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

Lysergic acid diethylamide (LSD) is the prototype of classic serotonergic psychedelics, a group of substances which unfold their psychoactive properties predominantly via the serotonin 2 A (5-HT_2A) receptor [1]. Psychedelics embrace structurally heterogenous subgroups like phenethylamines (e.g., mescaline) and tryptamines (e.g., psilocybin, N,N-DMT) [2], as well as substances from the ergoline subgroup (e.g., LSD) which have been characterized as “rigidified tryptamines” [3]. The term ‘psychedelics’ is also used in a broader sense, including non-serotonergic drugs like ketamine, PCP or MDMA. The use of the term ‘psychedelic experience’ is not limited to the effects induced by specific substances, instead refers to a group of psychological effects. However, there is no clear definition on the exact set of consciousness alterations that define a psychedelic experience. Here, we will refer to classic serotonergic psychedelics and the effects they induce when using the term ‘psychedelics’ or ‘psychedelic experience’. Several studies suggest that qualitatively, LSD might not be differentiated from other psychedelics with regard to the induced psychologic effects [4–6]. On the other hand, anecdotal reports mention differences in subjective experiences regarding different substances [7], and LSD somewhat differs from pharmacodynamical profiles of other 5-HT_2A agonists, including a broader variety of receptor targets [3].

After its initial synthesis in 1938, LSD’s psychedelic properties have accidentally been discovered in 1943 by the Swiss pharmacologist Albert Hofmann [8]. Ever since, LSD has been the most extensively investigated psychedelic from the 1950s to the 1970s, with more than 1000 scientific papers published in the context of basic science, as well as in clinical research as a therapeutic tool [9–11]. Most intensively studied indications included, among others, alcoholism [12] and existential distress in life-threatening physical illness [13]. After a hiatus of more than 20 years, during which regulatory hurdles prevented research on psychedelics, research eventually resumed in the 1990s. Also, in recreational underground use LSD is by far the most frequently used psychedelic worldwide [14], and doses of psychedelics are often compared to LSD equivalents by users.

LSD has recently been re-evaluated for the treatment of different mental health conditions, like anxiety and depression in patients with [15, 16] and without life-threatening illness [15]. Psychological effects of psychedelics underlie specific temporal dynamics [9], including (acute) psychedelic experiences, subacute effects, and long-term (enduring) effects [17]. There is some evidence that the quality and intensity of acute psychedelic effects might predict therapeutic outcome [18]. Thus, the classification and description of acute psychedelic experiences appear to be of high importance when it comes to optimizing treatment interventions regarding efficacy and safety. In order to determine the optimal dosing ranges of LSD for future clinical studies, first and most importantly the influence of the LSD dose on the nature and intensity of acute subjective effects needs characterization. Over the last decades some gold standards for the assessment of altered states phenomena have been established in terms of several well-validated questionnaires for a retrospective assessment of altered experiences [9, 19–21]. Such standardized assessment allows meta-analytic comparisons, as recently presented to establish dose-response relationships for altered experiences induced by psilocybin [20].

The data to establish dose-response relationships for LSD is limited by the availability of studies that cohere to current research standards, as LSD has only recently returned to basic and clinical research. The range of LSD doses employed in current research is on average markedly below the doses administered in studies from the 1950 to 70 s. At that time, high doses mostly ranging between 400 and 600 μg were applied, especially in the treatment of alcohol use disorder, whereas in recent studies doses of 200 μg have not been exceeded, neither in healthy subjects [22], nor in clinical samples [15, 16, 23]. Regarding the lower end of the dose spectrum, applications of so-called microdosing are gaining increasing interest both in recreational and scientific settings. It is thought that such low doses, which do not induce subjectively identifiable drug effects, can positively affect cognitive processes and improve mental health indicators [24–27]. The exact dose range of LSD microdosing is still debated; a recent literature review suggests a “plausible microdose range” for LSD of 6–20 μg [24]. Only few studies directly tested the dose-dependency of LSD’s effects [22, 26, 28]. With the application of within-subject designs, these studies provide well-controlled experimental conditions, while they do not allow to assess study-specific impacts on participants experiences. Such non-pharmacological effects appear, however, particularly pronounced for psychedelic substances, and are generally referred to as set and setting effects [29, 30]. Cross-study comparisons are essential to obtain data-based estimates of the variability of subjective LSD experiences across research sites and experimental protocols, as no corresponding meta-analysis is available to date.

With the present meta-analysis, we aim to obtain estimates of the relationship between LSD doses and the quality of psychedelic experiences. The data stem from the Altered States Database (ASDB, http://alteredstatesdb.org↗, [31]), which is a regularly updated database with questionnaire data extracted from articles identified by a systematic literature review, adhering to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 Statement Guidelines [32]. Available data was found for the Altered States of Consciousness Rating Scale (5D-ASC, 11-ASC) and the Mystical Experience Questionnaire (MEQ30) for a dose range up to 200 μg.

