Chris A. Roe & Nicola J. Holt
Centre for the Study of Anomalous Psychological Processes
University of Northampton, Northampton, UK
In recent work we have been concerned to evaluate whether the sender plays any active role in successful ganzfeld GESP experiments (e.g., Roe, Holt & Simmonds, 2003; Roe & Holt, in press) by using a random number generator (RNG) as a "virtual receiver" in a ganzfeld-like experiment. During the sending period descriptive statements were "selected" from among a pool of 768 items to give a 20-item "RNG mentation" that could be used by an independent judge. After early success in demonstrating the basic effect, later work considered the effects of varying the lability of the target selection method (Holt & Roe, 2005) using a simplified protocol in which senders were accurately briefed and attempted to influence the RNG; this allowed us to offer immediate feedback in the form of an on-screen display of the selected statements. Target selection method lability was manipulated to give three within-subject conditions: a random number table; a pseudo random process; and a live RNG. Participants were classified high, intermediate or low lability based on a combination of personality and experiential measures. Significant psi hitting was not obtained in any of the randomness conditions, although there was a significant interaction effect between target and sender lability, which emerged for both independent judges (F4,37 = 2.891, p = .028 [JW]; F4,37 = 4.536, p = .002 [LS]). The present study was designed to confirm that finding and to extend it by considering the possible interaction effects of sending strategy (active/willing versus passive/absorbed) upon feedback type (delayed versus immediate). Forty participants were randomly allocated to one of four conditions differing in sending strategy/feedback type. Each was presented with 24 statements as a virtual reading, consisting of 8 selected using each of the three randomness sources. The direct hit rates for all target systems are at or below mean chance expectation (MCE = 25%). A mixed 3x3 ANOVA found no significant main effects, neither the degree of target lability (F2,74 = .074, p = .929), nor the lability of the sender (F2,37 = .387, p = .651) significantly impacted upon psi-success. However, there was a significant interaction between target lability and sender lability (F4,74 = 2.747, p = .034). This replicates the interaction effect found by Holt and Roe (2005). An unrelated 2 x 2 ANOVA was conducted, using just the mentations produced by the Live RNG, to consider the two factors of sending strategy (absorbed versus willing) and feedback type (immediate and delayed). There was no main effect of sending strategy (F1,36 = .029, p = .865 ) nor of feedback type (F1,36 = 2.101, p = .156), and only a suggestive interaction effect (F1,36 = 2.310, p = .137).