Discovery Lab Answers

Emily Williams Kelly, PhD — Experiments and Research

Emily Williams Kelly, PhD, is a researcher at the Division of Perceptual Studies (DOPS), University of Virginia Health System. The sources retrieved for this question cover several distinct lines of her work: controlled mediumship studies, theoretical analysis of mediumship methodology, near-death and deathbed vision research, and editorial scholarship on survival-related phenomena.

Controlled Mediumship Studies

The most directly experimental work in the retrieved sources is a controlled investigation of mediums conducted by Kelly and co-author Dianne Arcangel ^[4]. The study examined mediums who claimed to provide information about deceased persons, motivated in part by growing public interest in mediumship among bereaved populations. The design used a blinded ranking procedure: sitters received multiple readings and were asked to identify which had been intended for them, without knowing in advance which was theirs.

The study reported quantitative outcomes — including average ranking figures and comparisons across conditions — but the retrieved excerpts contain partially damaged text around the specific statistics, so those individual figures should be read from the primary source directly ^[4] rather than taken from an excerpt. What the text does show intact is that the protocol included conditions designed to control for cold reading and normal information transfer: sitters were told only the day and time their reading would be held and were instructed to remain at a distance from the mediums throughout.

Methodological and Historical Analysis of Mediumship Research

Kelly’s 2010 essay in the Journal of Scientific Exploration ^[1] is not itself a primary experiment but is substantively analytical — it surveys the history of proxy sitting research and proposes directions for future controlled work. Proxy sittings are sessions in which a stand-in (the “proxy”) attends in place of the person for whom the reading is intended, specifically to block the possibility that the medium is reading the actual sitter’s demeanor, grief cues, or prior knowledge.

The essay reviews work by C. Drayton Thomas, who conducted 24 proxy cases with medium Mrs. Leonard beginning as early as 1917 ^[1], and discusses the Haraldsson protocol in which 10 sitters each received all 10 readings and were asked to rank them — four ranked their own reading first (p < .01), though Kelly notes that more recent studies have been less successful ^[1]. She also examines Stevenson’s method of classifying medium statements into objective (scorable as accurate or not) versus subjective dimensions, and his comparison of accuracy rates across the target sitter, a close relative, and unrelated controls ^[1].

The essay’s main argument is that the proxy design, rigorously executed, remains one of the strongest available tools for ruling out conventional explanations of medium success, and that this line of research warrants renewed systematic attention ^[1].

Deathbed Visions and Near-Death Research

Kelly is cited in Bruce Greyson’s 2010 analysis of “Peak in Darien” experiences — deathbed visions in which dying persons report seeing individuals they did not know had recently died ^[2]. Greyson notes that Kelly (2001) specifically contrasted the expectation hypothesis (dying persons anticipate seeing deceased relatives) against a survival hypothesis, and identified features of these cases that bear differently on each interpretation ^[2]. This places Kelly’s work squarely in the empirical analysis of anomalous deathbed phenomena, not merely in the philosophical literature.

Her book review work ^[5] also engages critically with near-death experience claims, noting where authors in that space make statements she regards as unsubstantiated — for instance, assertions about EEG monitoring during NDEs — indicating she applies evaluative scrutiny to the literature rather than simply reporting it favorably.

Editorial Scholarship

Erlendur Haraldsson’s review ^[3] documents Kelly’s role as editor of Science, the Self, and Survival After Death: Selected Writings of Ian Stevenson (Rowman & Littlefield, 2013), a 424-page volume collecting the work of Ian Stevenson, professor of psychiatry at the University of Virginia. Haraldsson describes the volume as giving “an excellent overview of Stevenson’s work” and notes its importance for transmitting that research to subsequent generations of researchers ^[3].

Skeptical Critiques

What critics argue. The retrieved sources surface one direct critical exchange. Lange and Houran published a response in the Journal of Nervous and Mental Disease to a book review Kelly wrote of Hauntings and Poltergeists: Multidisciplinary Perspectives — a volume Houran co-edited with Lange (2001). The ESP-Nexus page on Houran’s work cites this as Lange & Houran (2002), indicating Kelly’s review generated a formal rebuttal from the volume’s editors.

What the experimental data show. The mediumship ranking study ^[4] used blinded procedures intended to address the most common methodological objections to medium research (sitter contact, cold reading). Kelly’s own methodological essay acknowledges that more recent proxy studies have been “less successful” than earlier ones, and she discusses the difficulty of achieving the conditions needed to rule out subliminal cuing ^[1].

The mediumship study ^[4] is the clearest empirical test in the retrieved material; its ranking protocol is a recognized control design in this literature, though the excerpt text around the key statistics is not fully intact, and the primary source should be consulted directly for the figures.

For broader context on mediumship research methodology and the DOPS research program, the ESP-Nexus Scientists section and Studies section carry additional material on this line of work.