Discovery Lab Answers

Serena M. Roney-Dougal, PhD is a British parapsychologist whose research centers on two main questions: whether advanced meditation or yogic practice correlates with measurable psi performance, and whether geomagnetic and environmental factors interact with psychic phenomena. Her profile page is at Serena M. Roney-Dougal.

Experiments

Ashram ESP experiments (2003–2004) — with Jerry Solfvin

Roney-Dougal and Solfvin ran free-response precognition trials at an ashram in India, testing whether yogic attainment — measured by a Yogic Attainment Questionnaire (YAQ) — correlated with psi scoring ^[3]. Participants were divided into groups: swamis (advanced practitioners), student-novices, and student-teachers. The first study (2003) used static picture targets; the second (2004) built on those findings.

Tibetan Buddhist studies — with Solfvin and Fox

A further study reported in 2008, co-authored with Solfvin and Fox, tested Tibetan Buddhist monks and students using a computerized free-response protocol (PreCOG software), again examining whether degree of meditation attainment — indexed by monastic rank, years of practice, and type of meditation — predicted psi performance ^[6].

Extended Tibetan study (2011) — with Solfvin

Roney-Dougal and Solfvin published a follow-up exploring the same relationship with a larger sample of Tibetan practitioners, incorporating additional measures including an attempted Stroop Effect attentional-focusing task ^[5].

International collaborative replication studies (2021) — with a large multi-site team

Roney-Dougal was one of 19 co-authors on two pre-registered replication studies of Daryl Bem’s time-reversed priming paradigm, involving 32 experimenters and over 1,000 participants across multiple countries ^[1]. Her institutional affiliation listed in that paper is the Psi Research Centre, Glastonbury, UK.

She has also written reviews for the Journal of Scientific Exploration, including a review of a book on Medjugorje apparitions ^[2] and one on Sathya Sai Baba ^[4], but these are not empirical studies.

Methodology

Free-response precognition protocol

The ashram and Tibetan studies all used free-response designs in which a participant attempts to identify a randomly selected target image before it is revealed by a computer. The PreCOG software handled target selection via a pseudo-random number generator, removing the need for an assistant and allowing Roney-Dougal to run sessions one-on-one with participants at mutually convenient times ^[6]. The normalized rating assigned to the target image defined the “psi” score, with zero as chance expectation ^[5].

Meditation attainment measurement

Attainment was operationalized in multiple ways across studies: years of meditation practice, monastic rank (student, monk, Rinpoche, Geshe, Lama), type of meditation practice (mantra-based vs. visualization-based), and a Yogic Attainment Questionnaire ^[3][5][6]. This multi-faceted approach allowed correlational analyses across several dimensions of practice.

Blinding and controls

Because targets were computer-selected, the experimenter had no knowledge of the target during the session. Participant mentation was recorded throughout, permitting both quantitative and qualitative analysis ^[5].

Pre-registered multi-site design (2021)

The Bem replication studies used pre-registered confirmatory analyses and a mixed-model statistical framework, with intraclass correlation coefficients used to assess experimenter-level variance ^[1].

Data

Ashram studies (2003–2004)

In 2003, the advanced swami group scored significantly better than the student-novice group (p = .05) and the student-teacher group (p = .04), in line with the hypothesis ^[3]. In 2004, these between-group differences became non-significant ^[3]. Overall results in both years did not differ significantly from chance expectation, and gender differences in scoring did not approach significance ^[3].

Tibetan Buddhist study (2008)

Advanced meditators showed a consistent psi-hitting direction across both studies, with the advanced group recording an overall 33% direct hit rate ^[6]. Younger participants showed a psi-missing tendency that distorted the correlation between years of practice and psi scores ^[6]. One participant — an ex-abbot who had practiced for 35 years — recorded a 50% direct hit rate individually, while three other participants (one Geshe and two Rinpoches) scored most strongly in the psi-missing direction ^[6].

Tibetan study (2011)

Overall psi scores did not exceed chance expectation: t(79) = 0.70, p = 0.49 (2-tailed), effect size r = 0.08 ^[5]. Meditation type (mantra vs. visualization) did not significantly differentiate psi performance ^[5]. The correlation between years of meditation practice and psi scores was in the predicted direction but not significant ^[5]. However, the two Lamas in the study scored significantly above chance: t(15) = 2.25, p = 0.04 (2-tailed), with an effect size reported in the source ^[5]. Average psi scores by rank were: Lamas 0.39 (SD = 0.69), Rinpoches 0.15 (SD = 0.88), Geshes −0.03 (SD = 0.93); the main effect for monastic rank was not significant (F(2,7) = 0.07, p = 0.93) ^[5].

Bem replication (2021)

The two pre-registered confirmatory analyses did not support the psi hypothesis ^[1]. An exploratory analysis in Experiment 1 found a significant effect among English-language participants but not among participants using translated versions; Experiment 2 produced mixed results ^[1]. Experimenter-level variance was near zero (ICCs < 1.88 × 10⁻¹⁵) ^[1].

Skeptical critiques

What critics argue.

Wagenmakers et al. (2011) argued that Bem’s original time-reversed priming experiments — the paradigm Roney-Dougal’s collaborative replication attempted to test — involved analyses that were partly exploratory rather than strictly confirmatory, and were underpowered in ways that could produce false positives ^[1]. Schimmack (2012) raised related concerns about statistical power ^[1]. These methodological objections motivated the pre-registered multi-site replication design that Roney-Dougal co-authored.

What the experimental data show.

The 2021 pre-registered confirmatory analyses by Schlitz, Bem, Roney-Dougal, and colleagues did not replicate the primary psi effect ^[1]. The mixed-model analysis found that almost zero variance resided at the experimenter level, and neither psi belief nor psi experience among experimenters or participants predicted outcomes ^[1]. The authors themselves note that the results are open to two competing interpretations: that a genuine psi effect existed in Bem’s original studies and something in the replication disrupted it, or that the original findings were artifacts of the statistical environment ^[1].

Within Roney-Dougal’s own meditation-attainment program, the 2011 Tibetan study found null results at the overall level ^[5], and in the 2008 study a psi-missing tendency among younger practitioners partially drove the observed correlations ^[6] — a pattern that the authors acknowledged as a complication in interpreting positive findings among advanced practitioners.

Analysis.

The null overall result in the 2011 study and the non-replication of 2003 between-group differences in 2004 ^[3] sit alongside the subgroup finding for Lamas in 2011 ^[5] and the 33% hit rate pattern in 2008 ^[6]. The Lama subgroup finding in 2011 involves a very small number of participants (two individuals), and no independent replication of that specific subgroup effect outside Roney-Dougal and Solfvin’s own program appears in the sources retrieved for this question. The 2021 collaborative replication, though negative on its confirmatory pre-registered outcome, was itself a methodological contribution to the field’s effort to subject Bem’s paradigm to rigorous multi-site testing ^[1].