Debunking the Mysterious: The Ganzfeld and the Paranormal

Parapsychologists assert that they've found evidence for ESP through an innovative technique called the Ganzfeld.
Their results may be winning some converts.

by William Dalrymple

    He sees red; light from a red incandescent bulb shines through the two ping-pong ball halves he wears over his eyes. He hears static; 92 decibels of white noise flows through the headphones he wears in the soundproofed room. He feels almost weightless. Reclining in a Barca-Lounger, he rests comfortably in equilibrium with the chair. He thinks nothing; fifteen minutes of relaxation exercises have slowed down his mind and quieted his thinking. He is the receiver, and his senses calmed, he speaks vivid pictures.

    Ferris wheel, carriage wheel. Spokes .... Train entering a tunnel. Boxcars .... A submarine or a shark or a zeppelin, dirigible .... An eye like the window in Captain Nemo's submarine .... There are a lot of cartoon animals doing things .... A lot of cartoon images. Cartoon fish with sunglasses....

     Three doors away in room G75A, one of the many cubicle-like rooms in Uris Hall's social psychology lab, a sender watches a computer-selected video clip on a TV. Graduate student and researcher David Sherman describes the excerpt the sender sees. "The clip is called 'telepathic fish'; it's taken from The Hitchhiker's Guide to the Galaxy [television series]. It shows a miniature cartoon fish being placed into a person's ear. The fish's eyes are very prominent."

    The receiver's comments eerily clash with current cognitive psychology; he could not have seen what the sender saw _ and what he seemed to describe _ through any mechanism that science accepts. Later in the experiment, this receiver makes the correct match between his thoughts and the picture. The researchers score the transmission as a "hit."

    This experimental technique, called the Ganzfeld - a German term translated as "whole field" - creates a setting that lowers the volume of internal noise originating from thought, sight, and audition so that the receiver can pick up any weaker signals. The receivers select the "sent" visual target from three other red herrings, by chance a one-quarter hit rate.

    Critics and scientists in the field and parapsychologists have come to an accord, states parapsychologist Charles Honorton. Both agree that receivers hit their targets more often than chance would allow in Ganzfeld experiments, but they disagree whether the results prove the existence of psi. Cornell Psychology Professor and parapsychologist Daryl Bem defines psi as "anomalous processes of information or energy transfer. By anomalous, I mean processes that remain unexplained by currently known physical and biological mechanisms." Honorton adds that critics "...no longer claim to have demonstrated a relationship between methodological flaws [how tightly controlled the experiment is] and study outcomes [positive results]." In the past, critics had argued that as controls increased, psi results withered.

    Techniques like the Ganzfeld comprise a major part of the controversial field of parapsychology, which claims to explain a small fraction of yet a larger whole; the paranormal. By definition, the realm of the paranormal is everything the normal and well-explained is not; it includes UFO sightings, extra-sensory perception, psychokinesis (moving objects through non-normal ways), auras, astral projection, ghosts, magic, psychics, reincarnation, and precognition (knowing of events before they happen).

    Parapsychology officially began in February 1882, when a London group of believers, spiritualists, and scientists - refugees from the 'Darwinian Revolution,' which replaced the powers of God with the processes of evolution - founded the Society for Psychical Research (SPR). In its first years, the Society took on thought-transference (today psi), hypnosis, haunted houses, and mediums, among other studies. Psychologist William James was an early and active member in the organization.

    Many of the SPR's first investigations were tainted with fraud; critic C. E. M. Hansel interprets the history of the paranormal as a history of deception. "To the skeptic, psychical research seems as much a history of how the artful can mislead the innocent as it is a reflection of any more esoteric activity," he wrote. Trickery was especially prevalent in early tests of mediums and psychics. One particularly dexterous medium, Eusapia Palladino, could reach behind her body with her foot, kicking cabinets and knocking small tables over, all while held sitting on a chair by two men, with her feet supposedly planted.

Repetition and Experimental Problems
    Science will eventually overcome a cheat or a mistake through independent replication. Repeating other scientists' work plays a key part of accepting a phenomenon as genuine. A real phenomenon will continue to occur regardless of the "experimenter effect" - where the results seem to correlate with peculiarities of the experimenter conducting the study, and not with the experimental variables. "If other people, using the [the experimenter's] methodology, can independently produce the same results, it is much more likely that error and dishonesty are not responsible for them," critic James Alcock wrote.

    The Achilles' heel of parapsychology has been replication, and the experimenter effect rests at its core. The effect is significant because of the field's lack of diversity: only a handful of scientists study parapsychology; they are concentrated in a few large labs, and most of them believe in psi anyway. Replication of psi related experiments loses validity because of the homogeneity of the experimenters.

