LSD, madness and healing: Mystical experiences as possible link between psychosis model and therapy model

Since its discovery, LSD has been studied both as an experimental model of psychosis and as a therapeutic agent, creating a longstanding paradox in psychedelic research. Wießner and colleagues summarise similarities between psychedelic and psychotic phenomenology (altered perception, self and time, impaired cognition, magical thinking) and highlight aberrant salience—the inappropriate assignment of importance to stimuli or internal representations—as a central mechanism proposed in schizophrenia research that may underpin hallucinations and delusions. On the therapeutic side, the authors emphasise suggestibility (responsiveness to suggestions) and mindfulness (intentional moment-to-moment attention) as candidate mechanisms by which psychedelics may facilitate clinical change; prior work had shown that several psychedelics can increase suggestibility and that compounds such as psilocybin and ayahuasca can enhance mindfulness-related capacities, but the effects of LSD on mindfulness had not been examined. This study set out to test whether LSD at a low dose (50 µg) would increase aberrant salience (testing the psychosis model) and increase suggestibility and mindfulness (testing the therapy model), and to examine how those changes relate to the subjective psychedelic experience. The investigators hypothesised that LSD would increase aberrant salience, increase suggestibility and mindfulness, and produce positive correlations among LSD-induced changes in aberrant salience, suggestibility, mindfulness, and dimensions of the psychedelic experience (e.g. mystical experiences, ego-dissolution).

The researchers used a randomized, double-blind, placebo-controlled, crossover design with two treatment sessions (50 µg oral LSD and inactive placebo) separated by a 14-day washout. Participants were randomly assigned to treatment order. Ethical approval and safety procedures for human psychedelic research were in place. A convenience sample of 25 healthy adults was recruited; one participant withdrew after the first session, leaving 24 completers. Inclusion criteria included age over 21 and at least one prior LSD experience; exclusions covered current psychiatric symptoms, personal or first-degree family history of psychotic disorder, psychiatric medication, substance use disorder, serious medical conditions, pregnancy, and non-native Brazilian Portuguese. Self-reported prior drug use was recorded. The absolute LSD dose corresponded to a mean relative dose of 0.69 ± 0.18 µg/kg (range 0.45–1.11). Subjective drug effects were monitored over the session using visual analogue rating scales for intensity and valence (repeated at frequent intervals) and by standard questionnaires administered at 7 hours post-dosing: the Altered State of Consciousness Questionnaire (ASC), Mystical Experience Questionnaire (MEQ), Challenging Experience Questionnaire (CEQ), and Ego-Dissolution Inventory (EDI). Aberrant salience was measured with an adapted state version of the Aberrant Salience Inventory (ASI), translated into Brazilian Portuguese (validation ongoing). Suggestibility was assessed with a translated and split-version Creative Imagination Scale (CIS), and mindfulness was measured with Brazilian Portuguese versions of the Five Facet Mindfulness Questionnaire (FFMQ), Mindful Attention Awareness Scale (MAAS), and Experiences Questionnaire (EQ) at baseline (T0), 24 hours (T1) and two weeks (T2) after dosing. Study sessions were supervised by a psychologist and a psychiatrist; the same investigators attended both sessions for each participant. LSD or placebo was given at 9:30 a.m.; suggestibility testing occurred at c. 2:15 p.m., and the psychedelic questionnaires at c. 4:30 p.m. Follow-up measurements were completed the next morning (including ASI and mindfulness at T1) and online at two weeks (T2); a qualitative contact at four months noted no persisting side effects. Statistical analyses used repeated-measures General Linear Models (GLMrep) with treatment (LSD, placebo) as a within-subject factor and treatment order as a between-subject factor, complemented by within-subject factors where relevant. Effect sizes were reported as partial eta squared (ηp2). Spearman rank correlations examined relationships among LSD-induced changes (Δ = LSD − placebo) on psychedelic experience indices, ASI, CIS, and mindfulness measures. A two-tailed α = 0.05 was used; pairwise comparisons and correlations were Bonferroni-corrected as described in the paper.

