Discovery Lab Answers

Deborah L. Delanoy, PhD — Experiments and Data

Before drawing any conclusions, a coverage qualification is needed: the studies currently in the ESP-Nexus library represent an unknown share of the published literature on Delanoy’s work. The library’s coverage of that literature has not been measured, so what follows is a summary of what the library holds, not a settled account of the field.

The evidence base for this question consists of two audited study rows, reporting results on different metrics — one a Pearson r effect size from a DMILS study, the other Cohen’s h from a ganzfeld study. Because the metrics differ, no pooled figure is meaningful; the studies are described separately below.

Experiments

Delanoy’s experimental record spans several distinct lines of work, each contributing to her core concern: whether ESP and remote-influence studies can be made secure and repeatable enough to support process-oriented research.

Free-response ESP training. With Robert L. Morris and Caroline A. Watt, Delanoy ran a multi-session training study examining whether participants given access to relaxation, imagery, and focusing techniques could improve free-response ESP performance. The design included preliminary, training, and follow-up sessions, with formal ESP testing at each training session.

Out-of-laboratory free-response testing. With Watt, Morris, and Richard Wiseman, Delanoy developed and tested a new free-response methodology allowing participants to conduct trials outside the laboratory, using tamper-evident security bags and independent randomization. The study targeted experienced participants who had previously taken part in pilot or training work.

DMILS with shielded environments. With Morris, Delanoy ran a 36-session DMILS (direct mental interaction with living systems) study at the University of Edinburgh and the Institut für Grenzgebiete der Psychologie und Psychohygiene (IGPP) in Freiburg, using two shielded rooms to prevent sensory contact between agent and receiver. The study included a training component for the experimenters themselves, who also served as agents and receivers.

Ganzfeld: sender/no-sender comparison. Morris, Kathy Dalton, Delanoy, and Watt ran a study at the Edinburgh automated ganzfeld facility comparing three sender conditions: no sender, one sender, and two senders, across 97 pre-specified trials. Delanoy served as both experimenter and sender.

Target emotionality and DMILS physiology. With Sunita Sah, Delanoy tested whether remotely held positive emotional states produced different physiological (electrodermal) and cognitive psi responses in a receiver compared with neutral states.

Psi-conducive practices review. Delanoy conducted a survey of six parapsychological laboratories, examining conditions and practices associated with positive psi results, with a particular focus on environmental and interpersonal factors.

Methodology

The training study used a within-subjects design with twelve training sessions framed around individual differences in learning style, alternating imagery-based and concentration-based exercises. Randomization was performed prior to the study by an independent party using published weather data as an entry point into a standard random number table. Lab and remote ESP tests were included, with both holistic and atomistic judging.

The out-of-laboratory methodology was designed around tamper-evident security bags, allowing participants to work in their own environments and at times of their choosing. Independent randomization governed target assignment. Non-staff participants’ data were pre-designated as the primary outcome measure; experimenter-contributed trials were reported separately for completeness.

The DMILS study used shielded environments at two institutions to prevent any sensory contact between agent and receiver. Agents followed computer-generated instructions to alternate between “activate” and “calm” mental states directed at the receiver, whose electrodermal activity (EDA) was continuously recorded. A release-of-effort window (the period immediately following each instruction period) was also analyzed. The study distinguished between sessions run by trainee experimenters acting as their own agents/receivers and sessions involving non-experimenter participants.

The ganzfeld study used the Edinburgh automated ganzfeld system, with computer-controlled target selection, randomization, and feedback delivery. Three sender conditions were randomly assigned, with 32 pre-specified trials per condition (yielding 97 trials total due to random assignment of the first two conditions). The automated design was intended to shield against sensory cues and experimenter bias. Delanoy, Morris, and Dalton each served as experimenters and senders; Watt joined as a sender for the final portion.

The target-emotionality study used computer-generated pseudo-random assignment of the order in which positive and neutral emotional periods were presented to the agent. EDA was the primary physiological measure; a separate conscious response measure was also collected. The study pre-specified the EDA comparison between positive and neutral periods as a primary outcome.

Data

The two audited rows in the evidence base carry different metrics and must be read separately.

Morris, Dalton, Delanoy, and Watt (1995) found an overall direct-hit rate that was just statistically significant across all 97 trials. Breaking down by experimenter, however, the picture is uneven: one experimenter produced strong results while the others performed at or below chance — a pattern the authors describe as an unexpected experimenter effect. Delanoy’s own sessions as experimenter yielded results at or below chance. Condition-level comparisons (no sender, one sender, two senders) were non-significant.

Delanoy and Morris (1998) found a non-significant overall difference between activate and calm EDA periods across all 36 sessions. The sessions involving non-experimenter agents and receivers showed a result approaching significance, and a release-of-effort effect was found for that subgroup — but these are secondary analyses and the primary outcome is null.

The UNSETTLED SIGNALS in this evidence base are real and must be stated plainly: the two studies point in different directions (one positive, one null), and within the ganzfeld study the results diverge sharply by experimenter. These are not resolved by the current library holdings.

For the additional studies — the free-response training, the out-of-laboratory methodology, and the target-emotionality work — the retrieved excerpts contain figures, but those statistics are not in the audited evidence rows for this question. The retrieved excerpts do contain results (for example, the training study’s primary ESP measure and the out-of-laboratory study’s hit data), but per the grounding policy only the two structured evidence rows above can be stated as settled figures here. Readers wanting those details should go directly to the cited papers.

Skeptical critiques

Morris et al. (1995) themselves record that the experimenters and senders became very familiar with the target pools over the course of the study — a condition they flag as a departure from standard procedure that could affect interpretation of the results. The paper also notes that Delanoy had tended in past ganzfeld work to obtain overall chance results as an experimenter, while finding significant internal effects, raising the question of what drove the overall positive outcome when experimenter results diverged so sharply.

On the DMILS study, Delanoy and Morris (1998) note that the sessions run by trainee experimenters acting as their own agents/receivers — a design feature introduced for training purposes — may have confounded the primary effect, since the trainee-experimenter subgroup showed slightly greater EDA during calm periods rather than activate periods, contrary to the hypothesis. The non-significant primary outcome and the divergence between subgroups are acknowledged by the authors as limiting the conclusions that can be drawn.

Independent replication of the specific out-of-laboratory free-response methodology introduced in Delanoy, Watt, Morris, and Wiseman (1993) has not been documented in the sources retrieved for this question.

For a fuller account of Delanoy’s work and the security architecture underlying these studies, the Deborah L. Delanoy hub page and its spoke pages on free-response ESP training and target characteristics and automated ganzfeld methodology and security provide additional depth.