The Rising Star of Parapsychology
Reviewing 150 years of research
I’ve written previously about a communications anomaly called the psi effect, where information is found to pass—imperfectly, though with better than chance accuracy—through channels that can’t be explained through conventional means. Writing about psi is actually something I’ve been doing for some years, now; in the summer of 2012, I took extensive notes on my readings of the parapsychological research, and this post is essentially an update on my findings from that time. Readers may be interested not only to see the state of research today, but also to learn about recent advances in the field and changes in the social acceptance of parapsychology.
The general public doesn’t seem to know much about psi, and academics widely regard investigation into paranormal phenomena as nonscientific. For example, Wikipedia has:
Claims for the existence of paranormal and psychic abilities such as clairvoyance have not been supported by scientific evidence.[5] Parapsychology explores this possibility, but the existence of the paranormal is not accepted by the scientific community.[6] The scientific community widely considers parapsychology, including the study of clairvoyance, a pseudoscience.[7][8][9][10][11][12]
But writing eleven years later in 2023, it’s already getting harder even to substantiate the claim that parapsychology is widely dismissed as a pseudoscience. That quote above appears only in the Wikipedia article on clarvoiance, but if you check the article on parapsychology itself you’ll see it’s now careful not to refer to the field as a pseudoscience outright.
Poking around with web searches or browsing through YouTube makes the change seem even more striking. Quantifying the degree of change is difficult, but a quick search engine survey can help to give an idea of current opinion. I like using duckduckgo because (among other reasons) it’s unfiltered by my personal preferences—duckduckgo gives a sense of what people are interested in and talking about today. And the first six articles I get from a search on parapsychology pseudoscience are:
Reflections on pseudoscience and parapsychology: From here to there and (slightly) back again: Parapsychology is a science, “even if paranormal forces do not actually exist.”
Parapsychology: Science or Pseudoscience? - TCU Magazine: “[M]uch of the skeptics’ point of view is not coming from a scientific perspective. ‘They often are trying to embarrass believers or convince people how absurd it is. It becomes clear that the nonbelievers are often as biased as the people they’re trying to discredit…”
Longtime Skeptic Now Accepts Parapsychology as Science (Basically what the title says)
Parapsychology at Wikipedia (Discussed above)
Parapsychology: Science or Pseudoscience? A 2003 article published in the Journal of Scientific Exploration stating “qualitative distinctions were not found that could justify classification of parapsychology as pseudo-science. To warrant that, other criteria to define science would need to be established.”1
The legacy of the Duke Parapsychology Laboratory: “[T]he claim of parapsychology as pseudoscience has been paradoxically refuted by meta-analyses. One analysis looked at 309 precognition studies, conducted by 62 experimenters published in the Journal of Parapsychology between 1935 and 1987. The cumulative probability associated with the overall results was p = 10-24 (equivalent to .000000000000000000000001, where .05 is considered statistically significant). Other analyses support the idea that alleged experimental design errors alone cannot account for the uniqueness of parapsychology experimental results.”
So Wikipedia looks like a holdout on this issue, preserving older skeptical attitudes which are disappearing elsewhere. Not only is parapsychology no longer broadly regarded as a pseudoscience, but it seems many people are even beginning to realize that psi is a real, replicable effect.
That isn’t to say that mounting data has created anything like scientific consensus. Around twenty years ago, Luis Cordón’s (2005) Popular Psychology: An Encyclopedia, took the position that parapsychology is pseudoscientific, because,
Ordinarily, when experimental evidence fails repeatedly to support a hypothesis, that hypothesis is abandoned. Within parapsychology, however, more than a century of experimentation has failed even to conclusively demonstrate the mere existence of paranormal phenomenon [sic], yet parapsychologists continue to pursue that elusive goal.2
And the same attitudes are still around; as recently as 2020, the American Psychologist published a dismissive review by Reber and Alcock (the first of whom has been an active psychologist since before I was even born). In case it isn’t clear from their title, “Searching for the Impossible: Parapsychology’s Elusive Quest,” Reber and Alcock clarify that they absolutely aren’t going to listen to parapsychologists, no matter how much evidence turns up:
Our position is straightforward. Claims made by parapsychologists cannot be true. The effects reported can have no ontological status; the data have no existential value.3
But they also say,
What we find particularly intriguing is that, despite the impossibility of the existence of psi phenomena and the nearly 150 years of efforts during which there has been, literally, no progress, there are still scientists who continue to embrace the pursuit.4
So when they tell us “Claims made by parapsychologists cannot be true,” that’s just an admission of their epistemological position. I believe them when they tell me that evidence is meaningless when it contradicts their personal beliefs. But the idea that parapsychologists haven’t made any progress is a more objective claim, something which might or might not be true.
So is it true? Were there “nearly 150 years of efforts during which there has been, literally, no progress?”
Parapsychology in its Infancy
If we wanted to decide when parapsychology as a science officially began, we might very well choose 1882, the year when The Society of Psychical Research was established in London.5 That wasn’t exactly 150 years before 2020, but parapsychology before 1882 had accomplished essentially nothing, and the field was still a pretty mixed bag for a few decades after that. As in, I get the impression that the main problems old school parapsychologists would’ve had with Mystery Incorporated wouldn’t have been their lack of credentials, difficulty cooperating, or tendency to include their dog in strategic discussions, but rather their garish mode of dress:
Picking through the writings of early parapsychologists invokes a mixture of gothic weirdness and farce. They produced tracts with charming titles like Psychical Research and Coincidences, H. P. Blavatsky and the SPR: An Examination of the Hodgson Report of 1885, The Alleged Haunting of B—House, or Thirty Years of Psychical Research. The opening paragraphs of the latter text have:
[I]n spite of the great advances in physics, chemistry, and physiology, the laws of these sciences, as at present known, do not account for certain exceptional phenomena, and these phenomena being inexplicable by orthodox science, it has been found convenient to ignore them. But these strange occurrences, whether they be accepted or denied, still remain facts; their actuality is unaffected whether we find a place a for them in recognized science or not.
