Discovery Lab Answers

Julie Beischel, PhD is a parapsychologist and co-founder of the Windbridge Institute for Applied Research in Human Potential (later the Windbridge Research Center), whose work centers on controlled laboratory testing of mediums. Her ESP-Nexus profile is at Julie Beischel, PhD.

Experiments

Beischel’s core experimental program tests whether mediums can report accurate, specific information about deceased individuals under conditions that rule out conventional explanations — cold reading, fraud, and experimenter cueing. The main lines of work retrieved here are:

• Blinded mediumship accuracy studies — a series of controlled experiments in which Windbridge Certified Research Mediums (WCRMs) provided readings under triple-blind conditions. An initial study and its replication and extension are reported in Beischel (2015) ^[5] and summarized in Beischel (2021) ^[3]. Collaborators across the series include Mark Boccuzzi, Michael Biuso, and Adam J. Rock ^[5].

• Survival psi vs. somatic psi phenomenology study — Beischel, Rock, Pekala, and Boccuzzi (2021) compared mediums’ psychological state scores (using the Phenomenology of Consciousness Inventory, PCI) across four counterbalanced conditions: baseline, control, readings for living targets, and readings for deceased targets, to test whether distinct phenomenological signatures differentiated the conditions ^[2].

• Crisis impressions review — Beischel co-authored (with Patricia Pearson and Julia Mossbridge) a 2023 historical and conceptual review of crisis impressions — perceptual anomalies occurring at the time of a distant person’s death or emergency — published in Threshold: Journal of Interdisciplinary Consciousness Studies ^[1].

• WPRPAS instrument development — Beischel and Boccuzzi (2020) developed and deployed the Windbridge Psi and Related Phenomena Awareness Scale (WPRPAS), a self-report instrument for measuring awareness of and experience with psi and related phenomena, administered to samples of mediums and non-mediums ^[4].

• Bigelow Institute essay — Beischel (2021) contributed a formal evidence review, Beyond Reasonable: Scientific Evidence for Survival, to the Bigelow Institute for Consciousness Studies, synthesizing results across the Windbridge experimental program ^[3].

Methodology

The central methodological contribution of the Windbridge program is the triple-blind protocol for mediumship research. Its key design features, as described across the retrieved sources ^[3][5]:

• Blinding at the reading stage: mediums are given only a discarnate’s first name. They have no contact with the sitter and no access to identifying information. Readings are conducted by phone, separating medium, experimenter, and sitter across different locations ^[5].
• Blinding at the scoring stage: sitters score their own reading against a randomly selected decoy reading from another sitter’s discarnate, without knowing which is which. Raters are thus blind to target identity ^[3][5].
• Blinding at the analysis stage: in at least some analyses, raters scoring free-form content were also blinded to condition ^[5].
• The WCRM certification program: mediums are pre-screened through a multi-stage process before participating in research, standardizing the participant pool and reducing variability attributable to medium selection ^[3][5].
• Counterbalancing: in the survival psi vs. somatic psi study, the four conditions (baseline, control, living-target reading, deceased-target reading) were administered in counterbalanced order across participants to control for order effects ^[2].
• PCI administration: the Phenomenology of Consciousness Inventory was administered at baseline and after each condition, enabling quantitative comparison of altered-state dimensions (hypnoidal state, imagery vividness, absorption, volitional control, self-awareness, love/caring) across conditions ^[2].
• Bonferroni correction: the 2021 phenomenology study applied Bonferroni-corrected alpha (α = .002) to control for multiple comparisons across PCI dimensions ^[2].
• Item-level and global scoring: the accuracy studies used both fine-grained item-level scoring (Five-Questions and Free-Form sections) and holistic global scores (0–6 scale), and analyzed hits-to-misses ratios via chi-squared tests ^[3][5].

For the WPRPAS instrument ^[4], Beischel and Boccuzzi intentionally avoided loaded or leading terminology (e.g., “alleged,” “purported”) in item wording, and compared medium (n = 316) and non-medium (n = 1,030) samples matched for demographic characteristics.

