Michael A. Thalbourne experiments and data

Michael A. Thalbourne, PhD

Michael A. Thalbourne was an Australian parapsychologist whose career centered on developing psychometric instruments to measure individual differences in paranormal belief and cognition, then systematically examining how those differences predicted psi-task performance. His most influential contributions were the Australian Sheep-Goat Scale (ASGS) and the Transliminality Scale, and he used both as independent variables in correlational and experimental designs across several lines of work. His ESP-Nexus profile is at Michael A. Thalbourne, PhD.

Experiments

Australian Sheep-Goat Scale studies. Thalbourne developed the ASGS beginning in the early 1980s and ran or analyzed a substantial corpus of studies using it as a predictor of psi performance [1]. The most consistent finding across that corpus was the sheep-goat effect (SGE): participants who believe in ESP (sheep) tend to score above chance in forced-choice psi tasks, while disbelievers (goats) tend to score at or below chance. Within the studies he reviewed, the effect appeared most reliably in telepathy paradigms — experiments using pairs of people attempting to transmit information — where 6 out of the relevant studies reached significance compared with 1 out of 15 in other modalities [1]. He also ran specific ASGS investigations with clinical populations, comparing manic-depressive and schizophrenic participants against community samples [1].

Gambling task / ESP study with Storm. Thalbourne co-investigated, with Lance Storm at the University of Adelaide’s Anomalistic Psychology Research Unit, a study examining whether intuition type and attitudes toward gambling predicted ESP performance in a card-locating gambling paradigm [2]. The design separated participants into pro- and anti-gambling attitude groups and tested whether compliant versus noncompliant orientations modulated psi scoring.

I Ching divination studies. Thalbourne collaborated with Storm on I Ching divination tasks, using the Revised Transliminality Scale (RTS) as a predictor variable. Storm (2013) directly builds on the Storm and Thalbourne (2001) protocol and scoring format, making the earlier joint work the methodological foundation for subsequent replication attempts [5].

Creative personality and paranormal belief. Thalbourne conducted and synthesized a series of studies examining the correlation between creative personality and paranormal belief, assembling data from eight studies including his own collaborative work with Delin (1994, 1995) and others [3].

Unified ESP-PK theory (psychopractic model). Thalbourne developed and published a theoretical monograph, The Common Thread between ESP and PK (2004), arguing that ESP and psychokinesis are manifestations of a single underlying process he termed “psychopractic causation” [4]. This was primarily theoretical, drawing on accumulated experimental evidence rather than reporting new data.

Transliminality Scale development. Thalbourne initiated the Transliminality Scale in 1991, running validation studies across multiple samples, and later collaborated with Lange, Houran, and Storm on the Revised Transliminality Scale (RTS) through Rasch analysis. He also co-investigated correlates of transliminality including temporal lobe lability, creativity, mystical experience, and manic-depressive traits across numerous published studies.

Methodology

Thalbourne’s experimental approach was primarily correlational and psychometric. He designed questionnaire instruments — the ASGS and the Transliminality Scale — as independent variables, then either ran psi tasks himself or examined how scores on those instruments correlated with performance in tasks run by other researchers.

ASGS design. The scale measures belief in telepathy, clairvoyance, precognition, and psychokinesis. It was validated for test-retest reliability over an approximately 11-month interval (r = 0.66, p < 0.0005, n = 29) and showed a parallel-form reliability of 0.85 across 301 participants when forced-choice and visual analogue versions were compared [1]. Response-bias investigations using the Marlowe-Crowne Social Desirability Scale found zero correlation, addressing the concern that sheep-goat scores simply reflect acquiescence or social desirability [1].

Gambling task protocol. The Storm and Thalbourne (2005) study used a forced-choice card-locating task embedded in a simulated gambling context: participants tried to locate the Ace of Spades in a five-card face-down hand (chance = .20), with an Ace of Clubs serving as a decoy to test whether significant hitting on the target was specific rather than general [2]. Participants were pre-classified on gambling attitude (GAS subscales) and on intuition type. Omega-squared (ω²) was used to estimate effect sizes, with the threshold for practical significance set at 6% of variance explained [2].

I Ching protocol. The Storm-Thalbourne I Ching paradigm required participants to generate a hexagram by throwing three coins six times, with a hit defined as matching 1 of 16 possible hexagrams (chance = .25). Experimenters and participants were classified by transliminality and paranormal belief, creating High-High, High-Low, Low-High, and Low-Low conditions [5].

Psychometric validation standard. Thalbourne consistently applied factor analysis (with 100 participants recommended as a minimum) and Rasch analysis to scale development, and he collaborated with Lange and Houran on the RTS to improve measurement precision.

Data

No structured quantitative evidence block was provided for this question, so the figures below come directly from the retrieved primary sources and are cited accordingly.

Sheep-goat effect corpus — ASGS studies [1]:

ComparisonSignificant studiesNonsignificant studies
Telepathy paradigms68
All other modalities combined114

Thalbourne’s analysis of this pattern in [1] notes that the SGE was most reliably found in telepathy experiments; formal χ² testing was not applied due to small expected frequencies, so the comparison is descriptive.

