The paradoxical psychological effects of lysergic acid diethylamide (LSD)

Psychedelic lysergic acid diethylamide (LSD) produces profound, unusual alterations in consciousness and historically has been described both as a model of psychosis and as a therapeutic adjunct. Early clinical use in the 1950s and 1960s explored LSD for 'analytical' psychotherapy and for experimentally modelling psychoses, but research largely ceased in the late 1960s. More recent controlled studies with other classic psychedelics have revived interest by reporting a paradox: acute experiences often contain distressing or psychosis-like elements yet can be followed by durable improvements in wellbeing and personality (for example, increased openness and optimism in healthy volunteers and clinical benefits in addiction and anxiety contexts). Population studies and meta-analyses have not found strong evidence that psychedelic exposure increases long-term mental-health problems when screening and set/setting are adequate. Carhart-Harris and colleagues designed the present study to examine this apparent paradox directly for LSD in healthy volunteers. The stated aim was to measure both acute subjective effects and mid-term (2-week) psychological outcomes after a single controlled administration, using validated instruments to quantify altered states, psychotomimetic symptoms, optimism, personality traits and delusional thinking. The authors hypothesised that LSD would produce emotional lability and psychosis-like symptoms acutely, but would increase psychological wellbeing and trait openness at the 2-week follow-up.

The study used a placebo-controlled, within-subjects crossover design with balanced dosing order. Twenty healthy volunteers were recruited and each attended a screening visit and two dosing sessions separated by at least 2 weeks. Participants were blind to session order; the research team was not, and dosing order was assigned sequentially so that odd-numbered recruits received LSD in session 1 and even-numbered recruits in session 2. The study was conducted under institutional and regulatory approvals and followed Good Clinical Practice. Eligibility screening excluded people younger than 21, those with diagnosed psychiatric illness or a first-degree family history of psychosis, individuals without prior experience of classic psychedelics, pregnancy, problematic alcohol use, or other medically significant conditions. Screening procedures included routine blood tests, ECG, vital signs and a brief neurological exam. All enrolled participants were judged physically and mentally healthy and had no histories of dependence. On dosing days participants arrived in the morning, provided urine samples and were prepared for administration; a cannula was placed. The dose described in the extracted text was 75 µg of LSD; however, the extraction contains inconsistent statements about route and infusion (some passages state intravenous administration and infusion over 2 minutes, others state oral administration). Because the extraction is unclear on this point, the exact route and infusion procedure cannot be definitively reconstructed from the provided text. After dosing, participants underwent a neuroimaging protocol (mock scanner habituation, MRI at ≈1 h post-dose, then MEG at ≈3.5 h post-dose) followed by behavioural and cognitive testing later in the day. A psychiatrist and a dedicated researcher attended each participant throughout dosing; the same researcher accompanied each volunteer for both sessions. Participants completed acute subjective measures at the end of each dosing day and mid-term measures electronically at baseline and 2 weeks after each session. Primary acute outcomes were the Altered States of Consciousness questionnaire (ASC) and the Psychotomimetic States Inventory (PSI), completed with reference to peak effects or overall experience. Mid-term outcomes at 2 weeks included the Revised Life Orientation Test (LOT-R) for optimism, the Revised NEO Personality Inventory (NEO PI-R) for personality traits, and the Peters' Delusions Inventory (PDI) for delusional thinking. Additional baseline measures included the Beck Depression Inventory, QIDS, STAI, Ruminative Response Scale, Dysfunctional Attitudes Scale and a modified Tellegen Absorption Scale. For analysis, repeated-measures ANOVAs tested between-condition differences on multi-factor instruments (ASC, PSI, PDI). Predicted changes in NEO openness were tested with a paired t test; other NEO domains were examined with paired t tests and Bonferroni correction. The LOT-R was analysed with a paired t test. Order was included as a factor where relevant, effect sizes (Cohen's d) were reported for primary outcomes, and Pearson correlations (two-tailed) were used for predictor analyses, with Bonferroni correction applied for multiple comparisons.

