Alternative beliefs in psychedelic drug users

Lebedev and colleagues situate the study within a context of growing policy liberalisation and scientific interest in classical psychedelics (LSD, psilocybin, DMT), noting prior small-scale clinical findings that motivated wider attention to these substances. Earlier research has documented links between psychedelics and changes in personality, wellbeing and life attitudes, but the authors identify a gap concerning how psychedelic use relates to individuals' epistemic styles and acceptance of "alternative facts" or conspiratorial beliefs. They also note debate in the literature about whether non-conformist attitudes associated with psychedelic culture reflect drug effects or pre-existing subcultural traits. The present cross-sectional study therefore aimed to test associations between patterns of drug use, with a focus on classical psychedelics, and measures of conspiratorial thinking and epistemic belief tendencies. Specifically, the investigators examined relationships between psychedelic exposure and (1) general conspiracy mentality, and (2) three EBS (Epistemic Belief Scale) facets: Faith in Intuition for facts, Need for evidence, and Truth is political. Secondary aims included exploring links with personality and psychopathology-related traits to contextualise any associations observed.

The study used a cross-sectional, web-based survey administered in Swedish and English between January and May 2019, prior to the COVID-19 pandemic. Participants (N = 392) were recruited via e-mail invitations to a prior study pool and through social media and forum postings targeting groups likely to include people with experience of psychedelic drugs. No direct compensation was provided. The sample was young on average (mean age about 28 years), predominantly reporting lifetime use of common substances (e.g. alcohol, cannabis, MDMA), with 233 participants reporting at least one lifetime use of psychedelics. The study had ethical approval from the Swedish Ethical Board and obtained informed consent from all participants. Primary measures were the Conspiracy Mentality Questionnaire (CMQ; five items, 0–100% scale) to operationalise beliefs in alternative facts, and the Epistemic Belief Scale (EBS; 12 items divided into Faith in Intuition for facts, Need for evidence, and Truth is political). The investigators also collected detailed self-report measures of drug use (frequency and recency across substance classes), personality (BFI-S, five factors), and psychopathology-related traits (Peters Delusions Inventory, O-LIFE schizotypy, ASRS for ADHD symptoms, RAADS-R for autism spectrum-related symptoms). Swedish translations of CMQ and EBS were psychometrically evaluated (details reported in Supplement). Analyses focused on stepwise linear regression models testing recent and frequent psychedelic use as predictors of CMQ total and EBS facets. The modelling strategy introduced demographic controls (age, sex), psychiatric diagnoses, and concomitant drug use in successive steps. Between-group comparisons (psychedelic users vs non-users) employed independent samples t-tests. The investigators also computed correlation matrices and cross-correlation network plots among drug-use measures, personality, and psychopathology traits. Data screening included Cook's distance for influential outliers (none detected) and inspection of residuals for approximate normality; the Shapiro-Wilk test indicated non-normality for CMQ but residuals were judged adequate for linear modelling. Variance inflation factors were inspected for multicollinearity. Multiple comparisons were corrected using the false discovery rate procedure.

