Richard Wiseman’s “Experimenter Effect” Examined


A Reply to O’Neill


The O’Neill – Wiseman Controversy

by Richard Wiseman


My recent mass ESP experiment was designed to test the notion that a large number of senders might enhance psi communication. The study also aimed to inform the public about one of the techniques often used by parapsychologists to investigate ESP. The experiment was run over the course of one day and comprised ten trials. For each trial a ‘receiver’ described their thoughts aloud. In a distant location a group of senders were asked to concentrate on a randomly selected photograph. After approximately ten minutes the receiver was presented with four photographs (the actual target and three decoys) and was asked to choose the photograph that they believed best matched their thoughts. The first eight trials planned to involve a small number of senders looking at a target being projected onto a screen inside the Museum Of The Unknown on London¹s Southbank. The last two trials were designed to attract a larger number of senders, and involved them concentrating on a photograph projected across the entire outside wall of the Museum.

I went to considerable lengths to ensure that the experiment was methodologically sound. For example, to minimise the potential for sensory leakage we obtained special permission to have the receiver on the nineteenth floor of the IPC Tower directly behind the Museum. The number of trials was specified in advance of the experiment, along with the statistical tests that would be used to assess significance. The experimental design also incorporated several factors often seen as being psi conducive. For example, all of the receivers underwent a kind of mild sensory deprivation known as the ‘ganzfeld’ procedure – previous experiments using this technique have produced impressive evidence of psi communication. We selected the type of target images and receivers who had scored well in past ganzfeld trials (i.e., visually compelling photographs and people with a background in the visual and performing arts). Also, specially installed radio link between the IPC Tower and the Museum ensured that the receivers’ comments were broadcast live to senders to help act as feedback.

Only two of the trials resulted in direct hits, thus providing no support for the notion that a group of senders significantly enhanced psi communication. However, the study did succeed in communicating the ganzfeld methodology to an international audience via widespread media attention, including television (GMTV, BBC Newsnight, Channel 4 News, Japanese Television, CNN), radio (Radio 4 Midweek, Radio 1, Radio 5, Irish Radio), and broadsheet newspapers (The Daily Telegraph, The Guardian and The Independent).

SPR member Mick O’Neill joined us for many of the trials. Unfortunately, his resulting commentary is both factually inaccurate and conceptually confused.

For example, O’Neill first points out that I have a background in conjuring and states that although magicians have a role to play in psychical research, “it seems dubious that they should be performing experiments themselves”. O’Neill fails to mention that my background is not just in conjuring, but also in academia, including a Ph.D. from Edinburgh University in parapsychology. I have carried out many previous psi experiments, have published widely in parapsychology journals and have authored two textbooks on methodological issues within parapsychology. The mass ESP experiment utilised my background in experimental design and statistics, not my training as a magician. It is difficult to understand why O’Neill believes that I was not qualified to run such a study.

O’Neill stated that I ‘didn¹t want psychic investigators to attend’. This is absurd. The experiment was widely publicised throughout the mass media, including Radio Four, Radio Five, The Daily Telegraph, Time Out, The Independent and The Guardian. Details about the study were circulated on a well-known psi related internet group, and the Museum distributed hundreds of leaflets about the event. All of this pre-experiment publicity made it quite clear that everyone was welcome to attend and that we wanted as many people to participate as possible.

O’Neill also noted that, apart from technicians and the press, there was “absolutely no-one there”. In fact, each trial involved between approximately 10 and 100 senders. The first trial took place at 10 a.m., before the Museum was usually open. We knew that this trial was likely to attract a small number of people, but also knew that the experiment was technically very demanding, and were glad of an opportunity to conduct the first trial without the pressure of a large audience. However, the number of senders grew steadily throughout the day, and the last two outdoor trials attracted a relatively large crowd of approximately 100 people.

O’Neill criticises the method used to randomly choose the targets. The procedure involved three bags, each containing table tennis balls numbered between 1 and 10. For each trial, three members of the public were each asked to place their hand inside a bag and remove a ball. The first ball was used to select one of ten pages of random numbers from a book of 1 Million Random Digits (a source of randomness widely used in ESP experiments). Another ball was used to access one of ten rows on the page, whilst the third ball was used to choose one of ten columns. This process resulted in a block of 25 numbers being randomly selected. The experimenter then moved through the block until he came to a number between 1 and 4. This process resulted in one number being randomly chosen from a huge number of digits. The process was carried out by myself in front of each group of senders. During the inside trials the sheets of random numbers were projected onto a large screen to help people follow the process. O’Neill described the bags as ‘magician’s bags’ when, in reality, they were simply cloth bags. He also stated that I should not have known the identity of the potential targets in advance of the experiment. However, he doesn¹t state how such prior knowledge could have significantly biased the target selection and, until O’Neill presents such information, it is impossible to assess whether any bias actually took place.

O’Neill also makes much of my comments to the crowd of senders in the final trial. He refers to this trial as the “crucial final trial”. It was not. For the experiment to have been considered a success, we required six or more hits. During the first nine trials we had only obtained two hits. The outcome of the trial could not have affected the outcome of the experiment, and it is difficult to understand why O’Neill believes that it was crucial.

I would be the first to admit that this experiment, like any piece of scientific research, was not perfect. For example, it contained a small number of trials, and the outside trials were conducted under unexpectedly adverse weather conditions. However, O’Neill’s criticisms are inaccurate, ill-informed and do little to further psi reseach. Readers interested in a more balanced and accurate description of the experiment are referred to The Fortean Times for February 2000.

New Browser Icon

© 2014 The Association for Skeptical Investigation. All rights reserved.