RetroPsychoKinesis Bell Curve Experiment

This page allows you to perform various kinds of “runs” to determine if you are able to retroactively influence the output of a hardware random number generator based on radioactive decay. Your browser must support HTML5 in order to run this experiment. If you're using an older browser which supports Java™ applets, try the Java page for this experiment.

Running an Experiment

The following paragraphs will walk you through the process of running your first experiment, explaining what options you can choose for the various parameters. Once you're familiar with the options and settings you prefer, you can bypass this wordy experiment setup and use the Express Setup page, which also allows you to save a custom request as a bookmark in your browser.

Experiment Type

Select the type of experiment you want to run by checking one of the buttons below:

Simulates an experiment using pseudorandom data generated on your own computer, rather than using data from the Fourmilab radioactive decay random number generator. This mode allows demonstrating the experiment without consuming genuine random data (which are generated rather slowly), and avoids the delay due to obtaining random data over the Internet. It allows you to see what the experiment will look like without actually running it.
Runs an experiment using random data, but only displays the score on your own machine. A record of the run is made in the RPKP Project experiment database for quality control purposes, but will not be used for statistical analysis of experimental results.
Performs an “on the record” experiment, in which your results will be recorded permanently in the RPKP Project experiment database under the identification you entered in the “E-mail address” field. The mode settings you chose for the experiment are recorded along with your score, to permit analysis of any influence they may have had on the results.

E-mail Address (Record experiments only)

(E-mail address is required only for Record runs)

If you're running a Record experiment, enter the E-mail address under which your results will be recorded. Filing results by E-mail address allows statistical analysis of results by a given individual in a number of separate runs over an extended period of time. Without the identification of an E-mail address, there's no way to tie together runs made by the same person. This E-mail address is used solely for identification, and will not be published or disclosed in way. No unsolicited E-mail will be sent to the address you enter. If you remain worried about disclosing your address but still want to participate in Record mode experiments, make up a random password and enter that instead of your actual E-mail address, something like “asus78tupo”, for example, and use that password for all your Record runs. If the results of your runs show outstanding performance and you later decide to get in touch directly to participate in other experiments, you can, at that time, identify yourself as the person who made the runs in question by giving the password you made them under.



You can try to make the pointer move toward the right of the bell curve or toward the left. A purely random data stream will result in the pointer spending most of its time near the centre of the bell curve: your goal is to influence the pointer to move away from the centre, in the chosen direction. Choose the goal you prefer by checking the button to its left.


If you'd like audible feedback when a new high or low score is reached, check the Sound box. Some people consider this useful feedback; others just find it irritating. It's up to you. (Sound may not be available on some browsers.)

Let's Go!

Once you've selected the parameters for the experiment, press the button above to run it. You can restore all the entry fields to their default settings with the button below. If you'd like to make a number of experimental runs with the same parameters, use the Express Setup page to create a bookmark for your browser which launched the experiment directly.

Other Experiments

Table of Contents

by John Walker