Nelson’s theoretical argument in the 1999 PEAR Technical Report draws on the observation that REG anomalies in laboratory settings consistently appear when human operators focus intention on the device. Healing modalities, intercessory prayer, therapeutic touch, qigong, shamanic healing, share the structural feature of a practitioner directing focused intention toward a target system (the patient) with the explicit goal of reducing disorder or restoring homeostasis. If the mechanism underlying REG anomalies involves some form of consciousness-correlated order-induction, then healing sessions constitute a naturalistic instance of the same process. The report surveys biological-target healing studies alongside REG studies, noting that both literatures show small but statistically detectable effects. The proposed mechanism remains contested; Nelson frames it as a possible information-theoretic or negentropic process rather than a conventional energetic one, and explicitly acknowledges that no consensus physical mechanism has been identified.¹

The concept of negentropy (negative entropy, or increased order) appears in Nelson’s theoretical framework as the measurable correlate of healing intention. In the REG context, negentropy is operationalized as a departure from the expected random distribution, a statistically significant clustering or ordering of bit sequences that would not be expected by chance. Nelson’s 1999 review notes that this framing connects the REG anomaly literature to healing traditions across cultures, all of which describe the healer’s role as restoring order, balance, or homeostasis to a disordered system. The review explicitly cautions that this is a conceptual model, not a confirmed mechanism, and that the laboratory-to-clinic translation involves important differences in experimental control, blinding, and outcome measurement.¹ The term “energy” in “energy medicine” is noted as a misnomer from the physicist’s perspective, practitioners describe felt sensations that do not correspond to any currently measurable energetic field, and the REG approach is proposed as a potentially more sensitive detector of whatever process is actually occurring.⁴

The Crawford et al. 2003 study (published in the Journal of Alternative and Complementary Medicine) used two REGs running in parallel: one in the bioenergy healer’s office in Bethesda, MD, and one at a local library as a control. Two multiday data sets were collected in each setting, plus a third set in which the healer deliberately reduced attention to the REG. The primary finding was that the office REG produced excursions outside the 95% CI on 47 of 51 days (92%), compared with 35 of 61 days (58%) for the library REG, a mean difference of 34% (95% CI: 18%–49%; χ²=16.3, 1 df, p<0.0005). However, within the healer’s office, the comparison of healing phases (0.6496 excursions/segment) versus non-healing phases (0.6548 excursions/segment) showed no significant difference (t=−1.3, p=0.182). A comparison with Monte Carlo chance expectation showed significantly fewer mean excursions during healing phases (t=−7.8, p<0.0005) and no difference during non-healing phases. The high-attention versus low-attention comparison also showed no significant difference. The authors note that the overall office-vs-library effect is robust, but the within-office phase comparisons are inconsistent, a pattern that complicates causal attribution to healing intention specifically versus some ambient property of the healer’s environment. Sensory leakage between conditions was mitigated by physical separation of the two REGs (different buildings), but experimenter expectancy effects on data reduction were not fully addressed. Multiple comparisons across the three data sets were partially addressed by the authors but not corrected with a pre-specified family-wise error rate.²

The Lumsden-Cook et al. 2006 study (Journal of the Society for Psychical Research) tested individual izangoma who held the REG and focused healing intention on it in one condition, and held the device without attending to it in another. Control periods and experimenter-effect trials were also conducted. The healing condition produced a non-directional REG anomaly: χ²(80)=113.023, p=.009. None of the other conditions, non-attended holding, control periods, or experimenter-effect trials, reached significance. The non-directional test was pre-specified (the researchers did not predict the direction of the REG deviation, only that a deviation would occur), which is appropriate given the theoretical uncertainty about whether healing intention would increase or decrease the REG’s mean output. The study’s sample size (number of healers and sessions) is not fully specified in the available extract, which limits power assessment. The specificity of the effect to the healing condition, with null results in the matched non-healing and control conditions, partially addresses the alternative explanation of a general experimenter or environmental effect, though demand characteristics (healers knowing the purpose of the study) were not fully controlled. The study is cited in the Nelson pool as a conceptual replication of the broader healing-REG paradigm.³

The Carpenter et al. 2020 study (EXPLORE, Institute of Noetic Sciences) used a custom quantum noise generator (QNG) recording 16 independent channels of random samples produced by electron tunneling and avalanche effects in Zener diodes, digitized at 1 kHz. Two metrics were computed: a temporal-dependency metric (autocorrelation structure across the 16 channels over time) and a spatial-dependency metric (cross-channel correlations at a given moment). These were combined into a single “spacetime” negentropy variable. Across 110 half-hour energy medicine sessions, the QNG data showed significant deviations from chance expectation, with a peak at 24 minutes (z=4.24, p<0.00003, two-tailed). Deviations associated with p<0.05 were observed from 20 to 29 minutes into the sessions, after correction for multiple comparisons across the time-series. Control recordings taken eight hours after each session, in the same laboratory, with no one present, showed uniformly null results. The multiple-comparisons artifact was partially addressed by applying a correction to the time-series analysis, though the specific correction method and its family-wise error rate are not detailed in the available extract. The alternative explanation of a circadian or diurnal environmental artifact (e.g., electromagnetic interference patterns that differ between session time and the 8-hour-later control) was noted by the authors as a counterargument requiring future investigation. The study was described as a conceptual replication of earlier reports, including the Crawford et al. 2003 bioenergy healer study, and the authors recommend future studies with tighter environmental controls and preregistered hypotheses.⁴

Nelson’s 2025 paper on the subterranean chamber of the Pyramid of Khufu reports a FieldREG recording in the “Pit” chamber that showed a significant anomalous response, interpreted as consistent with the chamber’s possible function as a ritual or sacred space.⁷ This finding is methodologically adjacent to the healing-REG literature: both involve a REG placed in a context of focused intention or ritual activity, and both interpret deviations as possible consciousness-correlated effects. The Egypt FieldREG study (1997 data, published 2024) used pre-planned hypotheses predicting anomalous deviations at sacred sites and during coherence-building activities; both hypotheses were confirmed with a combined probability of 2.7×10⁻⁶ across the full two-week study.⁵ Nelson notes in the 2024 publication that he is not aware of direct replications of the sacred-site FieldREG paradigm, though related applications, including healing-session REG monitoring and group-meditation REG studies, have been conducted by other researchers.