Dose-response relationships of LSD-induced subjective experiences in humans

Prugger and colleagues situate LSD within the class of classic serotonergic psychedelics, noting that these substances principally act at the 5-HT2A receptor but differ in molecular structure and broader pharmacology. Earlier research has produced extensive descriptive and clinical literature on LSD from the 1950s–1970s, followed by a research hiatus and a more recent focus on other psychedelics such as psilocybin and N,N-DMT. The authors highlight that acute psychedelic effects are temporally dynamic and that the intensity and quality of the acute experience may predict therapeutic outcomes, so characterising dose-dependent subjective effects is clinically and experimentally important. They also note that contemporary experimental doses are generally lower than many historical studies and that no prior meta-analysis has quantified dose-response relationships for LSD-induced subjective experiences using modern psychometric instruments. The present study therefore aims to estimate dose-response relationships between LSD dose (within a range of 25 µg to 200 µg) and the intensity and quality of acute subjective experiences in healthy, highly selected participants. Using psychometric data from validated questionnaires aggregated in the Altered States Database (ASDB), the investigators seek to provide reference dose-response estimates for commonly used scales measuring altered states of consciousness, to support future experimental and clinical research on low to moderate LSD dosages.

This work is a meta-analysis drawing on psychometric questionnaire data stored in the ASDB (version ASDB-v2.0_12-2021) and updated literature through June 2022. Included sources were MEDLINE-listed studies published between 1975 and 2022 that reported standardized retrospective ratings of LSD-induced subjective effects. Data were excluded if the experimental conditions involved co-administration of other substances (for example ketanserin pre-treatment), if LSD was administered as microdoses (< 25 µg), if dosage was unclear, or if data reflected recreational use. Two reports were excluded as duplicates of previously published data, and one patient dataset (N = 11) was identified but not analysed. Where necessary, the study team obtained unpublished questionnaire ratings from authors and adjusted dose values in some reports following consultation, reflecting issues with unstable capsule formulations. The meta-analysis used ratings from three questionnaires: the Altered States of Consciousness Rating Scale in its 5D-ASC and 11-ASC formats, and the MEQ30 (Mystical Experience Questionnaire). The 5D-ASC organises items into five primary dimensions, while the 11-ASC and MEQ30 provide more granular scales; the MEQ30 comprises four subscales (Mystical, Positive Mood, Transcendence of time and space, Ineffability). Reliability estimates for these instruments are reported as acceptable to very good in the source text. For statistical analysis the investigators applied linear meta-regression models with random effects to approximate dose-response relationships across the available dose range. Because the full sigmoid functional form could not be estimated given limited dose extremes, linear models were used as approximations. To accommodate multiple effect sizes from within-subject designs and repeated measures, the robust variance estimation (RVE) framework with small-sample adjustment was employed, assuming a common within-study correlation p (default p = 0.8) and conducting sensitivity analyses to assess dependence of estimates on p. Weights were computed using a correlated effects model with inverse-variance weighting and a method-of-moments estimator. Heterogeneity was assessed with I2 and Tau2. Analyses were performed in R (version 4.2.1) using the robumeta package; visualization included radar charts and dose-response plots, and the analysis syntax was made available on Github.

Fourteen studies in healthy participants were included in the analysis, yielding an overall sample of N = 322. Available psychometric data covered LSD doses between 25 µg and 200 µg and contained questionnaire ratings from the 5D-ASC, 11-ASC and MEQ30, with some datasets providing multiple questionnaire applications and a few unpublished scale scores added after author correspondence. Across most questionnaire factors and scales, higher LSD doses were associated with increased ratings of subjective effects. The strongest dose-related increases were observed for visionary phenomena, notably audiovisual synesthesia and altered imagery. Measures reflecting positively experienced ego-dissolution phenomena (depersonalisation and derealisation components) also showed dose-related increases, albeit less pronounced than the visual phenomena. In contrast, scales indexing mystical or spiritual-type experiences (MEQ30 Mystical subscale and related measures) exhibited only weak modulation by dose within the examined range. Heterogeneity metrics varied by scale. Tau2, the between-study variance, was reported as rather small for most factors and scales, but I2 indicated considerable inconsistency (60%–91%) for many scales; only Anxiety on the 11-ASC and Mystical on the MEQ30 showed small-to-moderate inconsistency (< 60%). The investigators performed sensitivity analyses regarding the assumed within-study correlation parameter p, though the extracted text does not report the detailed outcome of those tests. Administration route varied across studies (mostly oral, with one intravenous and one sublingual study), but analyses did not stratify by route or formulation.

Prugger and colleagues interpret their findings as evidence that, in controlled laboratory settings with healthy, highly selected participants, small to moderate doses of LSD (25–200 µg) reliably increase most characteristics of altered states of consciousness measured by standard questionnaires. They emphasise that within this dose range LSD most potently modulated visual-perceptual phenomena and aspects of positively experienced ego-dissolution, while mystical-type experiences were less dose-sensitive and therefore may depend more on extra-pharmacological factors such as set and setting, personality, preparation and intentions. The authors compare their dose-response pattern to a previously published psilocybin meta-analysis and report broadly similar profiles, but note two notable differences: LSD produced stronger, more dose-sensitive effects for audio-visual synesthesia, and relatively weaker, less dose-dependent effects on mystical-type scales compared with psilocybin. They discuss the possibility that LSD's broader receptor interaction profile, beyond 5-HT2A, might contribute to its distinctive propensity to induce synesthesia, while acknowledging that extra-pharmacological factors could also explain differences between substances. The manuscript further raises the point that pronounced synesthesia with LSD has been discussed in relation to rare long-term perceptual disorders such as hallucinogen-persistent perception disorder, but the authors caution that prevalence data are confounded by LSD's widespread recreational use. Key limitations acknowledged by the investigators include reliance on self-report measures to quantify private subjective states, considerable unexplained variance likely driven by heterogeneous set and setting across studies, a relatively small number of studies and datasets limiting generalisability, and incomplete coverage of very low and very high doses that prevents fitting a full sigmoid dose-response curve. They additionally note that RVE allows inclusion of dependent effect sizes but is not optimised for precise estimation of heterogeneity parameters, and that varying administration routes and formulations were not modelled separately. Given these constraints and observed heterogeneity, the authors advise treating dose-response estimates cautiously and highlight the need for more standardised reporting of extra-pharmacological variables in future research.

LSD administration in healthy, carefully screened participants in controlled settings increased most measured dimensions of altered states of consciousness within the examined 25–200 µg range. The subjective profile of small to moderate LSD doses was dominated by visual alterations and positively experienced ego-dissolution, while mystical-type experiences showed weaker dose-dependence. Despite variability across studies, the authors present these dose-response estimates as a general reference for relating observed subjective effects to expected dose-specific effects in future experimental and clinical work.