Alternative beliefs in psychedelic drug users

Previous research has suggested that classical psychedelics can foster significant and enduring changes in personality traits and subjective wellbeing. Despite the lack of evidence for adverse effects on mental health stemming from psychedelic use, concerns persist regarding the capacity of these substances to modulate information processing and attitudes towards factual data. The aim of the present study was to investigate the propensity for accepting alternative facts and the general treatment of knowledge within a sample of 392 participants, 233 of whom reported at least a single incidence of psychedelic use in their lifetime. To do this, we leveraged step-wise methods of linear modelling investigating effects of demographics, psychiatric conditions and concomitant drug use. Our findings revealed a moderate positive association between psychedelic use and beliefs in alternative facts, as well as the specific belief that facts are politically influenced. However, no links were found for favouring intuition over evidence when confirming facts. Among other investigated drugs, only alcohol was negatively associated with beliefs in alternative facts. Taken together, our results support the link between psychedelic use and non-conformist thinking styles, which can be attributed to the psychological effects of the drugs themselves, but may also mirror a common trait related to unconventional beliefs and illicit substance use.


Procedure
Recruitment occurred through web-based announcements on social media services, forums which were expected to include our target population (Facebook and Reddit groups discussing scientific and recreational use of psychedelic drugs, drug policy, substance-related issues and disorders), as well as general platforms designed to recruit research subjects, as a part of a larger study at Karolinska Institutet focusing on schizotypy.The data from several assessments (see Supplement) were merged in order to maximize the sample size.Potential sampling bias in relation to different platforms was addressed elsewhere with no apparent evidence for such being identified 39 .However, due to the specific focus on drug use, it is important to recognize that the total study sample may not be representative of the general population.

Materials
In our endeavour to address the central research question, we employed two primary instruments: the Conspiracy Mentality Questionnaire and the Epistemic Belief Scale.We had a total of 392 participants, with a gender breakdown of 29% male, 41% female, and 30% who chose not to disclose their gender, complete the Conspiracy Mentality Questionnaire.The age range for this group was between 15 and 67 years, with a mean age of 28 years and a standard deviation of 7.2.
The Epistemic Belief Scale was completed by a sample of 305 participants.The gender distribution for this group was 23% male, 40% female, and 37% who did not provide their gender.The participants' ages ranged from 18 to 67 years, with a mean age of 28 years and a standard deviation of 7.4.
In addition to these primary measures, the participants also filled out a series of trait questionnaires relating to psychopathology and personality, along with questions pertaining to their usage of drugs.Comprehensive demographic data, including age, gender, and psychiatric diagnoses, were collected as well.Data on validation of the Swedish versions of the instruments can be found in the Supplement.
Assessment of the factor structure in Swedish-and English-speaking groups revealed that the strict invariance assumption is met for the EBS, while the metric invariance assumption (with constrained loadings) is met for the CMQ (Supplement).

