An "experiment" was recently reported in which paranormal signals were claimed to be observed.

The claim, this time, is that humanity's "collective consciousness" produces observable effects on some electronic devices situated at different locations, particularly at times when emotionally distressing events take place.

Probably the flaw is an incorrect handling of the statistics, where unintentional bias was difficult to exclude. In this particular case, the point is that whatever "emotionally distressing events" are, is defined while the experiment takes place.

I am prepared to challenge these and any other serious investigators with a bet:

If the experiment is repeated exactly in accordance with conditions to be formulated by me, and if the statistics is subsequently handled with mathematical precision, I am prepared to bet for €1000,- that the signal will be less than 5 sigma.

The point is that bias should be properly excluded. In my science, 5 sigma is considered to be significantly improbable if attributed to chance (though events as improbable as this do take place, occasionally). The amount of € 1000,- is admittedly modest -- I would have been happy to take bigger risks, but this is to avoid imposters who could attempt to trick me out of my money; after all, it is possible to cheat, and a certain amount of trust in honest play will be required from both sides. Also, paranormal investigators are not known for being rich; they may not be able to afford higher bets.

 The conditions on the experimental procedure depend on the nature of the experiment. If someone wants to take the challenge and bet with me concerning a certain kind of experiment, I will formulate the rules how such an experiment can be done without a priori bias one way or the other. In some cases it may be impossible to reach agreement. In the case mentioned above, however, it seems to be easy to formulate the rules in such a way that agreement can be reached. 

In the above example, my rules would be roughly as follows. The experiment is repeated, but this time the output of the electronic devices should be kept strictly secret, both for me and for my opponent(s). Their claim is that events at certain moments and at certain places will be  recognizable in the output. My claim is that this cannot be the case. Sealed boxes with the data will be shuffled in an arbitrary order using my random number generator. The times and places where the data were taken will be recorded, but not revealed to the person who claims to be able to correlate the data to the events.

Now, my opponent is asked to read and interpret the data. He or she may divide them into categories of two or more kinds, and try to correlate them to dates and places where events could have produced some signal. Any procedure now is acceptable to me, as long as it produces results that can be subject to statistical analysis.

Only after the boxes with data have been sorted in accordance with their paranormal messages, their true  origins in time and place will be revealed. We can then easily establish whether or not the correlations seen are statistically significant.

It will not be hard to agree on the mathematical calculation of the statistical probability that the result can be attributed to chance. In fact, if you take my bet, be prepared for the 50% chance that the outcome is in the wrong direction!

If any of the claims I have seen are true, then it should be easy for these investigators to give me signals with a sigma of 10 or much more.

Claims that paranormal effects have been detected are often based on large samples of data, which have been subject to statistical analysis. It is my impression that these results are due to involuntary selection mechanisms. In view of this, I will in general insist on conditions of the following kind: 

An experiment must be analyzed in one single procedure, immediately after disclosure of its results. 

Working with large samples, collected over many years, is acceptable, if my very severe conditions concerning the selection are met.

My conditions may be negotiable. If someone wants 3 sigma, I'll think about it.

If anyone takes the bet, the result will be reported here.

Important link:  you can aim for $ 1000,000.- if you take James Randi's bet!  So, anyone with paranormal gifts, here is the opportunity to become rich.

 Gerard 't Hooft

Back to homepage.