Skip to main content

Assessing the Evidence for Mind-Matter Interaction Effects

📄 Original study
Radin, Dean, Nelson, Roger, Dobyns, York, Houtkooper, Joop 2006 Modern Era psychokinesis

📌 Appears in:

Plain English Summary

Can your mind nudge a random number generator (basically a digital coin-flipper)? Critics said the evidence from decades of these experiments was inflated by publication bias -- researchers only publishing exciting results. This rebuttal punches back hard. The authors point out that critics assumed the mind would influence each individual random bit equally, regardless of whether the machine was spitting out ten bits per second or millions. That assumption spans six orders of magnitude and is, they argue, simply unjustified. The suspicious funnel-plot pattern (a graph used to sniff out bias) actually matches what you'd expect from genuinely varied experiments, not hidden negative results. A survey of researchers turned up roughly one unpublished study per person -- far fewer than the thousands needed to sink the findings. Even the critics' own stricter statistical model still yielded a significant result for pre-planned studies.

Research Notes

Most comprehensive single defense against simultaneous critiques of RNG/PK meta-analytic evidence. Companion to radin_2006_reexamining in JSE Vol. 20 No. 3. Introduces Collins's Experimenters' Regress into the MMI debate and proposes alternative effect-size models (constant-z, goal-oriented, time-normalized yield).

Responding to critiques by Schub, Scargle, Ehm, and Bösch et al. of the Radin & Nelson (1989, 2003) RNG meta-analyses, this commentary challenges the 'influence-per-bit' assumption — that mind-matter interaction (MMI) operates uniformly on each random bit regardless of bit rate, sample size, or psychological conditions. With bit rates spanning six orders of magnitude across experiments, the authors argue this assumption is unjustified. Funnel-plot asymmetry is shown to reflect genuine heterogeneity rather than publication bias via simulated constant-z data. A researcher survey found ~1 unreported study per investigator, far fewer than the thousands needed to explain the cumulative results. Bösch et al.'s own random effects model yielded z = 4.08 for pre-planned studies.

Related Papers

Also by these authors

More in Psychokinesis

📋 Cite this paper
APA
Radin, Dean, Nelson, Roger, Dobyns, York, Houtkooper, Joop (2006). Assessing the Evidence for Mind-Matter Interaction Effects. Journal of Scientific Exploration.
BibTeX
@article{radin_2006_assessing,
  title = {Assessing the Evidence for Mind-Matter Interaction Effects},
  author = {Radin, Dean and Nelson, Roger and Dobyns, York and Houtkooper, Joop},
  year = {2006},
  journal = {Journal of Scientific Exploration},
}