Content provided in partnership with
Why is PSI so elusive? A review and proposed model
Journal of Parapsychology, The, Sept, 2001 by J.E. Kennedy
Supporting Arguments
The hypothesis that psi is independent of task complexity and related information processing is supported by evidence (a) that the success rates are similar for normal PK and blind PK (in which the identity of the target must be obtained by paranormal means), (b) that psi can affect many different types of random processes, (c) that the psi effect does not depend on the inner workings of the random process, and (d) that the influence is not related to how well the psi source understands the random process (Kennedy, 1978, 1979, 1995; Stanford, 1977). The seminal experiments for this concept were by Foster (1940) and Schmidt (1974).
- Most Popular Articles in Reference
- The importance of understanding organizational culture
- Credit card attitudes and behaviors of college students
- What factors attract foreign direct investment?
- Libraries Need Relationship Marketing - mutual interest marketing concept, ...
- How to set performance goals: employee reviews are more than annual critiques
- More »
With goal-oriented experimenter effects, the z score for an experiment would be unrelated to the sample size. Several meta-analyses (Bem & Honorton, 1994; Honorton, 1983; May et al., 1986; Nash, 1989) have reported findings that are consistent with this hypothesis and are significantly different from the traditional assumptions for statistical research (Kennedy, 1994).
The hypothesis that psi is efficient is supported by studies that use majority-vote or repeated-calling techniques to try to enhance psi accuracy. The clearest example is the study by Brier and Tyminski (1970) that used the majority vote of five calls to predict the outcome of a random event. The most accurate predictions and entire statistical significance were due to cases with majorities of three out of five. The raw trials were not significant. The statistical theory underlying the majority-vote strategy predicts that the cases with majorities of five out of five will be the most accurate and that the raw trials will have a more significant p value than the majority-vote outcomes. The observed result suggests that psi achieved the goal of a significant majority-vote outcome in the most efficient manner. Several other majority-vote studies can be evaluated for efficiency and are consistent with the hypothesis (Kennedy, 1979, 1995). Unfortunately, it is not clear how to evaluate efficiency in the more compl icated majority-vote studies that involve index trials and extensive data manipulations, such as the studies by Carpenter (1991).
Situations when the experimental results appear to change when the experimenter's interest changes support the hypothesis of efficient, goal-oriented experimenter effects. For example, trials with still-picture targets were no longer significant in ganzfeld experiments after the experimenter began investigating the hypothesis that the video targets would give better results (Bem & Honorton, 1994; Hyman, 1994). Targ (2001) recently pointed out four other cases in which the experimental results changed when the experimenter's interests changed. In addition, the tendency for lines of research in parapsychology to have decreasing replication rates but overall statistical significance on meta-analyses may be consistent with the hypothesis of efficient goal-oriented experimenter effects (Kennedy, 1995).
Find Research Guides for:
Find Featured Titles for: Arts
Most Popular Articles in Reference
- The importance of understanding organizational culture
- Credit card attitudes and behaviors of college students
- What factors attract foreign direct investment?
- Libraries Need Relationship Marketing - mutual interest marketing concept, ...
- How to set performance goals: employee reviews are more than annual critiques
- More »