Among psychedelic-experienced users, only past use of psilocybin reliably predicts nature relatedness

Earlier research has reported a positive association between experience with classic serotonergic psychedelics and people’s sense of connectedness to the natural environment, commonly referred to as nature relatedness (NR). Such studies, however, often do not distinguish between different psychedelic compounds despite their distinct pharmacology and typical use contexts, and they are frequently unable to separate effects of substance use per se from confounding variables that differentiate users from nonusers (for example personality, values, or lifestyle). These limitations leave open whether reported links between psychedelics and NR reflect effects of particular substances, general characteristics of people who use psychedelics, or expectancy and setting influences. Forstmann and colleagues sought to address these gaps by reanalysing five independent datasets (combined N = 3817) to determine which specific psychedelic compounds, if any, reliably predict self-reported NR. The study focused primarily on participants who had used at least one psychedelic substance, allowing the investigators to control for many between-group differences that distinguish psychedelic users from nonusers, and tested both continuous and binary indicators of lifetime use for several substances (psilocybin, LSD, DMT/ayahuasca, mescaline, ibogaine, ketamine, and Salvia divinorum). The goal was to identify whether experience with particular compounds uniquely predicts overall NR and its subdimensions (NR-self, NR-experience, NR-perspective).

The investigators combined data from five independent samples collected online and offline across multiple countries and participant pools (total N after exclusions = 3817; demographic summary reported as 1836 male, 1854 female, 40 nonbinary/other/none; mean age = 30.60, SD = 11.51). Main analyses focused on samples 1–4 (combined N = 3076) because those samples provided a graded measure of lifetime experience with individual substances; sample 5 provided only binary (yes/no) indicators and was therefore reserved for secondary analyses. Across samples, 1887 participants reported lifetime experience with at least one psychedelic substance; the main regression analyses on psychedelic-experienced participants used N = 1289 (reported in the extracted text). Substance-use variables were harmonised across samples. For samples 1–4, original Likert-type responses (5– to 7-point scales) were rescaled to values from 0 (no previous experience) to 6 (extensive experience). Sample 5 used binary indicators and thus could not be rescored in the same way; for binary secondary analyses, the continuous indicators from samples 1–4 were dichotomised (0 = no experience, >0 = experience) and combined with sample 5. The seven substances analysed in all questionnaires were psilocybin, LSD, DMT/ayahuasca (composite), mescaline, ibogaine, ketamine, and Salvia divinorum. Nature relatedness was measured with subscales for NR-self (extent to which people incorporate nature into their self-construal), NR-experience (familiarity and preference for spending time in nature), and NR-perspective (worldview concerned with human impacts on nature). Scores were recorded on 1–5 agreement scales, then z-transformed within each sample prior to analysis. Age and gender (recoded to male, female, nonbinary/other/none and entered as two dummy variables) were included as covariates, and dummy variables for sample of origin were added to account for sample characteristics. Primary inferential approaches included Spearman correlations, linear regression models predicting NR from substance-use indicators and covariates, and a fully saturated structural equation model regressing the three NR subdimensions on the same predictors. Secondary analyses used binary coding to include sample 5 and examined access to nature and motives for nature connection using one-way ANOVAs on a subsample from sample 5.

