Stage 2 Registered Report: Anomalous Perception in a Ganzfeld Condition - A Meta-Analysis of More Than 40 Years Investigation
π Original studyπ Appears in:
Plain English Summary
For over four decades, parapsychology researchers have been running a peculiar experiment: put someone in a state of sensory deprivation called the ganzfeld (think halved ping-pong balls over the eyes, white noise in headphones, red light) and see if a distant sender can mentally transmit an image to them. This registered report β meaning the analysis plan was locked in before the authors touched the data, a gold-standard practice for preventing bias β gathered 78 of these studies spanning 1974 to 2020. The bottom line: participants picked the correct target image about 31% of the time when pure chance would predict 25%. That six-percentage-point bump may sound modest, but the statistics backing it up are genuinely impressive. The Bayes factor was 89.5, meaning the data is roughly 90 times more likely if the effect is real than if it is just noise. Both traditional and Bayesian statistical models agreed on an effect size of about 0.08 β small but stubbornly consistent. The authors threw four different publication bias tests at the data to check whether the result might just reflect journals only publishing positive findings. All four suggested the effect holds up. A cumulative analysis showed the effect has been remarkably stable since about 1997 β no sign of it fading away over time. Some people appear to be better at this than others. Pre-selected participants (often people who score high on certain personality traits or have prior experience) produced an effect of 0.13, roughly triple that of unselected volunteers. Telepathy-style tasks, where a real person actively sends the image, outperformed clairvoyance tasks where no sender is involved. The catch? The effect is small enough that you would need around 245 trials for an 80% chance of detecting it in a single experiment, and there is still unexplained variation between studies. One of three peer reviewers, a neuroscientist, declined to approve the report, while two others β including parapsychology researcher Dean Radin β gave it the green light.
Research Notes
In plain terms, this study looked at over 40 years of ganzfeld telepathy experiments and found a small but consistent effect β people in sensory deprivation identified the correct target image about 31% of the time when chance would predict 25%. The result held up across multiple statistical methods and tests for publication bias, with a Bayes factor of 89.5 β meaning the data is about 90 times more likely under the hypothesis that the effect is real than under the hypothesis that it is due to chance alone. However, the effect is small enough that a single experiment would need hundreds of trials to reliably detect it, and substantial variation between studies remains unexplained.
Definitive registered-report ganzfeld meta-analysis. Key strengths: pre-registered protocol, 4 publication bias tests, both frequentist/Bayesian models, comprehensive 46-year database. Heterogeneity was medium-large (IΒ² = 63.8%). Two influential outliers detected but removal had minimal impact. Statistical power analysis: 245 trials needed for 80% power with overall ES, but only 50 trials needed with selected participants in Type 3 (telepathy) tasks. Peer-review level did not moderate effects. One reviewer (neuroscience) did not approve; 2 approved (including parapsychology expert Dean Radin).
This Stage 2 Registered Report meta-analyzed 78 ganzfeld studies (113 effect sizes) from 1974-2020 involving 46 principal investigators. Both frequentist (REML with Knapp-Hartung adjustment) and Bayesian random-effects models yielded convergent estimates: ES β 0.08 (95% CI [0.04, 0.12]; BFββ = 89.5). Four publication bias tests (3PSM, p-uniform*, RoBMA, Mathur-VanderWeele sensitivity) indicated results are robust. Cumulative meta-analysis showed effect stabilization since 1997; meta-regression found no decline (slope = 0.0012, p = 0.53). Moderator analyses revealed selected participants produced ES = 0.13 vs. 0.04 for non-selected; telepathy-type tasks (Type 3) showed ES = 0.08 vs. 0.04 for clairvoyance (Type 2).
Links
Related Papers
Extends
- Does Psi Exist? Comments on Milton and Wiseman's (1999) Meta-Analysis of Ganzfeld Research β Storm, Lance (2001)
- Meta-Analysis of Free-Response Studies, 1992β2008: Assessing the Noise Reduction Model in Parapsychology β Storm, Lance (2010)
- Meta-Analysis of Free-Response Studies 2009-2018: Assessing the Noise-Reduction Model Ten Years On β Storm, Lance (2020)
Meta Analyzes
- Does Psi Exist? Comments on Milton and Wiseman's (1999) Meta-Analysis of Ganzfeld Research β Storm, Lance (2001)
- Meta-Analysis of Free-Response Studies, 1992β2008: Assessing the Noise Reduction Model in Parapsychology β Storm, Lance (2010)
- Meta-Analysis of Free-Response Studies 2009-2018: Assessing the Noise-Reduction Model Ten Years On β Storm, Lance (2020)
Same Research Program
Companion
More in Meta Analysis
π Cite this paper
Tressoldi, P.E, Storm, L (2024). Stage 2 Registered Report: Anomalous Perception in a Ganzfeld Condition - A Meta-Analysis of More Than 40 Years Investigation. F1000Research. https://doi.org/10.12688/f1000research.51746.4
@article{tressoldi_2024_ganzfeld_meta,
title = {Stage 2 Registered Report: Anomalous Perception in a Ganzfeld Condition - A Meta-Analysis of More Than 40 Years Investigation},
author = {Tressoldi, P.E and Storm, L},
year = {2024},
journal = {F1000Research},
doi = {10.12688/f1000research.51746.4},
}