Acute subjective effects: Repeated-measures ANOVA indicated a strong drug effect on altered states (F1,19 = 82.6, p < 0.001) and a significant drug × ASC interaction (F1,19 = 10.3, p < 0.001). Post-hoc tests showed all 11 ASC dimensions were significantly higher after LSD than after placebo (p < 0.01). Visual-perceptual changes (elementary and complex imagery, audio–visual synaesthesia) were prominent, but positive-affect ratings were also elevated: the ASC 'blissful state' factor increased by +0.37 (S.D. = 0.28, Cohen's d = 1.65), while the 'anxiety' factor increased by +0.15 (S.D. = 0.2, Cohen's d = 1.03). PSI scores indicated substantial psychotomimetic effects, with a mean total PSI score of 45.8 (S.D. = 21) under LSD. Mid-term effects (2 weeks): Optimism (LOT-R) increased significantly two weeks after LSD (t = 2.91, df = 18, p = 0.005, Cohen's d = 0.56). Trait openness (NEO PI-R) also increased (t = 1.95, df = 19, p = 0.03, Cohen's d = 0.16). No comparable changes in optimism or personality were observed after placebo. There was an exploratory trend towards increased agreeableness post-LSD (t = 2.2, p = 0.038, Cohen's d = 0.21) that did not survive correction for multiple comparisons (corrected p = 0.15). Peters' Delusions Inventory scores showed no significant change at 2 weeks (p > 0.05); exploratory trends suggested possible reductions in distress and preoccupation related to delusional thoughts (p ≈ 0.07), but these were not significant. Predictors: Several baseline and acute variables correlated with mid-term changes, but none of these correlations survived correction for multiple comparisons. Baseline agreeableness correlated negatively with increases in optimism (r = -0.56, p = 0.014), and baseline depression and anxiety correlated positively with increases in openness (r = 0.53, p = 0.024; r = 0.49, p = 0.04). Acute anxiety correlated negatively with post-LSD increases in optimism (r = -0.47, p = 0.04). Conversely, impaired/disorganized cognition scores on the ASC and PSI correlated positively with later increases in openness (ASC r = 0.44, p = 0.049; PSI r = 0.52, p = 0.019). The authors report that these associations did not meet significance after applying the stricter corrected thresholds.

Carhart-Harris and colleagues interpret the data as supporting the paradoxical profile of LSD: potent acute psychotomimetic and perceptual effects co-occurring with predominantly positive mood in many participants, followed by modest but significant mid-term increases in optimism and openness without increases in delusional thinking. The PSI results are emphasised as indicating stronger psychosis-like effects with LSD than have been observed with ketamine, THC or cannabis in prior comparisons cited by the authors; for context, the paper reports previously published mean PSI totals of 24 (ketamine), 15.9 (THC) and 33.3 (cannabis), versus 45.8 in the present LSD sample. Still, group-level mood effects skewed positive: the ASC 'blissful state' scores exceeded 'anxiety' scores, and previous controlled work with LSD has reported MDMA-like positive mood effects. The authors acknowledge several limitations. Follow-up duration was relatively short (2 weeks) though longer-term follow-up was planned. The study was single-blind: participants were blind to order but the research team was aware, and the authors note that double-blinding is challenging in psychedelic research because the blind is often ineffective. An inert placebo was used to provide a clear baseline for neuroimaging contrasts, but the lack of an active placebo may have reduced the effectiveness of blinding. The neuroimaging procedures (MRI and MEG sessions) required by the protocol could have been demanding and may have influenced acute subjective reports; the authors note that the scanner environment might have contributed to the especially high PSI scores. For mechanism, the discussion centres on serotonin 2A receptor (5-HT2A R) agonism as central to psychedelic pharmacology. The authors outline a hypothesis in which 5-HT2A R stimulation increases cognitive flexibility and induces a transient 'entropic' or disordered mode of brain function. This elevated cognitive entropy could underlie both the positive (e.g. enhanced imagination, euphoric states) and negative (e.g. anxiety, psychosis-like symptoms, impaired concentration) features of the acute state. They propose that an acute 'blast' of 5-HT2A R stimulation leaves a residual loosening of cognition and brain network dynamics that may facilitate improved psychological wellbeing in the mid to long term. The paper notes that 5-HT2A R antagonists (for example ketanserin) attenuate both the positive mood and the psychotomimetic effects of classic psychedelics, consistent with a central role for this receptor. Finally, the authors position their findings relative to previous literature showing enduring personality and wellbeing changes after psychedelics and suggest that deficient 5-HT2A R stimulation may relate to rigid, pessimistic modes of cognition observed in conditions such as depression. They indicate plans to report detailed neuroimaging results in subsequent papers and to pursue longer-term psychological and brain follow-up studies. The discussion closes by reiterating the main empirical outcomes: robust acute psychosis-like but also positive subjective effects after LSD, with increased optimism and openness and no rise in delusional thinking at two weeks, and by proposing the entropic brain/5-HT2A R framework as a mechanistic account.