In a placebo-controlled, within-subject neuroimaging study of 28 psychedelic‑naive participants, a single high (25mg) dose of psilocybin produced lasting functional and anatomical brain changes from 1 hour to 1 month and improved cognitive flexibility, psychological insight and well‑being at one month. Diffusion MRI showed decreased axial diffusivity in prefrontal–subcortical tracts that correlated with reduced brain network modularity (which correlated with improved well‑being), acute increases in cortical signal entropy predicted one‑month well‑being via next‑day psychological insight, and none of these effects were seen with a 1 mg control dose.
Psychedelics have robust effects on acute brain function and long-term behavior but whether they also cause enduring functional and anatomical brain changes is unknown. In a placebo-controlled, within-subjects, electroencephalography, and magnetic resonance imaging study in 28 healthy, entirely psychedelic-naive participants, anatomical and functional brain changes were detected from one-hour to one-month after a single high-dose (25 mg) of psilocybin. Increases in cognitive flexibility, psychological insight, and well-being were seen at one-month. Diffusion imaging done before and one-month after 25mg psilocybin revealed decreased axial diffusivity bilaterally in prefrontal-subcortical tracts that correlated with decreased brain network modularity over the same time period. Decreased modularity also correlated with improved well-being. Increased cortical signal entropy at 1– and 2-hours post-dosing predicted improved psychological well-being at one-month. Next-day psychological insight mediated the entropy to well-being relationship. All effects were exclusive to 25mg psilocybin; no effects occurred with a 1mg psilocybin ‘placebo’ dose.
Lyons and colleagues frame the study around the question of whether single high doses of psilocybin produce enduring anatomical and functional brain changes in humans. Previous work has established that psychedelics produce robust acute effects on brain function and can produce long-term psychological changes, and preclinical studies suggest 5-HT2A receptor agonism may be linked to structural plasticity. However, long-term in vivo human neuroimaging evidence is limited and inconsistent, especially in healthy, psychedelic-naive volunteers. To address this gap, the investigators performed a multi-modal within-subjects study combining electroencephalography (EEG), functional MRI (fMRI) and diffusion tensor imaging (DTI) in healthy volunteers receiving their first-ever high-dose (25 mg) oral psilocybin. A fixed-order repeated-measures design was used: participants first received a 1 mg psilocybin 'placebo' dose and, one month later, a 25 mg dose. Brain and behavioural outcomes were assessed at baseline, during dosing (EEG) and at one month post-dosing (MRI and behavioural measures), with the aim of identifying acute-to-longer-term neural changes and their relationships to psychological outcomes such as insight and well-being.
Papers cited by this study that are also in Blossom
Becker, A. M., Klaiber, A., Holze, F. et al. · International Journal of Neuropsychopharmacology (2022)
Vollenweider, F. X., Vollenweider-Scherpenhuyzen, M. F. I., Bäbler, A. et al. · NeuroReport (1998)
Ly, C., Greb, A. C., Cameron, L. P. et al. · Cell Reports (2018)
Shao, L-X,, Liao, C., Gregg, I. et al. · Neuron (2021)
Twenty-eight healthy, psychedelic-naive volunteers (mean age 41, SD 8.7; 43% female) participated in a fixed-order, within-subjects study. Each participant received two oral doses of psilocybin one month apart: a low-dose control (1 mg) followed by a high dose (25 mg). EEG was recorded at baseline and at 1, 2 and 4.5 hours post-dose during each session. Structural and functional MRI (including diffusion-weighted imaging and an 8-minute eyes-closed resting-state run) were acquired at baseline and at one-month after each dosing session. Behavioural measures included subjective intensity ratings during dosing, a psychological insight scale (PIS) measured sub-acutely, and the Warwick-Edinburgh Mental Wellbeing Scale (WEMWBS) assessed with reference to two-week periods. Analytical approaches varied by modality. EEG time-series were analysed using linear mixed-effects models testing Dose (1 mg vs 