Acute and post-acute neurobehavioral responses to lysergic acid diethylamide in healthy subjects: a randomized controlled study

LSD has re-emerged as a possible treatment candidate for several psychiatric disorders, and earlier work suggests that a single dose may produce benefits that last well beyond the acute drug state. Preclinical studies also indicate that serotonergic psychedelics can alter neuronal activity, synaptic signalling, functional connectivity and learning-related processes for days after dosing. However, human evidence for these post-acute effects remains limited, especially for objective measures of motor learning and perceptual processing. In particular, it has not been clear whether LSD produces lasting changes in these domains under controlled experimental conditions. Calder and colleagues therefore set out to examine acute and post-acute effects of a full psychedelic dose of LSD in healthy volunteers using a multimodal, repeated-measures design. They aimed to assess auditory processing, motor system excitability, motor learning, blood BDNF, and self-reported stress and cognitive flexibility over the week after dosing. The study was designed to characterise both immediate and longer-term effects of LSD on neurobehavioural functioning and to probe whether human measures of plasticity show signs of change after psychedelic exposure.

The researchers conducted a randomised, double-blind crossover trial in 45 healthy adults aged 21 to 55 years. Each participant received a single 100 µg dose of LSD and a placebo in separate sessions, with more than 4 weeks between visits. In total, 43 participants were included in the analysis. The study was approved by the relevant ethics committee and registered on ClinicalTrials.gov. Participants completed a screening visit and seven experimental sessions, including the two dosing days. Assessments were carried out in a quiet room with one or two investigators present throughout dosing. Baseline EEG and TMS measures were collected before dosing and then repeated at several points afterwards, including about 6 hours and 7 hours after dosing, the next day, and again 6 to 8 days later. Questionnaires were completed at baseline and one week after dosing. Motor learning was tested the day after dosing. The order of procedures was fixed, sessions were always held in the afternoon, and participants reported sleep and exercise in the previous 24 hours. To examine auditory processing and possible long-term potentiation-like effects, the researchers used auditory sensory tetanisation with EEG. They focused on the N1 and P2 event-related potential components over frontocentral electrodes and compared pre- and post-tetanisation responses. To probe motor cortical excitability and plasticity, they used paired associative stimulation (PAS) with transcranial magnetic stimulation, measuring motor-evoked potentials from the right abductor pollicis brevis muscle. Blood BDNF was measured in both plasma and serum before dosing and up to one week later. Motor learning was assessed with a sequence typing task using the left hand, separating online learning during practice from offline learning after rest. Subjective drug effects, mystical-type experiences, side effects, perceived stress and cognitive flexibility were also measured. Effects of LSD were analysed with linear mixed-effects models including fixed effects for drug, visit and treatment order, and random effects for participant. Baseline values and, for some outcomes, hours of sleep and exercise were included as covariates. Post-hoc contrasts were estimated at key timepoints, with adjustment for multiple comparisons. Sensitivity analyses examined whether results were driven by outliers or influential cases, and correlations used Spearman's coefficient with false discovery rate correction.

LSD produced the expected acute subjective effects: it significantly increased hourly ratings of drug effects and scores on the 5D-ASC and MEQ, and most participants reported positive effects on well-being. Side effects were more frequent during the acute and post-acute period, but were generally mild and transient. Blinding was weak: all participants correctly identified the LSD dose, although the placebo was successfully mistaken for a low LSD dose by 79% of participants. For the EEG sensory tetanisation paradigm, the researchers found no overall evidence that tetanisation increased N1 or P2 amplitudes, and the responder rates were 43% for N1 and 53% for P2. One week after dosing, P2 amplitude decreased after tetanisation in the LSD condition only, but the corresponding interaction did not survive multiple-comparison correction. When all post-tetanisation timepoints were examined together, P2 amplitude showed a steady linear decrease in the LSD condition compared with placebo. Independently of tetanisation, LSD acutely reduced both N1 and P2 amplitudes. The P2 reduction persisted one day after dosing, which was the only clear post-acute ERP effect. LSD did not affect evoked theta power. In the TMS/PAS experiment, PAS itself did not produce a significant overall increase in motor-evoked potential amplitude, and there was no evidence that LSD altered PAS-induced plasticity. However, LSD did affect corticospinal responses overall. Motor-evoked potential amplitude was higher under LSD on dosing days and lower one day later than under placebo. Motor-evoked potential latency was also shorter under LSD, with this effect driven by dosing days. During PAS on dosing days, participants under LSD counted slightly fewer pulses than those under placebo, and this difference approached significance. The number of pulses counted did not differ later. LSD did not alter BDNF levels in either serum or plasma at any measured timepoint. In the motor learning task, typing speed improved more in the rest periods than during active practice, and there was a significant drug-by-condition interaction. Post-hoc testing showed greater offline learning under LSD than placebo, mainly due to the second test block completed 80 minutes after practice. In that block, performance improved by 348 ms under LSD versus 262 ms under placebo, corresponding to a 32.8% larger gain. LSD did not affect online learning, accuracy, initial typing speed, or block-level accuracy. No meaningful correlations were found between neurophysiological measures and motor learning or questionnaire outcomes. One week after dosing, LSD increased scores on the Alternatives subscale of the Cognitive Flexibility Inventory and reduced Perceived Stress Scale scores. It did not change the Control subscale of the cognitive flexibility measure.

