The problem of replication and the "psi-conducive" experimenter - Brief Article

ABSTRACT: Experimenter effects continue to play an important role in parapsychological research, especially with regards to the issue of replication. In this paper, the concept of the "psi-conducive" experimenter is reconsidered, as are its implications for replicating psi effects. Research that has begun to explore allegedly psi-conducive experimenter variables such as the experimenter's expressed belief in psi, the experimenter's expectancy of a successful experimental outcome, experimenter "warmth," and experimenter personality is reviewed. It is concluded that there is evidence to suggest that the experimenter's expressed belief in psi, expectancy of success, and the warmth of the experimenter-participant relationship can distinguish psi-conducive experimenters from psi-inhibitory ones. However, evidence does not support a relationship between personality characteristics and success as a psi experimenter. Four unresolved questions regarding psi-conducive experimenter variables are considered. First, are p si-conducive experimenter variables independent of each other? For example, although some studies have typically attempted to manipulate an experimenter's belief in psi while other studies have attempted to manipulate the experimenter's expectancy about the outcome of the experiment, no studies have tried to manipulate these variables independently of each other within the same study. Such a study would allow one to assess the relative impact of these variables on psi scores and the extent to which they interact with each other. Second, are psi-conducive experimenter variables independent of other allegedly psi-conducive variables? For example, it is not clear how experimenter variables interact with other variables relating to the psi-testing situation (e.g., forced-choice vs. free-response methods). A strong case can be made that experimenter variables are likely to be more important in free-response methods, especially those using receiver optimisation methods such as the ganzfeld, given that participants are typically tested individually and there is considerably greater sustained interaction between the experimenter and participant. Third, how do psi-conducive experimenter variables affect psi scores? Two general hypotheses have been put forward in response to this issue. One is that experimenter variables such as belief in psi and expectancy of success are communicated to participants and so influence the participants' beliefs and expectancies regarding the experiment. A second hypothesis that has received considerable interest from parapsychologists is that, as the source of psi cannot be clearly delineated in psi experiments, experimenter effects are mediated by experimenter psi. It is suggested that greater replicability of psi effects is to be achieved once the relative importance of psychological and psi factors is better understood. Research that helps to clarify the limits of psi, and therefore experimenter psi, will help resolve this issue. Finally, can psi-conducive experimenter variables be exploi ted to enhance psi effects? It is suggested that psi research should draw on research into placebo effects and social psychological research on persuasion and attitude change to identify the most effective ways of exploiting psi-conducive experimenter variables.

COPYRIGHT 2001 Parapsychology Press
COPYRIGHT 2002 Gale Group