Discovery Lab Answers

James Houran, PhD is a research psychologist whose work spans psychometric measurement of paranormal belief and experience, haunt and poltergeist phenomenology, anomalous experience classification, and science communication. His profile page covers the full arc of his career; what follows works through his experiments, methods, data, and the critical record the sources support.

Experiments

Houran’s empirical output across the retrieved sources falls into several distinct lines:

Trickster-like experiences during paranormal investigation. Houran led a survey study — co-authored with Rense Lange, Brandon Massullo, Ciarán O’Keeffe, and David Schumacher — examining anomalous experiences reported by people who document the paranormal, framed around the concept of the “hitchhiker effect” and experimenter-related phenomena ^[1]. The study asked whether such experiences vary systematically by investigator type, cognitive profile, or demographic characteristics.

Haunted People Syndrome (HP-S) validation. Houran is a core contributor to the theoretical and empirical development of the HP-S model, which proposes that haunt-related experiences cluster around psychological characteristics of the percipient rather than properties of a location. A 2025 paper by O’Keeffe, Massullo, Laythe, Dagnall, Drinkwater, and Houran tested the concurrent validity of HP-S against an independent spontaneous case, comparing it to the AECKO framework ^[3].

Psychometric assessment of paranormal belief and experience. Houran has developed and validated several instruments in this space, including the Survey of Strange Events (SSE) and work on the Revised Transliminality Scale with Lange, Thalbourne, and Storm. The psychometric assessment spoke page covers this line in detail.

Quality-checking a “fact sheet” on ghostly episodes. A 2025 Frontiers in Psychology study — led by Massullo with Houran as co-author alongside Escolá Gascón, O’Keeffe, Drinkwater, and Dagnall — evaluated whether a publicly accessible fact sheet on ghostly episodes was accessible and useful across lay percipients, clinicians, ghost hunters, and lay non-percipients ^[6].

Poltergeist and haunting phenomenology. Earlier work includes a 1997 study examining the ambiguous origins and indications of poltergeist phenomena, and ongoing classification work comparing haunt and poltergeist categories. The poltergeist and haunting spoke page covers this line.

Editorial and science communication work. Houran serves as editor of the Journal of Scientific Exploration and has co-authored editorial pieces on science communication ^[4] and on rediscovering cryptozoology as an integrative science ^[2]. These are not primary empirical studies but reflect an active role in framing research questions at the field level.

Methodology

Rasch scaling is the methodological signature across Houran’s measurement work. Rather than classical sum-score approaches, he applies Rasch modeling — specifically a top-down purification procedure — to establish unidimensional, interval-level scales for anomalous experience and paranormal belief. This framework allows differential item functioning (DIF) analysis to detect whether scale items perform differently across subgroups.

In the trickster-experiences study ^[1], the methodology involved:

• A survey administered to two pre-specified subgroups: Paranormal Investigators (amateur/self-styled ghost hunters, n = 160) and Credentialed Scientists (professional researchers);
• A 14-item instrument subjected to Rasch analysis to produce a “Trickster-Like Experiences” scale score;
• DIF analysis across grouping variables including investigator type (amateur vs. scientist), a Cognitive Self-Check index (mental health/substance use history), and gender;
• Respondents comprised 68 men, 95 women, and four identifying as “other.”

In the HP-S concurrent validity study ^[3], the approach was content analysis of a spontaneous case (the Auerbach et al. case), with two independent coders applying the HP-S model’s recognition patterns and comparing the case’s anomaly frequency distribution against narrative archetypes from the SSE instrument using correlational analysis.

The fact-sheet quality-check study ^[6] used a multi-group design with four populations rating the instrument on accessibility, usefulness, and overall impression, with inferential statistics (Fisher’s F, chi-square, Bayes factors, and effect size estimates using ω² and ε²).

Data

No structured evidence block was retrieved for this question, so no pooled or comparative statistics are available here. What the numbered sources do support is summarized below, keeping figures only where the surrounding text is unambiguous.

Trickster-like experiences study ^[1]:

The retrieved excerpt notes that no notable DIF effects emerged at the scale level, meaning the 14-item Trickster-Like Experiences score functions consistently across subgroups despite some item-level variation.

HP-S concurrent validity study ^[3]:

• The SSE instrument’s Rasch-scaled scores carry a reliability of.87, ranging from 37.1 to 71.2 (mean = 50, SD = 10).
• Attenuation-corrected correlation between the SSE and a related measure: r = 0.78, p <.001.
• Correlational analysis of the case’s anomaly frequency distribution showed its closest narrative match was a “Primed” context (r =.43, p <.01); other narrative comparisons were not statistically significant.
• A significant decline in recorded disturbances over months was reported from the underlying Auerbach et al. case data: r = −0.397, p =.001 (two-tailed) — this figure is from Auerbach et al. as cited within ^[3], not from Houran et al.’s own primary analysis.
• The case’s SSE mean score most closely matched an “Illicit narrative” archetype (M = 45.90) compared to Spontaneous (M = 51.70), Primed (M = 52.30), Fantasy (M = 49.43), or Lifestyle (M = 50.60).

Fact-sheet quality-check study ^[6]:

Overall group differences on accessibility and usefulness were not statistically significant (Fisher’s F for accessibility = 0.252, p =.860; Bayes factor BF₁₀ = 0.0623, P(H₁|D) = 5.9%, favoring the null). Post-hoc comparisons on Overall Impression yielded two significant mean differences (3.56 vs. 3.13, p =.029, standardized ES = 0.756; and 3.73 vs. 3.56, p =.032, standardized ES = 0.779). The maximum explained variance across effect size estimates (ω² and ε²) was 12%. The authors interpreted the absence of broad group differences as evidence that the fact sheet has low inter-population variability — a property they treat as desirable for a public-facing tool.

Skeptical critiques

The sources retrieved for this question do not include published criticism directed specifically at Houran’s empirical work. The measurement and validation spoke page and the poltergeist and haunting spoke page are the most likely locations on the site where critical engagement with his classification frameworks and psychometric instruments is addressed in fuller detail.