Abstract

The aim of this study was to explore the role of the sender in a ganzfeld ESP task; more specifically it was a conceptual replication of an experiment by Raburn and Manning (1977) that manipulated both the presence of a sender (sender, no sender) and the receiver's expectation concerning sender's presence. There were 40 trials overall involving three experimenters and an opportunity sample of 40 pairs of participants consisting mainly of friends, acquaintances and staff and students at UCN. The ganzfeld sessions were run by an automated free-response testing system that selected one of the four possible conditions at random. Both the experimenter and the receiver were blind to the true nature of the condition. During the no-sender trials the designated sender was engaged in an alternative psi task deluding the use of a computerised PK greyhound task. Overall the mean z score based upon ratings of the target relative to the dummy video-clips was below chance expectations (- 0.10, r = - 0.10). There were no significant main effects of either sender role (p = .676) or receiver expectancy (p = . 734) and no significant interaction effects (p =.978) and the associated effect sizes for these were small. Contrary to Raburn and Manning (1977), performance was not significantly better in the sender than the no-sender conditions, although did give a trend in support of their finding that conditions in which the receiver expected a sender resulted in better performance than when the receiver did not. Post hoc
analysis of absolute z score for ratings suggested that the difference between the expectancy conditions was approaching significance (p =.057) with a medium effect size. These findings are discussed in elation to Palmer's (unpub) interaction model, which suggests that some variables might influence the direction of psi effects and some might influence the magnitude. An investigation of sender and receiver covariates of performance suggested moderate positive correlations between both senders' (r = .229) and receivers' (r = .242) expectations of being able to demonstrate psi in the experiment and task performance. There was also a modest positive correlation (r =.236) between receiver's having practised a mental discipline and performance, which was not found for senders. Overall, the results suggest that, at least in the ganzfeld context, receiver expectancy might be more important than whether or not there is a sender actually present, although the relatively small effect sizes, the lack of apparent psi effects in the study and the low statistical power mean that this remains uncertain.