A meta-analysis on psychometric data was performed to estimate dose-response relationships of subjective effects of LSD. The analysis included data from applications of doses from 4 to 200 μg LSD base, in healthy and clinical study participants. Given that the data is based on retrospective subjective reports relying on introspection, we did not employ assumptions on the specific shape of the fitted model (e.g., not applying a linear or a sigmoidal model); instead, we opted for fitting restricted cubic splines that allow for flexible model fits. Most of the fitted dose-response functions closely resemble sigmoid curves, which is expected and plausible for biophysiological processes. We established robust dose-response relationships, in which LSD dose increased effects on all scales and factors of the included questionnaires. On the 5D- and 11-ASC, the strongest dose-modulation was found for items referring to changes in perception and imagination (Visionary Restructuralization, Changed Meaning of Percepts) as well as positively experienced dissolution of boundaries between self and surroundings (Oceanic Boundlessness), whereas Anxiety, Auditory Alterations and Dread of Ego Dissolution were barely modulated. On the MEQ30, the strongest dose-modulation was found for Ineffability and the weakest for Mystical.

Most factors and scales exhibit a sigmoid-like shape of dose-effects that either flattens out or reaches a plateau at around 100 μg, while scales and factors with relatively small dose-modulation (Auditory Alterations, Dread of Ego-Dissolution, Anxiety) increase rather linearly. This is in line with a dose-response, within-subject study by Holze et al. (2021) that found similar rating scores for 100 µg and 200 µg for Blissful State, Insightfulness, and Changed Meaning of Percepts [28], but significantly higher rating scores for 200 µg compared to 100 µg on the scale Dread of Ego Dissolution (reflecting disembodiment and fear) [28], indicating a ceiling effect regarding positive subjective effects [65]. Despite considerable variability of effects, the present meta-analysis supports the suggestion that doses between 50 and 100 µg LSD base induce substantial psychological effects in most subjects—a dose range currently used in most clinical trials for therapeutic applications [65]. Accordingly, a dose of about 100 μg LSD base most likely induces a pattern of subjective experiences that resembles what is often called a “(full) psychedelic experience”, while the likelihood of adverse reactions is kept low. Please note that the term “psychedelic experiences/state” is not used consistently and no agreement on defining criteria of a psychedelic experience have been settled on.

LSD doses below 10 μg [65] or 20 μg [24] have previously been proposed as the “microdosing” range of LSD. The fitted curves in our analysis show an onset of effects below 20 μg LSD base. For instance, visual phenomena (11-ASC: Audio-Visual Synesthesia, Complex and Elementary Imagery) show an x-axis intercept at around 10 μg, and at 20 μg reach around 10% of the maximum score. These effects are unlikely due to placebo effects, as those of the given studies which included a placebo condition reported only negligible effects [22, 26, 28, 35, 39, 43–45, 66–69]. However, these estimates are strongly influenced by the fact that multiple studies investigated doses below 20 μg, while there are no studies available in the dose range from 20 to 50 μg, resulting in an overestimation of the curves slope at their onset. Most studies below 20 μg effectively show null-effects and make the microdosing range of 0–20 μg plausible, but more data is needed to determine the average onset of effects more precisely.

A recent systematic review indicates a mediating role of mystical-type experiences for treatment outcomes of psychedelic therapy [70], so special attention is drawn on the dose range to predict such experiences. Barrett et al. (2015) suggest that complete mystical-type experiences are reached with scores of ≥60% on each of the four factors of the MEQ30 [56]. Applying this criterion, the present analysis suggests that such experiences are rather unlikely to occur in the given dose range. Similarly, Liechti et al. (2017) report that full mystical-type experiences occur rather rarely at LSD doses of 200 μg. The authors suggest that mystical-type and spiritual experiences might highly depend on participant characteristics and experimental setting [39], which might explain that 25% and 35% of participants in a study by Family et al. (2022) met criteria for a full mystical experience after 50 µg and 100 µg, respectively [71]. However, considering the results of the MEQ30 factors determined in this meta-analysis, datapoints of mentioned study partially lie outside or in the margins of the calculated confidence intervals and thus may not reflect general outcomes. Overall, full mystical-type experiences seem unlikely to be induced with doses below 200 µg and their occurrence appears to be strongly influenced by non-pharmacological factors. Targeted therapeutic implementation of other factors than dose need to be further investigated.