It therefore seems desirable to present the mass of these phenomena methodically. However unusual in their occurrence they must, as facts, be subject to laws and therefore be accessible to study, i.e., to science. A science, or at any rate an orderly survey, of the supernatural and the occult is at least possible.6
There’s more there—much more, but no science as I recognize it. Some just might say that the habit of assuming the existence of controversial phenomena and then attempting to survey or categorize them is, at best, overly optimistic. Speaking more frankly, I suspect that this Panglossian attitude made such a terrible impression on the public that it saddled parapsychology with a reputation for lunacy that lingers to this day.
But it’s difficult to be harsh on them, especially after reading about the skeptical movement of the day, which responded to the early parapsychologists with fulminations like The Intellectual Underworld:
An intellectual underworld of supposed parapsychological phenomena threatens our time, giving a sham fulfillment to our human longing for knowledge and truth. The desire to master the world by a disciplined knowledge and to think the universe in ideas of order and law cannot go together with a real satisfaction and belief in the chaotic superstitions of mediumistic humbugs. Whatever the sources of this prevailing superstition may be, there ought to be no disagreement about its intellectual sinfulness and its danger to society… The most important part of the fight, however, is to recognize the danger clearly, to acknowledge it frankly, and to see with open eyes how alarmingly the evil has grown around us. No one will fancy that any social schemes can be sufficient to bring superstition to an end, any more than any one can expect that the present fight against city vice will forever put a stop to sexual immorality.7
As they say, the past is a foreign country. In light of opinions like this, one wonders how difficult it was for the early parapsychologists to function; in 1895, one of the more substantive articles, What Psychical Research Has Accomplished, opened by implying that the political climate of the day made parapsychological research politically dangerous:
For centuries America has given unstinted welcome to all comers, and if her oft-abused hospitality is now extended only to the immigrant who can show his credentials, we citizens of States too old and too prudent to be so open-handed have surely no cause to complain. Seeing that with us newcomers in the world of thought plead their cause with a rope round their neck, it is matter for thankfulness that America is still willing to hear the cause first, and adjust the rope afterwards.8
For all its informality, this article is much more respectable than most; later on, a number of experiments are mentioned, and we find discussion on the familiar subjects of study design, sensory contamination, collusion between subjects, and objections and suggestions from skeptics.
Most interesting was a report of some of their results. These early parapsychologists carried out a series of remote viewing studies where subjects tried to guess whole numbers drawn from a bag; the experimental protocols are clearly recorded, along with extensive remarks made during one of the sessions. Essentially the subjects had to guess a randomly selected integer from 10 to 90, so each guess had a 1 in 81 chance to be correct. But they report that,
[i]n the whole series of these experiments 644 trials were made, the number being correctly named (i. e., with both digits in their proper order) 117 times, and with digits reversed 14 times.
It is clear that chance cannot explain these results. It is almost equally clear that they cannot be attributed to fraud, unless, indeed, we suppose that not one, but all the experimenters, were in collusion. There still remains the possibility that the information was given unconsciously, and, probably, received unconsciously. Apparently, under the conditions described, the only normal channel of communication would be by the ear. It may be suggested, for instance, that Mr. Smith muttered the word audibly to the percipient. This hypothesis must, indeed, be regarded as extremely improbable, for various reasons : (1) Mr. Smith himself and the other experimenters were fully aware of this danger and on their guard against it. (2) No movements of Mr. Smith's lips were observed by the two trained and vigilant witnesses. (3) An analysis of the failures does not show that there was any tendency to mistake one number for another similar in sound.
Nevertheless it may be admitted, especially in view of the possible hyperaesthesia of hypnotized subjects, that if these experiments stood alone the hypothesis that the information was actually conveyed by auditory means might be preferable to the hypothesis of a new mode of communication. But they do not stand alone. They are but one of many groups of experiments conducted by different observers and under varying conditions, and no one hypothesis will cover them all. Mrs. Henry Sidgwick again, assisted by Mr. Smith, Miss Alice Johnson and others, conducted a further series of experiments in which the agent and percipient were in different rooms. In some of these experiments the agent and percipient were on different storeys of the house, separated by a wooden floor covered with a thick Axminster carpet. In others the percipient was in a room with the door closed and the agent, Mr. Smith, was outside in the passage, the distance between them varying from 10 to 15 feet and upwards. Both agent and percipient were under close observation throughout the trials; and it seems incredible that any sounds which escaped the notice of the observer who sat close to the agent and watched him continuously could have been perceptible to the percipient sitting at a considerable distance, and with a closed door or a ceiling and carpet intervening. In these experiments out of 252 trials, the number was guessed correctly 27 times, and with digits reversed 8 times (Proc. S. P. E., Vol. VIII., pp. 536-596)…. Men who have really examined the subject will, I think, admit that the evidence accumulated during the last ten years, if it does not demonstrate the reality of thought-transference, at least establishes its claim to our consideration.9
I was pretty surprised to discover this. Their methodology had certain weaknesses—the sender and receiver were in the same building, subjects were told whether their guesses were correct before further numbers were drawn, the randomization process was questionable, the total number of trials and replications was modest, and so on. But these early results were definitely intriguing.
