In a previous research project, we presented a telepathy study which we hoped would be both psiconducive and efficient in terms of data-collection; it involved coupling a multiple trial forcedchoice protocol with participant optimization procedures involving immersive hardware and software. Overall results were not significant, though they did show significant variability, possibly suggesting a combination of psi hitting and psi missing.
The current exploratory study is broadly inspired by the previous one, but introduces several improvements in the optimization procedures, as well as a shift from a dyadic-subject approach, to a single-subject protocol focused on precognition. This shift was motivated in large part by the rather positive track record of recent precognition research – as witnessed by physiologically based presentiment studies, and by the reverse-causality social psychology studies recent introduced by Daryl Bem. We thus decided to opt for a single-subject precognition protocol, both because it is comparatively easier to implement than telepathy protocols, and because it seems promising in terms of replicability.
A second, more process-oriented objective for this study, was to better grasp the role of feedback in multiple-trial psi tasks. Both theoretical and psychological reasoning would suggest that trial-by-trial feedback should contribute to scoring, either by injecting outcome information into earlier choices (according to a retro-causal model), or by helping subjects to zero in on productive mental strategies as they proceed through an experimental session. On the other hand, as some subjects in our previous study indicated, it may be that trial-by-trial outcome information induces stressful, performance-oriented mindset, offsetting the learning gains that might come with feedback. Thus, as part of our objective of producing a multi-trial psi-conducive protocol, we sought to assess feedback vs. non-feedback scoring under motivationally equivalent conditions.
Finally, this study was also intended to provide us with an empirical effect-size estimate, which in turn would allow us to specify power requirements for future replications. A pre-set total of 3000 binary choice trials were collected from 82 subjects over a 7-week testing period, each subject contributing either 20 or 40 trials, in accordance with pre-set criteria. 26 of these subjects were known to be regular meditation practitioners, invited by one of us (PAB) who is affiliated with a local Buddhist center. All subjects were tested either by MV or PAB at the Institute Métapsychique in Paris, using a specially constructed immersive environment and audio-visuel sequences used for the psi task. The task consisted of selecting and opening spherical image containers that emerged out of a hypnotic animated starfield. Following the subject’s choice, a binary random decision would decide whether the subject would obtain a miss or a hit, and whether the hit-miss feedback would be shown. A hit resulted in the emergence of the portrait of an animal, or an interesting or famous personality staring directly at the subject and growing in size. Misses were
associated with a noisy withdrawal of the sphere back into the starfield and no-feedback produced a simple fadeout of the sphere container.
The results for the 3000 trials were non-significant. The overall hit rate came in at null expectation (50.1%), and although feedback trials did have a higher hit rate than no-feedback trials (51% vs. 48.6%) the t-score for the difference was only 1.29 (one-tailed P-value=0.10). Nevertheless, these results, as well as several exploratory analyses, suggest directions for further
study. First, the difference between feedback and no-feedback hit rates indicates that a 0.05-level replication at 80% power could be achieved with only a modest increase in the number of trials. A positive replication would encourage the idea that feedback may be useful to subjects, provided it does not disrupt the “flow” state and induce a performance mindset.
Second, hit rates tended to increase over the course of a 20-trial series. A regression of trialordered hit rates found higher hit rates for later trials (P-value=0.04, two-tailed). In other words, subjects may have progressively found a mental strategy or a state that produced better scoring, which would accord with subject responses in a post-session questionnaire, where 89% agreed “scores would improve with continued training”.
Finally, the 26 Buddhist meditators had a group hit rate of 52.1% (one-tailed P-value=0.09) and the subgroup judged to be the most experienced had a hit rate of 54.7% (one-tailed P-value=.012). This finding contributes to the growing literature that suggests that experienced meditators are particularly good subjects for psi research. We are thus planning to follow up with our protocol this year, focusing more specifically on an experienced-meditator population. We would like to express our deep gratitude to the Bial Foundation for its support of the Selfield project.
|Varvoglis, M., Bancel, P., Ahmed, D., Boban, J., & Bailly, J. -P. (2017). The Selfield : A precognitive study using an immersive display system. In Abstracts of Presented Papers, 60th Annual Convention of the Parapsychological Association (pp. 31-33). Athens, Greece: Parapsychological Association.|