The authors interpret the study as the first randomised controlled crossover trial to show improved motor learning after a psychedelic drug, and as the first to report acute and post-acute LSD effects on motor-evoked potentials and auditory ERPs at a fully psychedelic dose. They also argue that the reductions in perceived stress and some aspects of cognitive flexibility extend previous reports of positive subjective and psychological effects after LSD. In their view, the acute ERP findings are consistent with earlier research showing that serotonergic psychedelics reduce sensory-evoked potentials, possibly through 5-HT2A receptor-mediated changes in early sensory processing, attentional allocation and neural synchronisation. The authors note that the persistence of reduced P2 amplitude one day later, and the more exploratory pattern at one week, are harder to interpret. They suggest that the delayed P2 changes may reflect altered attention or habituation, but emphasise that these findings require replication and are difficult to explain without concurrent physiological or subjective measures during EEG testing. They also highlight that tetanisation and PAS did not reliably produce the expected long-term potentiation-like changes, which mirrors the mixed or null findings reported in other recent psychedelic studies and raises doubts about how robust these translational paradigms are in humans. For TMS, Calder and colleagues suggest that the acute increase in motor-evoked potential amplitude and shorter latency may reflect increased corticospinal excitability, peripheral muscle tension, or both. They say they cannot distinguish central from peripheral mechanisms because muscle contraction was not quantified before TMS pulses. They also discuss the possibility that the next-day reduction in motor-evoked potential amplitude could relate to fatigue, since participants reported more fatigue in the LSD condition. They conclude that further work is needed to clarify whether LSD has any clinically useful effects on motor function or neurorehabilitation. Regarding BDNF, the authors interpret the null findings as consistent with a recent meta-analysis showing no peripheral BDNF effect for serotonergic psychedelics in humans. They caution that peripheral BDNF may not be a sensitive biomarker of brain plasticity. For motor learning, they propose several possible explanations for the selective offline improvement after LSD, including altered functional plasticity, or lingering effects on motivation, attention and arousal. They note that the lack of correlation with neurophysiological measures does not resolve the mechanism. For stress and cognitive flexibility, they suggest that positive expectancy and unblinding may have contributed to the self-report findings, especially because baseline stress was already low and flexibility was already high. The main limitations acknowledged by the authors are the modest sample size, the exploratory nature of some findings, the absence of plasma drug concentrations, the lack of repeated baseline covariates at each visit, and the fact that task performance and EEG were not accompanied by direct measures of motivation, attention, arousal or peripheral muscle activity. They also stress that all participants correctly identified the LSD dose, so failure of blinding may have influenced subjective and behavioural outcomes. They argue that future studies should include later learning timepoints, more precise measures of confounding factors, and alternative biomarkers of neuroplasticity, such as positron emission tomography imaging of synaptic proteins.

The authors conclude that a single dose of LSD improved next-day motor learning, reduced perceived stress one week later, and increased aspects of cognitive flexibility. They also state that LSD altered motor-evoked potentials and reduced auditory ERP amplitudes, with some effects persisting after the acute drug state. Overall, they argue that LSD’s effects on learning warrant further study and that stimulation-based measures of long-term potentiation are methodologically challenging, making alternative markers of neuroplasticity worth pursuing.