Discovery Lab Answers

Robert L. Morris, PhD was a central figure in experimental parapsychology across the final three decades of the twentieth century and into the early 2000s, holding the Koestler Chair of Parapsychology at the University of Edinburgh. The sources retrieved for this question cover several distinct lines of his work — ganzfeld research, direct mental interactions with living systems (DMILS), ESP training, associational psi, and methodological advocacy — and these are addressed across the four dimensions below.

Experiments

Ganzfeld and autoganzfeld work. Morris co-authored a ganzfeld comparison study (Morris, Dalton, Delanoy, and Watt, 1995) examining sender versus no-sender conditions ^[5]. A later ganzfeld study with a creative participant population was run by Morris, Summers, and Yim, presented at the 2003 Parapsychological Association convention, testing whether artists and other creative participants would outperform a general sample under ganzfeld stimulation ^[4]. He also referenced an Edinburgh laboratory replication of earlier autoganzfeld work with improved separation between sender and receiver and experimenter exclusion from the video-equipment room ^[5].

ESP training study. Delanoy, Morris, and Watt conducted a free-response ESP training study designed around individual differences in learning style, incorporating both lab and remote ESP tests across preliminary, training, and follow-up sessions ^[1].

DMILS (Direct Mental Interactions with Living Systems). Morris contributed to the cognitive DMILS paradigm. Brady and Morris (1997) conducted a DMILS study that is cited as a direct replication of an earlier finding; Morris also participated in the design of two cognitive DMILS studies carried out in Bali by Edge, Suryani, Tiliopoulos, and Morris ^[3].

Psi and associational processes. Wilson, Morris, Tiliopoulos, and Pronto conducted a word-association study testing whether psi might manifest by influencing ongoing cognitive processes ^[2].

Methodological writing. Morris authored reflective pieces on research methods in experimental parapsychology ^[5] and on the continuing evolution of the field ^[6], which function as meta-level critiques of and prescriptions for the discipline rather than primary data studies.

Methodology

Ganzfeld protocol refinements. In the Edinburgh ganzfeld work Morris described, several deliberate design upgrades over earlier autoganzfeld procedures were implemented: greater physical separation between sender and receiver; two sets of video targets rather than one; and complete exclusion of the experimenter from the room containing the video equipment, eliminating the possibility of inadvertent cuing ^[5]. Target selection for the ganzfeld study with the creative population used a computer to select from 100 video clips grouped into 25 four-clip target sets, with clips chosen to be as dissimilar as possible ^[4].

ESP training study design. In the Delanoy, Morris, and Watt study, randomization of lab and remote ESP tests, and designation of specific targets, was carried out by a person not otherwise involved in the study, prior to the study’s commencement. Published weather data provided an entry point into a standard random number table to generate randomization sequences ^[1]. Judging used both holistic and atomistic approaches, and planned primary analyses were pre-specified, distinguishing them from exploratory measures ^[1].

DMILS protocol. The Bali DMILS studies used automated computer software designed by Paul Stevens — an updated version of the program used in the Edinburgh laboratory — with Help and Control periods randomized via a pseudo-randomized algorithm ensuring equal numbers of each type per session ^[3]. One of the Bali studies incorporated meditation training for half the participants to examine whether it moderated psi performance ^[3].

Associational psi design. The Wilson et al. study used word-association tasks split by which of two experimenters served as agent, allowing examination of experimenter-as-agent effects as well as response-bias effects (minority versus majority responses) ^[2].

Methodological advocacy. Morris explicitly argued for learning from negative results, evaluating accumulated findings more completely, and addressing the challenge of theory-based hypothesis testing — framing methodological improvement as a community-level obligation rather than a per-study fix ^[5][6].

Data

The sources retrieved for this question contain several specific findings. Because the studies use different, non-comparable metrics, they are not pooled; each is reported separately.

Several unsettled signals in this data deserve explicit mention. The ESP training study’s primary measure was nonsignificant, and the concentration-group participants showed a significant decline in ESP scoring posttest — a psi-missing result the authors themselves noted alongside a first-session effect and a across-session decline trend ^[1]. The DMILS Bali studies showed significant effects overall, but the result in Study 2 was driven entirely by a single run-set and by experimenter-attitude differences rather than the pre-planned design, raising the question of what the driving variable actually is ^[3]. The word-association study overall produced no significant deviation from chance, with only sub-group splits approaching significance ^[2].

Skeptical critiques

What critics argue. Morris himself documented the skeptical challenge to the autoganzfeld procedure in his 2001 methods paper, noting that the automated ganzfeld procedure developed at the Psychophysical Research Laboratories “was challenged” — a challenge addressed in the same symposium at which he was presenting ^[5]. He also acknowledged in the same paper that parapsychology faces the charge of being a pseudo-science from “some public commentators,” and that the field struggles with difficulties in generating and testing theory-based hypotheses ^[5].

Regarding the DMILS paradigm specifically, the Bali studies’ own authors recorded that three of four run-set results in Study 2 were at chance, with the significant result emerging only in a condition involving varied experimenter attitudes — raising the question of whether the effect reflects psi or uncontrolled experimenter-related variables ^[3]. The authors note this explicitly rather than attributing it to external critics.

What the experimental data show. Across the studies in these sources, results are mixed in direction: the primary measure of the ESP training study was nonsignificant (z = 1.07) and included a significant posttest decline in one subgroup; the ganzfeld creative-population study reached the pre-specified significance threshold (p = 0.05) but only at the boundary; DMILS results in Bali were significant overall but driven by specific subgroup conditions and experimenter variables; the word-association study found no overall significant effect ^[1][2][3][4].

Analysis. Morris consistently engaged with methodological critique from within — his advocacy for improved controls, separation of confirmatory from exploratory analyses, and learning from negative results was part of his own published program ^[5][6]. The pattern across these studies illustrates the tension he identified in his methodological writing: positive results appear in some conditions, null and below-chance results in others, and the conditions that distinguish them — experimenter attitude, participant subgroup, session order — are often identified post-hoc or in exploratory analyses rather than confirmed by pre-specified replication. Independent replication of several specific Edinburgh-protocol findings by groups outside that laboratory is not documented in the sources retrieved for this question.

For broader context on the ganzfeld literature Morris contributed to, the Precognition page covers related meta-analytic findings, and the Scientists section is the starting point for his full ESP-Nexus profile.