Factor analysis of the mystical experience questionnaire: A study of experiences occasioned by the hallucinogen psilocybin

Mystical experiences are reported across religious traditions and historical accounts, with prior questionnaire research identifying core features such as unity, sacredness, ineffability, peace and joy, transcendence of time and space, and a noetic sense of objective truth. Although lifetime measures of mysticism (for example, Hood's Mysticism Scale) have been studied psychometrically, relatively little empirical work has focused on single, discrete mystical events or on mystical experiences occasioned specifically by hallucinogens such as psilocybin. Recent laboratory studies indicate that psilocybin can reliably occasion mystical-type experiences that in some cases predict lasting increases in well-being, yet the principal self-report instrument used in that literature, the Mystical Experience Questionnaire (MEQ), had not been formally validated or factor-analysed for single event assessment. Maclean and colleagues set out to evaluate the psychometric properties and factor structure of the MEQ. Using a large web-based convenience sample of individuals who reported having had a profound experience after ingesting psilocybin mushrooms, the study aimed to (1) explore the latent factor structure of the 43-item MEQ via exploratory factor analysis (Study 1), (2) assess reliability and convergent and construct validity against the 32-item Mysticism Scale, and (3) test the resulting factor model in an independent sample using confirmatory factor analysis (Study 2). The goal was to produce a validated, reliable scale for quantifying single, hallucinogen-occasioned mystical experiences.

Participants were recruited primarily through web-link advertisements, e-mail invitations, and online communities interested in spirituality and psychoactive substances. Study 1 data collection occurred during two recruitment rounds (May–June 2010 and July 2010). Inclusion criteria required participants to be 18 years or older, fluent in English, to have taken an active dose of psilocybin mushrooms producing moderate to strong effects, and to report a profound, personally meaningful psilocybin session. Of 1,720 respondents, 1,602 (93%) met all inclusion criteria and completed the survey. Study 2 used an available convenience sample collected in a similar manner (N = 440 complete MEQ responses). The online survey (administered via SurveyMonkey) collected demographics, lifetime hallucinogen use, and details about a single chosen psilocybin session. For Study 1 the 43 MEQ items were intermixed with 57 distracter items; in Study 2 only the 43 MEQ items were presented. Each MEQ item was rated on a six-point intensity scale (0–5). Participants also completed a modified, single-event version of the 32-item Mysticism Scale (items rated on a nine-point scale) to permit convergent validity analyses. Additional session-specific items included three questions from a Persisting Effects Questionnaire and an optional written description of the experience. Statistical analysis focused on the MEQ factor structure, reliability, and validity. Because item responses were sparse and non-normally distributed, polychoric correlations were used. In Study 1 the polychoric correlation matrix of 43 items underwent parallel analysis (1,000 draws) to guide factor extraction, followed by exploratory factor analysis (EFA) with promax rotation to allow correlated factors. Items were eliminated if they failed to load ≥ .40 on any factor, loaded ≥ .40 on more than one factor, or had high uniqueness (> .60). A single round of item elimination was predetermined; a second parallel analysis was run on the reduced item set to confirm factor number. Standardised factor scores were computed and group comparisons used t tests and ANOVA. Study 2 performed confirmatory factor analyses (CFA) in Mplus v.6.1 using categorical responses and a robust weighted least-squares estimator to compare three models: two established three-factor models derived from prior mysticism research (Hood and Stace variants) and the four-factor model emerging from Study 1. Fit indices reported included RMSEA, CFI and TLI, with RMSEA < .10 and CFI/TLI > .90 considered suggestive of acceptable fit given sample size considerations.