Primary outcomes
The Conspiracy Mentality Questionnaire (CMQ) is a measure of individual differences in conspiratorial thinking and ideation 40 , which for the purpose of this study was used to measure beliefs in alternative facts.CMQ consists of five statements (Table 1) that participants can rate their likelihood of agreeing with on a 11-point Likert-type scale, ranging from 0% (certainly not) to 100% (certain).For the main analysis, a mean score for all five items was calculated, while the exploratory analysis included the individual scores for each subitem.The subitems will be referred to as (1) misinformation, (2) politics, (3) monitoring, (4) hidden connections and (5) secret organizations.
The CMQ has previously been validated in England, Ireland, Germany and Turkey 40 .Examinations of the factorial structure supports that conspiracy mentality is a one-dimensional construct that is stable over time and has been shown to have convergent, discriminant, and predictive validity 40 .
The present study used a previously constructed, but unvalidated Swedish version of the CMQ (Supplement).For this reason, a psychometric evaluation was performed (Supplementary Materials).Construct reliability and validity could be confirmed for a modified model, where several of the observed variables were allowed to correlate.
The Epistemic Belief Scale (EBS) consists of twelve statements (Table 2) that assess how people treat knowledge and process information, an important foundation of beliefs 41 .The twelve statements are divided into groups of four, generating three facets of epistemic beliefs; Faith in Intuition for facts (FI-facts), Need for evidence and Table 1.Conspiracy mentality questionnaire.
1.I think that many very important things happen in the world, which the public is never informed about 2. I think that politicians usually do not tell us the true motives for their decisions 3. I think that government agencies closely monitor all citizens 4. I think that events which superficially seem to lack a connection are often the result of secret activities 5.I think that there are secret organizations that greatly influence political decisions Truth is political.The two first facets (FI-facts, Need for evidence) relate to how much bearing an individual puts on feelings compared to evidence when forming opinions.The third facet (Truth is political) includes items regarding the conviction that facts are influenced by politics 41 .Participants rated their likelihood of agreeing with each statement by using a 5-point Likert-type scale ranging from disagree completely to agree completely.All the three EBS facets have shown acceptable scale reliability and test re-test reliability 41 .Sufficient convergent and discriminant validity has been confirmed using structural equation modelling with several other measures, including conspiracist ideation, which was found to be positively associated with FI-facts and Truth is political, and negatively with Need for evidence 41 .For the present study, we translated the EBS scale to Swedish and performed a validation analysis confirming its construct validity and a good fit with the data.
Traits related to psychopathology Peters et al.Delusions Inventory (PDI) was designed to measure delusional ideation in the normal population 42 .It is comprised of 21 items with three domains that measures different levels of Distress, Preoccupation and Conviction.Each item is a yes or no question, however, if someone answers yes, they are asked to mark their answer for the amount of Distress, degree of Preoccupation and the level of Conviction that a certain belief/experience may have, on a 5-point scale 42 .The scale has adequate internal consistency, concurrent validity and test-retest reliability.In the present study only yes or no alternatives were used (no subitem rating was performed).
Oxford-Liverpool Inventory of Feelings and Experiences (O-LIFE) consists of four subscales and 104 items in total.Each subscale is thought to measure different facets of the schizotypy construct in the normal population: Unusual Experiences (30 items) measures different types of cognitive and perceptual disturbances such as magical thinking and hallucinations.Cognitive Disorganization (24 items) relates to executive functioning, such as attention-deficits, disorganized speech as well as social anxiety.Introvertive Anhedonia (27 items) measures dissatisfaction or lack of interest in exciting, social and physical engagements.Impulsive Nonconformity (23 items) taps into odd behaviours that relate to both impulsivity and antisocial behaviour.All items are yes or no questions.All scales have been found to have adequate internal consistency and test-retest reliability 43 .
The Adult ADHD Self-Report Scale (ASRS) was used to evaluate (subclinical and clinical) symptoms of attention deficit hyperactivity disorder (ADHD) over the full range in the population.The ASRS consists of 18 items that are divided into two parts that are consistent with the Diagnostic and Statistical Manual for Mental Disorders (DSM-IV) criteria for ADHD in adults.Each question can be answered on a 5-point Likert-type scale that ranges from never to very often in frequency.Answers can be scored as either positive or negative and each question has a different threshold.Four or more answers coded as positive in Part A (items 1-6) is indicative of occurrence of ADHD.The ASRS was developed by the World Health Organization (WHO) 44 .ASRS is normally distributed in the population 45 and has been shown to have good internal reliability, test-retest reliability, as well as concurrent validity with a rater-administered ADHD-scale 46 .
The Ritvo Autism and Asperger Diagnostic Scale-Revised (RAADS-R) was used to appraise symptoms of autism spectrum disorder (ASD).RAADS-R is a 50-item scale that has been developed to diagnose ASD in adults.It is normally distributed in the population and has demonstrated excellent psychometric properties 47 .

Drug use
The participants were asked a number of questions related to the use of different drugs; alcohol, cannabis, MDMA, opiates (e.g.heroin, morphine, opium), psychedelics (e.g.LSD, magic mushrooms/psilocybin, ayahuasca/DMT), stimulants (e.g.amphetamine, ephedrine, cocaine) and tobacco.For each drug, the subjects were required to answer how many times they had used it (frequency) by selecting one of 5 scale points; never, 1-3 times, 4-10 times, 11-50 times or more than 50 times.Additionally, participants were asked when their last time of intake had been (recency) by selecting one of 8 scale points; never, less than 24 h ago, less than 3 days ago, less than a week ago, less than a month ago, less than a year, less than 5 years ago or more than 5 years ago.The analysis was specifically focused on the variables that represent a more nuanced understanding of exposure to better tease out effects in a manner that transcends lifetime use and is not confounded by age.However, the participants were also asked to submit at what age they used a drug for the first time.

