Six experiments were performed to confirm published reports of mental influence on random events. A microprocessor-based PK test machine used gamma radiation to produce two types of random events: "direct" events, generated whenever a Geiger tube detected gamma particles, and "seed" events, based upon a string of pseudo-random numbers determined from a truly random seed number and a mathematical algorithm.
Feedback was provided by clockwise (CW) and counterclockwise (CCW) illumination of 16 lamps arranged in a circle. As each random event was detected, the illumination pattern was reversed. The first event caused a CW motion, the second CCW, and so on, until one "run" of 16 CW-CCW pairs had occurred. Additional feedback was provided by a musical-tone generator that sounded a bell whenever the lights stepped CW Hits could be defined as CW or CCW steps. Hit and miss counts were automatically recorded and stored in a memory chip, enabling independent verification of the data.
The first experiment consisted of one subject (the author) using the direct mode for 110 runs. (The task for all direct-mode experiments was to produce more CW steps than CCW steps.) An equal number of runs was performed with the machine left unattended as a control condition. Results were significant in the experimental condition (p < .007) and nonsignlficant in the control condition.
The second experiment consisted of ten unselected subjects performing 150 direct-mode runs. Results were marginally significant (p < .07) In the experimental condition. A control study was not performed. A third experiment, consisting of five subjects performing 75 runs, was nonsignificant in experimental and control conditions.
In the fourth experiment, the author performed 50 runs in the seed mode. The task was to produce as many CW steps as possible. Results in both experimental and control conditions were nonsignificant.
In the fifth experiment, the author performed 200 runs in the seed mode; only this time, the task (to produce more CW or CCW steps) was randomly determined for each run after the seed number had been generated. Results of the experimental condition were highly significant (P < .0005); the control condition was nonsignificant.
In the last experiment, the author performed 50 runs in the seed mode. The task was randomly determined as in Experiment 5 but was not revealed until after the run was complete. Results of the experimental condition were significant (P < .05, two-tailed), but with more misses than hits; the control condition was nonsignificant cant.
Post-hoc analysis of the environmental and psychological conditions of the experiments indicated that the unsuccessful experiments were performed under conditions generally considered to be unfavorable to psi For example, Experiments 2 and 3 were performed in a rather cold (65 degrees) laboratory with unfortunate distractions, such as telephones ringing and the sounds of people talking in nearby rooms. In addition, subjects could devote only a short amount of time to the experiment, and could not fully relax in the laboratory setting. In Experiment 4, the author felt that the task was extremely difficult, since he knew that after a seed was generated the rest of the run would be predetermined. We may speculate the lack of confidence was related to the overall chance performance. The author experienced a similar psychological block in the last "hidden-task" experiment (although in this experiment the results showed significant psi-missing rather than nonsignificance) Successful Experiments 1 and 5, on the other hand, were performed under more favorable, psi-conducive conditions, in which uninterrupted meditative states were achieved during the experiments.
In summary, of six experiments testing a mental influence on machine-generated random events, four showed some evidence of PK. Two one-subject experiments showed significant psi-hitting when the subject was both relaxed and confident; a third one-subject experiment showed significant psi-missing when the subject was relaxed but not confident; a fourth experiment with ten unselected subjects showed marginal psi-hitting in an unfavorable environment; and the remaining two experiments were nonsignificant; one in an unfavorable environment, the other when the subject was doubtful of the outcome. All control studies were nonsignificant.