Abstract

The experimenter, acting as his own subject, released 30 dice on each trial to roll down an inclined plane on to a horizontal surface while intending that they should fall to show a preponderance of either high faces (4,5,6) or low faces (1,2,3). The same number of throws were made for each target. A total of 400 trials (12,000 single die-throws) were made, the first half of them with the subject acting as his own recorder and the second half with another person present as a second observer and recorder. The total results showed a positive deviation of 287 hits (i.e., 287 more dice falling with the target faces uppermost than the most likely number expected on a chance basis), which gave a C.R. of 5.24 (P = 2 x 10-super( -7)). In view of the exclusion of possible counterhypotheses by the experimental conditions, the results support the PK hypothesis.