They also cast their nets much more broadly than later parapsychologists, considering apparitions, magic, and witchcraft. And there seems to have been some rhetorical benefit to this, at least at the time: a very sensible a priori argument against the paranormal is the way in which people have often circulated questionable accounts of phenomena which turned out not to exist. The members of the Psychical Society considered this with a brief investigation:
At the outset of the work Mr. Gurney was confronted by an argument seemingly of some weight, which he states as follows: “All manner of false beliefs have in their day been able to muster a considerable amount of evidence in their support. What right had we then to assume that our evidence was more trustworthy than that by which these admittedly false beliefs were supported? In order to test the validity of this argument Mr. Gurney decided to examine the grounds for the belief in witchcraft. He chose this partly because it was the classic instance of a false belief alleged to have been amply supported by evidence; partly because there is an immense and comparatively recent literature dealing with the subject, most of it written by men of eminence in their day. Mr. Gurney succeeded in showing that there is practically no first-hand evidence for the more extreme marvels, except that of occasional drunkards, or confessions extracted under torture; and that the only phenomena fully attested by eyewitnesses were such as could readily be explained by hallucination, hysteria, trance and other known causes.
In other words, “we’ve seen psi directly, but nobody really saw witches doing magic first-hand (except some drunks and other people who don’t count).” There may be something to this; in general I tend to resist being swayed by gossip about people, and prefer to form opinions on the basis of things I have directly observed. In terms of parapsychology, however, it might tend to devalue the products of research in favor of direct exposure, or lack thereof, to unusual phenomena.
I don’t present any of these early arguments to suggest they ought to be convincing to anyone. Still, I think it’s pretty clear that parapsychology made some progress even in its infancy. They explored various phenomena in different ways, learned about some of the problems to be overcome, and began to listen seriously to skeptical objections. This wasn’t progress in terms of making meaningful discoveries, but it was definitely progress in the sense that they laid the groundwork for others to begin more systematic investigations.
There was a strong current of naïveté at the time, and even the better work tended to proceed under the assumption that extraordinary claims might only require ordinary evidence. But the stimulus of skeptics and disbelievers taught them that they had better try to be convincing. Gradually, gropingly, they refined their methods, and began to consider the weaknesses in their program.
For better or for worse, investigation of the psi effect had begun.
The Early Years: 1930 to 1940
It’s commonly claimed (by wikipedia, and the Duke University) that parapsychology was actually founded by botanist Joseph Rhine in the early 20th century. This really isn’t true, but his founding of the Parapsychology Laboratory at Duke University of North Carolina, in 1930, seems to have marked an upswing in research.10
From reading through their published papers, my impression of parapsychologists as a whole is that they’d become subtly more serious around then. And my general reading of the literature suggests that by 1930, parapsychologists realized that they needed to definitively establish the reality of their subject matter by ruling out alternative causes for so-called psychical phenomena.
They were still interested in other phenomena, like ghosts,11 or psychic healing and near death experiences, of course.12 But now parapsychologists were aware that they had something to prove, and they were savvy enough to know that they needed to choose the ground of the debate. Ghosts, seances, levitation, the kind of things that excite the press were pretty hard to substantiate. So the goal now was mainly just to establish the existence of psi, either in terms of extrasensory perception (ESP) the ability to receive anomalous information, or in terms of psychokinesis (PK), the ability to influence the environment through unexplained means. If you poke around in the literature, you’ll find studies using the terms ESP and PK were generally concentrated around this era.
Part of the reason for the narrowing of focus may have had to do with parapsychologists disbelieving in spirits; a lot of them seemed to prefer using psi as an explanation for everything; for example, claims about hauntings could be beexplained in terms of ghosts, but parapsychologists often preferred to explain hauntings as a manifestation of PK from living people.13
But why were they so interested in ESP and PK to begin with? My sense is really that they were learning to be pragmatic about concentrating their efforts on phenomena that could be studied under laboratory conditions. Proving ghosts exist is extremely difficult if you have to chase all around trying to find them, but getting volunteers to come down to the laboratory is pretty easy. (It worked for the Ghostbusters, anyway.) If the parapsychologists could conclusively establish that there was (or wasn’t) such a thing as an anomalous transfer of information which could not be plausibly explained through conventional means, then they could move on to the other stuff later.
So to this end, several experiments were designed. The most common of the era used a set of 25 Zener cards, with one of five different symbols on the face of each card:
The cards were laid out face-down, subjects were asked to guess what lay underneath, and their guesses were compared to chance expectation. Viewed by the standards of the day, the experimental paradigm had problems. For example, even though Rhine eventually moved to a mechanical method for shuffling, cards being cards, there will always be the possibility for subjects to notice subtle marks on them; even a single barely-recognizable card in a deck of 25 could shift the chance for guessing correctly from 20% to as high as 24%.
Rhine also looked at PK through experiments where subjects tried to influence the roll of a die, first by rolling themselves, and then ultimately by using a machine.14 Here obviously problems in the randomness of the dice could come into play, and while there are ways of dealing with this—say, by requiring subjects to roll a series of numers in a sequence—I’m not aware of Rhine having taken those steps to eliminate conventional explanations for his results.
