« Comparing Cabinets and Classes: Harding, Obama, and Trump | Main | Orbits in Strongly Curved Spacetime: HTML5 Update »

Friday, January 27, 2017

Retropsychokinesis Project: Twentieth Anniversary Update

The Retropsychokinesis Project (RPKP) celebrates its twentieth anniversary this month. Launched in January of 1997, the project may be the largest and longest-running parapsychology experiment in history, having recorded over 880,000 experiments performed by more than 34,000 volunteers around the globe. (Participants can designate experiments as “practice” or “for the record”. The total above includes both categories, while the Experiment Summary, updated daily, includes only for the record experiments, which currently number more than 380,000.) In addition, control runs are automatically run and reported daily: almost 150,000 control runs are in the database.

Retropsychokinesis tests the hypothesis that users are able to exert an influence upon random data which have been pre-recorded but not examined prior to their experimental runs, and that the statistics of these data will differ from those of unexamined data. Please see the Background to the experiments for additional details, the retrocausality bibliography for additional reading, or the main project page for further references, some of which are online. “But that's crazy!”, you exclaim. Yes, I know, but previous experiments, notably those conducted by the physicist and parapsychologist Helmut Schmidt, have suggested that such an effect, albeit very small, may exist. I have proposed the outline of a crackpot theory which might explain such seemingly impossible experimental results.

The great advantage of experimenting with retropsychokinesis is that most of the problems which plague investigations of other claims of paranormal phenomena do not occur. Data are pre-recorded before the user examines and attempts to influence them. There is no opportunity for fraud by modifying the experimental data, nor by reporting only successful experimental runs. Random data are generated by the Fourmilab HotBits radioactive random number generator, which has been in operation for more than twenty years and passed all statistical tests of randomness. The data collected by the RPKP are completely transparent, and the complete database of experiment and control runs, updated daily, may be downloaded by anybody with access to the Internet. Individuals who participate in the project by running experiments may consult a list of their experiments and results, updated in real time. A number of other resources are available, including an introduction to probability and statistics and statistical analysis tools, all accessible from the RPKP experiments Table of Contents.

The RPKP experiments deliver a stream of 1024 random bits to the user, who then attempts to influence them as they are presented via a visual feedback program, with a statistical report of the result at the end. The visual feedback programs were originally written as Java applets, which were the going thing in the 1990s when the project was launched, and the only game in town at the time for providing interactive animation in Web pages. Unfortunately, the “write once, run anywhere” promise of Java now deservedly receives the rejoinder, “yeah, right”, and two decades of incompatibilities, security flaws, and browser integration problems have brought Java, at least for Web page applets, to the brink of being pitched into the dustbin of computer history—sad.

The end of Java has made the RPKP experiments less accessible to users, and reduced the rate at which experiments are run. Fortunately, there is now an alternate to Java, HTML5 canvas animations with JavaScript (which, despite the name, has nothing whatsoever to do with the Java language): Web standards implemented in all modern browsers, including those on mobile devices, which require no third-party plug-ins to be installed by users.

The Twentieth Anniversary Update of the Retropsychokinesis Project implements all of the visual feedback programs in HTML5/JavaScript, rendering them accessible to anybody with a modern browser. (The Java applet versions of the programs remain available for users with older browsers, or who do not wish to change to the new versions.) There is no change in the functionality of the experiments; the results recorded do not depend upon which version of the visual feedback programs are selected. As has been the case with the RPKP since inception, all of the source code of these programs is transparent and may be downloaded, and users are free to create their own visual feedback programs submit them for inclusion in the Project.

So what about the results? I don't comment on that. I collect the data and publish a daily statistical analysis of the aggregated data. I leave to individual investigators the analysis of the database and formulation and testing of hypotheses based upon the experiments it records.

This update should dramatically increase the number of users who can run RPKP experiments, as the struggle to download, install, enable, and incessantly update Java is no longer a prerequisite. I hope this will contribute to the continued growth of the RPKP database and its value to investigators as the project enters its third decade.

Meanwhile, try running some experiments yourself!

Posted at January 27, 2017 00:04