Discovery Lab Answers

Rick E. Berger, PhD — Experiments and Data

Before drawing any conclusions from the studies below, one important qualification: the ESP-Nexus library’s share of the published literature on Rick Berger’s work has not been measured. What follows reflects the studies currently held in the library — treat this as a summary of what we hold rather than a settled account of his complete research record.

Berger’s work falls into three distinct lines: building and running the automated ganzfeld system at PRL, conducting forensic audits of existing psi datasets, and analyzing geophysical correlates of psi effect sizes across six decades of published experiments. The evidence base in the library covers all three, and the results point in genuinely different directions across them.

Experiments

Autoganzfeld (PRL, 1983–1989). Berger designed and built the automated ganzfeld testing system at the Psychophysical Research Laboratories (PRL), with Charles Honorton as principal investigator. The system placed target selection, stimulus delivery, blind judging, and data logging under computer control, allowing a single target-blind experimenter to run a session. The full 11-experiment series is reported in Honorton, Berger, Varvoglis, Quant, Derr, Hansen, Schechter, and Ferrari (1990). A subsequent forensic audit of the target and target-set distributions — co-authored by Berger, Bierman, and Broughton (1998) — revisited the randomness properties of the PRL target pool.

Forensic audit of the Blackmore ESP corpus (1989). Berger obtained Blackmore’s unpublished dissertation and compared it line-by-line against her published reports, finding pervasive discrepancies. This work is published in the Journal of the American Society for Psychical Research.

Forensic audit of the Spinelli dissertation data (1989). During the Blackmore audit, Berger examined Ernesto Spinelli’s ESP experiments with children, published in the Journal of the Society for Psychical Research. The Spinelli audit was a direct extension of the same methodology.

Geomagnetic correlates of psi effect sizes (1991). With Michael Persinger, Berger analyzed standardized effect sizes calculated from all psi experiments published between 1882 and the study’s close, examining whether annual geomagnetic activity (the aa index) predicted experimental outcomes. Published in Perceptual and Motor Skills.

Methodology

Autoganzfeld protocol. The PRL system addressed the main methodological objections to earlier ganzfeld work itemized in the Hyman–Honorton joint communiqué. Computer-controlled randomization handled target selection; a single target-blind experimenter ran each session; the receiver’s blind judging and subject feedback were under computer control; and the series sample sizes were pre-specified (except two pilot series). Computer-accessed videotape introduced dynamic (film clip) targets alongside static picture targets. The full database included every session since the system’s 1983 inauguration — pilot and ongoing alike — removing file-drawer concerns by construction.

Forensic audit methodology. Berger’s approach to both the Blackmore and Spinelli audits involved obtaining unpublished primary materials (dissertations, original data logs) and comparing them against published versions, tracking discrepancies in sample sizes, reported statistics, study order, and analytical choices. This is a form of data provenance verification applied at the source-document level.

Geomagnetic analysis. Berger and Persinger (1991) calculated standardized effect sizes across the published psi literature spanning six decades, then ran lag/lead correlations between those annual median effect sizes and the aa geomagnetic index across a ±3-year window to identify the dominant temporal relationship.

Data

The three studies in the library yield results across different, non-comparable metrics — Pearson r correlations and a game Z-score shift — so they cannot be pooled into a single overall effect. The full table is rendered below this answer; here is the pattern across the three results.

Two unsettled signals run through this set that must be stated plainly:

• Directions disagree. The geomagnetic study (1991) and the FSS game study (1987) report positive associations; the Blackmore chronological-decline analysis (1989) reports a below-chance, psi-missing result.
• One result is marginal. The decline-over-time correlation (r = −0.80, p = .056) does not reach conventional significance; Berger himself notes in that the rearrangement of study order in Blackmore’s publications obscured this pattern.

The autoganzfeld 11-series outcome from Honorton et al. (1990) — 34% direct hits across 358 sessions, z = 3.90, Cohen’s h = 0.21, results homogeneous across all eight experimenters — is the primary outcome of the PRL program Berger built and co-authored, but it is not currently in the structured evidence rows for this question and so its statistics are cited here only from source rather than from the evidence table.

Skeptical critiques

What critics argue. No published critique of this work appears in the sources retrieved for this question.

What the experimental data show. The corrected z = 2.47 reported in remains above the conventional significance threshold, so the recalculation does not eliminate the overall above-chance result — it reduces and reframes it. The target-type differential, however, loses significance under correction, which bears directly on Bem and Honorton’s (1994) secondary hypothesis about dynamic targets.

Analysis. This is an unusual case: the critique of the PRL target distributions was produced by Berger as a co-author, alongside Bierman and Broughton — making it a self-critical forensic re-examination of a database Berger himself built and contributed to. The specific question of whether the corrected z = 2.47 sustains the original conclusions, and whether the dynamic-target effect survives in any form, has not been resolved by an independent replication using a fully uniform target pool.

For the Blackmore and Spinelli forensic audits, Berger documented that Blackmore’s published corpus invoked methodological flaws selectively — flaws in significant studies were not reported while the same types of flaws in null studies were cited as grounds for dismissal. An independent re-examination of the Spinelli data by a party other than Berger has not been published in the sources retrieved for this question.

For the full profile including Berger’s role in building the autoganzfeld system and the broader context of his forensic audit work, see Rick E. Berger, PhD and the spoke page Forensic critique of psi datasets.