And it goes without saying that other parapsychologists continued their own investigations elsewhere, like Stanford University,15 Clark University,16 or (because hey, why not) Hollywood.17
But frankly the nature of the studies themselves is not what is most interesting to me about these experiments. What’s interesting is this:
In 1940, parapsychologists Pratt, Rhine, and others reviewed all experimental studies on the psi effect they could find in scientific journals and other published sources.18 They rated the 33 studies they could locate on quality, and calculated an overall effect size. Forget for a moment about whether they found a statistically significant effect—this was the first meta-analysis to have ever been carried out.
Parapsychologists literally invented the highest standard of experimental evidence seen in science today. Not chemists, not anthropologists, not even physicists. Parapsychologists.
So now for the outcome of the meta-analysis—you probably won’t be surprised to learn that the results supported the existence of psi. Considering the problems that are understood to plague the sciences today, there would have been many reasons to doubt the result. But the meta-analysis was evidently somewhat convincing by the standards of the day; though only three critics agreed to offer new comments, two of these changed their position on seeing the studies compiled together.19
In many scientific domains, clear effects established across a wide array of studies by unrelated researchers in multiple laboratories would be regarded as definitive. Psi’s existence would be called “settled science.” But here again, parapsychology was ahead of its time. Psychology wouldn’t reach its replication crisis for another half-century, but questions of fraud, sensory leakage, and improper analysis were raised even before the meta-analysis was published.20 Such criticism was extremely important for the field to progress—remember that psi is defined as an anomaly that cannot be explained through conventional means; so long as conventional explanations remain plausible, an effect cannot be called psi.
Now we probably shouldn’t be too quick to dismiss the studies in the 1940 meta-analysis as explainable by conventional means. Many of the experiments did make use of double-blind recording techniques and took precautions against sensory leakage.21 The rate at which different laboratories returned statistically significant experiments showed no statistical difference from one another, which is not what would be predicted by fraud; it would be easy to claim, say, a 92% success rate, but across 17 studies from other laboratories, roughly 60% returned significant results, just as the 33 from Rhine’s laboratory did.22
Nevertheless, the problems weren’t fully accounted for by every study. Protocols for experimentation were not standardized, many studies definitely did allow for sensory leakage or subject cheating by allowing them to shuffle decks, and overall, experimental quality was not such that the criticism of skeptics could be fully addressed.23 And even deeper problems remained in the form of the file drawer.
The File Drawer Problem
Simply accepting the results of published studies as representative of the state of research is problematic, due to something called the “file-drawer problem.” The idea is simple: maybe the published studies were cherry-picked, leaving the null results to gather dust in a proverbial file drawer.
As we can see, there have been a great many people carrying out parapsychological research. But if studies demonstrate anything at all, it is something statistical rather than deterministic. Some subjects, on some trials, would score a successful hit; other times they would miss. Entire studies might or might not even meet the threshold for statistical significance. Many studies could very well exist which failed to demonstrate the existence of paranormal phenomena, and so were never published.
On the one hand, it’s easy to dismiss this argument as every bit an appeal to the unseen as parapsychology itself. We imagine such studies exist. But we have no real evidence for them. Shouldn’t we prefer the simpler explanation which says no file drawer exists?
A skeptic could easily respond that a world without paranormal phenomena is (to them) simpler than one with such phenomena. But ultimately whatever explanation is simpler isn’t really the point—an effect which could be plausibly explained through conventional causes was not psi. The file-drawer problem had to be overcome before claiming results demonstrate the existence of a communications anomaly.
The best way to overcome such a problem was continued research. Fortunately, the classic experiments in card-guessing continued, now with more careful methodology. And they returned a few interesting results along the way:
The Sheep-Goat Effect
A consistent finding in parapsychology studies is for believers in psychic ability, termed “sheep,” to outscore unbelievers, termed “goats.” The first study to find this effect was carried out by Gertrude Schmeidler. In a series of studies beginning in 1943, she found that, while the sheep scored above chance, the goats actually scored below chance.24
This by itself was a significant advance—to my knowledge, this was the first point where parapsychologists were able to say something about psi, rather than just saying “psi exists (if you don’t mind the possibility for cheating, experimental errors, problems in experimental methodology, etc).”
But the sheep-goat effect was also interesting because of the way it eroded the plausibility for the most obvious problems in parapsychological studies.
If psi existed, the obvious explanation was that goats could sometimes unconsciously sense the correct answer to a question, but miscall the answer. But what if psi didn’t exist? In that case, the sheep-goat effect was hard to explain by pointing to problems at the level of the subjects themselves. Sensory leakage and subject cheating seemed like reasonable explanations for scoring above chance, sure. But were goats scoring below chance on purpose? And whether or not they were, wouldn’t eliminating goats from studies then boost the overall hit rate of the samples, making the psi effect even stronger? Even bumbling errors in scoring and record-keeping on the part of the researchers didn’t seem to provide a plausible explanation. From a skeptical standpoint, if people who didn’t believe in psi scored worse on parapsychological tests than people who did, that was really rather difficult to explain away.
The best skeptical explanation for the sheep-goat effect was probably just that there was no sheep-goat effect, which left file drawer problems or fraud as the most attractive explanations, at least at the time.
The Ganzfeld
Now move forward to the 1970s.
