Roger D. Nelson, PhD Sources:
FieldREG at Sacred Sites and Group Coherence
The FieldREG program brought portable random event generators (REGs) out of the laboratory and into the world, monitoring group gatherings, ceremonies, and sacred sites to test whether collective human consciousness could produce measurable deviations in random physical processes. Nelson and colleagues at the Princeton Engineering Anomalies Research (PEAR) lab developed this approach in the mid-1990s, extending individual operator-intention research toward a passive, ambient model of group consciousness effects.
Deeper dives — Nelson:
Key findings
- Ten initial FieldREG deployments across diverse group venues produced a collective probability against chance of 2×10⁻⁴, with the most anomalous data segments correlating with observer-reported periods of high group coherence.1
- A formal replication series of 21 hypothesis-based FieldREG applications in “resonant” environments yielded a composite probability of 2.2×10⁻⁶ against chance, while a matched “mundane” subset produced a probability of 0.91, near chance expectation.2
- FieldREG deployments at Egyptian sacred sites during group meditation and chanting confirmed both pre-planned hypotheses, with a combined probability of 2.7×10⁻⁶.3
- The pattern across studies suggests that the quality of group engagement, characterized by shared attention, emotional intensity, or spiritual resonance, is a stronger predictor of REG deviations than the mere presence of a group.2
- Logbook notes and participant reports consistently associated statistically unusual REG sequences with moments of high attentional cohesiveness, intellectual engagement, or shared emotion, providing contextual triangulation for the quantitative results.1
Overview
The FieldREG program represents a conceptual extension of the PEAR laboratory’s individual operator-REG research into naturalistic group settings. Rather than asking a single participant to intentionally influence a REG device, FieldREG studies deploy portable REG equipment passively in environments where groups gather, conferences, ceremonies, sacred sites, theatrical performances, and ask whether the ambient quality of group consciousness leaves a detectable signature in the random data stream. A key vulnerability of this research design is that effect sizes are expected to be small and state-dependent: only environments characterized by genuine group coherence are predicted to show deviations, making null results in non-resonant settings an expected and informative outcome rather than a disconfirmation.2 This means that dismissing the FieldREG literature on the basis of any single null result would be premature without first characterizing the coherence quality of the group environment tested.
The Consciousness Field Concept
Nelson and colleagues proposed that certain states of group consciousness, characterized by shared attention, emotional resonance, or spiritual engagement, might manifest as small but statistically detectable changes in sensitive physical systems such as REGs. This “consciousness field” framing draws on earlier PEAR work showing individual operator effects on REG outputs, and extends it to the hypothesis that groups, even without directed intention toward the device, might produce analogous effects through ambient coherence. The researchers explicitly acknowledged that this interpretation remains speculative and that the data, while statistically significant, resist explanation via canonical scientific models.2 Alternative interpretations, including undetected environmental correlates, equipment sensitivity to physical vibration during chanting, or post-hoc selection of significant segments, are discussed in the skeptical critiques section below.
Origins of the FieldREG Approach
The FieldREG methodology grew directly from the PEAR laboratory’s development of portable REG hardware in the early 1990s. Once compact, battery-operated REG devices became available, Nelson and colleagues recognized that the same technology used for individual operator studies could be deployed passively in field environments to monitor ambient group states. The name “FieldREG” was intentionally a double entendre: the device operates in field (outdoor or naturalistic) settings, and it was hypothesized to respond to changes in a consciousness “field” of the kind proposed in various theoretical frameworks.2
From Laboratory to Field: Equipment and Protocol Development
The portable FieldREG systems used microelectronic binary generators producing continuous sequences of random bits, recorded with time-stamped computer indices and accompanied by onsite logbook notes of relevant observations and activities. The shift from laboratory to field introduced new methodological challenges: unlike controlled laboratory sessions with a single operator, field deployments involved multiple people, variable physical environments, and no direct participant attention to the device. The PEAR team addressed the multiple-comparisons artifact by pre-specifying temporal segments (sessions, presentations, or days) and applying corrections for multiple sampling before compounding probabilities across venues.1 The device’s nominal randomicity was established through extensive laboratory calibration prior to field deployment, partially addressing the artifact of equipment malfunction or systematic bias, though physical vibration from chanting or drumming during ceremonies remained an unresolved potential confound in some applications.
