sensitive persons (HSPs), possessing a foreknowledge of a coming event is thought to be the product of one’s hyper-sensitized cognizance of their […]

Share Your Thoughts