By this point, parapsychologists were starting to realize there was no such thing as being too careful. They could complain all they liked about unfairness and bias in skeptical reviewers, they could rail against double standards which shouldn’t exist in science, but complaining wasn’t going to be of any use. Ultimately they needed to either do their absolute best to obviate any possible criticisms throughout all stages of their experimentation, or else, resign themselves to impotence and irrelevance.
Studies had to be designed and executed so as to absolutely eradicate any possibility for sensory leakage and subject cheating. At every step of the way, safeguards had to be put in place to rule out all conventional explanations—sensory leakage, collusion, error, and fraud—for any communications anomaly which they might be lucky enough to capture in an experiment.
But there was also increasing confidence in the character of psi. Studies had begun to note that psi scoring seemed enhanced under altered states of consciousness like hypnosis.25 And evidence had continued to accumulate for the sheep-goat effect, spurring parapsychologists to claim that “statistically significant results on this general pattern have been obtained by about a half dozen experimenters in the United States, Europe, South America, and India.”26
Such large numbers of positive results began to give researchers some reason to doubt the file-drawer and fraud as an explanation for anomalous information transfer. The more positive results piled up, the more null results were needed in file drawers, or, the more researchers had to be committing fraud; the scale of the problem had to be larger.
There still remained, however, other possible confounds—sensory leakage, lack of standardization in experimental design, errors in data-handling, and subject cheating. Few provisions had been made by that time to deal with the possibility of deliberate cheating, particularly by a small percentage of skilled conjurers and stage magicians. Likewise problematic was the idea that certain procedural designs may be more open to small-scale fraud than others. Even a tiny fraction of false positives could have been enough to skew statistical results from null to positive. But even worse is the fact that, with nonstandard design and analysis procedures, study authors can “shop” among different measures of performance, and report only that one which is most congenial to their hypotheses—something well enough known today as p-hacking or data dredging.
So in retrospect, it seems the time was ripe for a change in the field. And, noting that believing respondents in states of hypnosis tended to score well, parapsychologists began to think that the key was getting selected subjects into as suggestive a state as possible, in order to enhance the psi effect and improve experimental clarity. By the start of the 1970’s, a new type of study put this line of reasoning to the test: The ganzfeld.
As described by the parapsychological association, the
[g]anzfeld (“whole field”) technique was developed to quiet… external noise by providing a mild, unpatterned sensory field to mask the noise of the outside world. In the typical ganzfeld experiment, the telepathic “sender” and “receiver” are isolated, the receiver is put into the ganzfeld state, and the sender is shown a video clip or still picture and asked to mentally send that image to the receiver.
The receiver, while in the ganzfeld state, was asked to continuously report aloud all mental processes, including images, thoughts, and feelings. At the end of the sending period, typically about 20 to 40 minutes in length, the receiver was taken out of the ganzfeld, and shown four images or videos, one of which is the true target and three are non-target decoys. The receiver attempted to select the true target, using perceptions experienced during the ganzfeld state as clues to what the mentally “sent” image or video might have been. With no telepathy, chance expectation allowed researchers to predict that the correct target would be selected about 1 in 4 times, for a 25% “hit rate.”
Ganzfeld protocols were not only attractive because they made use of computers for randomization, completely isolated the receiver from any possible sensory leakage, and removed the chance for subjects to cheat. They were attractive precisely because all of this made them expensive and time consuming to conduct.
A single ganzfeld session is orders of magnitude more difficult than a typical test of psi where one (or more!) subjects show up off the street and guess cards or numbers in rapid succession. Rather than being able to obtain a hit or miss in a matter of seconds, half an hour and more time in a carefully controlled environment was required for a single guess. And this evident weakness contained a positive side effect—the effort required to carry out such studies made it difficult for ganzfeld experiments to be undertaken quietly and discarded easily. This dramatically reduced the problem of the file-drawer by itself.
But the file-drawer question was nevertheless investigated in a meta-analysis of 28 studies for which a uniform measure of significance (direct hits) could be applied: a powerfully significant effect was found (z = 6.6, p < 10–9) requiring 15 unpublished null-result studies sitting in file-drawers for each of the 28 under consideration to nullify the result.27
The Honorton-Hyman Protocols
By 1986, the two figures emerged as the most active debaters in the field. The first was Charles Honorton, promoting belief in the psi effect, with skeptic Ray Hyman on the other side.
Honorton is probably better remembered by parapsychologists, but the real progress is ultimately due to Ray Hyman. Honorton, after all, was a parapsychologist doing what parapsychologists did. Maybe he was better than most researchers. Maybe he was better at communicating with skeptics than other parapsychologists. Still, it’s pretty easy to guess that Charles Honorton was following through on his personal interests.
On the other hand, well, it doesn’t take much imagination to realize Ray Hyman’s friends were looking at him pretty strangely for hanging out with parapsychologists, or to guess that, on some level, Hyman really figured that the entire thing was just a complete waste of time. Hyman was a skeptic, and remained a skeptic long after his association with Honorton. But whether out of curiosity or common decency, he still took the parapsychologists’ efforts seriously enough to engage with them.
