This document provides an introduction to the subject of retropsychokinesis, a historical review of the evidence from previous experiments, and an overview of the on-line retropsychokinesis experiments underway at this site. This is an adaptation and elaboration, by John Walker, of the original proposal for Web-based retropsychokinesis experiments by Matthew R. Watkins. Text from the original proposal appears in dark brown; new material describing the current implementation of the experiments is set in black. Some minor editing of text quoted from the proposal has been done to better describe the present state of the experiments.
A complete stranger hands you a sealed envelope and asks you to choose an number between one and fifty. A bit puzzled, perhaps, you think for a moment and announce “twenty-eight”. The stranger scribbles this in a notebook, tells you to open the envelope in two weeks, smiles, and disappears.
Two weeks later you open the envelope to find a piece of paper with “28” printed neatly in the centre. Your mind swims with possible explanations, including the possibility that it was merely a coincidence. But a few days later, the stranger reappears with another envelope, you choose another number, and the sequence of events repeats. How many times would this have to occur before you accepted that something very strange indeed was going on?
The idea of the Project is to make use of the Internet in order to explore the purported anomalous effect known as retropsychokinesis (from now on we shall refer to this as “RPK”). The existence of such an effect has such profound philosophical implications that, despite repeated and well-regulated demonstrations carried out for over 20 years, it has remained an obscure matter of parapsychological controversy. However, the emergence of the World-Wide Web has created an exciting new possibility.
Although attempts to create “online” interactive parapsychological experiments have already appeared on the WWW, these are in early stages of development and published results have not been extensive. These experiments are increasing awareness of the claims and methods of parapsychology research. But the collection of data for serious research purposes is obviously restricted, as subjects cannot be supervised, and the lack of control in the experiments jeopardises the credibility of any results obtained.
However, the “retrocausal” or reverse-time nature of RPK is such that problems of this nature can generally be overcome. The proposed experiment(s) would, in fact, bypass most of the usual obstacles which occur in parapsychological research. These include attracting and motivating appropriate subjects, the limitations on the number of subjects which can be tested in any reasonable length of time, the elimination of all possible fraud, and the difficulties subjects face in performing in unfamiliar laboratory settings or in the presence of sceptical observers. The difficulties in publicising and gaining acceptance for the results obtained has perhaps been the most significant obstacle. Experiments which yield significant results have generally been accepted by the “believers” and rejected by the sceptics as insufficiently well-regulated (a claim which is often justified, but which can never be overcome in the existing research format). However, this too could change, as we shall see.
The more generally-defined effect known as psychokinesis (PK) has been studied extensively since the 1930's when J.B. Rhine (Duke University) began systematically testing claims that seemingly random events such as dice and coin throws are subject to subtle psychic influences. His successor, Helmut Schmidt, increased the rigour and efficiency of these experiments by introducing an electronic random number generator which used subatomic decay processes in order to generate data of the most fundamentally “random” nature. One must keep in mind that the issue of randomness is a difficult one—the interpretation of probabilities, the seeming effectiveness of statistical methods, and the fundamental role which probabilities play in quantum mechanical models of objective reality are all problems closely related to this work. Schmidt went on to run experiments similar to the original ones (PK effects on electronically generated random data), the only difference being that the data were prerecorded (and, importantly, unobserved), rather than generated in “real time” as the experiments were carried out. Despite being extraordinarily counterintuitive, the results suggested strongly that unobserved random events which occurred in the past are subject to psychokinetic influence—in other words, the human mind can in some (limited) sense “influence” or at least “select” the past.
Schmidt and others have spent many years repeating the experiments, refining the techniques, and gathering valuable data, despite the general lack of public awareness and academic acceptance which they have received. Various acausal models of reality, and appropriate modifications of quantum theory have been suggested in order to account for the phenomenon, yet many fundamental questions regarding the nature of time, consciousness and causality remain largely unanswered.
The experiments underway at this site involve subjects attempting to influence random data aided by a visual display appearing on their computer screen. One of a variety of display programs, chosen by the subject, is sent to the subject from the Project, accompanied with (unobserved) a randomly-generated data file containing a stream of bits from the Fourmilab HotBits radioactive random number generator. Before the data stream is transmitted to the subject, a secure copy is made in a log file on the www.fourmilab.ch server, which also records the subject's identity and which feedback program was chosen and the parameters the user selected for it. Since we are attempting to detect influence on already recorded data, there is no need to rely on the honesty of the subject. An on-line database retrieval program allows subjects to review their own results and anybody on the Web to produce reports of all experiments conducted by subjects around the globe (subjects' identities are obscured in these reports to protect their privacy). Researchers who wish to verify the analysis of the raw experimental data with programs of their own can download it in verbatim form, including the entire random data file sent to each subject.