One ASC dataset was lost (n = 23 for ASC analyses); otherwise 24 participants completed both sessions. Subjective intensity and valence measures and the psychedelic questionnaires showed robust LSD effects. ASC total and nearly all ASC factors were significantly higher under LSD than placebo (GLM: F(1,21) = 99.01, p < 0.001, ηp2 = 0.83). MEQ total and all MEQ factors were also markedly increased under LSD (F(1,22) = 137.61, p < 0.001, ηp2 = 0.86). CEQ showed a weaker but significant main effect, indicating only modestly increased challenging experiences. Aberrant salience (ASI) increased significantly after LSD relative to placebo (F(1,22) = 55.00, p < 0.001, ηp2 = 0.71), with pairwise increases in the Total score and all five ASI factors (Increased Significance; Senses Sharpening; Impending Understanding; Heightened Emotionality; Heightened Cognition), the last reaching p = 0.002. Suggestibility (CIS) was higher under LSD than placebo (F(1,22) = 12.03, p = 0.002, ηp2 = 0.35); exploratory modality analyses suggested increases in Extern Ambience, Weight, Sensation and a marginal effect for Taste, whereas Intern Ambience did not increase (modal analyses were exploratory and not corrected for multiple comparisons). Mindfulness measures (FFMQ, MAAS, EQ) showed no significant main effects at the group level. Correlation analyses examined relationships among LSD-induced changes (Δ scores). ΔASI Total correlated strongly with several psychedelic indices: ΔASC Total (rs = 0.71, p = 0.002), ΔMEQ Total (rs = 0.72, p = 0.001), and ΔEDI (rs = 0.82, p < 0.001). Multiple ASI factor scores also showed moderate-to-high correlations with ASC, MEQ and EDI subscales. ΔCIS (suggestibility) did not correlate with other LSD-induced effects. Although no group-level mindfulness change was observed, individual changes in mindfulness showed relationships with psychedelic measures: short-term (T1) increases in mindfulness decentering (EQ) correlated positively with ΔASC and ΔMEQ, whereas at two weeks (T2) there were moderate negative correlations between psychedelic indices (ΔASC, ΔMEQ, ΔEDI, ΔASI) and mindfulness measures (ΔEQ and FFMQ Nonreact). The authors report several moderate-to-high inter-correlations among ASC, MEQ and EDI, fewer correlations involving valence and CEQ, and none for peak intensity (ΔInt).

Wießner and colleagues interpret the findings as evidence that a low (50 µg) dose of LSD produces an identifiable psychedelic state characterised by altered consciousness, mystical-type experiences and ego-dissolution while eliciting only mild challenging effects. The increase in aberrant salience supports the proposition that LSD can model psychotic-like meaning attribution processes; ASI increases spanned sensory sharpening, heightened perceived significance and cognitive and emotional aspects associated with salience attribution. The authors emphasise that the pattern of correlations—robust links between aberrant salience, complex imagery, mystical experiences and ego-dissolution—highlights meaning attribution and visionary phenomena as central to the overlap between psychedelic and psychotic-like states. On the therapeutic side, LSD increased suggestibility even at a low dose, which the investigators suggest may point to a ‘‘late therapeutic window’’ when intensity and cognitive impairment are reduced but receptivity to therapeutic suggestions remains elevated. By contrast, mindfulness did not increase at the group level; individual-level analyses indicated short-term positive correlations between mystical/ego-dissolution experiences and decentering but unexpected negative associations at two weeks, leading the authors to urge caution in interpreting mindfulness effects. Suggestibility did not correlate with other psychedelic measures, suggesting that therapeutic suggestions may exert effects independently of the phenomenology captured by the psychedelic questionnaires. The authors note several limitations that constrain interpretation: participant blinding was difficult despite the low dose, the randomisation produced imbalances in education levels, and the crossover design may carry undetected carryover effects. The convenience sample was not representative (underrepresentation of women, non-Caucasian and lower-education groups), abstinence relied on self-report without drug screening, and the ASI adaptation to a state measure had not completed validation. Wießner and colleagues also acknowledge that ASI does not directly measure psychotic symptoms and that complementary pharmacological or neurophysiological assessments would be needed to clarify mechanisms. Despite these limitations, the investigators propose that mystical experiences and ego-dissolution may functionally link the psychosis and therapy models—i.e. that salience-driven reductions in ego-boundaries and heightened meaning attribution could facilitate therapeutic perspective change—and recommend future studies to test whether therapeutic suggestions that foster mystical experiences can improve outcomes in psychedelic-assisted therapy.

The authors conclude that a low dose of LSD elicits psychotic-like features (increased aberrant salience) while also enhancing suggestibility and producing mystical and ego-dissolution experiences that are relevant to therapeutic processes. They argue that mystical experiences may bridge the psychosis model and the therapy model of psychedelics, and suggest that guided suggestions that promote such experiences could boost the efficacy of psychedelic-assisted therapy. The paper calls for further research to replicate findings in clinical samples, to validate measures (including the state ASI), and to investigate dose, substance, setting and placebo influences as well as underlying neurobiological mechanisms.