Study 1 sample characteristics: N = 1,602, 50% female, wide age and country representation, generally well educated and reporting satisfactory life satisfaction. When asked to define their religiousness/spirituality, 70% identified as “spiritual but not religious,” 13.5% as “spiritual and religious,” 16% as “neither,” and <1% as “religious but not spiritual.” Exploratory factor analysis of the 43 MEQ items indicated that a four-factor solution was most appropriate (parallel analysis and scree inspection). Thirteen items were removed for low primary loadings, cross-loadings, or high uniqueness, leaving 30 items in the final model. The promax-rotated four-factor solution explained on average 57% of the variance across items (communalities range = 39–78%). Factor 1 comprised items related to internal and external unity, noetic quality and sacredness; Factor 2 reflected positive mood; Factor 3 represented transcendence of time and space; and Factor 4 captured ineffability. Reliability estimates were high for the 30-item revised MEQ (Cronbach's alpha = .933) and acceptable to good for the four subscales (Factor 1 = .926; Factor 2 = .831; Factor 3 = .810; Factor 4 = .800). Convergent validity with the Mysticism Scale was strong: total Mysticism Scale correlated r = .81 (p < .001) with total MEQ, and MEQ Factor 1 correlated r = .79 (p < .001) with the Mysticism Scale. Correlations were more modest for the other MEQ factors (Factor 2 r = .55; Factor 3 r = .47; Factor 4 r = .41; all p < .001). For construct validity, 1,410 participants (88%) endorsed having had a mystical experience during a psilocybin session; these participants had significantly higher MEQ factor scores and higher Mysticism Scale scores than those who did not endorse a mystical experience (p < .001). Study 2 (N = 440) characterised a different demographic profile (predominantly male 88%, mean age ≈ 28 years, lower formal education levels); 88% again endorsed a spiritual experience after psilocybin. CFA compared the Hood and Stace three-factor models with the four-factor “Psilocybin” model. The Hood and Stace models failed to meet fit cutoffs (Hood: RMSEA = .110, CFI = .88, TLI = .88; Stace: RMSEA = .109, CFI = .89, TLI = .88). The four-factor Psilocybin model showed acceptable fit (RMSEA = .074, CFI = .95, TLI = .94). Factor loadings in the Psilocybin model were high (> .65) and factors were intercorrelated; on average the model explained 64% of the variance per item (residual variances range = 16–54%). Reliability estimates in Study 2 were consistent with Study 1 (30-item alpha = .957; Factor alphas = .946, .831, .887, .864).

Across two web-based samples, the investigators found that a 30-item revision of the MEQ is psychometrically robust for measuring single, psilocybin-occasioned mystical experiences. The EFA in Study 1 identified four correlated factors — a primary “mystical” factor combining unity, noetic quality and sacredness, plus separate factors for positive mood, transcendence of time and space, and ineffability — and the four-factor solution was confirmed in an independent convenience sample by CFA. Both total and subscale MEQ scores demonstrated excellent internal consistency and convergent validity with the established Mysticism Scale, and participants who self-reported having had a mystical experience after psilocybin scored higher on the MEQ, supporting construct validity. The authors note that the principal novel finding is the combination of internal and external unity with noetic quality and sacredness on a single factor, a pattern that departs from some earlier lifetime-focused analyses that distinguished introvertive and extrovertive unity. They propose several possible explanations: the MEQ's particular item wording and structure, measurement of a single time-limited event rather than lifetime experience, and the characteristics of the self-selected sample (predominantly “spiritual but not religious”). The discussion situates findings relative to previous factor-analytic work, noting both consistencies with more recent studies that also found unity and noetic quality to co-load, and contrasts with older models that separated phenomenology from interpretative elements. Key limitations acknowledged by the study team include the self-selected, convenience sampling that may bias results toward positive spiritual experiences, and the need for further confirmatory work to test reliability (including test–retest) and validity in other populations and contexts. The authors caution that while the revised MEQ appears suitable for laboratory studies of classic hallucinogens and informal naturalistic settings, its appropriateness for other populations or modalities is not established. They recommend additional confirmatory factor-analytic studies across diverse samples and complementary laboratory and survey research to further characterise hallucinogen-occasioned mystical phenomena.