Recent use of psychedelics was associated with higher scores on the CMQ in a single-predictor model (standardised β = 0.174, t = 3.49, p < .001; R2 ≈ .032). This association persisted within the subsample of psychedelic users (N = 233; β = 0.147, p = .024). After adjusting for age, sex and psychiatric conditions, recent psychedelic use remained a significant predictor in the full sample (β = 0.212, p = .001) and in psychedelic users (β = 0.195, p = .022). In a fully adjusted model including concomitant drug use, recent psychedelic use (β = 0.187, p = .027) and alcohol use (β = −0.142, p = .023) were differentially associated with beliefs in alternative facts in the whole sample, although these effects became marginal when the analysis was restricted to psychedelic users. Introducing the Peters Delusions Inventory into models reduced some effects to marginal significance (psychedelics p ≈ .05). A separate model analysing frequent use across drug classes found stimulants to be the only significant predictor of CMQ scores (standardised β ≈ 0.222, p = .028). Analyses limited to participants without psychiatric diagnoses (n = 200, 73.26%) showed broadly similar patterns for psychedelic and alcohol effects, though some estimates became marginally significant in fully adjusted models. The Swedish subsample scored lower on CMQ than the non-Swedish sample (47.18 ± 20.98 vs 53.08 ± 22; t = 2.71, p = .007); adjusting for mother tongue did not materially change results. Group comparisons indicated that lifetime psychedelic users (N = 233) had higher mean CMQ scores than non-users (M = 52.42, SD = 21 vs M = 46.44, SD = 22; t ≈ 2.69, p = .008, Pcorr = .0076), with a small effect size d = 0.28 (95% CI 0.075–0.48). Among CMQ subitems, the difference for the "secret organisations" item remained significant after multiple comparison correction (Pcorr = .001); differences for "misinformation" and "politics" were of marginal significance after correction. For EBS facets, no association was found between recent psychedelic use and Faith in Intuition for facts. Need for evidence was not predicted by recent use in primary models, but a single-predictor model found frequent psychedelic use predicted lower Need for evidence (β = −0.130, p = .024); this effect remained after demographic adjustment but was lost when controlling for concomitant drug use. By contrast, Truth is political showed a consistent association with recent psychedelic use: single-predictor model β = 0.176 (p = .002), remaining significant after demographic controls and marginally significant when adjusting for other drug use (psychedelics β = 0.197, p = .048). Including PDI reduced this effect to marginal significance in some models. Between-group comparisons for EBS showed psychedelic users scored higher on Truth is political (M = 2.85 vs 2.53; t ≈ 2.58, p = .006, Pcorr = .0177) and marginally lower on Need for evidence (Pcorr ≈ .075). Regarding personality and psychopathology correlations, agreeableness showed a small positive correlation with recent stimulant use (r = .116, p = .021). Neuroticism displayed small negative correlations with recent and frequent use of psychedelics, MDMA and tobacco. Delusion-proneness correlated positively with recent and frequent psychedelic use (recent r = .169, p = .005). The authors report broader correlation matrices linking CMQ and EBS facets with O-LIFE, PDI, RAADS and ASRS, indicating associations between conspiracy mentality and schizotypy-related traits, but full matrices and exact values are presented in supplementary material.

Lebedev and colleagues interpret their findings as evidence of an association between psychedelic use and a greater propensity to endorse non-conventional beliefs, particularly general conspiracy mentality and the view that factual information is politically influenced. They note that these associations remained after controlling for demographic variables and other drug use in several models, though some effects attenuated to marginal significance when additional covariates (for example PDI) or when analyses were restricted to users only. The study did not support the hypothesis that psychedelic use is associated with a stronger preference for intuition over evidence (Faith in Intuition for facts). The authors situate the results against prior literature linking unconventional beliefs with personality traits such as high openness and low conscientiousness, and with schizotypy-like traits; they confirm similar patterns in their sample and argue that openness and non-conformity may partly explain the association between psychedelic use and beliefs. Two alternative explanations are advanced: (a) pre-existing differences lead certain individuals both to use psychedelics and to endorse alternative views; or (b) psychedelic experiences causally alter epistemic style, increasing receptivity to non-conventional ideas. The authors emphasise that the present cross-sectional design cannot distinguish these possibilities and suggest longitudinal or experimental studies assessing beliefs before and after psychedelic administration. Several limitations are acknowledged. The data are self-reported, which may introduce reporting bias for both drug use and beliefs. The sample was recruited from groups with expected higher recreational drug use and was largely Swedish; this restricts generalisability and may reflect cultural or platform-specific biases. The investigators did not collect robust quantity measures for substance exposure and observed little heterogeneity in age at first use, limiting analyses of dose–response or developmental timing. Finally, data were gathered before the COVID-19 pandemic, which the authors consider both a strength (avoiding pandemic-related confounds) and a contextual caveat for replication. In closing, the authors present the study as an initial step addressing an understudied question: whether and how patterns of drug use relate to information processing and epistemic beliefs. They call for multidisciplinary, longitudinal, and experimental research to disentangle causal mechanisms and to inform policy and public-health considerations without promoting stigma towards users of illicit substances.