Personality
The short version of the Big Five Personality Inventory (BFI-S) incudes 15 items and measures the five factor model constructs: openness, conscientiousness, extraversion, agreeableness and neuroticism.Participants rate how much they agree with each statement using a 7-point scale ranging from 1 (strongly disagree") to 7 (strongly agree).Both convergent and discriminant validity with other personality measures has been demonstrated for the present scale 48 .

Statistical analyses
Prior to statistical analysis, all data was screened for outliers with high influence, using Cook's Distance.A Cook's distance > 1.0 was considered large.No outliers were detected.Normality of the distributions was analysed for the total score of CMQ and the three facets of EBS: 1-Faith in Intuition for facts (FI-facts), 2-Need for evidence and 3-Truth is political).The Shapiro-Wilk test for normality indicated that the data for the total score of CMQ was not normally distributed (p < .001)although an overview of histogram and distribution of residuals revealed sufficient normality for the analysis employing linear modelling.Similarly, regarding the EBS facets, normal distribution of residuals was considered sufficient to support the planned analysis, and no additional transformations were carried out.
In order to investigate the relationship between recent psychedelic drug use, alcohol use and beliefs in alternative facts, as measured by the total score of CMQ, three linear regression models were fitted stepwise.The first model included recent use of psychedelics as a single-variable predictor of beliefs in alternative facts (the outcome variable total score on the CMQ).Next step included verification of significant findings by means of additional adjustments for demographics, psychiatric conditions, and concomitant drug use.In the last step, an exploratory analysis was performed investigating the relationship between frequent use of the different drugs of reference and epistemic beliefs.
To test the hypotheses regarding the effect of recent use of psychedelics on the three facets of epistemic beliefs (FI-facts, Need for evidence, Truth is political), as measured with EBS, several linear regression models were fitted stepwise, employing a similar strategy.The first step included three single-variable models with recent use of psychedelics, as predictors, as measured by the latest intake of the drug.Each model had a different EBS factor as an outcome variable.Next steps followed the same procedure as for the CMQ analysis.
To check for multi-collinearity, the variance inflation factors (VIFs) were inspected.A VIF < 1 was considered satisfactory.
To further investigate the associations between the use of psychedelics and beliefs in alternative facts, a comparison between two groups was also carried out.The group of psychedelic users (N = 233) included people who had reported having used psychedelics at least once in their lifetime.The second group consisted of subjects that reported not using psychedelics (non-users, N = 159).In order to establish if there were any significant differences between the groups in relation to the total score on CMQ and its subitems, independent samples t-tests were performed.The same procedure was performed for the three facets of the EBS (N total = 305, N users = 180, N non-users = 125).
Following the main analyses, we also reported inherent associations among the variables under investigation within the sample of psychedelic users (n = 180), leveraging cross-correlation network plot.
Finally, two correlation matrices were estimated.The first included measures of recent use and the different subscales of BFI-S as well as the composite score delusion proneness, in order to investigate the associations between different drug use patterns and personality.The second matrix included raw-scores for psychopathology (O-LIFE, PDI, RAADS, ASRS) as well as the total score of CMQ and the three facets of EBS, in order to explore the associations between our main measures and psychopathology.
To correct for multiple comparisons, the false discovery rate procedure was employed, as represented in-text with the variable Pcorr.
The factor analyses were performed using Lavaan package version 0.6-5 49 in R Studio version 3.6.2 50.All other statistical analyses were carried out in Jamovi (Version 1.1.9.0) and replicated independently in R Studio version 3.6.2.