So in response to the new studies, and the problems with the older research, Hyman and Honorton published a joint agreement on standard protocols for parapsychological research. Their agreement included strict security precautions against sensory leakage, protocols on randomization, and full documentation of the analyses used with indication of whether tests were planned, or post hoc.28
Interestingly, both authors explicitly agreed that the cost and difficulty of ganzfeld research rendered the file-drawer explanation for any observed effect unlikely, and further agreed that the balance of the ganzfeld studies already in existence suggested there was some effect to be explained, even though a conventional explanation might still be found. Given the way parapsychological research has proceeded, the idea that the file-drawer effect, an important possible conventional explanation for the results, could be largely ruled out, already represented clear and unambiguous progress in the field of parapsychology.29
A furious spate of publications followed the 1986 standardization protocols of Hyman and Honorton. The ponderous rate of data collection under ganzfeld protocols meant that information built up very slowly, but there was a lot of excitement, and people started publishing meta-analyses prematurely. To get a sense of what I mean by “prematurely,” the first meta-analysis in 1994 contained only 11 studies complying with the accepted guidelines. They found an overall hit rate of 32% (p = .002)30 but a later meta-analysis in 1999, including 30 studies, reduced the rate to nonsignificance (p = .24) and reported a significant declining effect size with study year, suggesting the possibility for fraud in the earlier studies.31 Rebuttals showing a significant effect made use of studies which were criticized as being of poor quality and probably shouldn’t have been included.32 By 2010, an analysis of 29 studies published after 1997 revisions found a fairly strong signal (p < 10–7).33
All of this was hotly contested on technical grounds. But rather than going into these details, I want to point out the location where these last meta-analytical studies on the ganzfeld experiments are published. Earlier research is generally found where we might expect, in the Journal of Parapsychology, the Journal of the Society for Psychical Research, and similar parapsychological venues which most academics regarded as controversial, or even slightly embarrassing for the scientific community as a whole. But check the last four references in the previous paragraph: these studies and communications mentioned above are all found in the classic journal of psychology, Psychological Bulletin.
Now, my readers may not hold psychology in especially high regard. And whether parapsychologists had indeed eliminated conventional explanations for the above-chance results obtained in their studies—whether they really demonstrated psi or not—may have still been open to debate at that time. But the positive results obtained over hundreds of independent experiments across the globe, combined with fact that the groundbreaking parapsychological studies of the last few decades have been published in the pages of mainstream scientific journals, throws a harsh light on claims that parapsychology is a pseudoscience, or has made, “literally,” no progress over the last 150 years.
After 2012
After weighing all the research above in 2012, I was pretty well convinced that the psi effect was real. There were plenty of caveats, of course, the worst of these being that I’d never been able to carry out a serious replication of my own. I made an effort around 2016, but my experimental protocols were rough, and skeptics in my subject pool refused even to comply with those. I dislike being in the position where I have to take other people’s word for things, and I vastly prefer to hedge rather than commit to believing something I haven’t thoroughly investigated myself, but I haven’t been able to properly investigate any of this on my own. Sometimes things are this way, and I can only really defer to what other people said they’ve done.
But evidently there was a great deal that went on over the last eleven years to further strengthen the parapsychologists’ position. Two items are of particular interest:
Continued research into the sheep-goat effect, and
Something totally new, a completely new way of investigating the psi effect, relating to subjects’ ability to predict the future.
1. The sheep-goat effect, revisited
At the time the sheep-goat effect had been discovered in the middle of the 20th century, it was only supported by a few isolated studies by Gertrude Schmeidler. By the 1990s, a meta-analysis by Tony Lawrence 34 covering 73 experiments by 37 different researchers, found a strongly significant difference between the hit rates of sheep and goats (z = 8.17, p < 10-15). The number of unpublished studies returning a null result which would be needed to reduce this finding below significance was 1726, which wasn’t a bad argument against the file drawer.35
This was nice so far as it went, but it would really have been more convincing to have ganzfeld studies showing the sheep-goat effect. And a meta-analysis came out in 2017 reporting findings that were “generally comparable to Lawrence's, altogether indicating a ‘belief-moderated communications anomaly’ in the forced-choice ESP domain that has been effectively uninterrupted and consistent for almost 70 years.”36
So what we have is not only evidence that the sheep-goat effect is found on high-quality studies, but also that the early studies were every bit as good as the ganzfeld. In other words, now that psi has been pretty well established as a genuine phenomenon, it doesn’t really look like a researcher needs to use ganzfeld protocols except as a safeguard against criticisms about the file-drawer.
2. Feeling the Future
But a more striking confirmation of the existence of psi comes from emotional reactions to stimuli seen before the stimulus appears. The best known researcher in this vein is probably Dean Radin, a Baby Boomer with training in electrical engineering and educational psychology.
I had never followed these studies in previous years, but they open up a completely different research paradigm. The idea is simple: Subjects sit before a screen with sensors reading their skin conductance level (SCL). The screen is blank; then, an emotionally arousing image is shown (often either violent or erotic), or a bland control image. But studies find that the subject’s skin conductance level begins responding a few seconds before the image is shown. The following is an example from Radin’s work.37
The nature of these experiments seems extremely convenient to the study of psi, since the readings are all involuntary and recorded in real time. In 2011 there was already a report of nine experiments showing psi in this way;38 while we’ve seen that early meta-analyses can be premature, a 2015 meta-analysis had “90 experiments from 33 laboratories in 14 countries which yielded an overall effect greater than 6 sigma, z = 6.40, p = 1.2 × 10-10” and was, evidently, the last straw for skeptics, some of whom appeared to simply give up and lose all trust in science when they found out that reality doesn’t conform to their ideas about what is rational.39
Conclusion
As it psi usually defined, it will probably never be technically possible to demonstrate that anything must have truly been an example of psi. After all, maybe we’re all very confused. Maybe the entire world we think we’re living in is an elaborate hoax, like from They Live. Maybe parapsychologists actually get paid way more than we might imagine, as part of a nefarious plan carried out by the Illuminati. Maybe we’re just really unlucky, and by the sheerest coincidence, and all of these studies returned significant results even though the possibility for that is astronomically improbable. These explanations may not be reasonable, and they may not be even remotely conventional in the sense of what most people believe, but they don’t challenge conventional notions of the way information can and cannot be transferred. Will it ever really be possible to address them all?