Data

The table below consolidates the quantitative outcomes reported in the retrieved sources. The reader should note that these sources represent a sample of Beischel’s published work retrieved for this question, not a complete accounting of the Windbridge program’s output. Because the metrics across studies are not comparable (item accuracy percentages, global scale scores, chi-squared statistics, PCI dimension scores, and a binomial proportion), they are presented separately and not pooled.

Accuracy studies (blinded readings — target vs. decoy comparisons)

The Five-Questions estimated accuracy comparison (47.4% vs. 35.6%) did not reach significance at p = .09 ^[5] — this is a null result within the series and should not be overlooked.

Survival psi vs. somatic psi study (phenomenological conditions)

From Beischel, Rock, Pekala, and Boccuzzi (2021) ^[2], under blinded, randomized, counterbalanced conditions with Bonferroni-corrected α = .002:

• In 74% of readings, the blinded WCRM’s impressions accurately reflected whether the target was living or deceased (p = .032) ^[2].
• Hypnoidal state scores were significantly higher for both reading conditions (living and deceased) than for baseline and control conditions ^[2].
• Scores for the PCI dimension love/caring were significantly higher for deceased-target readings than for living-target readings: 3.65 ± 0.63 vs. 2.75 ± 1.18; t(9) = 2.78, p = .02 (two-tailed) ^[2].
• Volitional control scores were lower for both reading conditions than for baseline ^[2].
• Global accuracy scores for living-target readings were significantly negatively correlated with imagery vividness (r^[6] = −0.76, p = .027) and absorption (r^[6] = −0.74, p = .037) ^[2].
• Global accuracy scores for deceased-target readings were not significantly correlated with any PCI dimension ^[2].
• The correlation between positive affect and accuracy differed significantly between the deceased-target and living-target conditions (z = 1.92, p = .027, one-tailed) ^[2].
• The primary hypothesis — that PCI scores would differ significantly across all four conditions — was not uniformly supported across all dimensions; significant differences were found for specific PCI dimensions (those in bold per the paper’s Table, corrected α = .002), not globally ^[2].

WPRPAS instrument study

From Beischel and Boccuzzi (2020) ^[4], comparing mediums (n = 316) and non-mediums (n = 1,030) with no significant demographic differences:

• More than 80% of both samples reported awareness of 8 of the 10 phenomena assessed.
• The proportion of mediums aware of microPK and macroPK was significantly larger than that of non-mediums (each p < .00001); no significant differences existed for the remaining eight phenomena ^[4].

Skeptical critiques

What critics argue. Muir (2015) published a critical Amazon review of Beischel’s book Among Mediums, characterizing the science as “seriously flawed” ^[3]. The citation for this appears in Beischel’s own Bigelow essay, where it is listed alongside a skeptical popular work by Hyman and a work by Moncrieffe (2000) titled The Naked Quack ^[3] — indicating that Beischel herself engages with critical assessments in her published writing.

What the experimental data show. The blinded protocols in the Windbridge series were specifically designed to address the most common methodological objections to mediumship research — cold reading, fraud, and experimenter cueing — by separating medium, sitter, and experimenter across locations and conducting all communication by phone ^[5]. The replication study ^[5] reproduced the significant hits-to-misses ratio across both Five-Questions and Free-Form item categories, though the estimated item-accuracy comparison in that same study did not reach significance (p = .09) ^[5]. The phenomenology study ^[2] used Bonferroni correction specifically to guard against false positives arising from multiple comparisons across PCI dimensions.

Analysis. The specific content of Muir (2015)’s methodological objections and of the other critical works cited in ^[3] is not reproduced in the retrieved excerpts, so the details of those arguments cannot be stated here. What the sources do show is that Beischel treats this critical literature as part of the relevant evidence landscape, engaging with it in her published work. The question of whether the blinding protocols fully exclude all alternative explanations — including what the sources call “somatic psi” (retrieval of information about the deceased through psi applied to living persons, rather than communication with the deceased) — is framed by Beischel herself as unresolved by the accuracy data alone ^[2][3]. The 2021 phenomenological study was explicitly designed to generate evidence relevant to the survival-psi vs. somatic-psi distinction, and its authors characterize the findings as exploratory rather than confirmatory ^[2].