Ethical consideration
The study was approved by the Swedish Ethical Board (Etikprövningsmyndigheten, DNR: 2018/1040-31) and fully adhered to the Declaration of Helsinki principles 51 .To our knowledge, there were no physical risks associated with participating in this study and steps were taken to ensure the participants' psychological well-being.Prior to starting the survey, informed consent was given by all participants.This included information that described the purpose of the study, which was formulated as an investigation of the relationship between cognitive styles and the perception of one's environment in relation to different kinds of drug use patterns.
A separate model included beliefs in alternative facts as the dependent variable and all drugs of reference, as measured by frequent use of the drug, and control variables was fitted.The individual predictor stimulants (β std = .2215,p = .028)was the only significant predictor of beliefs in alternative facts (F(10, 262) = 2.57, p = .006,R 2 = .0893,R 2 Adjusted = .0545,see Supplement).
A similar pattern of associations was observed when Swedish and Non-Swedish groups were analysed separately.Of note, Swedish sample scored significantly lower on CMQ compared to the non-Swedish sample (47.18 ± 20.98 and 53.08 ± 22, respectively; t 382 = 2.71, p = .007).See Supplement for details.
Adjustment of the models for mother tongue did not result in any substantial deviations in the reported results.
A similar pattern of associations was observed when Swedish and Non-Swedish groups were analysed separately.See Supplement for details.

Difference between psychedelic drug users and non-users (group comparisons): EBS
In order to further explore the differences between psychedelic users (N = 180) and non-users (N = 125), the same approach as the one used for the CMQ subitems was employed, for the different facets of EBS.As can be seen in Fig. 4, the results showed that there was a difference in the score for EBS factor Need for evidence, where psychedelic users scored lower (non-transformed M = 3.77, SD = 0.859) than non-users (non-transformed M = 3.96, SD = 0.72).This difference, however, was only marginally significant when adjusting the result for multiple testing (t(292.67)= 2.08, p = .0376,p corr = .075).

Exploratory cross-correlation network analysis
To further scrutinize the inherent associations among the variables under investigation within our psychedelic user sample (n = 180), we conducted a cross-correlation matrix analysis leveraging the Pearson r coefficient.The resultant patterns were found to be congruent with those deduced from our regression analyses and group comparisons (Fig. 5).

Personality and drug use
In order to investigate the relationships between drug use and personality, two correlation matrices containing the measures of recent and frequent use of all drugs of reference were created, both containing variables for the subscales of BFI-S (openness, conscientiousness, extraversion, agreeableness and neuroticism).We have also explored corresponding associations of the investigated variables with the psychopathology-related traits delusion proneness and schizotypy, as measured by PDI and O-LIFE, respectively.Significant findings will be reported here, the rest can be found in the Supplement.

Agreeableness
Agreeableness was found to be significantly associated with recent use of stimulants (r(392) = .116,p = .021).

Traits related to psychopathology
In order to investigate the relationship between our primary measures (CMQ, EBS) and psychopathologyrelated traits, a second correlation matrix was created (Table 3).This included the total score for CMQ, the three facets of EBS (FI-facts, Need for evidence, Truth is political), the total score of ASRS, RAADS, total score of PDI, and the different subscales of O-LIFE (Unusual Experiences, Cognitive Disorganization, Introvertive Anhedonia, Impulsive Nonconformity).