Ultimately, I think a sensible person shouldn’t have to. The evidence is so strong at this point that I don’t object to the psi effect except on grounds that
I’ve learned it’s safer to hedge on controversial matters, particularly when I’m in the minority, and
I haven’t personally investigated it.
But I haven’t investigated the theory of relativity either, except on mathematical grounds, and I have no real doubt about that. And we’re all probably in the minority in believing in a round earth that revolves around the sun, given the numerous generations who came before us with no idea about even what lay on the other side of the nearest mountain range.
If we were forbidden from tentatively agreeing with anything that wasn’t socially approved or personally verified, would most of us even be able to speak about the deleterious effects of inbreeding, or the dangers of hard drugs? I’ve never personally known anyone who used heroin, cocaine, or methamphetamines; I’ve never personally known anyone who married her brother. Yes, having first-hand experience is very helpful to providing strong and convincing evidence for things, but frankly I don’t feel the need to marry my sister and develop a heroine addiction in order to investigate the biological downsides everyone is always talking about. Sometimes it really is worth it to trust the evidence scientists are giving you.
So rather than being concerned about whether we can really trust the scientists (when we do trust them in general on most everything else), I think it’s much more useful to ask: How might we make any sense of all of this? As I’ve detailed in my previous post, there are many good ways.
To be totally fair, however, the issue of fraud does remain somewhat troubling. I don’t personally think fraud is a good explanation for these communications anomalies appearing in the research. But it is still at least a plausible explanation for the anomalies appearing in the research.
Still, I must also admit that fraud is an issue in every scientific field. Indeed, motivation to commit fraud in a low-prestige, minimally funded area like parapsychology is much less than in other areas such as medicine which generate strong reputational and financial incentives to falsify results. And the same remedies for fraud applied throughout other fields—multiple experimenters per study, verification across independent labs, strict experimental procedures, and sustained criticism over time—do apply here, giving reason to doubt fraud as an explanation.
At this point, given the vast extent of the results, what fraud requires is for many, many researchers to have been leaning on the experimental scales across a dozen countries and across decades of time. Is fraud plausible on such a scale? And doesn’t the assumption of fraud here prove too much? Pick any scientific result you like, any field that makes you slightly uncomfortable, and just accuse everyone in it of being a pack of frauds. Mathematicians are frauds! The doctors who say you can’t lose weight eating doughnuts and watching YouTube are frauds! Believe whatever you want with zero accountability! It’s very freeing.
For everyone else who isn’t so wedded to a specific perspective of reality that it can’t accommodate the psi effect, the exciting question is, what’s next? What might such a communications anomaly mean? And what is the underlying cause for the psi effect? To date, there is no one clear framework for explaining it. While this may be dissatisfying, it’s actually quite typical of science.
An example may be helpful here. Clearly, we know that gravity exists. Yet gravity is an extremely weak effect, and only becomes measurable with very large masses, such as the mass of the earth which pulls us toward it. Moreover, we have no scientifically settled explanation for why gravity exists. Einsteinian relativity proposes gravity is caused by the bending of spacetime near massive objects, while quantum theory postulates the existence of gravitons—tiny particles emitted from massive objects which pull other objects in the direction from which the gravitons came. Some support for the former explanation comes from its ability to explain Mercury’s orbit (which deviates slightly from Newtonian predictions). But even here, the explanation appears to be a redefinition of gravity rather than an explanation for it—it says gravity is the warping of spacetime, but does not explain how matter is able to cause this warp. For now, it seems we must content ourselves with using gravity, without knowing how it works.
Perhaps, in years to come, similar things will be said of the psi effect. The successes of recent research should be provocative enough to attract the interest of many independent researchers, and, hopefully, much needed funding for further experimentation. A day may even be coming when psi grows to be so popular that people start to blame everything around them on psi; this would be a scientific tragedy, if it meant the quality of research in the field dropped back to that of the late 19th century. But in the interim, the simple but fascinating mystery of psi cries out for science to explore and understand it.
All in all, it’s not a bad time to be a parapsychologist.
Mousseau, M. C. (2003). Parapsychology: Science or pseudo-science. Journal of Scientific Exploration, 17(2), 271-282.
Cordón, Louis (2005). Popular Psychology: An Encyclopedia. Greenwood.
Reber, A. S., & Alcock, J. E. (2020). Searching for the impossible: Parapsychology’s elusive quest. American Psychologist, 75(3), 391–399.
Ibid.
Parapsychology | psychokinesis, telepathy, ESP. (n.d.). Britannica. Retrieved November 7, 2023, from https://www.britannica.com/topic/parapsychology
Richet, C. (1923). Thirty years of psychical research: Being a treatise on metapsychics. Macmillan.
Münsterberg, H. (1914). The intellectual underworld.
Podmore, F. (1895). What Psychical Research Has Accomplished. The North American Review, 160(460), 331-344.
Podmore, F. (1895). What Psychical Research Has Accomplished. The North American Review, 160(460), 331-344.