25 mg) by Time (0–4.5 h) effects; informational entropy was quantified as Lempel–Ziv complexity (LZc) and spectral power was examined in canonical bands. DTI analyses used ANOVA across three timepoints (baseline, one-month post-1 mg, one-month post-25 mg) to test for Timepoint interactions with axial diffusivity (AD); significant effects were followed by post-hoc t-tests and hemisphere-specific tests with Bonferroni correction. Free-water correction was applied and fractional anisotropy (FA) was additionally reported. A tract-based spatial statistics (TBSS) analysis was also run and reported as non-significant. fMRI analyses included whole-brain BOLD contrasts to emotional faces (happy, fearful, neutral), region-of-interest (amygdala) analyses and generalized psychophysiological interaction (gPPI) connectivity. Resting-state functional connectivity (RSFC) used seed-based ANOVAs for predefined seeds (parahippocampus, amygdala, vmPFC, sgACC) plus hippocampus and dACC added after peer review; network modularity and mass-univariate network metrics were also examined. Correlational and cluster-corrected matrix methods assessed relationships among contemporaneous change scores. Predictive analyses used cross-validated sensor selection for EEG–behaviour regressions and linear mediation models to test whether sub-acute psychological insight mediated the relationship between acute entropy and one-month well-being.
EEG: Robust acute changes occurred after 25 mg psilocybin but not after 1 mg. Linear mixed-effects modelling showed significant increases in cortical informational entropy (LZc) at 1 and 2 hours post-25 mg (p < 0.001), accompanied by decreases in alpha power, an increase in gamma power at 2 hours, and decreases in theta power at 1 and 2 hours. No significant EEG changes were detected under 1 mg. DTI / anatomical: ANOVA revealed significant Timepoint-by-AD effects for two tracts after the high dose: a bilateral prefrontal cortex–striatum tract (PFC–STR; F(2,48)=10.30, p=0.005) and a PFC–thalamus tract (PFC–THA; F(2,48)=9.51, p=0.008). Post-hoc t-tests showed decreases in axial diffusivity one-month post-25 mg versus one-month post-1 mg for PFC–STR (t(24) = -3.72, p = 0.006) and PFC–THA (t(24) = -3.85, p = 0.005). Hemisphere-split tests indicated decreases in both hemispheres but, after Bonferroni correction, left-hemisphere effects remained significant. Free-water correction produced consistent AD effects and revealed significant FA changes; TBSS produced no significant clusters. fMRI — emotional faces and amygdala: Whole-brain BOLD contrasts showed a large effect for fearful faces in the key contrast (one-month post-25 mg vs one-month post-1 mg), with effect size reported as d = 0.9 (95% CI [0.33, 1.34]). An inclusive whole-brain interaction between face type and Timepoint was not significant. Exploratory amygdala ROI analyses and gPPI functional connectivity results (reported in the supplement) showed left amygdala effects and gPPI findings that survived false discovery rate correction. Resting-state functional connectivity and modularity: Seed-based RSFC analyses found a pronounced increase in amygdala RSFC one-month post-25 mg compared with one-month post-1 mg (mean difference = 3.68, SE = 0.64, p < 0.0001; d = 1.2) and when comparing post-25 mg to pre-dose baseline (M diff = 1.49, SE = 0.56, p = 0.046; d = 0.6, Bonferroni-corrected). Across-network measures showed no statistically significant group-average changes in network modularity or consistent within- or between-network RSFC changes. However, when controlling for pre-25 mg well-being, a significant negative correlation was observed between modularity change and well-being change for the key contrast (r = -0.40, p = 0.04), i.e., decreased modularity correlated with improved well-being. Modularity change also correlated with aggregated AD change (PFC–STR and PFC–THA merged). Subjective and psychological outcomes: Subjective intensity showed a strong Dose-by-Time interaction (F(8,197) = 57.73, p < 0.0001), with markedly greater intensity reported after 25 mg versus baseline and versus 1 mg; 94% of participants rated the 25 mg session as the most unusual conscious state of their lives. The 1 mg session produced no significant subjective-intensity elevation. Psychological insight (PIS) showed a