Group Situations: The 1996 Foundational Study
The foundational FieldREG paper, authored by Nelson, Bradish, Dobyns, Dunne, and Jahn, reported results from ten separate field deployments across a range of group environments, including professional conferences, creative workshops, religious gatherings, and theatrical performances. The study found that the most anomalous data segments from each venue, after correction for multiple sampling, compounded to a collective probability against chance of 2×10⁻⁴.1 Logbook notes and participant reports consistently associated the statistically unusual sequences with moments of high attentional cohesiveness, shared emotion, or intellectual engagement.
Ten-Venue Results and Multiple-Comparisons Correction
Across the ten venues, the FieldREG system operated under formal protocols in which each application subdivided naturally into temporal segments such as sessions, presentations, or days. The most extreme data segment from each of the ten applications was identified, and a correction for multiple sampling was applied before compounding. The resulting collective probability of 2×10⁻⁴ (N=10 venues) was treated as exploratory, providing the empirical basis for formulating a testable general hypothesis for subsequent formal replication. The researchers noted that interpretation remained speculative at this stage, and that the study’s primary contribution was establishing the plausibility of group-consciousness REG effects and identifying the environmental conditions, high group coherence, shared attention, emotional intensity, that appeared to correlate with anomalous deviations.1 The San Francisco Bay Revels application, a community performance event, was among the venues contributing to this initial dataset.4
FieldREG II: Replications and Explorations
Building on the exploratory 1996 findings, Nelson and colleagues formulated a testable general hypothesis: that REG data collected in environments fostering relatively intense or profound subjective resonance would show larger mean deviations from chance expectation than data from more pragmatic or mundane assemblies. This hypothesis was then tested in a formal replication series of 21 applications, with an additional 40 exploratory deployments designed to map the boundary conditions of the effect.2 The formal replications strongly confirmed the hypothesis, and the contrast between resonant and mundane subsets provided an internal control that partially addressed the artifact of general environmental sensitivity in the REG devices.
Resonant vs. Mundane Subsets: Composite Statistics
The FieldREG II paper (Nelson, Jahn, Dunne, Dobyns, and Bradish) reported results from 61 applications in total: 21 formal hypothesis-based replications and 40 further explorations. The formal replications were divided a priori into “resonant” environments (those predicted to foster intense subjective engagement) and “mundane” environments (more pragmatic assemblies used as a comparison class). The resonant subset yielded a composite probability of 2.2×10⁻⁶ against chance expectation, while the mundane subset produced a probability of 0.91, statistically indistinguishable from chance. This internal contrast between resonant and mundane conditions partially addressed the artifact of general equipment sensitivity or environmental electromagnetic interference, since both subsets were collected with the same hardware under similar physical conditions. The exploratory work additionally identified venues, including certain business meetings and sporting events, that did not appear conducive to anomalous REG responses, providing boundary-condition information.2 A companion paper in the same EXPLORE issue provided additional detail on the anomalies in group situations.5
Proposed Mechanism and Competing Interpretations
Nelson’s preferred interpretation is that certain states of group consciousness, characterized by shared attention, emotional resonance, or spiritual engagement, produce a “consciousness field” that manifests as small but detectable deviations in sensitive physical systems. This interpretation remains contested. The researchers themselves acknowledged that the effects resist explanation via canonical scientific models.2 Competing non-psi explanations include: (1) post-hoc classification of venues as “resonant” or “mundane” introducing selection bias, partially mitigated by the pre-stated hypothesis derived from the 18 prior exploratory applications, but not fully eliminated; (2) physical vibration from group activities such as chanting, drumming, or applause affecting REG hardware, unresolved, as the papers do not report vibration-isolation tests; (3) experimenter expectancy effects in logbook annotation influencing which segments are flagged as coherent, partially addressed by the quantitative pre-specification of temporal segments, but unresolved for the qualitative coherence ratings. No consensus interpretation has emerged in the literature.
Sacred Sites in Egypt: A Case Study
In 1997, Nelson conducted a two-week FieldREG deployment at Egyptian sacred sites, temples, pyramids, and tombs, accompanying a group engaged in informal ceremonies including chanting and meditation. Originally documented as a PEAR technical report, the study was published in the Journal of Scientific Exploration in 2024, making it publicly available for the first time to a broad academic audience.3 Both pre-planned hypotheses were confirmed: anomalous REG deviations occurred during visits to sacred interior spaces, and coherence-building activities such as chanting and meditation showed additional correlation with deviations.
