There are claims that this experiment yields results that deviate significantly from randomness, and represent some of the strongest experimental evidence for psi phenomena to date. As with all purported psi phenomena, these claims are highly controversial within science in general and even within parapsychology. The debate is well-documented in parapsychology journals and in articles such as those referred to below.
The ganzfeld experiments are among the most recent in parapsychology for testing the existence of and affecting factors of telepathy, the ability to communicate information from one person's mind to that of another without resorting to normal means. After the initial success of the forced-choice card-guessing telepathy experiments of J. B. Rhine and his associates, researchers noted a "decline effect" where the accuracy of card guessing would decrease over time for a given subject, which some parapsychologists attributed to boredom. Some parapsychologists turned to free response experimental formats where the target was not limited to a small finite predetermined set of responses (e.g. zener cards), but rather could be any sort of picture, drawing, photograph, movie clip, piece of music, etc.
Recent surveys and anecdotal reports over the centuries had indicated that spontaneous psi phenomena are frequently reported in states other than normal consciousness, such as dreaming, hypnosis, meditative states, etc. In the mid-1970s, from these reports several parapsychologists, Charles Honorton, William Braud, and Adrian Parker, independently hypothesized that psi signals may in fact be available most of the time, but are masked for most people by all the cognitive or environmental "noise" that we commonly experience during the normal conscious state, while the noise is reduced in these altered states, allowing better access to the psi signals. The "ganzfeld" or "total (sensory) field" experiments they developed aim to reduce such noise through a mild form of sensory deprivation. This is done by covering the subject's eyes with a translucent mask of ping-pong ball halves, placing headphones over the subject's ears and playing a mild white noise through them, and having the subject relax on a couch or waterbed. Sometimes additional relaxation exercises are incorporated.
Once the subject has been immersed in the ganzfeld for some minutes, then the subject, acting as "receiver", is asked to access through psychic means some target, typically a picture or video clip selected randomly from a large pool which is being viewed in a remote location by another subject acting as "sender". The subject then verbalizes their impressions which are recorded and transcribed. Then the transcript and the actual target along with three other randomly chosen dummy targets are submitted in random order to independent judges who are asked to rank-order the four targets against the transcript for best match. A "hit" is recorded if the judge matches the transcript to the actual target (i.e. actual target ranked 1), otherwise a "miss" is recorded. Over a series of such trials, a random ranking (null hypothesis) would result in an expected hit ratio of 1 in 4, or 25%. In these experiments, the receiving subject, the judges, and the experimenter(s) working with them are kept blind as to what the actual targets are for any transcript until after judging is complete.
Over many such trials conducted in many labs around the world, the average hit rate has been closer to 35% rather than the 25% expected by null hypothesis, a result which is highly significant statistically. In 1982, Charles Honorton presented a meta-analysis of the studies to date at a meeting of the Parapsychological Association, concluding that this was sufficient to establish the existence of psi phenomena. Prominent critic and skeptic Ray Hyman disagreed. This led to separate meta-analyses by each of them published in 1985 and 1986 respectively in the Journal of Parapsychology covering 42 ganzfeld studies conducted in 10 labs around the world. Of the 28 studies that reported actual hit rates, 23 reported higher-than-chance hit rates, for a combined odds against chance of ten billion to one.
Honorton and Hyman's respective analyses agreed on some points, such as that something interesting was going on, and that selective reporting (the "file drawer problem") could not account for the results. However they disagreed on other points, most importantly as to whether the studies established the existence of psi phenomena. Specifically, Hyman pointed to potential design flaws in the experiments. This led to a joint communique between the two published in 1986 outlining more stringent standards for experimental protocols and their reporting and evaluation that they agreed should be met in order for a definitive statement could be made regarding the existence of psi phenomena in the ganzfeld.
In 1983 preceding the joint communique, Honorton and his colleagues, notably Rick Berger, designed a new series of largely automated experiments, now called the "autoganzfeld", that adhered to the new Honorton-Hyman guidelines. The experimental design was reviewed by two independent observers, both stage magicians who specialized in the simulation of psi phenomena, and they gave their stamp of approval to the security of the design. These experiments continued until 1989, when funding dried up for Honorton's lab. Over the six-year period, 11 series of tests totaling 354 sessions were conducted with a variety of subjects, with overall hit rate of 34% and odds against chance of 45000 to 1.
Since then, ganzfeld studies have continued at many labs, with meta-analyses published from time-to-time, both favorable and unfavorable to the psi hypothesis. One of the most recent such reviews is that of Bem, Broughton & Palmer cited below.
See also: Remote viewing