The visual feedback programs are written in Java and run locally on the subject's computer as “applets” within the RPK experiment Web page. This allows the feedback programs to provide smooth animation regardless of the speed of the subject's Internet connection. The feedback programs exploit the object oriented features of the Java language to allow implementation of other forms of feedback without extensive programming. Developers are encouraged to create new feedback programs and make them available on this or other sites.
If significant results are obtained by one or more subjects in these experiments, it is essential that other independent investigators verify the validity of the experiments running at this site and independently repeat the experiments themselves. All of the software components used in the experiments have been placed in the public domain and may be downloaded, by anybody, with complete source code, from this site. Other sites which conduct their own experiments using these programs are also encouraged to establish their own hardware-based random number generators, using a variety of technologies, to preclude potential defects in our HotBits generator as a source of error.
Our experience with the Internet suggests that a huge number of potential subjects could quite easily be reached in a relatively short time. The fact that they will be free to perform in their own environment, rather than in a (possibly distracting) laboratory, may make a significant difference. Furthermore, the almost unlimited numbers of people who could potentially participate in the experiment simultaneously, without the need for careful supervision, should speed the research process up considerably. Dr. Schmidt has cautioned us, however, that the key to getting results in these experiments may be the effort we make to “maintain some semblance of personal contact between subject and experimenter”. Also, Charles Tart, the well-known U.C. Davis consciousness researcher has suggested that “there is an important experimenter effect in all psi research; some people have the ‘magic touch’ and regularly get results, others don't and we have little idea as to why.”
Our making everything needed to conduct these experiments freely available on the Web will permit a variety of researchers to perform these standardised experiments on subjects of their own choosing, at this site or on Web sites of their own.
The existence of this effect (if in fact it does exist) raises some very deep questions concerning the nature of time, the relationship between consciousness and objective material reality, the concept of causality, and the concept of randomness. The much misunderstood “multiple (or parallel) universes” interpretation of quantum mechanical phenomena has been suggested as part of a model which encompasses the RPK phenomenon. This in itself raises many important questions. The idea of “will” is certainly related, as this is the best existing description of that which the subject uses in order to exert an influence. The will's influence has been discussed in some details by the eminent parapsychologist J. Beloff in an article on teleology vs. mechanism.
Many peripheral issues are also worthy of consideration. For example, there is the reliability of experimental data in particularly sensitive areas of subatomic physics (largely based on subtle statistical assumptions, and therefore possibly subject to PK or RPK influences). Perhaps more controversially one might want to contemplate the supposedly “random” nature of the genetic mutations which are axiomatic in the Darwinian model of biological evolution.
Perhaps most profoundly, the very idea of “similar” measurable events, which constitutes the starting point of all theories of probability and statistics (keeping in mind that all descriptions of the physical universe are inherently statistical), and the (related) ubiquity of certain statistical distributions in the physical world, particularly the Gaussian (or normal) distribution, may need serious re-evaluation. For if the phenomenon is real, it suggests that such distributions can effectively be warped through the exertion of poorly understood psychic faculties. This then suggests that the perception of the physical universe as proceeding according to statistically regular (i.e. “normal”, in the truest sense) patterns may be linked to some ongoing, collective, and/or self-reinforcing psychic projection or interaction.
Finally, the purported results can be very simply adapted to demonstrate what can only be interpreted as an instantaneous transmission of information. This violates the speed-of-light limit set by Einstein's Theory of Relativity, and is related to the Einstein-Podolsky-Rosen paradox and the concept of quantum nonlocality. This in itself, we believe, is sufficiently important to justify the experiments.
The scenario described at the beginning of this document, involving strangers with envelopes, numbers printed on pieces of paper, etc. was only intended as a metaphor. However, we aim to determine if equivalently “impossible” things can happen, using the slightly less “physical” medium of the Web. But it should be pointed out that if RPK really exists (and the best existing database suggests that the odds are in the order of 1 in 630 thousand million that the experimental evidence is the result of chance), then such a stunt could theoretically be carried out. The number in the envelope would be the “compression” of a large block of unobserved random numbers, calculated and printed via computer, and sealed in the envelope without being observed. After a number was chosen, the “stranger” would “unravel” it according to a certain algorithm, and set up a series of RPK trials with talented subjects who would in effect be “encoding” subtle biases into the original block of random data. This seemingly “retrocausal” action should result in the printed number “having been” the number which you (at a later time) selected.