Egypt Study: Hypotheses, Data, and Combined Probability
The Egypt study used a portable REG and palmtop computer to generate and record ongoing random sequences with time-stamped indices and onsite logbook notes. Two pre-planned hypotheses were specified before data collection: (1) that anomalous deviations would occur during visits to sacred sites, including the inner sanctum or Holy of Holies in each temple and all interior chambers of the pyramids; and (2) that resonance- or coherence-building activities of the group, specifically chanting and meditation in these locations, would also correlate with anomalous deviations. Both hypotheses were confirmed, with a combined associated probability of 2.7×10⁻⁶.3 Additional data categories provided contextual information to help distinguish the sources of the anomalous effects. Nelson noted in the 2024 publication that he was not aware of independent attempts to replicate this specific sacred-sites paradigm in the intervening quarter-century, making the Egypt study an isolated positive result without independent replication as of the publication date. The original 1997 PEAR technical report is the primary archival source.6
Physical Vibration as an Unresolved Confound
The Egypt study’s most methodologically challenging artifact is the potential for physical vibration from group chanting to affect REG hardware outputs. Chanting in stone chambers produces acoustic energy that could mechanically perturb electronic components. The published paper does not report vibration-isolation tests or comparisons between chanting and non-chanting periods with the REG physically isolated from the floor or group. This artifact is unresolved. The pre-specified hypothesis structure partially addresses the concern, the hypotheses were stated before data collection, reducing post-hoc flexibility, but the absence of a vibration-control condition means the physical mechanism cannot be ruled out for the chanting-correlated deviations specifically. The sacred-site hypothesis (deviations in interior chambers regardless of chanting) is less vulnerable to this confound, since the group was not always chanting during chamber visits.3
Modern Context
The FieldREG-at-sacred-sites paradigm sits at the intersection of two mainstream methodological literatures: detection of small expected effects in noisy field environments, and the broader replication-and-power crisis in the behavioral sciences. Nelson’s pre-planned hypotheses and small-N field contexts (two-week Egypt study, ~125 ritual epochs) place this work directly in the regime where Cohen’s classical statistical-power framework is most consequential: small effects times small samples produce low power, and low-powered studies that do reach nominal significance carry inflated effect-size estimates and elevated false-positive rates.78 The Open Science Collaboration’s landmark 2015 replication effort, while focused on psychology rather than physical-systems anomalies, established the now-standard expectation that field-based behavioral findings carry meaningful replication uncertainty until independently reproduced — an expectation that applies to the sacred-site FieldREG paradigm, which Nelson notes has not yet been directly replicated.9 Environmental-confound literature in field-data analysis (vibration, ambient electromagnetic noise, temperature drift in REG hardware) is methodologically adjacent and addressed by Nelson’s pre-planned hypothesis structure and group-state-contingent analysis windows, though formal independent replication remains the standing methodological gap.
Skeptical Critiques and Discussion
Critique 1: Post-hoc classification of “resonant” vs. “mundane” venues introduces selection bias that inflates apparent effects
Skeptic source: A recurring methodological concern in the FieldREG literature is that the categorization of venues as “resonant” (predicted to show effects) versus “mundane” (predicted not to) was derived from the same exploratory dataset used to formulate the hypothesis, creating a circularity that could inflate the apparent effect in the formal replication series.1
Response: Nelson and colleagues addressed this concern directly in FieldREG II by explicitly deriving the formal hypothesis from 18 prior exploratory applications and then testing it in 21 new, independent formal replications, venues not included in the hypothesis-derivation set. The resonant subset of these new replications yielded a composite probability of 2.2×10⁻⁶, while the matched mundane subset produced 0.91, providing an internal control that would be unlikely under a pure selection-bias account, since both subsets were collected with the same hardware and experimenters.2 The selection-bias artifact is mitigated but not fully eliminated, since the a priori classification of new venues as resonant or mundane still required experimenter judgment before data analysis.
Analysis. The selection-bias critique concerns whether the resonant-versus-mundane venue classification, derived from the original exploratory dataset, biases the apparent effect in the formal replication series. Nelson and colleagues’ FieldREG II protocol cites a formal test of 21 new venues independent of the 18 used to derive the hypothesis, yielding composite probabilities of 2.2×10⁻⁶ for the resonant subset and 0.91 for the mundane subset. Whether independent researchers not involved in the original classification scheme would produce the same resonant/mundane assignments — and whether the contrast survives an independent replication — remains a methodological question that has not been addressed in the published literature.