Duke University Library Exhibits | About the Exhibit · Early Studies in Parapsychology at Duke. (n.d.). Online Exhibits. Retrieved November 7, 2023, from https://exhibits.library.duke.edu/exhibits/show/parapsychology/about-the-exhibit
Reeves, M. P. (1944). Tyrrell's Study of Apparitions: A Review. The Journal of Parapsychology, 8(1), 64.
LA, D. (1946). Spontaneous experiences reported by a group of experimental subjects. The Journal of the American Society for Psychical Research, 40, 55-93.
Reeves, M. P. (1944). Tyrrell's Study of Apparitions: A Review. The Journal of Parapsychology, 8(1), 64.
McVaugh, M., & Mauskopf, S. H. (1976). JB Rhine's extra-sensory perception and its background in psychical research. Isis, 67(2), 161-189.
Kennedy, J. L. (1939). Experiments on the nature of extra-sensory perception. The Journal of Parapsychology, 3(2), 194.
McDougall, W. (1934). Psychical research as a university study. In W. McDougall, Religion and the sciences of life, with other essays on allied topics (pp. 64–81). Methuen & Co.
Garland, H. (1936). Forty years of psychic research: a plain narrative of fact. Macmillan.
Pratt, J. G., Rhine, J. B., Smith, B. M., Stuart, C. E., & Greenwood, J. A. (1940/1966). Extra-sensory perception after sixty years. Boston: Bruce Humphries.
Bösch, H. (2004). “Reanalyzing a meta-analysis on extra-sensory perception dating from 1940, the first comprehensive meta-analysis in the history of science.” In S. Schmidt (Ed.), Proceedings of the 47th Annual Convention of the Parapsychological Association, University of Vienna, (pp. 1–13).
For example, Harold Gulliksen. (1938). Extra-Sensory Perception: What Is It?. American Journal of Sociology. Vol. 43, No. 4. pp. 623-634.
Honorton, Charles (1993) “Rhetoric over Substance: The Impoverished State of Skepticism.” The Journal of Parapsychology, 57.
Honorton, C. (1975). “Error some place!” Journal of Communication, 25, 103-116.
Honorton, Charles (1993) “Rhetoric over Substance: The Impoverished State of Skepticism.” The Journal of Parapsychology, 57.
Schmeidler, G. R. (1943). Predicting good and bad scores in a clairvoyance experiment: a preliminary report. Journal of the American Society for Psychical Research.
Honorton, C., & Krippner, S. (1969). Hypnosis and ESP performance: A review of the experimental literature. Journal of the American Society for Psychical Research, 63(3), 214–252.
Dean, E. D. (1970). Techniques and Status of Modern Parapsychology. Science, 170(3963), 1237-1238.
Honorton, Charles (1985). Meta-analysis of psi ganzfeld research: A response to Hyman. Journal of Parapsychology, 49, 51-91.
Hyman, Ray and Honorton, Charles (1986). “A Joint Communiqué: The Psi Ganzfeld Controversy.” Journal of Parapsychology, 50, 351-364.
Ibid.
Bem, Daryl J. and Honorton, Charles (1994). “Does psi exist? Replicable evidence for an anomalous process of information transfer.” Psychological Bulletin, 115, 1: 4–18.
Milton, Julie and Wiseman, Richard (1999). “Does Psi exist? Lack of replication of an anomalous process of information transfer.” Psychological Bulletin, 125, 4: 387-91.
Milton, Julie and Wiseman, Richard (2001). “Does Psi exist? Reply to Strom and Ertel (2001).” Psychological Bulletin, 127, 3: 434-438.
Strom, L., Tressoldi, P.E., and Di Risio, L. (2010). “Meta Analysis of Free Response Studies, 1992-2008: Assessing the Noise Reduction Model in Parapsychology.” Psychological Bulletin, 136, 4, 471-485.
Lawrence, T. R. (1993). “Gathering in the Sheep and Goats. A Meta-Analysis of Forced-Choice Sheep-Goat ESP Studies, 1947-1993.” In Schlitz, M.J. The Parapsychological Association 36th Annual Convention. Proceedings of Presented Papers (pp. 75-86). Toronto.
Lawrence, T. R. (1993). “Gathering in the Sheep and Goats. A Meta-Analysis of Forced-Choice Sheep-Goat ESP Studies, 1947-1993.” In Schlitz, M.J. The Parapsychological Association 36th Annual Convention. Proceedings of Presented Papers (pp. 75-86). Toronto.
Storm, L., & Tressoldi, P. E. (2017). GATHERING IN MORE SHEEP AND GOATS: A META-ANALYSIS OF FORCED-CHOICE SHEEP-GOAT ESP STUDIES, 1994-2015. Journal of the Society for Psychical Research, 81(2).
Radin, D. (2003). Time-reversed human experience: Experimental evidence and implications. Journal of Nonlocality and Remote Mental Interactions, 2, 256-279.
Bem, D. J. (2011). Feeling the future: experimental evidence for anomalous retroactive influences on cognition and affect. Journal of personality and social psychology, 100(3), 407.
Bem, D., Tressoldi, P., Rabeyron, T., & Duggan, M. (2015). Feeling the future: A meta-analysis of 90 experiments on the anomalous anticipation of random future events. F1000Research, 4.
Another substacker ("Roger’s Bacon") writing about similar things -
https://www.secretorum.life/p/the-most-dangerous-idea
https://www.thestudiesshowpod.com/p/episode-15-halloween-special-on-parapsychology