In cases of alleged hauntings, a large number of seemingly trustworthy witnesses consistently report experiencing unusual phenomena (e.g. apparitions, sudden changes in temperature, a strong sense of presence) in certain locations. The two studies reported here explored the psychological mechanisms that underlie this apparent evidence of 'ghostly' activity. The experiments took place at two locations that have a considerable reputation for being haunted-Hampton Court Palace (Surrey, England) and the South Bridge Vaults (Edinburgh, Scotland). Both studies involved participants walking around these locations and reporting where they experienced unusual phenomena. Results revealed significantly more reports of unusual experiences in areas that had a reputation for being haunted. This effect was not related to participants' prior knowledge about the reputation of these areas. However, the location of participants' experiences correlated significantly with various environmental factors, including, for example, the variance of local magnetic fields and lighting levels. These findings strongly suggest that alleged hauntings may not necessarily represent evidence for 'ghostly' activity, but could be, at least in part, the result of people responding to 'normal' factors in their surroundings.
Results revealed significantly more reports of unusual experiences in areas that had a reputation for being haunted.
two-site descriptive convergence with limited control of spatial exposure and historically defined area characteristics
conscious ‘prior knowledge’ does not account for the clustering of experiences in certain locations within the two test sites.
underpowered subgroup contrasts and exclusion of knowledgeable participants do not establish that prior knowledge has no explanatory role
the location of participants' experiences correlated significantly with various environmental factors, including, for example, the variance of local magnetic fields and lighting levels.
small observational environmental comparisons with uncontrolled multiplicity and magnetic nonreplication
Taken together, these findings strongly suggest that these alleged hauntings do not represent evidence for ‘ghostly’ activity, but are instead the result of people responding—perhaps unwittingly—to ‘normal’ factors in their surroundings.
observational environmental correlations cannot identify the proposed causal explanation or exclude competing accounts
the variance of the local magnetic field in the 'haunted' areas was significantly greater than of the 'control' areas
nonrandom area comparison whose printed summary statistics yield p≈.059 rather than the reported p=.04
Derived from the full evaluation — not a separate score.
Strengths
Current visitor reports are compared with independently established historical locations at a palace and underground vaults, extending the cited fieldwork beyond retrospective testimony.
↳ Experiment 1 framing and Table 1; Experiment 2 Hypotheses
Experiment 1 concealed historical spatial classifications during data collection and aggregated participant responses within groups. Experiment 2 used numbered clipboards for randomized vault allocation.
↳ Experiment 1 Method and Participant grouping; Experiment 2 Procedure
The paper reports that the magnetic result did not replicate and sets out alternative interpretations of lighting and physical-space associations, providing a basis for subsequent experiments.
↳ General Discussion, pp. 208–209
Limitations
The printed means, SDs, and six areas per arm give t≈2.13 and p≈.059, rather than the reported t(10)=2.34 and p=.04. This changes the haunted-versus-control magnetic-variance comparison to nonsignificant at .05.
↳ Experiment 1 Results, Magnetic fields; footnote 3
Table 2 examines 16 environmental correlations across ten vaults without multiplicity adjustment. Experiment 1 also excludes 216 of 678 recruits for incomplete questionnaires without analysing the associated selection.
↳ Experiment 1 Participants; Experiment 2 Table 2
Fixed environmental characteristics were not experimentally isolated. Although the discussion acknowledges competing interpretations, the final paragraph attributes the experiences to environmental factors.
↳ General Discussion, environmental-interpretation and final paragraphs
The two-site design, concealed historical classifications, and group-level analysis support a useful descriptive contribution. Environmental explanations carry less weight because the measured characteristics were observational, Table 2 contains multiple uncorrected tests across ten vaults, and substantial questionnaire exclusions were not analysed. The Experiment 1 magnetic-variance comparison also loses conventional significance when recalculated from its printed summary statistics. The discussion's engagement with alternatives does not resolve the contradiction between acknowledged causal ambiguity and the final environmental conclusion.
Nabu’s assessment, alongside the field’s view.
Are you an author of this paper?
Sound3.0
Confidence mediumThe two-site field data extend earlier work by comparing current visitor reports with independently established historical locations. The advance is primarily descriptive because prior-knowledge nulls and environmental associations do not establish the proposed explanation.
Blinding, group aggregation, and randomized vault allocation support execution, while unanalysed exclusions and uncontrolled environmental testing limit inference. The merit-based score is separate from the Red reliability finding that the printed Experiment 1 magnetic-variance comparison does not reproduce.
The descriptive sequence and presentation are understandable, but the final environmental-causation claim materially exceeds the observational evidence. The score includes the downward pricing for that central mismatch.
The discussion engages competing accounts, site differences, and magnetic nonreplication. Its explicit recognition of unresolved causal alternatives conflicts with the final environmental explanation, binding the score at 2.8.
Concerns4 of 4 checks
The supplied text includes participants who were children but reports no ethics approval or explicit consent procedure. Withdrawal rights and post-study contact arrangements are described; the omission does not establish that ethical review or consent was absent in practice.
Ethics approval: not stated · Preregistration: not stated · Data availability: not stated
22 of 29 checkable references verified
53 references in manuscript 23 have no canonical index record — counted, but not index-checkable 1 could not be checked — not the same as no match found 17 references confirmed by manual review
References not verified automatically are resolved manually within 24 hours.
No retraction or expression of concern found in Crossref (includes Retraction Watch).
Sources: Crossref (incl. Retraction Watch) ✓
Where this paper’s evidence sits on the path from initial observation to real-world use.
These are naturalistic observations rather than a validated diagnostic or intervention. Environmental mechanisms require further isolation and manipulation before practical use.
AI-generated, human-governed. Something look off? Contact us to request a review.
Keep reading
Other papers Nabu has evaluated blind, starting with the same field.
Daryl J. Bem · 2011 · Psychology
Zhang, Zhao, Hancock, et al. · 2025 · Psychology
Our findings suggest that socially oriented chatbot use is consistently associated with lower well-being, particularly among users who engage more intensively or exhibit higher levels of emotional self-disclosure.
TentativeDaniël Lakens · 2010 · Psychology
stick figures waving in synchrony were found to be rated higher on entitativity than stick figures waving in different rhythms
Well-supported