robust Dose-by-Time interaction (F(2,44) = 14.05, p < 0.0001), with higher insight scores after 25 mg; details on exact sub-acute scores are provided in the supplement. Well-being (WEMWBS) analyses indicated significant improvements that were exclusive to the 25 mg condition, with authors noting robust one-month well-being increases compared with the 1 mg control. Predictive and correlational findings: Acute LZc under 25 mg predicted next-day psychological insight and one-month well-being. LZc increases in posterior electrodes (P3, P4, O1) at 2 hours post-25 mg predicted next-day insight (Spearman r_s = 0.59, p = 0.006). Whole-head LZc at post-dose timepoints predicted one-month well-being change (r_s = 0.66, p = 0.006). Cross-validated analyses using all sensors produced consistent results. A mediation analysis indicated that 2-hour LZc significantly predicted next-day insight (β = 197, SE = 67, p = 0.008, normalised β = 0.55); both LZc and insight predicted one-month well-being, but the direct effect of LZc on well-being became non-significant when insight was included (β = 29.6, SE = 17.2, p = 0.10), consistent with statistical mediation. A cluster-corrected correlation matrix of contemporaneous change scores revealed a non-spurious structure among brain and behavioural changes (cluster p = 0.006). All acute and enduring effects reported were exclusive to the high-dose (25 mg) condition.
Lyons and colleagues interpret their findings as evidence that a single high-dose psilocybin session in psychedelic-naive healthy volunteers produces measurable changes in brain function and microstructure from as early as one hour to at least one month post-dosing, changes which were not observed after a 1 mg control dose. They emphasise the predictive role of acute increases in cortical signal entropy (LZc) for later mental-health-relevant outcomes: higher entropy during the acute session predicted next-day psychological insight, which in turn statistically mediated improvements in well-being at one month. On anatomical findings, the authors note decreased axial diffusivity in prefrontal–subcortical tracts one month after high-dose psilocybin and discuss this cautiously as a possible marker of microstructural plasticity. They highlight several ambiguities: axial diffusivity can change for many biological reasons (neurofibril or glial growth, myelination, axon density, membrane permeability, extracellular fluid) and can also reflect pathology in other contexts. The DTI effects were robust to free-water correction and correlated with modularity changes, lending support to their potential functional relevance, but the authors call for multi-shell diffusion acquisitions in future work to disambiguate underlying tissue processes. Functional imaging effects were more modest than the EEG and DTI findings. Reduced alpha power and increased cortical entropy during the acute state are described as robust and dose-dependent markers of the psychedelic experience. fMRI changes at one month were subtler in this healthy sample than reported in clinical trials with depressed patients; nevertheless, reductions in amygdala/salience responses to fearful faces and increases in amygdala RSFC were observed and a modularity change correlated negatively with well-being change. The authors suggest functional brain changes may be easier to detect in clinical populations with greater baseline atypicality, though they also acknowledge that assay sensitivity, metric choice and parameters could account for variability across studies. Limitations acknowledged by the study team include interpretational uncertainty around DTI metrics, the fixed-order design (1 mg then 25 mg) which leaves open the possibility of order confounds despite the functional inactivity of the 1 mg dose, and modest sample size. The authors recommend between-subject confirmatory studies, larger samples, and more advanced diffusion imaging to clarify the biological underpinnings and replicability of the observed effects. They conclude that acute entropic brain activity and sub-acute psychological insight appear important to the mechanisms linking the psychedelic experience with subsequent well-being improvements, while cautioning against over-interpretation of the microstructural findings until further evidence is available.