Discussion
The primary aim of the present study was to investigate the relationship between patterns of drug use and users' beliefs, with a particular emphasis on classical psychedelics.The focus was to examine potential associations between psychedelic exposure and the inclination to affirm alternative views related to three distinct epistemic beliefs, as measured by the Epistemic Belief System (EBS): (1) reliance on intuition for fact formation (Faith in intuition for facts), (2) tendency to corroborate assertions with empirical data (Need for evidence), and (3) perception of factual information as politically driven (Truth is political).In a similar way, we wanted to elucidate their relation to non-conventual beliefs (using CMQ).To further understand the complex associations between drug use patterns and beliefs, we also wanted to look at these relationships in connection to different personality and psychopathology-related traits.
The obtained results aligned with our expectations, showing a statistically significant association between recent and frequent psychedelic use and affirmations of alternative views, as assessed by the Conspiracy Mentality Questionnaire 40 .Correspondingly, recent alcohol consumption exhibited an inverse relationship with the endorsement of alternative views.Also noteworthy is the significant positive association between frequent stimulant use and alternative views.The yielded results retained significance after controlling for demographic factors and concomitant drug use, indicating that certain drug use patterns might be associated with accepting non-conventional beliefs, such as conspiracy theories, which previous research has tied to intuition-based rather than evidence-based beliefs 41 .
Contrary to our hypothesis, the study did not establish any association between psychedelic use and a preference for intuition over facts.Collectively, these findings suggest that psychedelic use may be linked to cognitive styles that question mainstream information sources rather than to outlandish conspiracy ideation 52 .
Previous literature has found connections between the endorsement of alternative facts and certain psychopathology-related traits, including facets like paranoia and schizotypy 53,54 .This finding has been confirmed in the present study as well showing significant associations between beliefs in alternative facts and facets of schizotypy.Of note, our group previously explored the link between psychedelic use and schizotypy and found that this association is largely explained by use of other drugs like stimulants 39 .However, the association with unconventional beliefs remained significant even after adjusting for other drug use, suggesting a specific link to psychedelic use.Some 26,29 , but not all 55,56 studies identified associations between alcohol use and pro-social behaviours and beliefs.In this context, our results showing negative association between alcohol use and beliefs in alternative facts provide certain support for this link.However, they must be interpreted with caution as only a minority of www.nature.com/scientificreports/ the participants reported to never have used alcohol.Of note, we also confirm previous findings demonstrating that alcohol use is associated with extraversion 7,9 .Furthermore, the use of all investigated drugs except opiates and alcohol, revealed significant associations with high scores on the personality trait openness.The use of MDMA and stimulants showed negative associations with the trait conscientiousness.These results are in line with previous literature 8,9,57 .Scoring high on openness and low on conscientiousness is not merely associated with drug use but may also reflect openness toward unconventional views and non-conformity 33,34 .
It is again important to acknowledge that most (73.26%) of our subjects were healthy, normally functioning adults and that higher scores on the CMQ do not equal pathological conspiracy ideation.In fact, it could be hypothesized that the challenging of social and ideological norms can be triggered by real societal phenomena such as the revealing of widely spread surveillance programs, e.g. the leak of the NSA files 58 , or the lack of adequate legal frameworks protecting personal integrity and privacy on the internet 59 .In this perspective, a theoretical separation between the pathological grounded conspiracist beliefs and the unconventional healthy sceptic must be assumed.This may indicate that the results, in fact, rather reflect an existing spectrum of non-conformist mentality, which to a degree is dependent on contextual factors.This view is supported in our exploration of between-group differences in main facets of the CMQ.In line with this line of thought, psychedelic users scored significantly higher than non-users on two items (secret organizations, misinformation), however, the item hidden connections ("I think that events which superficially seem to lack a connection are often the result of secret activities.") was not significant between the groups.Of note, it can be speculated that this particular item reflects an integral aspect of the pathological conspiracy construct, namely, the tendency to see patterns of extraordinary phenomena in everything 60 or pareidolia 61 .The absence of between-group differences in this domain support the idea that the results may reflect differences in non-conformity and general scepticism toward official sources of information, rather than a conspiracy ideation per se.
The presence of a difference between users and non-users of psychedelics in relation to beliefs in alternative facts suggest a least two possible explanations: (a) individuals inclined towards psychedelic use might inherently affirm alternative views more than others, irrespective of the drug's effects, due to shared traits or environment, or (b) these individuals might develop such beliefs following psychedelic experiences that enhance their openness to experience and novel, unconventional ideas [35][36][37] .Further experimental investigations assessing non-conformist ideation and epistemic beliefs before and after the administration of psychedelics could provide further clarity on these drugs' potential to significantly alter how people process information.
