The effects of microdose LSD on time perception: a randomised, double-blind, placebo-controlled trial

Perception of time is central to moment-to-moment consciousness and to cognitive functions that rely on precise interval timing, and distortions of interval timing are common across altered states of consciousness and some psychiatric disorders. Earlier human studies report that lysergic acid diethylamide (LSD) and related serotonergic psychedelics alter subjective time perception and performance on behavioural timing tasks, but many of these studies lack placebo control, randomisation, sufficiently small doses or adequate trial numbers, which limits interpretation. A further unresolved issue is whether drug-induced distortions of timing arise from direct neurochemical effects on timing systems or indirectly from broader alterations in conscious experience produced by psychedelic doses. Yanakieva and colleagues addressed this gap by using microdoses of LSD (< 20 µg) to attempt partial dissociation of direct neurochemical effects from overt changes in consciousness. The study used a randomised, double-blind, placebo-controlled design in healthy older adults to test whether single microdoses of LSD (5, 10 and 20 µg) affect interval timing on a temporal reproduction task spanning subsecond and suprasecond intervals (800–4000 ms), administered roughly 3 h after dosing. The authors contrasted behavioural timing outcomes with repeated self-report measures of subjective drug effects to determine whether any timing changes were independent of altered conscious states.

This was a randomised, double-blind, placebo-controlled clinical trial conducted in an inpatient Early Phase Clinical Unit. Forty-eight healthy, native English-speaking older adults (age range 55–75 years; M = 62.92, SD = 5.65; 44% female) were randomised in equal-sized cohorts (n = 12 per cohort) to receive one of four treatments: placebo (distilled water) or LSD tartrate at 5, 10 or 20 µg. The dosing regimen for the larger safety trial involved six single doses administered every 3 days; the temporal reproduction task reported here was completed on the fourth dosing day. Dosing and drug manufacture were performed under regulatory and good manufacturing practice conditions, and experimenters were masked to condition and hypotheses. Eligibility criteria selected medically healthy, postmenopausal female participants and males who accepted contraception; major exclusions comprised recent LSD use, substance dependence, relevant psychiatric history or family history of psychotic disorders, cardiovascular or relevant systemic disease, positive infectious serology, and recent medication that could confound outcomes. Participants completed medical screening up to 28 days before first dosing and attended a 4-week follow-up. The primary behavioural measure was a temporal reproduction task in which participants viewed a target blue circle presented for one of nine stimulus intervals (800, 1200, 1600, 2000, 2400, 2800, 3200, 3600, 4000 ms) and subsequently reproduced that duration by holding down a space bar. After practice, each participant completed 108 experimental trials (12 trials per interval) on a laptop while seated in bed. Subjective drug effects were assessed repeatedly with a 22-item Visual Analogue Scale (SDEVAS); for the present report five SDEVAS items were analysed (e.g. "Do you feel a drug effect?", "Do you feel high?", concentration), scored 0–100. The temporal task was typically completed around 170 min post-dose (mean task completion time M = 170 min, 95% CI 156–183). Statistical analyses used mixed-model ANOVAs to test drug (placebo vs LSD) and dose (placebo vs 5 vs 10 vs 20 µg) effects with stimulus interval as a within-subject factor; Greenhouse–Geisser corrections were applied when sphericity was violated. Primary behavioural dependent variables were median reproduction time and coefficient of variation (CV = SD/median) per interval; extreme outliers in reproduction times (two participants) were excluded. Within-participant regressions of reproduction time on stimulus interval produced slope (beta) coefficients that indexed how reproduction scaled with stimulus duration; these betas were compared across groups with ANOVA. Hierarchical regression analyses (two-stage) tested whether drug or dose predicted reproduction times after accounting for self-reported drug effects at 180 min (entered in stage 1), with changes in R2 (ΔR2) used to assess independent predictive value.