The present work sheds important new light on human brain changes after first-time high-dose psilocybin. The high-dose session constituted the first-ever psychedelic experience for all of the study participants. All except one participant rated their high-dose experience with psilocybin as the single most unusual conscious state of their entire lives, and the single exception rated it within their top-five most unusual. Significant changes in brain function and anatomy were evident from one-hour to one-month after high-dose psilocybin. No acute or long-term changes were apparent after a 1mg 'placebo' dose. Empirical modeling highlighted the predictive power of increased brain entropy under high-dose psilocybin, implying its role in manifesting not just the characteristic 'psychedelic' effects of psilocybin, but also its longer-term mental-health impact. More specifically, improved well-being could be predicted directly from acute increases in brain entropy as early as one-hour post dosing. Prediction could also be done through a sequence where increased brain entropy first predicted next day psychological insight, which then predicted the one-month improvements in well-being. If non-pharmacological variables are standardized as 'psychologically supportive', these results imply that human brain changes as early as one-hour into a 25mg psilocybin experience-and that seem closely related to the subjective 'trip'-can predict mental health improvements one-month later. Decreased axial diffusivity in prefrontal-subcortical tracts one-month after participants' first-ever high-dose of a psychedelic could be construed as potential anatomical 'neuroplasticity', echoing earlier in-vivo animal studies following single-dose psilocybin that reported increases in synaptic spine formation in female miceand synaptic density in pigs. However, interpretation of the axial diffusivity changes is complex, not least due to crossing fiber-related confounds. The diffusion-weighted signal can change due to neurofibril or glial growth, altered myelination, axon density, membrane permeability or extracellular fluid. Decreases in axial diffusivity have been observed with meditation, healthy neurodevelopmentand learning, but also with axonal injury, ageing and related pathology. 5-HT2AR agonism has been associated with in vivo dendro-architectural changes in adult mouse brain, axonal development in embryonic mouse brain, and oligodendrocyte changes in rodent brain tissue in vitro. The high density of 5-HT2ARs in the human prefrontal cortexand greater number of cortical compared with subcortical terminalssuggest a PFC 5-HT2A receptor locus of action for the observed diffusivity changes. The DTI results were robust to free-water correction, and the AD decreases correlated with decreased modularity at one-month post-25mg psilocybin, lending support to the inference that functionally relevant microstructural changes had occurred. Further research using multi-shell sequences is needed to disambiguate the current findings and inform on their robustness and replicability. Until then, serious caution is advised when interpreting the biological basis of these DTI findings. If the effects are microstructural, however, then decreases in AD and FA exclusively after the 25mg dose might tentatively relate to two types of change: 1) the pruning of weak or redundant connections; 2) neurogenesis with under-myelinated axons. Decreased EEG alpha power, a well-replicated effect of psychedelics, is linked with cortical disinhibitionand increased brain entropy under psychedelics appears to be dose-dependentand has been associated with acute and sub-acute psychological changes. Decreased cortical alpha power and increased signal entropy are robust and reliable markers of the acute action of psychedelics in humans. Future research in larger samples could test whether acute EEG changes, induced by 5-HT2AR agonist psychedelics, can predict downstream anatomical neuroplasticity 3 or therapeutically-relevant functional brain changes. However, we failed to find a relationship between acute increases in entropic brain activity and the downstream DTI changes. Compared with other outcomes in this study, the fMRI-measured brain changes assessed onemonth post 25mg were relatively modest; for example, explorative ROI and network-based RSFC analyses (see supplement and methods), including network modularity, yielded largely non-significant results, albeit with some exceptions, including brain responsivity to facial expressions, amygdala RSFC, and a statistically significant modularity change to well-being improvement correlation (Figure,D,F and Figures S8-17 in the supplement) as well as modularity change to axial