    If the experimenter effect does not invalidate the results of experimental parapsychology, cheating might. Cheating consists of intentional transfer via trickery or through unconscious subtle cues, both of which scientists describe as "sensory leakage." Because there is no theoretical understanding of ESP transmission, and because scientists generally do not accept it, everything in such experiments must be controlled so that subjects do not receive information through normal channels. So "psi ability" has become a "trash can" hypothesis that scientists attempt to isolate only by process of elimination. According to Alcock, "[w]e can never prove that there are not sensory channels of which we are not aware." Every small logical or procedural flaw can invalidate the data in these so-called alternative (as opposed to null) hypotheses.

    In addition, even if every single "normal" sensory channel were eliminated, the results could just be the vagaries of chance. Since Ganzfeld experimenters rely on their base-chance rate of 25%, they are vulnerable to chance fluctuations because they idealize the chance rate. Although the average over an infinite number of trials should work out to be exactly one in four, who is to say chance did not deviate slightly - or a lot - from the norm for a given experimental trial? Alcock wrote, "It is impossible to tell which of the various hits are the 'extra-chance' ones (due to ESP) and which are just 'chance.'"

The Ganzfeld Up Close
    Parapsychologists adapted the Ganzfeld method from Gestalt psychologists in the 1970s to generate a more psi-friendly experimental technique. "Historically, psi has often been associated with meditation, hypnosis, dreaming, and other naturally occurring or deliberately induced altered states of consciousness," Bem said. The experiments themselves are meticulous and time-consuming. One Ganzfeld session takes an hour and a half. "It's very slow; four people have to be involved," Bem said. "If you run more than two sessions a day, it's pushing it, everyone gets tired and unexcited and it doesn't work," he said.

    The Ganzfeld suffers from other problems as well.

    What is social atmosphere? Bem and Honorton did create an unfalsifiable hypothesis in their attempt to help psi ability by providing a sense of warmth and friendliness in their experiments. They call the social atmosphere "a critical determinant of [psi experiments'] success or failure." But they could not quantify all the dimensions of the social atmosphere. The danger comes when a replication experiment found no evidence of psi; the experimenters could always attribute the failure to a deficient social environment. "The best the original researchers can do," they respond, "is to communicate as complete a knowledge of the experimental conditions as possible in an attempt to anticipate some of the relevant moderating variables."

    Critic Ray Hyman questioned the randomization of the targets in one of the sets of Ganzfeld data, arguing that guesses could become lucky more often than chance. With non-random targets, chance deviates more often from the one-in-four statistical baseline. But according to Bem, incomplete randomization would not affect the results because the Ganzfeld has such a large target pool and because each subject usually acts as a receiver for only one session. But "the possibility remains open that an unequal distribution of targets could interact with receivers' content preferences to produce artifactually high hit rates [that is, through the peculiarities of the experiment]," Bem said.

    At Cornell, Bem and his research assistants tested effects of belief and non-belief on 50 Cornell student receivers. Although the receivers who reported high belief did not achieve significance over chance, they were significantly superior to the non-believer sample. "The striking thing is that non-believers hit 8%, in other words, significantly below chance [25%] which means that they must have some ESP ability that [makes their score] differ from chance," Bem said. Believers, on the other hand, had a 34% hit rate in choosing the correct target video.

    In order to compile data from studies like these, social scientists use a technique called meta-analysis to determine if many studies together result in meaningful data. Meta-analysis standardizes the results from a series of similar experiments, weighting better-quality ones based on preset criteria. Honorton's meta-analysis of 28 Ganzfeld studies found an average hit rate of 35%, fully 10 points above chance, with a margin of error of plus or minus eight percentage points. Honorton's analysis of 350 sessions of a variant of the Ganzfeld methodology yielded a hit rate of 32%, with a margin of error of plus or minus three percentage points.

    In these Ganzfeld experiments, parapsychologists perhaps finally have significant results. However, caution is certainly needed; parapsychology has made claims of evidence to overthrow the establishment since 1882. It hasn't yet. Until these studies have been better replicated, the experimental evidence does not support a paradigm shift. They do, however, warrant increased interest by other skeptical psychologists.

The Problem with Psi
    Parapsychologists investigating psi have the difficult task of demonstrating the existence of a process deemed impossible by science today. The proof of psi would require flawless experimentation and analysis. Unfortunately, the work done by today's researchers investigating psi is unconvincing. In their analysis, Bem and other parapsychologists use a process called meta-analysis to standardize results from different studies. In this process, the results from experiments that met certain criteria are weighed more heavily than the results of experiments that did not meet these criteria. How can the data from less stringently controlled or poorly performed experiments be given any weight in scientific analysis? The use of meta analysis also points to the fact that there are simply not enough studies of one type that have been performed. Instead, the results from different studies are collated together to quantify the amount of psi activity in these experiments. In addition, parapsychologists admit that factors such as "social atmosphere" are critical to the results. Such vague and imprecise definitions of experiments make them difficult to replicate in large numbers by a diverse group of scientists. Thus, while today's parapsychologists have made progress over many of the cheats and frauds of the past, there still remains a need for more rigorous experimentation and analysis in their work for it to be considered believable.
--V.J. & K.W.