What’s in the box? An ESP test with Chris Robinson

Journal of the Society for Psychical Research, 60, 322-324

Acknowledgements: I would like to thank Prof A.J.Ellison, Dr A.Hart-Davis, and Prof R.L. Morris for help with running the experiment and Dr R.Matthews for help with the analysis.


Chris Robinson is a well known English psychic who has appeared recently on many television and radio programmes. He claims to have precognitive dreams which have enabled him to predict, among other things, air crashes, traffic accidents, murders and sexual attacks. Perhaps his most well-known claim is that he has worked with the police to track down criminals and to predict IRA bombings.

On television he has often been seen guessing what object is concealed inside a closed box. In at least two cases he was apparently successful at guessing the object. On Wire TV (in Bristol) he guessed the object was a circle or disk with a whole in the middle, and added that it might be a telescope mirror or a compact disk. It was a record. On Granada TV’s “This Morning” he guessed that it might be a dolly or some kind of child’s toy and it was a teddy bear. The people who hid the objects claimed that Robinson could not possibly have seen the object in advance.

In October 1994 on an “Esther” television programme for BBC1, Robinson made further claims about his psychic abilities and challenged me to test him under controlled conditions. I welcomed this suggestion and contacted him to arrange for tests. He suggested that the “box test” gave him the best chance of success and that we should do several trials of this type.

There are a number of problems to be solved if this test is to be made worthwhile. When done as a TV demonstration there are three main weaknesses. First there is no adequate control over the choice of objects. Usually someone involved in the programme simply chooses any object that comes to mind (or is handy) and hides it in a box. It is likely that some objects are chosen far more frequently than others. Robinson might know what they are and therefore strike it lucky more often than viewers would expect. Second there is inadequate control over the box itself and over who knows its contents. Television studios are often chaotic and, although the people in charge told me that there was no possibility of Robinson seeing into the box, this cannot be entirely replied upon. Nor can we entirely exclude the possibility of other people in the studio finding out and (consciously or unconsciously) cueing him. Third, it is not clear what counts as a success. Robinson records long complex dreams and there is always the chance that something in the dream will correspond to the chosen object. The cases above seem impressive but are impossible to evaluate statistically.

A method was needed that would get round all these problems and provide adequate controls over error, fraud and artefacts.

We agreed that I would conceal an object in a box in my home and change the object twice a week for six weeks. Meanwhile Robinson would record his dreams. At the end of the six weeks I would give him a list of the 12 objects and a photograph of each one. He would then, on the basis of the dreams, guess which object was in the box when – that is, he would match up the objects with the dates. It might have been possible to use an independent judge to read the dreams and match them to the objects, but it was clear from watching Robinson work, that he alone understands the intricacies of his own dreams. It therefore seemed better to get him to do the matching himself. We carried out the experiment during December 1994 and January 1995, during a period when Robinson was out of the country for much of the time.

Method

At Robinson’s suggestion I took a strong cardboard box and printed his name in large letters on all sides of it. I then selected 12 household objects that would fit easily into the box. An assistant (Adam Hart-Davis) then photographed all the objects, and prepared a random list with one of the 12 objects matched against each of the 12 dates. He made copies of this list, sealed them and sent one each to Professors R.L.Morris and A.J. Ellison (both Vice-Presidents of the SPR) for safe keeping. This precaution was to prevent me from changing the target objects after I heard what Robinson’s guesses were.

The assistant took charge of the objects and was responsible for placing the correct one in the box on the correct dates. He fixed a padlock to the box so that any attempt to look inside could be detected. I did not know the order of the objects until after the experiment was completed. There were some days in which he could not change the objects as planned. The few slight changes this necessitated were sent to Morris and Ellison before the completion of the experiment.

At the end of the planned period Robinson sent me, as arranged, copies of all his dream records. This amounted to 87 hand-written pages. I then sent him (1) photographs of all 12 objects, and (2) a list of the dates in which the objects were hidden. His task was to match the objects to the dates on the basis of his dream records. He returned the completed list to me. I then compared it with the assistant’s target list.

Note that Robinson did not know the identity of the objects when he was recording his dreams. Only when the trials were complete, and I had received his dream records, was he given the list of objects to match to the dreams.

Results

Robinson made two lists of his guesses. First he made a list of only seven objects about which he said he was certain. On my request he made a second list including a guess at all 12 objects, as planned. These guesses can be seen in Table 1.

 

Date Target Object Guess (certain) result Guess (forced) result
Dec 9 Pot Plant Bottle of Wine    x
Dec 12 Teddy Bear Cereal Packet    x
Dec 16 Bottle of Wine       – Light Bulb     x
Dec 19 Coal Tongs  Coal Tongs    3
Dec 23 Cereal Packet       – Bowler Hat     x
Dec 27 Watering Can Watering Can    3
Jan 3 Bowler Hat Phrenology Head    x
Jan 6 Boot        – Teddy Bear     x
Jan 9 Toilet Roll        – Boot     x
Jan 15 Phrenology Head        – Pot Plant     x
Jan 16 Light Bulb Boot    x
Jan 21 Bible Toilet Roll    x
Total Correct    2     0

Table 1.   Results of Matching.

The experiment was designed to have 12 trials. By chance one hit in 12 would be expected. Robinson obtained 2 hits which is not significant (using the Bernouilli distribution p = 0.26). It could be argued that he did better on his initial 7 guesses and that it was unfair to force him to guess at the rest. An unplanned, post hoc, analysis was therefore carried out on just these 7 guesses, but this too is not significant (p = 0.11). There is therefore no evidence that Robinson was able to identify the objects hidden in the box more often than would be expected by chance.

Discussion

Robinson was co-operative and helpful at all stages of the experiment and I much appreciated the chance to put his claims to the test. However, under controlled conditions, in a test of his own choosing, he showed no evidence of psychic ability.

Susan Blackmore,
Department of Psychology,
University of the West of England,
St Matthias College,
Bristol BS16 2JP.                                                                             May 1995