Sample and timing: Forty-eight participants were randomised evenly across four conditions; groups did not differ significantly by age or gender distribution. The temporal reproduction task was completed at similar post-dose times across conditions (overall M ≈ 170 min), indicating comparable timing of behavioural assessment between groups. SDEVAS data were reported as missing for eight participants (n = 2 per dose condition); additionally, two participants were excluded from task analyses for extreme, uniformly fast reproduction times indicative of task failure. Subjective drug effects (SDEVAS): Across the five analysed SDEVAS items there were no robust, statistically significant main effects of LSD versus placebo or of dose on most measures. A time effect reflected increases in reported drug effects from baseline to 120–180 min followed by decline, F10,360 = 4.91, p < .001, ηp2 = .12. Exploratory contrasts at 180 min showed a suggestive main effect of drug on the single item indexing perceived "drug effects" (F1,38 = 5.18, p = .030, η2 = .04), but no reliable dose effects on SDEVAS scores. The authors conclude that microdose LSD produced little in the way of overt subjective alterations of consciousness as indexed by these self-reports. Temporal reproduction times (drug vs placebo): Reproduction times increased with longer stimulus intervals, as expected, F8,352 = 411.63, p < .001, ηp2 = .90. There was a main effect of drug on reproduction times, F1,44 = 6.18, p = .017, ηp2 = .12, and a drug × interval interaction, F8,352 = 5.13, p = .003, ηp2 = .10. Subsidiary analyses indicated significant over-reproduction in the LSD versus placebo condition for suprasecond intervals of 2800 ms (F1,45 = 7.53, p = .009, η2 = .15), 3200 ms (F1,45 = 8.34, p = .006, η2 = .16) and 3600 ms (F1,45 = 8.47, p = .006, η2 = .16). Effects at 2000, 2400 and 4000 ms were suggestive (p < .05) with comparable effect sizes, while intervals of 800–1600 ms showed no reliable drug effects. Hierarchical regressions that entered perceived drug effects at 180 min in the first step found that subjective ratings did not account for reproduction times; adding drug condition in the second step significantly improved prediction for 2000–4000 ms intervals (ΔR2 range .11–.20, p range .025–.002), indicating the LSD effect on timing was independent of self-reported drug effects. Reproduction slopes and variability: Within-participant regression betas (slope of reproduction time vs stimulus interval) were larger in the LSD condition, reflecting steeper scaling, F1,44 = 8.53, p = .005, η2 = .16. Coefficients of variation (CVs) showed a main effect of interval (variability decreased with longer intervals) but no main effect of drug, F1,44 = 0.70, p = .41, nor a drug × interval interaction, indicating LSD did not significantly affect response variability in timing. Dose-related analyses: Analysing dose as a four-level factor produced a main effect of dose on reproduction times, F3,42 = 3.22, p = .032, ηp2 = .19, and a dose × interval interaction, F24,336 = 2.15, p = .037, ηp2 = .13. Post hoc comparisons indicated that the 10 µg dose tended to produce longer reproduction times than placebo at several suprasecond intervals (notably 2000, 3200 and 3600 ms; Tukey post hoc p range .01–.029). Hierarchical regressions with dose added after perceived drug effects again showed dose improved prediction for 2000–3600 ms intervals (ΔR2 range .11–.20, p range .040–.004), consistent with a positive linear relationship between dose and over-reproduction in the suprasecond range. Re-analysis of slope betas by dose showed a modest dose effect, F3,42 = 2.88, p = .047, η2 = .17, with a suggestive increase in slope for 20 µg versus placebo (p = .055). CV analyses by dose replicated the absence of a dose effect on variability.

Yanakieva and colleagues interpret their findings as evidence that microdose LSD produces temporal dilation—increased reproduction of perceived duration—for suprasecond intervals, particularly those ≥ 2000 ms, in older adults. The effect was most pronounced around the 10 µg dose in these analyses and appeared independent of self-reported alterations in consciousness: subjective drug-effect ratings were small and did not account for the behavioural timing differences. The authors therefore argue that LSD-mediated distortions of interval timing can occur in the absence of overt psychedelic phenomenology, suggesting a degree of dissociation between direct neurochemical modulation of timing systems and broader, reportable changes in conscious experience. The results are discussed relative to prior human studies that reported psychedelic-induced timing distortions, noting apparent divergences in directionality. Whereas some previous reports with psilocybin or higher doses of LSD found temporal under-reproduction, the present microdose data showed over-reproduction. The authors suggest several possible explanations that are grounded in existing theory and animal work: (1) interval specificity, with suprasecond timing relying more on cognitive resources such as attention and working memory, which may be differentially affected by microdoses; (2) enhanced attentional focus to duration under microdose LSD producing over-reproduction; and (3) a speculative biphasic pharmacology in which late-phase dopaminergic agonism (D2 receptors) could drive over-reproduction, given dopamine agonists’ known tendency to increase perceived duration. The authors emphasise that these mechanistic accounts are tentative and require direct testing. Key limitations acknowledged by the investigators include a modest sample size that reduced statistical power, particularly for dose comparisons, and the between-group design, which has less internal sensitivity than a within-subject crossover. The absence of a baseline (pre-dose) temporal reproduction measure limits within-participant inference, and the exclusive enrolment of older adults may constrain generalisability because ageing is associated with altered interval timing and reduced striatal dopamine receptor availability. The authors also note a weak tendency for participants in LSD conditions to report greater subjective drug effects, raising a possibility of partial unblinding, although these reports did not explain the timing effects. For future research, the authors recommend systematically probing interval specificity and time course of effects, concurrent measurement of cognitive mediators (attention, arousal, working memory, affect) and physiological indices, repeated timing assessments across post-dose intervals, and pharmacological manipulations (serotonin and dopamine antagonists) to clarify neurochemical contributions. They conclude that further work is needed to replicate these microdose effects and to delineate their cognitive and neurochemical mechanisms.