Discovery Lab Answers

Robin K. Taylor, PhD is a parapsychologist and psychologist whose work spans laboratory psi research, cross-cultural fieldwork, and theoretical writing on evolutionary approaches to psi. He trained at the Koestler Parapsychology Unit (KPU) at the University of Edinburgh, completing his PhD under Robert L. Morris, and has since been based at the University of the South Pacific in Suva, Fiji, where he operates under the Oceanik Psi label,.

Before reading further: the studies currently in the ESP-Nexus library for this question represent a partial window on Taylor’s published output, and the library’s share of the broader published literature on this question has not been measured. What follows is a summary of what the library holds, not a settled account of the field.

His profile page is at Robin K. Taylor.

Experiments

Taylor’s two most directly relevant experimental contributions are both reported in a single 1996 paper in the European Journal of Parapsychology:

• Experiment 1 — Gymnasts (imagery training, PK): Gymnasts were assigned to more-imagery and less-imagery groups and performed a psychokinesis (PK) task using a random number generator (RNG) game engine with sport-themed feedback. The design tested whether imagery training could improve PK performance.
• Experiment 2 — Shooting athletes: A second experiment extended the same logic to a shooting-sport population, with the timing of imagery training relative to PK sessions differing from Experiment 1.

Taylor also co-authored a pilot replication study of remote mental influence on human electrodermal activity, published in the European Journal of Parapsychology in 1995, with Dean I. Radin and William G. Braud. That study tested whether one person’s mental intention could influence another person’s electrodermal (skin conductance) responses at a distance.

A third line of work is a security audit of the Edinburgh automated ganzfeld system, co-authored with Kathy Dalton, Robert Morris, Deborah L. Delanoy, Dean Radin, and Richard Wiseman — published in the Journal of Parapsychology — as part of the KPU’s response to methodological criticisms of earlier ganzfeld research [see the Delanoy page on Automated ganzfeld methodology and security].

Beyond the laboratory, Taylor has written on conducting parapsychological research in cross-cultural settings, drawing on his fieldwork in Fiji, and published a book review in the Journal of Scientific Exploration. A 2024 interview discusses his career trajectory from Edinburgh to Fiji.

Methodology

Imagery–PK studies: Both experiments used an RNG-based PK game engine — a well-documented approach in micro-PK research at the time — with animated sport-themed feedback. Athletes were divided into higher-imagery and lower-imagery groups. The pre-specified analysis was directional: the design committed in advance to comparing PK scores between groups, with the experimental group’s scores needing to exceed the control group’s to count as a positive result — independent of each group’s position relative to mean chance expectation. Scoring for statistical analysis differed from the feedback scoring shown to participants (which used FINA-style regulations for the gymnastics study). A notable design difference between the two experiments concerned when imagery training occurred relative to PK game sessions — in the gymnastics study, early PK sessions preceded any instruction to use imagery strategy.

The paper acknowledges that real-world demands on athlete participants (scheduling, attendance) had a marked effect on the proportion of sessions completed, which Taylor flags as a limitation of working with active competitive athletes.

Remote mental influence study: Radin, Taylor, and Braud used a 30-second data-collection window, with a 5-second analysis window chosen before examining window-size effects. They used a conservative “percentage score index” (PSI statistic) to enable direct comparison with prior published studies, notably Braud and Schlitz (1991). The study also examined a correlate — the relationship between geomagnetic activity (planetary Ap index) and session-level effect sizes.

Ganzfeld security audit: The Edinburgh automated ganzfeld paper addressed three artifact classes (sensory leakage and others) with specific procedural countermeasures, and offered recommendations for future automated ganzfeld settings.

Cross-cultural methodology: Taylor’s 2000 paper engages substantively with the practical and conceptual challenges of parapsychological research outside Western academic settings — participant worldviews, local belief structures, and their interaction with standard experimental protocols.

Data

The evidence table below this answer displays the four result rows the library holds. Key points in plain prose:

The direction of findings is not uniform across the available results. Taylor (1996) returned a positive direction in both the per-experiment result for gymnasts and the pooled result across both experiments. The remote mental influence replication (Radin, Taylor, and Braud, 1995) also returned a positive direction on a PSI-score one-sample comparison across 16 sessions. However, the pre-series calibration test reported by Dalton (1996) — a global randomness check of the target-generating system itself across a very large number of trials — returned a null result, as expected for a technical calibration rather than a psi test.

The metrics across rows are not comparable to one another: the Taylor (1996) results use Cohen’s d as a standardized effect size, while the Radin, Taylor, and Braud (1995) result uses Pearson’s r. These cannot be averaged or pooled into a single summary figure, and the table should be read row by row rather than as a single aggregate.

The gymnast experiment (Experiment 1) produced a Cohen’s d in the positive direction at a p-value that did not cross the conventional threshold. Taylor himself, in discussing the results, noted that imagery-group comparisons partially replicated findings from Braud (1983) and George (1982), and raised the interpretive question of whether motivation — rather than imagery skill as such — might be the active factor driving group differences,. Changes in self-reported visual imagery ability over the training period correlated with PK scores in the shooting study, partially replicating a Braud (1983) finding.

The unsettled signals the evidence table flags are real: directions disagree across rows, and one result is null (the calibration condition).

Skeptical critiques

The sources retrieved for this question do not include published criticism directed specifically at Taylor’s work. What the sources do record is that the imagery–PK paper situates itself explicitly against the possibility of a motivation confound — Taylor himself raises the question of whether group differences in PK scores reflect imagery skill or athlete motivation — and that the Edinburgh automated ganzfeld security paper in which he was a co-author was itself a response to methodological criticisms of earlier ganzfeld procedures raised in the early 1990s. Independent replication of the imagery–PK findings outside the original group has not been documented in the sources retrieved here.