Test-retest reliability for the ASGS: r = 0.66 over ~11 months (n = 29, p < 0.0005); parallel-form reliability r = 0.85 (n = 301) [1].

Creative personality and paranormal belief [3]:

StudyNr (or equivalent)p
Thalbourne & Delin (1994) — Students241.38< .001
Thalbourne & Delin (1994) — Manic-depressives86.35< .001
Thalbourne & Delin (1994) — Schizophrenics38.43.010
Thalbourne & Delin (1995) — Students123.42.001

These correlations between creative personality and paranormal belief were consistently positive and statistically significant across student and clinical samples [3].

Gambling task / ESP study [2]:

The effect of group affiliation (pro- vs. anti-gambling attitude) on ESP and intuition measures explained approximately 4% of variance (ω² ≈ .04), falling short of the pre-specified 6% threshold the authors treated as practically significant. Only the “moderate” gambling-attitude group showed notable patterns on retesting [2].

I Ching divination [5]:

The overall hit rate on First Hexagrams was 34 out of 60 (56.7%), exceeding the 25% chance level (p < .001). The High-High transliminality/paranormal-belief condition produced 13 hits out of 20 (65%, p < .001) [5]. Transliminality and paranormal belief were highly correlated in this sample: for participants, r(58) = 0.95 (p < .001); for experimenters, r(58) = 0.92 (p < .001), which the authors note creates an interpretive challenge in separating the two predictors [5].

Null and non-significant results. Within the ASGS corpus, the majority of non-telepathy studies were nonsignificant (14 out of 15) [1], and the gambling task failed to meet its pre-specified effect-size threshold [2]. These null and below-threshold findings are part of the record.

Skeptical critiques

What critics argue.

Roberts and Hume (2007) reviewed Thalbourne’s The Common Thread between ESP and PK monograph [4] and identified two potential problems Thalbourne himself acknowledged: the theoretical framework depends on the reality of psi phenomena that have not been established to mainstream scientific satisfaction, and the “psychopractic causation” model risks being unfalsifiable if the criteria for what counts as a “pro attitude” are not independently specified. The reviewers note these as the monograph’s most significant vulnerabilities [4].

On the sheep-goat effect more broadly, Thalbourne himself reported in [1] that the predictive validity of the ASGS holds most clearly for telepathy paradigms and is inconsistent across other ESP modalities, and that the 14 nonsignificant non-telepathy studies outweigh the 1 significant one — a pattern that raises the question of whether the SGE generalizes or is paradigm-specific.

The I Ching study [5] records a confound that Storm explicitly flags: transliminality and paranormal belief correlated at r = 0.95 in the participant sample, making it effectively impossible to determine which variable is driving any observed psi-scoring differences. This is a methodological limitation internal to the study, not merely an external critique.

What the experimental data show.

The ASGS corpus shows a consistent positive relationship between belief in ESP and above-chance scoring in telepathy tasks, with the opposite direction (below-chance) for goats [1]. The effect appears weaker or absent outside telepathy paradigms [1]. Creative personality correlates with paranormal belief across student and clinical samples at moderate magnitudes [3]. The gambling task did not reach the pre-specified effect-size threshold [2]. The I Ching paradigm showed above-chance hit rates, but the near-perfect correlation between the two primary predictors limits causal interpretation [5].

Analysis.

The core methodological questions that remain unresolved in the retrieved sources are: whether the sheep-goat effect is genuinely paradigm-specific (telepathy vs. other modes) or whether the pattern reflects sampling variability; whether transliminality and paranormal belief can be meaningfully separated as predictors in designs where they correlate at r > 0.90; and whether the psychopractic model generates independently testable predictions beyond the existing psi-task literature. Roberts and Hume (2007) raised the falsifiability concern directly [4], and Storm (2013) documented the collinearity problem empirically [5], but neither issue has been addressed in the retrieved sources through subsequent experimental work specifically designed to resolve it.

References
  1. Thalbourne, M. A. (2010). The Australian Sheep-Goat Scale: Development and Empirical Findings. Australian Journal of Parapsychology.
  2. Storm, L., & Thalbourne, M. A. (2005). The Effects of Intuition and Attitudes Towards Gambling on ESP Performance During a Gambling Task. European Journal of Parapsychology, 20, 22–49.
  3. Thalbourne, M. A. (2005). Creative Personality and Belief in the Paranormal. European Journal of Parapsychology, 20, 79–84.
  4. Roberts, B. L., & Hume, I. R. (2007). A review of “The Common Thread between ESP and PK (Parapsychological Monograph Series No. 19)”, by Michael A. Thalbourne (2004). European Journal of Parapsychology, 22, 193–199.
  5. Storm, L. (2013). Towards a Replicable Formula for Significant I Ching Outcomes. Australian Journal of Parapsychology.
Deeper dives on ESP-Nexus

Ask another question