diffusivity change (Figure). Previous work assessing post psilocybin therapy changes in brain function in individuals with depression have revealed more robust changes. For example, in two separate trials of psilocybin for depression, decreased brain network modularity was seen that correlated with or predicted improvements in symptom severity post-treatment(see also). Doss et al. found similar effects with psilocybin-therapy for depression that they labelled increased "neural flexibility". In the present study, changes in network modularity failed to reach statistical significance (Figure) but we did see a network modularity change vs mental health change relationship (Figure, 4B) that was directionally consistent with previous work, i.e., decreased modularity correlated with improved mental health. Reductions in amygdala and salience-network responses to fearful emotional stimuli post-25mg psilocybin were observed in the present study (Figureand S8-11) and these are broadly consistent with prior work in healthy volunteers. Psychophysiological interaction results were observed for the bilateral amygdala within the face paradigm (see S12-13). Since these analyses were included on request, we refrain from a detailed discussion here but point the reader to a relevant discussion in the supplementary file. Briefly, we note some similarities and differences between a prior study examining amygdala PPI responses to emotional faces post psilocybin-therapy for depression. Also note, S10&11 (happy > neutral and fear) and two prior psilocybin studies in depressionfor evidence of augmented brain reactivity to emotional face stimuli post-psilocybin-therapy, most reliably in the non-negative emotional domains -i.e., happy and neutral faces. These latter results could be interpreted as consistent with the re-opening critical periods for social learning. However, some have questioned the reliability of brain responsiveness to emotional faces as a biomarker of therapeutic action. The present study's fMRI results suggest that enduring functional brain changes post psilocybin are less robust and reliable in healthy versus mentally unwell populations e.g.,. If this principle is reliable, it could imply that greater baseline atypicality in the clinical populations, primes for a more robust remediating change via psilocybin. However, given the reliability of enduring psychological changes post psilocybin in healthy samples, it remains plausible-if not likely-that functional brain changes do occur in healthy populations but their detection is dependent on-or sensitive to-experimental modality, metric and parameter choices. In short, we may have not yet found a sufficiently sensitive assay to detect true functional brain changes. The present multi-modal neuroimaging study in healthy participants sheds new light on the brain effects of first-time high-dose psychedelic use and the therapeutic action of psilocybin-therapy, suggesting, for the first time, how therapeutically relevant effects (i.e., improved well-being) can be forecast via an acute human brain action, i.e., entropic brain activity, that is known to relate to the subjective 'trip'. Recent evidence suggests that this characteristic effect of psychedelics is dose-dependent and somewhat exclusive to this category of drug versus psychoactive stimulants and cannabinoids. Long-term improvements in well-being were predicted by the acute increases in entropic brain activity, temporally coinciding with the psilocybin's acute subjective effects. Results support a role for psychological insight in mediating the causal association between increased entropic brain activity and potentially enduring improvements in well-being. These and prior results highlight psychological insightand entropic brain effectsas important to the action of psychedelic therapy. Possible white-matter changes, as well as improvements in well-being and cognition were also observed in this study. All acute and enduring psychological and neurobiological effects were exclusive to the high-dose psilocybin condition. Justified by prior evidence of enduring psychological changes post-psychedelics, this study used a fixed-order cross-over design. Pairwise contrasts on all the study's main outcomes showed no evidence of acute brain or behavioural effects or changes one-month post-1mg; however, order confounds cannot be entirely discounted. Between-subject confirmatory studies are now required to examine the reliability of these novel findings. controlled drug oversight; Joe Peill, Gregory Cooper, Katie Trinci and Pablo Mallaroni for their support with data analysis; and Bruna Giribaldi for advice and support.
Create a free account to open full-text PDFs.