Critique 2: Physical vibration from group activities (chanting, drumming, applause) may mechanically perturb REG hardware, producing artifactual deviations
Skeptic source: Because FieldREG devices are portable and deployed in close proximity to group activities, physical vibration from chanting, drumming, or collective movement could mechanically affect electronic components and produce systematic deviations in the random output that mimic a consciousness effect. This concern is particularly acute for the Egypt sacred-sites study, where chanting in stone chambers was a primary coherence-building activity.3
Response: The published FieldREG papers do not report dedicated vibration-isolation tests or comparisons between physically active and physically quiet periods with the REG isolated from the floor or group. This artifact is unresolved for the chanting-correlated deviations specifically. The sacred-site hypothesis, predicting deviations in interior chambers regardless of chanting, is less vulnerable to this confound, since the group was not always chanting during chamber visits, and the two hypotheses were pre-specified and confirmed jointly.3 The broader FieldREG II dataset includes many venues (conferences, theatrical performances) where physical vibration is not a plausible mechanism, and these also contributed to the resonant-subset composite probability.2
Analysis. The physical-vibration critique tests whether chanting, drumming, or collective movement could mechanically perturb REG hardware and produce artifactual deviations. Dedicated vibration-isolation tests are not reported in the FieldREG literature; the published response notes that the sacred-site hypothesis predicts deviations in interior chambers regardless of chanting, and that the broader FieldREG II dataset includes conference and theatrical venues where mechanical vibration is not a plausible mechanism. Targeted vibration-control experiments — particularly for the chanting-correlated Egypt results — remain an outstanding methodological item in the pool.
References
- Nelson, R. D., Bradish, G. J., Dobyns, Y. H., Dunne, B. J., & Jahn, R. G. (1996). FieldREG anomalies in group situations. Journal of Scientific Exploration, 10(1), 111-141. https://global-mind.org/rdnelson/fieldreg.html R001 [Nelson 1996] ↩︎
- Nelson, R., Jahn, R. G., Dunne, B., Dobyns, Y., et al. (2007). FieldREG II: Consciousness Field Effects: Replications and Explorations. EXPLORE, 3(3), 279–293. https://doi.org/10.1016/j.explore.2007.03.020 R002 [Nelson 2007] ↩︎
- Nelson, R. (2024). FieldREG Measurements in Egypt: Resonant Consciousness at Sacred Sites. Journal of Scientific Exploration, 38(4), 686–697. https://journalofscientificexploration.org/index.php/jse/article/view/3393 R003 [Nelson 2024] ↩︎
- Nelson, R. D., Bradish, G. J., Dobyns, Y. H., Dunne, B. J., & Jahn, R. G. (1998). FieldREG II: Consciousness field effects: Replications and explorations. Journal of Scientific Exploration, 12(3), 425-454. https://www.scientificexploration.org/docs/12/jse_12_3_nelson.pdf R004 [Nelson 1998] ↩︎
- Nelson, R., Bradish, G., Dobyns, Y., Dunne, B., et al. (2007). FieldREG Anomalies in Group Situations. EXPLORE, 3(3), 278–278. https://doi.org/10.1016/j.explore.2007.03.013 R005 [Nelson 2007] ↩︎
- Nelson, R. D. (1997). FieldREG Measurements in Egypt: Resonant consciousness in sacred sites. Princeton Engineering Anomalies Research. R006 [Nelson 1997] ↩︎
- Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences (2nd ed.). Lawrence Erlbaum Associates. https://doi.org/10.4324/9780203771587 R007 [Cohen 1988] ↩︎
- Button, K. S., Ioannidis, J. P. A., Mokrysz, C., Nosek, B. A., Flint, J., Robinson, E. S. J., & Munafò, M. R. (2013). Power failure: why small sample size undermines the reliability of neuroscience. Nature Reviews Neuroscience, 14(5), 365–376. https://doi.org/10.1038/nrn3475 R008 [Button 2013] ↩︎
- Open Science Collaboration. (2015). Estimating the reproducibility of psychological science. Science, 349(6251), aac4716. https://doi.org/10.1126/science.aac4716 R009 [OSC 2015] ↩︎
Deeper dives — Nelson: