Considering the sender as a pk agent in ganzfeld esp studies

Journal of Parapsychology, The, Spring, 2003 by Chris A. Roe, Nicola J. Holt, Christine A. Simmonds

ABSTRACT

Previous ganzfeld research that has considered the role of the sender has been inconsistent in its findings and may be criticised for not taking sufficient account of the sender--receiver relationship. This study used a novel method for assessing any sender influence. A random event generator (REG) placed in the vicinity of the receiver acted as a proxy receiver during the sending period, "selecting" descriptive statements from among a pool of 768. Twenty statements were selected to produce an "REG mentation" that was compared with the performance of the "live" receiver. Forty ganzfeld trials were conducted with novice sender--receiver pairs using a standard protocol apart from the introduction of the REG. Receivers registered a 35% hit rate. The REG mentation was used by an independent judge to rank order the clips in each target set and gave rise to a suggestive hit rate of 32.5%. None of a battery of individual differences measures significantly predicted receiver performance but there was a sign ificant relationship between sender synaesthesia rating and REG performance. The effect size for the REG trials compares favourably with that for other micro PK protocols, and further research is encouraged.

**********

One theoretical question that remains to be answered satisfactorily is whether the sender plays any active role in successful GESP experiments. Given the relative success of clairvoyance designs (see, e.g., reviews by J. B. Rhine, Pratt, Stuart, Smith, & Greenwood, 1940/1966; Utts, 1996), it seems clear that a sender may not be necessary, but may make a positive contribution to ostensible psi. In favour of such a role, we could note that many of the most impressive spontaneous cases seem to involve an "agent" of some sort (cf. Beloff, 1993; L. E. Rhine, 1965). In the laboratory, a number of studies using a variety of methods have reported better performance when a sender or agent was involved compared with clairvoyance alone (e.g., Klein, 1972; Ullman, Krippner, & Vaughan, 1989), suggesting that the agent may be able to somehow gently direct the participants' mentation or choice of target. Further circumstantial evidence in favour of a role for the sender in ganzfeld ESP research can be claimed from Honorton' s (1995) meta-analysis, which found that those studies that included senders generated better performance than those without, although the effect seemed to be confined to those experimenters who had used both conditions at some time. These results are compromised, however, by receivers often knowing whether or not a sender would be present, so that any differences in performance may be a simple psychological effect. Studies that have controlled for this have had mixed results (see Palmer, 1978, for a longer discussion of the psychological sender effect).

Three previous studies involving ganzfeld stimulation have compared sender and no-sender conditions within the same study (Morris, Dalton, Delanoy, & Watt, 1995; Raburn & Manning, 1977; Williams, Roe, Upchurch, & Lawrence, 1994), but these have been inconclusive. Raburn and Manning (1977) did find better performance when a sender was present (particularly, but not exclusively where receivers knew that a sender was present). However, Morris et al. (1995) reported no difference in scoring between sender and no-sender conditions in which the receiver and experimenter were blind, nor surprisingly did they find better performance in a sender condition in which the receiver and experimenter were not blind (and which therefore might be expected to benefit from a psychological sender effect). Williams et al. (1994) reported overall psi missing, making interpretation difficult, although they did not register any hits at all in 12 no-sender trials, which was suggestively different from performance in the two conditions in which a sender was present.

The role of the sender has been considered more directly by Kreitler and Kreitler (1973), who found that trials on which the sender was actively sending the target yielded significantly higher ESP scores than those on which the sender simply thought about the target. Schmeidler (1961) reported better scoring in a clairvoyance task when the senders were instructed to wish for a positive outcome than when they wished for the trial to be a failure. However, Kanthamani and Palmer (1993) described an experiment in which senders were presented with the target image only briefly (ten 1-ms exposures during the sending period) and between exposures were engaged in a PK task so that they might adopt a "subliminal sending strategy." Independent judges' ratings of receivers' mentations gave only two direct hits in 22 trials, which was nonsignificantly below a mean chance expectation of 5.5. The apparent moderating effects of variables such as the agent-subject relationship are also relevant here. Schmeidler (1961) found that she was able to predict ESP performance at better than chance levels by using a rating of agent-subject compatibility. Similarly, Honorton et al. (1990) had noted an improvement in the hit rate of participants who brought in their own senders as opposed to those who had lab personnel assigned to be their sender, and Dalton (1994) found that different sender-receiver sex pairings gave rise to different effect sizes, together suggesting that the nature of the relationship may be an important factor. Unfortunately, such effects of sender-receiver relationship have not been considered in the ganzfeld work that has focused on the role of sender; for security reasons, Raburn and Manning (1977) and Morris et al. (1995) used only lab personnel as senders, and Williams et al. (1994) reported interpersonal dynamics problems that may not have made those sessions typical of the friend-as-sender condition.