Taken together, our findings should not be misconstrued as advocating restrictive drug policies.Conspiracy ideation is a complex societal issue, more nuanced than a direct consequence of drug use, and could even be fuelled by societal alienation and the ostracization of users of certain drugs 62 .Hence, the stigmatization of psychedelic drug users and the resultant social polarization may cultivate the conditions that foster conspiracy ideation.Moreover, the increasing prevalence of psychedelic culture and scepticism towards mainstream media might be components of the same generational shift in attitudes, as some scholars suggest 63 .
It is important to acknowledge that there are several limitations that restrict the generalizability of our findings.First, self-reporting is an inevitable limitation when assessing unobservable behaviours like beliefs and attitudes.All scales measuring such constructs possess inherent limitations in terms of accuracy and reliability that are often beyond the reach of current scientific methodologies 64 .Specifically, reporting bias in relation to drug use have previously been found 65 , highlighting the need for objective measures.Unfortunately, for this study, such measures were unfeasible, making self-reported measures the only accessible tool.As such, the presence of bias in this study is plausible, wherein participants could either underreport or exaggerate their drug use habits.
Moreover, the study's scope was primarily focused on the Swedish population, specifically within groups where recreational drug use was expected.This specificity inevitably limits the generalizability of our findings.Cultural, social, and political contexts play a significant role in shaping both drug use patterns and belief systems.Thus, the associations observed in this study, rooted in a particular socio-cultural context, may not hold true across different populations or cultures.There may be distinct patterns in drug use and alternative beliefs amongst diverse global communities, influenced by numerous variables such as societal norms, drug policies, and cultural attitudes towards unconventional beliefs.Consequently, our findings should be cautiously extrapolated beyond the studied demographics notwithstanding the confirmed invariance for the main study scales in Swedish and non-Swedish groups and equivalent main finding's effect-sizes.Future studies with more diverse and representative samples are necessary to verify and extend our results.
The COVID-19 pandemic is another important contextual factor.It should be noted that data for this study were collected before any general restrictions were implemented in Sweden.This may be seen as a strength, as subsequent restrictions and social distancing measures may have affected responses.It also suggests caution for future replication attempts and indicates a potentially interesting area of future research exploring whether there are subpopulations particularly sensitive to such events.
This study tried to tease out potential links between drug use and beliefs, which is why both measures of frequency and recency of use were collected.Including a measurement of quantity could have improved the validity of our measures.This would have refined our measure of exposure to the different drugs to a greater extent.However, estimating an average quantity of use over time may be problematic considering it being a heterogeneous and context-specific behaviour.Intra-individual differences regarding quantity might therefore be larger than the inter-individual differences, making it a poor measure in this case.We did include a measure regarding age and drug use.This was meant to complement the other measures with the purpose of evaluating differences between longer and heavier use, compared to shorter and more recreational forms of use.Unfortunately, this was not possible as the sample was quite homogenous.Most participants reported to have started using drugs at the same age, with a few "late starters" and no "early starters".Therefore, the present study was focused on variables that represent a more nuanced understanding of exposure in order to better tease out its effects in a manner that is not confounded by age.
In conclusion, our results suggest that psychedelic use is associated with non-conformist views, specifically, with beliefs in alternative facts and the belief that facts are influenced by politics.To the best of our knowledge, these findings are unprecedented in connection to current scientific research.As a result, this study represents a first step towards filling the knowledge gap regarding the associations between different drug use patterns and information processing.Many unanswered questions remain, and a continuation of this research in a multidisciplinary direction would be beneficial in order to better understand the underlying neural and behavioural mechanisms related to these constructs.Do certain drugs exhibit lasting effects on people's perception, beliefs and behaviour or do the discovered associations rather represent effects of non-conformist subpopulations that are more likely to question conventional socio-political norms, including use of illicit drugs?What is the role of alienation of illicit drug users on their non-conformist points of view?Answering these questions could provide valuable insights into factors surrounding drug abuse and inform evidence-based drug policy-making. https://doi.org/10.1038/s41598-023-42444-zwww.nature.com/scientificreports/

Figure 1 .
Figure 1.Coefficient plot of multiple regression models for the predictors of mean score on the CMQ.Estimate: non-standardized coefficients.Model1: minimal (psychedelic use and intercept), Model2: demographic adjustment, Model3: full model adjusted for concomitant drug use.Error bars denote 95% confidence intervals.

Figure 4 .
Figure 4. Differences in mean EBS scores between psychedelic drug users and non-users.FI-facts-Faith in intuition for facts.p/p corr -p-value, unadjusted and corrected for multiple testing: * < .05;** < .01.

Figure 5 .
Figure 5. Cross-correlation network for the main investigated variables in a sample of psychedelic users (n = 180).Edges with numerical values represent Pearson correlation coefficients.The abbreviations used are as follows: CMQ-Conspiracy Mentality Questionnaire, EBS f -Faith in intuition for facts, EBS e -Need for evidence, EBS p -Truth is political; ALCp/f-temporal proximity and frequency of alcohol use, PSYp/f-temporal proximity and frequency of classical psychedelic use.