Raw correlations across samples 1–4 (N = 3076) replicated previously reported positive associations between lifetime psychedelic use and overall NR, particularly the NR-self and NR-experience subscales. The strongest bivariate associations were observed for past experience with psilocybin and LSD, with weaker correlations for DMT, ketamine, and Salvia divinorum. The substance-use indicators, however, were moderately intercorrelated. Restricting analyses to participants who had used at least one psychedelic (reported N = 1280 in the correlation step; main regression reported N = 1289), only past use of psilocybin correlated significantly with overall NR (r = 0.15, p < 0.001) and with NR-self (r = 0.15, p < 0.001) and NR-experience (r = 0.14, p < 0.001). LSD use correlated significantly only with NR-self (r = 0.10, p < 0.05). To estimate unique predictive power while controlling for intercorrelations among substances and covariates, the investigators ran linear regressions. In the primary regression predicting z-transformed overall NR among psychedelic-experienced participants (N = 1289), gender and age were significant covariates: women scored higher on NR (b = 0.44, β = 0.22, SE = 0.06, t = 7.57, p < 0.001, 95% CI = 0.32 to 0.55), and older participants scored marginally higher (b = 0.01, β = 0.08, SE < 0.01, t = 2.43, p = 0.015, 95% CI = 0.00 to 0.01). Importantly, of the seven substance-use indicators, only psilocybin emerged as a significant unique predictor of overall NR (b = 0.10, β = 0.18, SE = 0.02, t = 5.34, p < 0.001, 95% CI = 0.06 to 0.13). In other words, when controlling for use of multiple psychedelics, demographic covariates, and sample origin, past psilocybin experience was the only substance reliably associated with higher NR. To address whether this association reflected broader interindividual differences among users of different substances, the researchers examined subsamples of exclusive users. Among 153 participants who had used psilocybin but no other psychedelic, higher self-reported frequency of psilocybin use positively predicted overall NR (b = 0.19, β = 0.24, SE = 0.07, t = 2.86, p = 0.005, 95% CI = 0.06 to 0.32). By contrast, in 96 participants who had used LSD exclusively, frequency of LSD use did not predict NR (b = 0.08, β = 0.12, SE = 0.08, t = 1.02, p = 0.312, 95% CI = -0.08 to 0.25). A structural equation model regressing the three NR subdimensions on the substance-use indicators, age, gender, and sample origin showed that the three subdimensions covaried and that only psilocybin significantly predicted all three subdimensions. The pattern of effects matched earlier reports: psilocybin most strongly predicted NR-self, followed by NR-experience, and then NR-perspective. No other substance showed a comparable multivariate prediction pattern. Secondary analyses using binary substance indicators (to include sample 5) focused first on psychedelic-experienced participants (N = 1887). Again, age and gender were significant covariates (women and older participants higher on NR). Psilocybin remained a significant predictor of overall NR (b = 0.33, β = 0.13, SE = 0.06, t = 5.75, p < 0.001, 95% CI = 0.21 to 0.44). In this binary model, mescaline (b = 0.15, β = 0.05, p = 0.024) and DMT (b = 0.17, β = 0.09, p = 0.002) also emerged as positive predictors, but after adjusting for multiple comparisons only DMT remained significant. Expanding to the full combined sample including nonusers (N = 3817) and using binary indicators, psilocybin again showed the largest effect (b = 0.39, β = 0.19, SE = 0.05, t = 8.64, p < 0.001, 95% CI = 0.30 to 0.48). Other substances (LSD, DMT, mescaline) showed smaller positive associations in this full-sample binary analysis, but the psilocybin effect was roughly three times larger than the next-largest effect (DMT). Finally, using a subsample from sample 5 that assessed situational variables around a recent psychedelic experience (741 respondents, with N = 681 reporting access-to-nature information and N = 653 reporting motives), the investigators tested whether access to nature during the experience or the motive to enhance connection with nature differed across substances. Access to nature did vary across substances (one-way ANOVA F(6,675) = 2.36, p = 0.029), but this effect was driven largely by mescaline users (82% had access to nature), which was significantly above the grand mean. Access for psilocybin (34%), LSD (35%), smoked DMT (25%), and ayahuasca (36%) did not differ from the grand mean. Motives to increase connection with nature did not differ between substances (F(6,646) = 0.50, p = 0.808), with an overall moderate endorsement (mean = 2.97, SD = 1.03). These results suggest that, in this subsample, neither greater access to nature nor stronger motive to connect with nature fully account for the stronger association between psilocybin use and NR.

Across a pooled reanalysis of five independent samples, Forstmann and colleagues report that among people who have used psychedelic substances, past experience with psilocybin is the only substance-level indicator that reliably and uniquely predicts self-reported nature relatedness. Effects were strongest for the NR-self subdimension (the degree to which people incorporate nature into their self-construal), and the association persisted when analyses controlled for age, gender, sample origin, and for experience with other psychedelic substances. Additional analyses indicated that, among people who had used only psilocybin, greater frequency of use predicted higher NR, whereas an analogous analysis among exclusive LSD users did not show such a relationship. The investigators interpret these findings as evidence that not all psychedelics are equally associated with subsequent reports of connection to nature. They suggest that the observed specificity for psilocybin might reflect pharmacological differences between substances, contextual differences in how they are used, or differences in user expectations and cultural associations. However, in the single sample that assessed situational factors around a recent experience, neither greater access to nature during the experience nor stronger motives to connect with nature were convincingly higher for psilocybin compared with other substances, which tentatively argues against explanations based solely on setting or deliberate motivation. The authors also note that DMT/ayahuasca showed a weaker and less consistent association with NR, and that collapsing inhaled DMT and ayahuasca into a single composite may have obscured important distinctions between modes of administration and cultural contexts. Key limitations acknowledged by the investigators include the correlational, retrospective nature of the data, which precludes definitive causal inference; heterogeneity across samples in measurement instruments and in how substance use was assessed (necessitating rescaling and, for secondary analyses, dichotomisation); and potential cultural or sample-specific effects (for example, other studies have found ayahuasca/DMT to be the primary predictor of NR in different populations). They further note that composite coding of DMT and ayahuasca prevented disentangling effects of mode of administration and brew constituents, and that the analyses relied on self-report measures that may be vulnerable to bias. Forstmann and colleagues recommend that future work focus on psilocybin when investigating psychedelic effects on nature relatedness, and that more rigorous designs are needed to establish causality and mechanisms. Suggested next steps include larger-scale double-blind, placebo-controlled trials, experiments that hold pharmacology constant while manipulating set and setting (for example indoor versus outdoor administration), more precise measurement of expectations and intentions, and investigation of whether mystical-type experiences or ego-dissolution mediate effects on NR. The authors also highlight the potential relevance of NR as a psychological pathway linking psychedelic experiences to mental health and pro-environmental behaviour, while noting that any potential benefits must be weighed against the known risks of psychedelic use in some contexts.