Vargas, M. V., Dunlap, L. E., Dong, C. et al. · Science (2023)
Moliner, R., Girych, M., Brunello, C. A. et al. · Nature Neuroscience (2023)
Griffiths, R. R., Johnson, M. W. · Journal of Psychopharmacology (2016)
Ross, S., Bossis, A. P., Guss, J. et al. · Journal of Psychopharmacology (2016)
Carhart-Harris, R. L., Giribaldi, B., Watts, R. et al. · New England Journal of Medicine (2021)
Carhart-Harris, R. L., Bolstridge, &. M., Day, C. M. J. et al. · Psychopharmacology (2017)
Goodwin, G. M., Aaronson, S. T., Alvarez, O. et al. · New England Journal of Medicine (2022)
Bogenschutz, M. P., Ross, S., Bhatt, S. R. et al. · JAMA Psychiatry (2022)
Griffiths, R. R. · Journal of Psychopharmacology (2008)
Doss, M. K., Považan, M., Rosenberg, M. D. et al. · Translational Psychiatry (2021)
Wießner, I., Olivieri, R., Falchi, M. et al. · European Neuropsychopharmacology (2022)
Mans, K., Kettner, H., Erritzoe, D. et al. · Frontiers in Psychiatry (2021)
Mason, N. L., Mischler, E., Uthaug, M. V. et al. · Journal of Psychoactive Drugs (2019)
Valle, M., Maqueda, A. E., Rabella, M. et al. · European Neuropsychopharmacology (2016)
Kringelbach, M. L., Cruzat, J., Cabral, J. et al. · PNAS (2020)
Madsen, M. K., Fisher, P. M., Burmester, D. et al. · Neuropsychopharmacology (2019)
Carhart-Harris, R. L., Muthukumaraswamy, S., Roseman, L. et al. · PNAS (2016)
McCulloch, D. E-W., Madsen, M. K., Stenbæk, D. S. et al. · Journal of Psychopharmacology (2021)
Barrett, F. S., Doss, M. K., Sepeda, N. D. et al. · Scientific Reports (2020)
Daws, R. E., Timmermann, C., Giribaldi, B. et al. · Nature Medicine (2022)
Roseman, L., Demetriou, L., Wall, M. B. et al. · Neuropharmacology (2018)
Carhart-Harris, R. L., Roseman, L., Bolstridge, M. et al. · Scientific Reports (2017)
Pasquini, L., Palhano-Fontes, F., Araújo, D. B. · Journal of Psychopharmacology (2020)
Peill, J. M., Trinci, K. E., Kettner, H. et al. · Journal of Psychopharmacology (2022)
Carhart-Harris, R. L., Roseman, L., Haijen, E. C. H. M. et al. · Journal of Psychopharmacology (2018)
Carhart-Harris, R. L., Nutt, D. J. · Journal of Psychopharmacology (2017)
Murray, C., Frohlich, J, Haggarty, C. J., Tare, I. et al. · Neuropsychopharmacology (2024)
Lebedev, A. V., Kaelen, M., L€ Ovd En, M. et al. · Human Brain Mapping (2016)
Timmermann, C., Roseman, L., Haridas, S. et al. · PNAS (2023)
Timmermann, C., Roseman, L., Schartner, M. et al. · Scientific Reports (2019)
Schartner, M., Carhart-Harris, R. L., Barrett, A. B. et al. · Scientific Reports (2017)
Mertens, L. J., Wall, M. B., Roseman, L. et al. · Journal of Psychopharmacology (2020)
Wall, M. B., Demetriou, L., Giribaldi, B. et al. · American Journal of Psychiatry (2025)
Nardou, R., Sawyer, E., Song, Y. J. et al. · Nature (2023)
Carhart-Harris, R. L., Chandaria, S., Erritzoe, D. E. et al. · Neuropharmacology (2023)
Carhart-Harris, R. L., Leech, R., Shanahan, M. et al. · Frontiers in Human Neuroscience (2014)
Carhart-Harris, R. L. · Neuropharmacology (2018)
Papers in Blossom that reference this study
Brendstrup-Brix, K., Ozenne, B., Fisher, P. M. et al. · Journal of Psychopharmacology (2026)
Kuc, J., McAlpine, R. G., Sellers, A. et al. · Journal of Psychopharmacology (2026)
Avram, M., Menegaux, A., Müller, F. et al. · Cell Reports (2026)
Simonsson, O., Lyons, T., Marks, J. et al. · Journal of Psychopharmacology (2026)
Pellegrini, L., Fineberg, N. A., O'Connor, S. et al. · Comprehensive Psychiatry (2025)
Egger, K., Meling, D., Polat, F. et al. · Imaging Neuroscience (2025)
Roseby, W., Kettner, H., Roseman, L. et al. · Frontiers in Psychology (2025)
Agnorelli, C., Spriggs, M. J., Godfrey, K. et al. · Preprints (2024)