Discovery Lab Answers

James C. Carpenter, PhD — Experiments and Data

James C. Carpenter is a parapsychologist whose research career spans roughly six decades, centered on the psychological and unconscious determinants of ESP performance, and culminating in the development of First Sight Theory (FST), a comprehensive theoretical framework proposing that psi operates as a continuous, preconscious process underlying ordinary perception and behavior. His ESP-Nexus profile is at James C. Carpenter, PhD.

Experiments

Carpenter’s experimental record covers several distinct lines of work, which the sources retrieved for this question illuminate at varying depth.

Mood-state and ESP variance studies (1960s–1990s). Beginning in the mid-1960s, Carpenter ran a series of forced-choice ESP experiments probing whether measurable psychological states — particularly mood — could predict ESP run-score variance and hitting rate. An early exploratory study on anxiety proneness appeared in 1965, followed by the development of a mood-adjective scale to predict run-score variance ^[5]. This program continued across multiple studies and produced the Part III series (1991), in which three separate attempts were made to retrieve coded information using mood reports and a repeated-guessing technique ^[5].

Practical application pilot studies (2010). Two pilot studies were designed to test whether laboratory-derived ESP effects could be applied to real-world prediction tasks — specifically financial or event-outcome forecasting. These studies used majority-vote analysis and repeated-guessing techniques as signal-amplification strategies ^[4].

Preference formation and unconscious ESP (2021). A two-study series conducted with Christine Simmonds-Moore, Steve Moore, and Ferrell Carpenter tested whether extrasensory stimuli presented subliminally could influence participants’ preferences — a direct empirical test of FST’s claim that psi shapes preconscious processing ^[2].

Group PK observation (2024). A later paper describes Carpenter’s participation as both experimenter and subject in a group context organized by James McClenon’s group, observing purported psychokinetic effects with paper objects named “Alice” and “Lefty,” and connecting those observations to FST predictions about affect and PK expression ^[1].

Book review / theoretical commentary (2020). Carpenter also contributed a review-essay on Richard Reichbart’s The Paranormal Surrounds Us, engaging psychoanalytic and literary perspectives on psi ^[3].

Methodology

Forced-choice ESP design with psychological predictors. The core methodology of Carpenter’s early work used standard forced-choice ESP tests administered to undergraduate participants, with pre-specified psychological scales — mood-adjective checklists, a Heuristic scale (H), a Variance scale (V) — used to predict, before scoring, whether a given participant’s run would show above-chance hitting or large run-score variance. The logic was to demonstrate not just psi but predictable psi: if psychological variables reliably sort participants into high- and low-scoring groups, that constitutes evidence for systematic rather than random anomalous effects ^[5].

Index target sampling and repeated guessing. A methodological innovation across several studies was the use of “index targets” — a subset of calls analyzed early — to predict the scoring direction of the remainder of a run. The repeated-guessing technique, drawn from Thouless (1960), involves having participants make multiple calls at the same target and aggregating them by majority vote, amplifying a weak above-chance signal into a more reliable decision ^{[4] [5]}.

Subliminal and extrasensory priming (2021). The 2021 preference studies presented stimuli either subliminally (below conscious threshold) or extrasensorially (outside normal sensory range altogether), with mood manipulated via implicit primes. Participants completed personality measures — Need for Cognition, Tolerance for Merger, NEO scales — which served as predictors of psi-relevant responses. Stepwise multiple regression was used post hoc in both studies to identify independent predictors of preference scores ^[2].

Group observational design (2024). The 2024 paper represents a departure from controlled laboratory protocol toward a naturalistic group setting, with Carpenter occupying a dual role as experimenter-participant. The paper is explicitly exploratory in character ^[1].

Data

No structured quantitative evidence block was supplied for this question, so the figures below come exclusively from the retrieved primary sources, reported only where the surrounding text is intact enough to make attribution unambiguous.

Mood-prediction / forced-choice studies ^[5]. In Study 1 of the 1991 series, extreme-quartile predictions on the V scale produced a predicted-large-variance group with a mean run score of 6.40 (N = 200 runs, z = 0.62) and a predicted-small-variance group with a mean of 4.94 (N = 170 runs, z = 1.32); the difference between the two was marginal (p = .06, one-tailed). On the H scale, the predicted-psi-hit group produced a mean run score of 12.36 (N = 280 runs, z = 2.12, p = .017, one-tailed) against a predicted-psi-miss group mean of 11.76 (N = 210 runs, z = 1.15, p = .125); the difference between the two group z-scores was significant (z_diff = 2.14, p = .016, one-tailed). Repeated-guessing analysis showed the V scale yielding 51.3% correct votes and the H scale 52.6% correct votes, with combined-vote correct decisions reaching 92% (χ² = 8.33, df = 1, p = .004) ^[5].

Practical application pilots ^[4]. The 2010 pilot studies demonstrated that a 60% base rate of above-chance scoring could be amplified by majority-vote analysis to 100% correct decisions in a simulated exercise, and to 92% correct (χ² = 8.33, p = .004) in the investment-prediction pilot. The paper explicitly flags that the first attempt failed because the overall scoring rate was not reliably above chance — a necessary condition the procedure requires ^[4].

Preference formation studies ^[2]. In Study 1, neither subliminal nor extrasensory primes produced an overall preference shift significant at p < .05 (both means near MCE). However, a “dynamic activation effect” emerged: t = 2.29, p = .012 (one-tailed), Cohen’s d = 0.52. The correlation between subliminal and extrasensory preference scores was virtually nil (r = −.004), and the predicted positive relationship was not confirmed. In Study 2 (N = 94), of 9 variables predicted to relate to extrasensory pre-exposure, 7 were significant at the specified threshold. A post-hoc stepwise regression identified Need for Cognition as the sole independent significant predictor of subliminal-pre-exposure scores (original r = .214, p = .03) ^[2].

Across the record, the pattern is one of modest positive effects conditional on psychological moderators, several significant findings that did not survive straightforward replication without additional variables, and at least one explicit null (no main effect of extrasensory priming on preference) alongside significant interaction and individual-difference effects. The 2024 group PK paper does not report quantitative outcomes in the retrieved excerpt ^[1].

Skeptical critiques

The sources retrieved for this question do not include published criticism directed specifically at Carpenter’s work. Carpenter’s own 1991 paper notes that an earlier attempt to cross-replicate the mood-variance relationship “failed to cross-replicate,” leading the research program to forswear theory and move toward “blind empiricism” for a period ^[5], and the 2010 paper acknowledges that the first practical-application attempt failed because the necessary above-chance base rate was absent ^[4] — both are limitations the author himself records. No named critic with a specific published objection appears in the retrieved sources.