Non-pharmacological factors that influence psychedelic experiences are often categorized in set (participant personality and preparation and expectation of substance use) and setting (environment of substance administration) [30, 72] and are thought to lead to considerable inter- and intra-individual variability of subjective effects [73, 74]. The studies included in our analysis vary greatly with regard to experimental setting. Bershad et al. (2019) reports of “living-room style” environments with possibilities to relax, read or watch movies between measurements [44], similar to Preller et al. (2017) (“in an aesthetic living-room-like room”) [35] and de Wit et al. (2022) [45]. In other studies, however, participants were also given tasks during LSD applications that involved greater effort, potentially inhibiting the manifestation of certain effects [75]. Previous work indicated that spatially constrained neuroimaging procedures may be demanding for some individuals and could increase the likelihood of challenging experiences [17]. Carhart-Harris et al. (2016) reported on fMRI and MEG measurements of over 60 min each, as well as previous MRI environment habituation and a subsequent battery of cognitive and behavioral tests [66]. From the studies included, also Holze et al. (2020) [67] and Liechti et al. (2017) [39] (fMRI data published in Müller et al., (2017) [41]) reported of fMRI measurements during the experimental session. Schmid et al. (2015) [22], Liechti et al. (2017) [39], Holze et al. (2021 [28], 2022 [5]) reported on “quiet standard hospital patient room” environments, and the study procedure by Family et al. (2022) [71] included a 60 min breathing exercise. Additional factors influencing the psychedelic experience and thereby increasing the variability within and between the given datasets may involve the subject’s age, previous experience with psychedelics or other mind-altering substances, as well as differences in individual pharmacokinetics [76, 77]. The study by Family et al. (2022) for instance reports of applications of 50 µg LSD to predominantly (9 of 12) psychedelic-naïve participants, as well as 75 and 100 µg LSD to psychedelic-non-naïve participants. They report of higher questionnaire ratings on all factors of the 5D-ASC and MEQ30 in the (naïve) 50 µg cohort compared to the (non-naïve) 75 µg cohort, as well as higher ratings on most factors in the (naïve) 50 µg cohort compared to the (non-naïve) 100 µg cohort [71]. The study by Wießner et al. (2021) also reports relatively high 11-ASC scores (especially for the factors Insightfulness, Spiritual Experience, Blissful State, and Complex Imaginary) [68] compared to other studies on the same LSD dose included in this analysis. This may result from relatively high lifetime use of other psychedelics or mind-altering substances among study participants (particularly ayahuasca, with a mean lifetime use of 69 ± 131 S.D.) and the fact that most participants (67%) identify themselves as spiritual. Other studies, however, have not observed any differences in subjective effects between experienced and non-experienced LSD users when compared within the same study [5, 22, 28]. Therefore, differences in extra-pharmacological factors likely led to the observed variance in the ratings of subjective experiences. The confidence intervals of the present study can yield as data-driven estimates for the expected variability of experiences, also in clinical samples (Note: The confidence intervals refer to the mean group effects, while the variance across participants can be substantially higher). Interestingly, factors assessing alterations of visual perception and cognitive abilities exhibit less variability (smaller CI range) than factors assessing experiences of unity, blissful state, and spirituality. This could be due to their highly personal character, which might be stronger influenced by non-pharmacological factors. Alternatively, variability might stem from less reliable assessment of these high-order concepts as compared to factors assessing perceptual alterations. Future studies with a more comparable assessment of non-pharmacological factors are needed to allow for systematic investigation of their exact contributions in shaping subjective experiences.

Further limitations should be considered when using the analysis at hand to determine LSD doses. First, administration methods varied across studies. Most studies administered LSD orally as capsules, while it was administered sublingually in two studies [44, 45] and intravenously in one study [66]. Moreover, LSD formulations differed across studies: most studies applied LSD base (or provided the dose calculated for LSD base), while some doses referred to LSD salt in the form of LSD tartrate—the typical formulation sold on the black market. No direct experimental comparison of the bioequivalence of the two formulations has been made to date. We transformed doses according to Holze et al. (2021) [46], who suggested that 1 µg of LSD base corresponds to ~1.46 µg of 1:1 LSD tartrate or 1.23–1.33 µg 2:1 LSD tartrate, depending on the salt form and amount of crystal water [65]. Moreover, the results reflect estimates on the variability of average effects in study samples based on doses of up to 200 µg, so the experiences of individual subjects may differ considerably. Finally, it should be noted that study samples are usually comprised of highly selected and well-prepared participants. For these reasons and because the quality and quantity of recreationally used products is often unclear, the present results do not necessarily apply to the general population or to recreational use outside controlled laboratory or therapeutic settings.

In conclusion, the present meta-analysis on subjective effects of LSD in study participants revealed dose-dependent alterations of consciousness with substantial heterogeneity of effects across laboratory and therapeutic settings on all scales and factors assessed by given questionnaires. The strongest dose-modulation was found for perceptual and imaginative changes followed by positively experienced dissolution of boundaries between self and surroundings with a limited spiritual or mystical-type character in the given dosage range. Dose-response relationships resemble sigmoid functions on most factors and scales of the questionnaires, with an observable ceiling effect at circa 100 μg LSD base. Results may be used as a general reference to relate observed with expected dose-specific effects.