Introduction
It is a truism that sensitives find testing by forced choice of 5 cards, e.g. ESP cards, boring and do not do as well as would be expected from their spontaneous psychic experiences. Harribance* prefers photographs but is unusual in doing well at ESP cards.
A long history has developed of ways of testing which allow the sensitive to respond freely when "reading" a person or object. The Saltmarsh-Soal*, the Pratt Birge* and Roll-Burdick extensions* are examples.
The person (target-person TP) wanting a reading submits an object. The sensitives make statements on holding the object. The statements are typed and submitted to the target person, who checks (/), double checks (//) or X’s the accuracy of the sensitives’ statement.
The statistical methods are used to decide whether more statements are accurate than would be expected ona basis of coincidence. If so, then the "sensitives" ability is confirmed under strict conditions. However, the decisions about the sensitives’ ability rests on how many correct checks (/) are made by the target person.
Too few and the results are non-significant. Too many and the results are swamped and non-significant. This is unfortunate and has made several sensitives very angry at scientists. The problem arises that the sensitive seems to know more about the target person than the target person does himself.
This can arise by the target person being extremely anxious, nervous, suicidal, disturbed, naïve, inhibited, scared, etc., as target persons often are, which is why they approach a sensitive in the first place.
Thus the target person misunderstands the statement made by the sensitive and marks it wrong when it should be checked correct, or marks it right when it should be wrong, or does not know himself as well as the sensitive does.
Thus the statistical procedure results in the sensitive being considered no good (rejected) when in fact, the sensitive is good (acceptable); in statistical words, it is a Type I error. Only a relative few mis-checked statements by two or even one out of six target persons can cause this change in result.
Method
What can be done to overcome this occasional break-down in Pratt-Birge-Roll-Burdick procedure? An interesting procedure has been proposed by Carol Ann Liaros, at that time an instructor at Human Dimensions Institute.
If we could convert the experiment to precognition, then we could safeguard against sensory cues. If we could take the decision about whether the sensitives’ statement is correct off the shoulders of the target person, then he could not influence the result incorrectly. A solution thus suggests itself. Throw the decision off on to a future event.
Suppose the target person is required to ask several, say three questions, about events coming up in his life soon, say within 6 weeks, but which at the moment he does not know how they will pan out?
Suppose further the questions must be written in such a form that a yes or no answer has to be given. Then all the sensitive does is go down the list of questions and provide yes or no to each.
This is not trivial and can provide some highly exciting situations. For example, "Will our Union contract be accepted or rejected, Sunday, October 24, 1971?" Reply: "Accepted" (Immediately on Tuesday, October 19, 1971).
He thought he would be on strike and out of work after the Sunday. He was very concerned. Then there was a complete change and the vote was 54 to 6 against striking on the Sunday.
Then, 4 to 6 weeks later the target person confirms whether the Yes or No answer was how the event did come out. Here, no matter how anxious or scared he is, the likelihood of marking this wrongly is minimized, the sensitive’s reputation is protected and scientific procedure is maintained. Then a simple chisquare, one degree of freedom, indicates statistical significance of the result.
Caution is required to guard against the possibility that all or nearly all the questions were so formed that a Yes, No answer was the result; and that the sensitive had a strong bias to provide a Yes (No) "reading" of the future event.
For such a possibility the result would be due not to the sensitive’s psychic ability but to response bias. A second caution is required, namely that the question is posed such that the answer is a foregone conclusion even six weeks into the future.
Here the result would be due to inference and not psychic ability. Such questions would need to be struck out of the results. Then all the remainder should be published so that the reader can judge for himself and realize the extraordinary psychic ability some sensitives have.
1) J. G. and W. R. Birge, "Appraising Verbal Test Material in Parapsychology" J. Parapsychol. 12, Dec. 1948, 236-256.
2) W. G. Roll and D. S. Burdick, "Statistical Models for the Assessment of Verbal and Other ESP Responses", J.A.S.P.R. 63, No. 3, July 1969, 287-301.
3) W. G. Roll, "Designs for Tests with Free Response Material", J.A.S.P.R. 56, Oct. 1962, 184-195.
4) H. F. Saltmarsh and S. G. Soal, "A Method of Estimating the Supernormal Content of Mediumistic Communications", Proc. S.P.R., 39, 1930-31, 266-271.
5) J. P. Stump, W. G. Roll and M. Roll, "Some Exploratory Forced Choice ESP Experiments with Lalsingh Harribance", J.A.S.P.R., 64, No. 4, Oct. 1970, 421-431.
